system

Abstract

Provide a system. A means for collecting past business data; means for pre-processing the collected data; means for training a machine learning model using the preprocessed data; a means for inputting a summary of a new business; A means for extracting necessary tasks based on the input outline of new work; A means for organizing the extracted tasks to generate a WBS; A means to display the generated WBS; A means of gathering feedback on the generated WBS; A means to retrain the machine learning model based on the collected feedback; and A system including:

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] Within companies, there is a need to respond quickly and efficiently to new tasks and inexperienced work resulting from transfers. However, manually identifying tasks for each project or task and creating an appropriate WBS (Work Breakdown Structure) requires a great deal of time and effort. Furthermore, because it relies on the experience and knowledge of each individual person in charge, there is a high possibility of missing tasks and inefficiencies. This can easily lead to project delays and a decline in quality. To solve these issues, a system is needed that can automatically extract the tasks required for new tasks using past business data and efficiently generate a WBS. [Means for solving the problem]

[0005] The present invention provides the following means to solve the above problems. First, it has a means for collecting past business data, including data from project management tools, emails, chat logs, documents, etc. Next, it has a means for cleansing the collected data and removing duplicates and noise data, thereby creating a dataset suitable for learning a machine learning model. It provides a means for training a machine learning model using the preprocessed data, learning task patterns and dependencies from past projects. It provides a means for inputting an outline of new work based on this model, and a means for extracting necessary tasks from the input information. It also provides a means for organizing the extracted tasks and generating a WBS. It has a means for providing the generated WBS to users through a display means and collecting feedback from users. It has a means for retraining the machine learning model based on this feedback, building a system that improves prediction accuracy for future projects. This enables quick and efficient response to transfers and the introduction of new projects, preventing project delays and quality degradation.

[0006] "Business data" refers to data related to projects and business carried out within a company, and includes information generated from project management tools, emails, chat logs, documents, etc.

[0007] "Preprocessing" is the process of removing duplicates and noise data from collected business data and converting it into a format suitable for machine learning models.

[0008] A "machine learning model" is a collection of algorithms and methods that analyze large amounts of past business data and learn task patterns and dependencies.

[0009] The "new business summary" is basic information about a new project or business entered by the user, including the type of business, goals, resources, and so on.

[0010] "Task extraction" is the process of automatically identifying the individual tasks required from the outline of a new job using a machine learning model.

[0011] "WBS" stands for Work Breakdown Structure, and is a hierarchical structure of the tasks required to complete a project.

[0012] A "task list" is a list of individual tasks that need to be performed for a particular project or job.

[0013] A "dependency" is a relationship of priority or execution order that exists between multiple tasks, and indicates a constraint such as one task cannot start until another task is completed.

[0014] "Feedback" refers to any corrections or additional information the user makes to the generated WBS, which is used to improve the system's forecasting accuracy.

[0015] "Retraining" is the process of retraining a machine learning model using collected feedback to improve its prediction accuracy and performance. [Brief explanation of the drawings]

[0016] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0017] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0018] First, the terms used in the following description will be explained.

[0019] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0020] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0021] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0022] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0023] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0024] [First embodiment]

[0025] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0026] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0027] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0028] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0029] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0030] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0031] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0032] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0033] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0034] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0035] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0036] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0037] This invention relates to a system that automatically identifies tasks required for new work by utilizing past work data and generates an efficient work breakdown structure (WBS). This system consists of three main elements: a server, terminals, and users.

[0038] First, the server collects past business data, including project management tools, emails, chat logs, documents, etc. During this collection process, APIs and data extraction scripts are used for each data source to collect information and manage it centrally.

[0039] The server then preprocesses the collected data, removing duplicates and noise and converting it into a unified format. Ensuring data quality at this stage creates a dataset suitable for training subsequent machine learning models.

[0040] Using the preprocessed data, the server trains a machine learning model. This model analyzes past project data and learns task patterns and dependencies. By using a large amount of past data to train the model, it is possible to make highly accurate predictions.

[0041] The user inputs an outline of a new job via a terminal. The terminal then sends this input data to the server. The server then analyzes the received data using natural language processing technology and uses a machine learning model to extract the necessary tasks.

[0042] Once the tasks are extracted, the server automatically generates a WBS based on them. The generated WBS includes the hierarchical structure, dependencies, and priorities of the tasks. The terminal visually displays this WBS to the user.

[0043] As a concrete example, consider the case of launching a new marketing campaign. When a user types "launch a marketing campaign for a new product," the server generates the following task list based on data from similar past projects:

[0044] Marketing strategy development

[0045] Target Audience Research

[0046] Creating advertising materials

[0047] Advertisement delivery plan

[0048] Campaign monitoring and evaluation

[0049] The user can then review the generated WBS and add, delete, or modify tasks as needed. The device then sends the user's feedback to the server, which then retrains the machine learning model based on that feedback. This improves prediction accuracy for future tasks.

[0050] This system enables companies to respond quickly and efficiently to new tasks within the company or tasks that employees have no experience with due to transfers, improving work productivity and the success rate of projects.

[0051] The processing flow will be explained below.

[0052] Step 1: Data collection

[0053] The server collects historical business data using APIs and scripts that retrieve data from project management tools, emails, chat logs, documents, etc.

[0054] Specific operation: Obtain information such as project names, task names, deadlines, and assignees from project management tools via API. Filter and obtain related messages from emails and chat logs, and extract text data containing related keywords from documents.

[0055] Step 2: Data Preprocessing

[0056] The server pre-processes the collected data, which is the process of cleansing the data, removing duplicates and noise data, and converting it into a unified format.

[0057] What it does: It removes duplicate entries, maps information from different data sources into one unified format, filters out noisy data, and fills in gaps and incomplete data.

[0058] Step 3: Data training

[0059] The server uses the preprocessed data to train machine learning models, which analyze past project data and learn task patterns and dependencies.

[0060] Specific operation: Input data into a machine learning algorithm (e.g., decision tree, neural network) to generate and learn task co-occurrence patterns and dependency graphs within the project.

[0061] Step 4: Enter an outline of the new job

[0062] The terminal provides an interface for users to input the outline of a new project or task. The user inputs basic information about the new project or task (type of task, goals, resources, etc.).

[0063] Specific operation: An interface is displayed through the terminal input form, allowing the user to enter the type of work, goals, resources, deadlines, etc., and enter the required information.

[0064] Step 5: Task extraction

[0065] The server uses a machine learning model to extract the necessary tasks based on the input summary of the new business.

[0066] Specific operation: The input data is analyzed using a natural language processing algorithm, and the necessary task list is generated based on task patterns from related past projects.

[0067] Step 6: Generate WBS

[0068] The server automatically generates a work breakdown structure (WBS) based on the extracted tasks, including the hierarchical structure, dependencies, and priorities of the tasks.

[0069] Specific operations: Organize task lists hierarchically, represent dependencies between tasks in a graph structure, and set priorities and deadlines to build a WBS.

[0070] Step 7: View the WBS

[0071] The terminal displays the generated WBS to the user, allowing the user to visually check the overall task structure and dependencies.

[0072] Specific operation: Renders the WBS in Gantt chart format and hierarchical display format, and provides an interface that allows interactive operation.

[0073] Step 8: User Feedback

[0074] The user can review the generated WBS and make adjustments as needed, such as adding, deleting, or reordering tasks.

[0075] The terminal collects user feedback.

[0076] Specific operation: Record user operations (addition, deletion, change) for each task in the WBS and send them as feedback data.

[0077] Step 9: Retrain the model

[0078] The server retrains the machine learning model based on the collected feedback, improving the accuracy of future predictions.

[0079] What it does: Analyzes collected feedback data and retrains machine learning models to learn feedback patterns.

[0080] Through this processing flow, the system generates an efficient and accurate WBS, enabling companies to respond quickly to new business and transfers.

[0081] Example 1

[0082] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0083] Modern business activities require the ability to quickly and efficiently handle new tasks and unfamiliar work resulting from transfers. However, conventional methods have struggled to effectively utilize past work data to automatically identify the tasks required for new work. Furthermore, there has been a lack of feedback mechanisms to ensure the accuracy and usefulness of the generated work breakdown structure (WBS). This has led to issues such as reduced work productivity and project success rates.

[0084] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0085] In this invention, the server includes a means for collecting past business data, a means for preprocessing the collected data, a means for training a machine learning model using the preprocessed data, a means for inputting an outline of a new business, a means for extracting necessary tasks based on the input outline of the new business, a means for organizing the extracted tasks to generate a WBS, a means for visually displaying the generated WBS, a means for collecting feedback on the generated WBS, and a means for retraining the machine learning model based on the collected feedback. This makes it possible to effectively utilize past business data and respond to new business quickly and efficiently. Furthermore, the accuracy and usefulness of the generated WBS are improved, thereby improving business productivity and project success rates.

[0086] "Historical Business Data" refers to all records and information relating to previously executed projects or business operations, including project management tools, emails, message logs, documents, etc.

[0087] "Collection methods" refers to the methods and tools used to obtain necessary information from various data sources and manage it centrally.

[0088] "Preprocessing means" refers to techniques and tools for removing redundant data and noise data from collected data and converting it into a unified format.

[0089] A "machine learning model" refers to an algorithm or system that learns and predicts task patterns and dependencies based on data.

[0090] "Training methods" refers to techniques and tools used to train machine learning models using preprocessed data to improve their predictive accuracy.

[0091] The "new business overview" refers to the basic content and goals of the new business or project that the user intends to implement.

[0092] "Task" refers to the specific work or activities required to accomplish new business.

[0093] "Task extraction means" refers to methods and tools for finding and extracting necessary tasks from the outline of new work.

[0094] "WBS (Work Breakdown Structure)" refers to a hierarchical organization and structure of the tasks required to achieve the project's goals.

[0095] "Organizational methods" refer to techniques and tools for organizing and categorizing extracted tasks into a rational order and structure.

[0096] "Visual display means" refers to graphical interfaces and tools that present the generated WBS to users in an easy-to-understand manner.

[0097] "Feedback methods" refer to methods and tools for collecting user opinions and corrections and reflecting them in the system.

[0098] "Retraining methods" refer to methods and tools used to retrain machine learning models based on feedback collected from users, further improving their predictive accuracy.

[0099] This invention relates to a system that automatically identifies tasks required for new work by utilizing past work data and generates an efficient work breakdown structure (WBS). This system consists of three main elements: a server, terminals, and users.

[0100] First, the server collects past business data. This data includes data from project management tools, emails, message logs, documents, etc. During the collection process, APIs and data extraction scripts are used for each data source (e.g., project management tools, email servers, chat tools, cloud storage) to obtain information and manage it centrally. The server then stores the collected data in a specific directory.

[0101] Next, the server preprocesses the collected data. Specifically, it uses Python's Pandas library to read the data and remove duplicates and noise. It then uses regular expressions to filter out the noise and extract only the necessary information. To standardize the data format, it uses preprocessing modules from Numpy and Scikit-learn. This stage ensures the quality of the data and creates a dataset suitable for training subsequent machine learning models.

[0102] Using the preprocessed data, the server trains a machine learning model. It uses Tensorflow (registered trademark) or PyTorch to train a recurrent neural network (RNN) or Transformer model. This allows it to learn task patterns and dependencies from past project data. It evaluates the model using cross-validation and tunes hyperparameters as needed.

[0103] A user inputs a summary of a new task via a terminal. For example, if a user inputs "Run a marketing campaign for a new product," the data is sent from the terminal to the server in JSON format. The server analyzes this input data using natural language processing technology (e.g., BERT, GPT-3 (registered trademark)) and extracts the necessary tasks using a machine learning model.

[0104] Once tasks are extracted, the server automatically generates a WBS based on them. The generated WBS includes the task hierarchy, dependencies, and priorities. This WBS is displayed on the device using visualization libraries such as D3.js and Highcharts. Users can review the generated WBS and add, delete, or modify tasks as needed.

[0105] User feedback is sent from the device to the server, which then uses this feedback to retrain the machine learning model, improving the accuracy of task extraction and WBS generation from the next time onwards.

[0106] Specific examples

[0107] For example, when a user is running a marketing campaign for a new product, they can input the following prompt: "What are the tasks required to run a marketing campaign for a new product?" Based on this, the server analyzes data from similar past projects and generates a task list like this:

[0108] Marketing strategy development

[0109] Target Audience Research

[0110] Creating advertising materials

[0111] Advertisement delivery plan

[0112] Campaign monitoring and evaluation

[0113] As users review and modify the generated WBS, feedback is collected and the machine learning model is retrained based on this data, improving the efficiency of new work and the success rate of projects.

[0114] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0115] Step 1:

[0116] The server collects past business data from various data sources. The collected data includes project management tools, emails, message logs, documents, and more. Specifically, project data is obtained using the project management tool's API, and email data is collected using the email server's IMAP protocol. Message logs are downloaded using the chat tool's API, and document data is downloaded using the cloud storage API. The input is raw data obtained from each data source, and the output is a data file in a unified format stored on the server.

[0117] Step 2:

[0118] The server preprocesses the collected data. It uses Python's Pandas library to read the data and remove duplicates and noise. It uses regular expressions to filter out unnecessary strings and noise and extract the necessary information. To unify the data format, it uses preprocessing modules from Numpy and Scikit-learn. The input is the collected raw data, and the output is clean data after preprocessing.

[0119] Step 3:

[0120] The server uses the preprocessed data to train a machine learning model. It uses TensorFlow or PyTorch to train a recurrent neural network (RNN) or Transformer model. The model is trained by learning task patterns and dependencies based on past project data. The input is clean data that has been preprocessed, and the output is a trained machine learning model.

[0121] Step 4:

[0122] The user inputs the outline of a new business via the terminal. For example, the user inputs "Run a marketing campaign for a new product" in text format. The terminal converts this data into JSON format and sends it to the server. The input is the outline of the new business entered by the user, and the output is the JSON format data sent to the server.

[0123] Step 5:

[0124] When the server receives the input outline of the new work, it analyzes it using natural language processing technology (e.g., BERT, GPT-3). It then uses a machine learning model to extract the necessary tasks. The input is new work data in JSON format, and the output is a set of necessary tasks. Once the tasks are extracted, a WBS is automatically generated based on them. The generated WBS includes the hierarchical structure, dependencies, and priorities of the tasks.

[0125] Step 6:

[0126] The server sends the generated WBS to the terminal, which uses a visualization library such as D3.js or Highcharts to visually display the WBS to the user. The input is the task group and the generated data of the WBS, and the output is the visualized WBS that is displayed to the user.

[0127] Step 7:

[0128] The user checks the displayed WBS and adds, deletes, or modifies tasks as necessary. The results of the user's operations are sent from the terminal to the server as feedback. The input is the user's feedback operations, and the output is the feedback data.

[0129] Step 8:

[0130] The server retrains the machine learning model based on the collected feedback. The feedback data is used as new training data to update the machine learning model. This improves the accuracy of task extraction and WBS generation from the next time onwards. The input is the feedback data, and the output is the updated machine learning model.

[0131] (Application example 1)

[0132] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0133] Optimizing production processes in traditional factories relies primarily on the experience and intuition of managers, making efficient project management difficult. Furthermore, identifying appropriate tasks and creating a work breakdown structure (WBS) is particularly difficult for new work. This results in lower productivity and project success rates. Furthermore, it is not easy to monitor the progress of each process in real time and make appropriate readjustments.

[0134] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0135] In this invention, the server includes means for collecting past work data, means for preprocessing the collected data, means for training a machine learning model using the preprocessed data, means for inputting an outline of a new work, means for extracting necessary tasks based on the input outline of the new work, means for organizing the extracted tasks to generate a WBS, means for displaying the generated WBS, means for collecting feedback on the generated WBS, means for retraining the machine learning model based on the collected feedback, means for assigning tasks to robots to optimize production processes in a factory, and means for monitoring and readjusting the progress of each process in real time. This enables automatic and efficient optimization of production processes, thereby improving project success rates and productivity.

[0136] "Past business data" refers to all information related to past projects or work, including project management tools, emails, chat logs, and documents.

[0137] "Preprocessing means" refers to the process of removing redundancy and noise from collected data and converting the data into a unified format.

[0138] "Means for training machine learning models" refers to running machine learning algorithms using pre-processed historical business data to learn task patterns and dependencies.

[0139] The "means for inputting the outline of a new business" refers to an interface through which the user inputs details of a new business to be started.

[0140] "Means for extracting necessary tasks" refers to the process of automatically identifying appropriate tasks from the input outline of new work.

[0141] "Method for generating WBS" refers to the method for organizing the extracted tasks into a hierarchical structure and creating a work breakdown structure.

[0142] "Means for displaying the generated WBS" refers to an interface that visually displays the generated WBS to the user.

[0143] "Means for collecting feedback" refers to an interface for collecting comments and correction requests from users.

[0144] "Means for retraining the machine learning model" refers to the process of updating the machine learning model based on collected feedback to improve the accuracy of task extraction and WBS generation from the next time onwards.

[0145] "Means for assigning tasks to robots to optimize production processes" refers to a system that automatically distributes appropriate work instructions to robots in a factory.

[0146] "Means for monitoring and readjusting the progress of the process in real time" refers to a system for monitoring the progress of the production process in real time and readjusting it as necessary.

[0147] This invention is a system that uses past business data to automatically identify tasks required for new work and generate an efficient work breakdown structure (WBS).Application examples include a system that assigns tasks to robots to optimize production processes within a factory, and monitors and readjusts the progress of each process in real time.

[0148] The server first collects past business data, including project management tools, emails, chat logs, documents, etc. Data collection is centrally managed using APIs and data extraction scripts.

[0149] The server then preprocesses the collected data, which includes removing duplicates and noise and converting the data into a unified format, ensuring data quality and creating a dataset suitable for training machine learning models.

[0150] The server uses the preprocessed data to train a machine learning model. The model analyzes past project data and learns task patterns and dependencies. By using a large amount of past data for this training, highly accurate predictions are possible.

[0151] The user inputs an outline of the new job through a terminal. This input data is sent to the server, which then analyzes the received data using natural language processing technology and extracts the necessary tasks using a machine learning model.

[0152] The server automatically generates a WBS based on the extracted tasks. The generated WBS includes the hierarchical structure, dependencies, and priorities of the tasks. The terminal visually displays the generated WBS to the user.

[0153] Furthermore, the system has the ability to optimize production processes within the factory and assign tasks to robots. Specifically, it lists the tasks for robots in each process, monitors their progress in real time, and adjusts them as necessary.

[0154] Users can check the generated WBS and add, delete, or modify tasks as needed. This feedback is sent to the server, which then uses it to retrain the machine learning model and improve prediction accuracy for future projects.

[0155] As a concrete example, consider the case where a new plan for setting up a new product line is entered into a "production process optimization app" on a smartphone. The server generates the following task list based on similar past data:

[0156] 1. Manufacturing line design

[0157] 2. Machine settings

[0158] 3. Production Test

[0159] 4. Quality check

[0160] An example of a prompt sentence is "Please generate the optimal production process for designing a new manufacturing line for airplane parts." By using this prompt sentence, the server generates the optimal task list and WBS, realizing the efficiency of the production process.

[0161] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0162] Step 1:

[0163] The server collects historical business data using APIs and data extraction scripts. It receives input from data sources such as project management tools, emails, chat logs, and documents. It integrates this data and stores it in a central database. The output is unified historical business data.

[0164] Step 2:

[0165] The server preprocesses the collected data, specifically removing duplicate and noisy data and converting the data into a unified format. The input for this process is historical business data, and the output is a high-quality, unified dataset.

[0166] Step 3:

[0167] The server uses the preprocessed dataset to train a machine learning model by running a machine learning algorithm that learns patterns and dependencies in the data. The input to this process is the preprocessed dataset, and the output is a machine learning model that has learned the patterns and dependencies for the task.

[0168] Step 4:

[0169] The user inputs the outline of the new business through the terminal. The details of the business to be newly started are input as input to the terminal. The output is the outline data of the new business sent from the terminal.

[0170] Step 5:

[0171] The server analyzes the received outline of the new business using natural language processing technology. The input is the outline data of the new business, and the output is data extracting the necessary tasks.

[0172] Step 6:

[0173] The server generates a WBS based on the extracted task data. The input is the analyzed task data. The output is a WBS that includes the hierarchical structure, dependencies, and priorities of the tasks.

[0174] Step 7:

[0175] The terminal visually displays the generated WBS to the user. The input is the generated WBS, and the output is a representation of the WBS that the user can visually confirm.

[0176] Step 8:

[0177] The user inputs feedback for the generated WBS. The input is the feedback entered by the user via the terminal, and the output is the feedback data sent from the terminal to the server.

[0178] Step 9:

[0179] The server retrains the machine learning model based on the collected feedback: the input is the user feedback data, and the output is the updated machine learning model.

[0180] Step 10:

[0181] The server assigns tasks to robots to optimize production processes in the factory. The inputs are the generated WBS and robot work information, and the output is specific work instructions delivered to the robots.

[0182] Step 11:

[0183] The server monitors the progress of each process in real time and adjusts it as necessary. The input is real-time progress data, and the output is optimized process information.

[0184] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0185] This invention combines a system that uses past business data to automatically identify tasks required for new business and generate an efficient WBS (Work Breakdown Structure), with an emotion engine that recognizes user emotions. This system consists of four main elements: a server, a terminal, a user, and an emotion engine.

[0186] First, the server collects past business data. This data includes project management tools, emails, chat logs, documents, etc. The server uses APIs and data extraction scripts for each data source to collect information and manage it centrally.

[0187] The server then preprocesses the collected data, removing duplicates and noise and converting it into a unified format. Ensuring data quality at this stage creates a dataset suitable for training subsequent machine learning models.

[0188] Using the preprocessed data, the server trains a machine learning model. This model analyzes past project data and learns task patterns and dependencies. By using a large amount of past data to train the model, it is possible to make highly accurate predictions.

[0189] The user inputs an outline of a new job via a terminal. The terminal then sends this input data to the server. The server then analyzes the received data using natural language processing technology and uses a machine learning model to extract the necessary tasks.

[0190] Once tasks are extracted, the server automatically generates a WBS based on them. The generated WBS includes the hierarchical structure, dependencies, and priorities of the tasks. The terminal visually displays this WBS to the user. Furthermore, by incorporating an emotion engine, the system analyzes the user's emotional state and adjusts the way the WBS is displayed.

[0191] As a concrete example, consider the case of a new project to launch a new product. When a user types "launch a new product," the server generates the following task list based on data from similar past projects:

[0192] Product packaging design

[0193] Package design approval

[0194] Selection of sales channels

[0195] Sales channel contract negotiations

[0196] Creating a marketing plan

[0197] Marketing plan approval

[0198] Creating advertising materials

[0199] Formulating an advertising distribution plan

[0200] If a user reviews the generated WBS and feels, for example, nervous or anxious, the emotion engine will recognize this and determine that more support is needed, providing more detailed explanations or additional guidance.

[0201] The user then provides feedback on the generated WBS. The device collects the user's feedback and sends it to the server. The server then retrains the machine learning model based on the feedback, improving prediction accuracy for future iterations.

[0202] This emotion engine recognizes the user's emotional state and adjusts the interface and feedback based on that state, providing a user-friendly system. This further strengthens rapid response to new tasks and transfers, further improving work productivity and project success rates.

[0203] The processing flow will be explained below.

[0204] Step 1: Data collection

[0205] The server collects historical business data using APIs and scripts that retrieve data from project management tools, emails, chat logs, documents, etc.

[0206] Specific operation: Obtain information such as project names, task names, deadlines, and assignees from project management tools via API. Filter and obtain related messages from emails and chat logs, and extract text data containing related keywords from documents.

[0207] Step 2: Data Preprocessing

[0208] The server pre-processes the collected data, which is the process of cleansing the data, removing duplicates and noise data, and converting it into a unified format.

[0209] What it does: It removes duplicate entries, maps information from different data sources into one unified format, filters out noisy data, and fills in gaps and incomplete data.

[0210] Step 3: Data training

[0211] The server uses the preprocessed data to train machine learning models, which analyze past project data and learn task patterns and dependencies.

[0212] Specific operation: Input data into a machine learning algorithm (e.g., decision tree, neural network) to generate and learn task co-occurrence patterns and dependency graphs within the project.

[0213] Step 4: Enter an outline of the new job

[0214] The terminal provides an interface for users to input the outline of a new project or task. The user inputs basic information about the new project or task (type of task, goals, resources, etc.).

[0215] Specific operation: An interface is displayed through the terminal input form, allowing the user to enter the type of work, goals, resources, deadlines, etc., and enter the required information.

[0216] Step 5: Task extraction

[0217] The server uses a machine learning model to extract the necessary tasks based on the input summary of the new business.

[0218] Specific operation: The input data is analyzed using a natural language processing algorithm, and the necessary task list is generated based on task patterns from related past projects.

[0219] Step 6: Generate WBS

[0220] The server automatically generates a work breakdown structure (WBS) based on the extracted tasks, including the hierarchical structure, dependencies, and priorities of the tasks.

[0221] Specific operations: Organize task lists hierarchically, represent dependencies between tasks in a graph structure, and set priorities and deadlines to build a WBS.

[0222] Step 7: Emotion Recognition

[0223] The device sends the user's input data and operation history to the emotion engine, which analyzes the user's emotional state.

[0224] Specific operation: Input data (keyboard input speed and pressure, screen operation history) is input into the emotion engine, and an emotion analysis algorithm is used to recognize the user's emotional state (e.g., stress, anxiety, concentration, etc.).

[0225] Step 8: Emotion-Based Display Adjustment

[0226] The server receives feedback from the emotion engine and adjusts the display of the WBS, providing an environment that is easy for users to understand and work in.

[0227] Specific behavior: If the user is stressed, simplify the WBS display and add specific guidelines and support messages. Conversely, if the user is focused, display more detailed information and options.

[0228] Step 9: Gather feedback

[0229] The user can review the generated WBS and make adjustments as needed, such as adding, deleting, or reordering tasks.

[0230] The terminal collects user feedback.

[0231] Specific operation: Record user operations (addition, deletion, change) for each task in the WBS and send them as feedback data.

[0232] Step 10: Retrain the model

[0233] The server retrains the machine learning model based on the collected feedback, improving the accuracy of future predictions.

[0234] What it does: Analyzes collected feedback data and retrains machine learning models to learn feedback patterns.

[0235] This emotion engine recognizes the user's emotional state and adjusts the interface and feedback based on that state, providing a user-friendly system. This further strengthens the ability to respond quickly and efficiently to new tasks and transfers, further improving work productivity and project success rates.

[0236] Example 2

[0237] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0238] In today's business environment, launching and executing new projects requires significant planning and management. However, traditional project management systems are unable to effectively utilize past work data, making it difficult to efficiently identify the tasks required for new projects. Furthermore, systems are not designed to take into account the user's emotional state, resulting in a suboptimal user experience. Furthermore, even if feedback on the generated WBS is collected, there is insufficient means for applying that feedback to future projects. These issues result in reduced work efficiency and project success rates.

[0239] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for collecting past business data, a means for preprocessing the collected data, and a means for training a machine learning model using the preprocessed data. This makes it possible to effectively utilize past business data and efficiently identify tasks required for a new project. The server also includes a means for inputting an outline of a new business, a means for extracting necessary tasks based on the input outline of the new business, a means for organizing the extracted tasks to generate a WBS, a means for displaying the generated WBS, a means for collecting feedback on the generated WBS, a means for retraining a machine learning model based on the collected feedback, a means for analyzing a user's emotional state, and a means for adjusting the interface and feedback based on the analyzed emotional state. This makes it possible to improve the user experience and utilize feedback to increase the system's prediction accuracy.

[0240] "Historical business data" refers to information about previously executed tasks or projects, and is data obtained from project management tools, emails, electronic chat logs, documents, etc.

[0241] "Preprocessing" is the process of organizing collected data by removing redundancy and noise and converting it into a unified format, making it suitable for training machine learning models.

[0242] A "machine learning model" is a set of algorithms that are trained using collected and preprocessed data to learn patterns and dependencies of tasks required for new business operations.

[0243] The "new business outline" refers to detailed information about a new project or task, and is input data from the user that includes specific goals and necessary requirements.

[0244] "Task extraction" is the process of using a machine learning model to identify necessary tasks from the outline of a new business and define them as specific activity items.

[0245] "WBS" stands for "Work Breakdown Structure" and is a visual representation of the hierarchical tasks required to complete a project.

[0246] "Feedback" refers to information including opinions and improvements provided by users regarding the generated WBS, and is used as data to improve the system's prediction accuracy.

[0247] "Emotional state" refers to the user's emotional response or mood, and is the psychological state that is analyzed while using the system.

[0248] An "emotion engine" is a program or set of algorithms for detecting and analyzing a user's emotional state, with the goal of improving the user experience.

[0249] "Interface" refers to the screen and operating means that users use to interact with the system, and must be designed to be easy for users to use.

[0250] The present invention provides a system that combines four main elements: a server, a terminal, a user, and an emotion engine. A specific embodiment of this system will be described below.

[0251] First, the server collects past business data. This involves obtaining information from data sources such as project management tools, email systems, electronic chat logs, and document management systems. Specifically, it executes APIs and data extraction scripts using tools such as JIRA, Slack, mail servers, and Google® Drive, and stores the necessary data in a centralized database.

[0252] The server then preprocesses the collected data. This includes removing duplicate data, reducing noise, and converting the data into a unified format, such as JSON, to ensure data quality. Specifically, the data is cleaned and centralized using an SQLite database and regular expressions.

[0253] The server uses the preprocessed data to train a machine learning model using a machine learning framework such as TensorFlow or PyTorch, which analyzes past project data and learns task patterns and dependencies.

[0254] Next, the user enters a summary of the new business via the terminal, for example, by entering "Implement market launch of new product" into a web form. This information is sent from the terminal to the server as an HTTP POST request.

[0255] The server analyzes the new business data it receives using natural language processing (NLP) technology and uses machine learning models to extract necessary tasks. Specifically, it uses NLP libraries such as spaCy and NLTK to tokenize and analyze the data. It then identifies tasks based on the analysis results and defines them as specific actions.

[0256] Once the tasks are extracted, the server automatically generates a work breakdown structure (WBS) based on the extracted tasks. The generated WBS includes the hierarchical structure, dependencies, and priorities of the tasks. The terminal visually displays this WBS to the user.

[0257] Furthermore, an emotion engine built into the server analyzes the user's emotional state. When the user checks the WBS, emotion analysis is performed based on their reaction. For example, if the user is feeling nervous or anxious, the emotion engine will recognize this and display additional guidance or detailed explanations. This analysis is performed using a cloud-based emotion analysis API (for example, IBM Watson (registered trademark)).

[0258] Users provide feedback on the generated WBS. The device collects the feedback and sends it to the server, which then retrains the machine learning model based on the feedback to improve prediction accuracy in the future.

[0259] As a concrete example, when a user starts a new project to launch a new product, they can input "launch a new product," and the server will generate the following task list based on data from similar past projects:

[0260] Product packaging design

[0261] Package design approval

[0262] Selection of sales channels

[0263] Sales channel contract negotiations

[0264] Creating a marketing plan

[0265] Marketing plan approval

[0266] Creating advertising materials

[0267] Formulating an advertising distribution plan

[0268] An example of a prompt sentence to be input to the generative AI model is, "Please tell me the tasks required to launch a new product to the market." By using this prompt sentence, the AI ​​model can efficiently extract the tasks required for the new business and propose an appropriate WBS.

[0269] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0270] Program processing flow

[0271] Step 1:

[0272] The server collects past business data.

[0273] Input: API requests from project management tools, email systems, electronic chat logs, and document management systems.

[0274] Data processing / calculation: Send API requests and extract the necessary information from each data source.

[0275] Output: Save the retrieved data to the database.

[0276] Specific behavior:

[0277] Get project ticket information from the project management tool API.

[0278] Email data is retrieved using the IMAP email system.

[0279] Extract chat logs via API.

[0280] Download relevant documents from your document management system.

[0281] Step 2:

[0282] The server preprocesses the collected data.

[0283] Input: Collected business data.

[0284] Data processing / calculation: Delete duplicate data, reduce noise data, and convert to a unified format.

[0285] Output: Preprocessed data converted into a unified format.

[0286] Specific behavior:

[0287] Execute a SQL query to remove duplicate data in an SQLite database.

[0288] Use regular expressions to filter out noisy messages.

[0289] All data will be unified and centralized in JSON format.

[0290] Step 3:

[0291] The server trains the machine learning model.

[0292] Input: Preprocessed business data.

[0293] Data processing / computation: Input the data into a framework such as TensorFlow or PyTorch and train a model.

[0294] Output: A trained machine learning model.

[0295] Specific behavior:

[0296] Input the dataset into the TensorFlow model and run the training loop.

[0297] Tune hyperparameters to improve model accuracy.

[0298] Save the trained model and use it in a later step.

[0299] Step 4:

[0300] The user inputs an outline of a new job via a terminal.

[0301] Input: Enter text about the new business.

[0302] Data processing / calculation: The data entered by the user is sent as an HTTP POST request.

[0303] Output: A summary of the new job is sent to the server.

[0304] Specific behavior:

[0305] A user fills out a web form with a project description (e.g., "Implement market introduction of a new product").

[0306] The device generates and sends an HTTP POST request to send the input content to the server.

[0307] Step 5:

[0308] The server analyzes the new business data and extracts tasks.

[0309] Input: New business summary, trained machine learning model.

[0310] Data processing / computation: Tokenize the data using natural language processing and use models to extract tasks.

[0311] Output: The extracted task list.

[0312] Specific behavior:

[0313] Use spaCy to tokenize the outline of the new business and perform semantic analysis.

[0314] Input data into a trained machine learning model to extract the required tasks.

[0315] Step 6:

[0316] The server automatically generates the WBS.

[0317] Input: The extracted task list.

[0318] Data processing / calculation: Calculate the hierarchical structure and dependencies of tasks and generate a WBS.

[0319] Output: The generated WBS.

[0320] Specific behavior:

[0321] Based on the extracted tasks, an algorithm is used to generate a hierarchical structure for the WBS.

[0322] Save the generated WBS in JSON format and use it in a later step.

[0323] Step 7:

[0324] The server uses an emotion engine to analyze the user's emotional state.

[0325] Input: User operation logs and input data.

[0326] Data processing / calculation: Estimate emotional state using emotion analysis API.

[0327] Output: Parsed emotion data.

[0328] Specific behavior:

[0329] Collects logs of user keyboard input and mouse operations.

[0330] Emotional states are estimated using a cloud-based emotion analysis API.

[0331] Step 8:

[0332] The user provides feedback on the generated WBS.

[0333] Input: User feedback.

[0334] Data processing / calculation: Generate an HTTP POST request to send the feedback to the server.

[0335] Output: The feedback is sent to the server.

[0336] Specific behavior:

[0337] The user fills out the feedback form and clicks the submit button.

[0338] The device generates and sends an HTTP POST request to send the feedback content to the server.

[0339] Step 9:

[0340] The server retrains the machine learning model based on the collected feedback.

[0341] Input: Feedback data, existing trained model.

[0342] Data processing / computation: Add feedback data and retrain the model.

[0343] Output: A machine learning model with improved accuracy.

[0344] Specific behavior:

[0345] Add new feedback data to the dataset.

[0346] Retrain your machine learning model and see improvements in accuracy.

[0347] Save the improved model.

[0348] (Application example 2)

[0349] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0350] Conventional work management systems have difficulty extracting appropriate tasks from past data and generating an efficient work breakdown structure (WBS) when starting a new task. Furthermore, they do not provide an interface or support that takes into account the user's emotional state, which can cause stress for users and lead to lower productivity and a deterioration in work efficiency. Furthermore, they lack a mechanism for appropriately collecting feedback and improving the accuracy of future predictions. These issues need to be resolved.

[0351] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0352] In this invention, the server includes means for collecting past business data, means for preprocessing the collected data, means for training a machine learning model using the preprocessed data, means for inputting an outline of a new business, means for extracting necessary tasks based on the input outline of the new business, means for organizing the extracted tasks to generate a work breakdown structure (WBS), means for displaying the generated WBS, means for collecting feedback on the generated WBS, means for retraining the machine learning model based on the collected feedback, and means for adjusting the WBS display method and support content based on the user's emotional state using an emotion engine that recognizes user emotions. This makes it possible to efficiently extract tasks necessary for new business based on past data, generate and display a WBS, and provide an interface and support that responds to the user's emotions.

[0353] "Past business data" refers to information about various business operations that have been carried out in the past, including data from project management tools, emails, chat logs, documents, etc.

[0354] "Preprocessing" refers to the process of removing duplication and noise from collected data and standardizing the format.

[0355] A "machine learning model" refers to an algorithm or computational model used to analyze and predict data based on collected data.

[0356] "Outline of new business" refers to basic information such as the purpose, scope, and conditions of the new business.

[0357] "Task extraction" refers to identifying and listing the necessary work items based on the outline of the new business.

[0358] A "Work Breakdown Structure (WBS)" is a tool used to hierarchically break down the work involved in a project or task and organize dependencies and priorities.

[0359] "Display" refers to visually presenting the generated information and WBS to the user.

[0360] "Feedback" refers to information collected from users, such as opinions and ratings, that is used to improve the system and retrain the model.

[0361] An "emotion engine" refers to a system that analyzes a user's emotional state and adjusts the system's behavior and display accordingly.

[0362] An embodiment of this invention will be described in detail. This system automatically generates work schedules within a factory and efficiently manages robot work. It generates new tasks based on past work data and automatically creates a work breakdown structure (WBS). Furthermore, it recognizes the emotions of factory staff and provides support and alerts according to their emotions.

[0363] Overview of program processing

[0364] The system starts by collecting historical business data and preprocessing it. The server collects data from project management tools, emails, chat logs, documents, etc. This data is collected and preprocessed using Python. Pandas is used to shape and clean the data, and Scikit-Learn is used to train machine learning models.

[0365] When a new task summary is input from a terminal, the server analyzes it and extracts the necessary tasks. It uses natural language processing technology to analyze the new task summary and generates appropriate tasks by referencing past data. To do this, it uses TfidfVectorizer to extract text features and RandomForestClassifier to predict tasks.

[0366] Next, a WBS is automatically created based on the generated tasks. The generated WBS is structured based on the time and dependencies of each task and displayed on the terminal. This display is provided as a user-friendly interface.

[0367] Additionally, the robot is equipped with an emotion engine that analyzes the emotional state of factory staff based on their input and behavior. The emotion engine uses the Hugging Face Transformer to tailor feedback and support content based on specific emotions. If the emotional state is likely to be negatively affected, the robot will provide special support or direct customers to a hotline.

[0368] Specific examples

[0369] For example, if a factory needs to perform a final inspection of a new part and reset the assembly line, enter "Final inspection of new part" and "Resetting the assembly line" as new tasks. The server generates the following tasks based on past similar projects:

[0370] Setting inspection standards for new parts

[0371] Inspection

[0372] Assembly line circuit settings

[0373] Test run of the reconfiguration

[0374] The WBS generated in this way is displayed visually on the terminal. If a factory worker types, "I'm feeling stressed today," the emotion engine analyzes it and provides an appropriate support message (e.g., "Attention, you need emotional support").

[0375] Prompt Sentence Examples

[0376] Example of new task input: "Final inspection of new parts"

[0377] Example of emotional state input: "I'm stressed today."

[0378] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0379] Step 1:

[0380] The server collects past business data. This data includes project management tools, emails, chat logs, and documents. Specifically, it collects information from each data source using APIs and data extraction scripts. The collected data is then stored in a database.

[0381] Inputs: Project management tools, emails, chat logs, documents

[0382] Output: Collected business data (stored in a database)

[0383] Step 2:

[0384] The server preprocesses the collected business data. Specifically, it uses Pandas to remove noise and duplicate data, and converts the data into a unified format. This step ensures the quality of the data.

[0385] Input: Collected business data

[0386] Output: Preprocessed business data

[0387] Step 3:

[0388] The server trains a machine learning model using the preprocessed data: it uses Scikit-Learn's TfidfVectorizer to extract text features for the task, and RandomForestClassifier to train a predictive model for the task.

[0389] Input: Preprocessed business data

[0390] Output: A trained machine learning model

[0391] Step 4:

[0392] The user inputs the outline of the new task via the terminal. For this, the task content is input in simple text format. When the outline of the new task is input, the terminal transmits the data to the server.

[0393] Input: New business summary (text format)

[0394] Output: Sent business summary data

[0395] Step 5:

[0396] The server uses natural language processing technology to analyze the outline of the new business and generate the necessary tasks based on past data. Specifically, it extracts text features of the new business using TfidfVectorizer and predicts tasks using a trained machine learning model.

[0397] Input: Summary data for new business

[0398] Output: Generated task list

[0399] Step 6:

[0400] The server automatically creates a work breakdown structure (WBS) based on the generated task list, including the task hierarchy, dependencies, and priorities.

[0401] Input: Generated task list

[0402] Output: Generated WBS

[0403] Step 7:

[0404] The terminal visually displays the generated WBS to the user, who can then check it and provide corrections or feedback as necessary.

[0405] Input: Generated WBS

[0406] Output: A visual representation of the WBS

[0407] Step 8:

[0408] The terminal collects user feedback on the generated WBS and transmits the information to the server.

[0409] Input: User feedback

[0410] Output: Feedback data sent

[0411] Step 9:

[0412] The server uses the collected feedback to retrain the machine learning model, which then improves its prediction accuracy in future runs.

[0413] Input: User feedback data

[0414] Output: A retrained machine learning model

[0415] Step 10:

[0416] The emotion engine analyzes the emotional state of factory staff from their input and actions, and uses the Hugging Face Transformer model to determine their emotional state, adjusting the interface and support content based on the user's emotional state.

[0417] Input: Text data expressing the emotions of factory staff

[0418] Output: Analyzed emotional state and support message based on it

[0419] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0420] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0421] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0422] [Second embodiment]

[0423] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0424] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0425] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0426] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0427] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0428] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0429] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0430] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0431] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0432] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0433] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0434] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0435] This invention relates to a system that automatically identifies tasks required for new work by utilizing past work data and generates an efficient work breakdown structure (WBS). This system consists of three main elements: a server, terminals, and users.

[0436] First, the server collects past business data, including project management tools, emails, chat logs, documents, etc. During this collection process, APIs and data extraction scripts are used for each data source to collect information and manage it centrally.

[0437] The server then preprocesses the collected data, removing duplicates and noise and converting it into a unified format. Ensuring data quality at this stage creates a dataset suitable for training subsequent machine learning models.

[0438] Using the preprocessed data, the server trains a machine learning model. This model analyzes past project data and learns task patterns and dependencies. By using a large amount of past data to train the model, it is possible to make highly accurate predictions.

[0439] The user inputs an outline of a new job via a terminal. The terminal then sends this input data to the server. The server then analyzes the received data using natural language processing technology and uses a machine learning model to extract the necessary tasks.

[0440] Once the tasks are extracted, the server automatically generates a WBS based on them. The generated WBS includes the hierarchical structure, dependencies, and priorities of the tasks. The terminal visually displays this WBS to the user.

[0441] As a concrete example, consider the case of launching a new marketing campaign. When a user types "launch a marketing campaign for a new product," the server generates the following task list based on data from similar past projects:

[0442] Marketing strategy development

[0443] Target Audience Research

[0444] Creating advertising materials

[0445] Advertisement delivery plan

[0446] Campaign monitoring and evaluation

[0447] The user can then review the generated WBS and add, delete, or modify tasks as needed. The device then sends the user's feedback to the server, which then retrains the machine learning model based on that feedback. This improves prediction accuracy for future tasks.

[0448] This system enables companies to respond quickly and efficiently to new tasks within the company or tasks that employees have no experience with due to transfers, improving work productivity and the success rate of projects.

[0449] The processing flow will be explained below.

[0450] Step 1: Data collection

[0451] The server collects historical business data using APIs and scripts that retrieve data from project management tools, emails, chat logs, documents, etc.

[0452] Specific operation: Obtain information such as project names, task names, deadlines, and assignees from project management tools via API. Filter and obtain related messages from emails and chat logs, and extract text data containing related keywords from documents.

[0453] Step 2: Data Preprocessing

[0454] The server pre-processes the collected data, which is the process of cleansing the data, removing duplicates and noise data, and converting it into a unified format.

[0455] What it does: It removes duplicate entries, maps information from different data sources into one unified format, filters out noisy data, and fills in gaps and incomplete data.

[0456] Step 3: Data training

[0457] The server uses the preprocessed data to train machine learning models, which analyze past project data and learn task patterns and dependencies.

[0458] Specific operation: Input data into a machine learning algorithm (e.g., decision tree, neural network) to generate and learn task co-occurrence patterns and dependency graphs within the project.

[0459] Step 4: Enter an outline of the new job

[0460] The terminal provides an interface for users to input the outline of a new project or task. The user inputs basic information about the new project or task (type of task, goals, resources, etc.).

[0461] Specific operation: An interface is displayed through the terminal input form, allowing the user to enter the type of work, goals, resources, deadlines, etc., and enter the required information.

[0462] Step 5: Task extraction

[0463] The server uses a machine learning model to extract the necessary tasks based on the input summary of the new business.

[0464] Specific operation: The input data is analyzed using a natural language processing algorithm, and the necessary task list is generated based on task patterns from related past projects.

[0465] Step 6: Generate WBS

[0466] The server automatically generates a work breakdown structure (WBS) based on the extracted tasks, including the hierarchical structure, dependencies, and priorities of the tasks.

[0467] Specific operations: Organize task lists hierarchically, represent dependencies between tasks in a graph structure, and set priorities and deadlines to build a WBS.

[0468] Step 7: View the WBS

[0469] The terminal displays the generated WBS to the user, allowing the user to visually check the overall task structure and dependencies.

[0470] Specific operation: Renders the WBS in Gantt chart format and hierarchical display format, and provides an interface that allows interactive operation.

[0471] Step 8: User Feedback

[0472] The user can review the generated WBS and make adjustments as needed, such as adding, deleting, or reordering tasks.

[0473] The terminal collects user feedback.

[0474] Specific operation: Record user operations (addition, deletion, change) for each task in the WBS and send them as feedback data.

[0475] Step 9: Retrain the model

[0476] The server retrains the machine learning model based on the collected feedback, improving the accuracy of future predictions.

[0477] What it does: Analyzes collected feedback data and retrains machine learning models to learn feedback patterns.

[0478] Through this processing flow, the system generates an efficient and accurate WBS, enabling companies to respond quickly to new business and transfers.

[0479] Example 1

[0480] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0481] Modern business activities require the ability to quickly and efficiently handle new tasks and unfamiliar work resulting from transfers. However, conventional methods have struggled to effectively utilize past work data to automatically identify the tasks required for new work. Furthermore, there has been a lack of feedback mechanisms to ensure the accuracy and usefulness of the generated work breakdown structure (WBS). This has led to issues such as reduced work productivity and project success rates.

[0482] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0483] In this invention, the server includes a means for collecting past business data, a means for preprocessing the collected data, a means for training a machine learning model using the preprocessed data, a means for inputting an outline of a new business, a means for extracting necessary tasks based on the input outline of the new business, a means for organizing the extracted tasks to generate a WBS, a means for visually displaying the generated WBS, a means for collecting feedback on the generated WBS, and a means for retraining the machine learning model based on the collected feedback. This makes it possible to effectively utilize past business data and respond to new business quickly and efficiently. Furthermore, the accuracy and usefulness of the generated WBS are improved, thereby improving business productivity and project success rates.

[0484] "Historical Business Data" refers to all records and information relating to previously executed projects or business operations, including project management tools, emails, message logs, documents, etc.

[0485] "Collection methods" refers to the methods and tools used to obtain necessary information from various data sources and manage it centrally.

[0486] "Preprocessing means" refers to techniques and tools for removing redundant data and noise data from collected data and converting it into a unified format.

[0487] A "machine learning model" refers to an algorithm or system that learns and predicts task patterns and dependencies based on data.

[0488] "Training methods" refers to techniques and tools used to train machine learning models using preprocessed data to improve their predictive accuracy.

[0489] The "new business overview" refers to the basic content and goals of the new business or project that the user intends to implement.

[0490] "Task" refers to the specific work or activities required to accomplish new business.

[0491] "Task extraction means" refers to methods and tools for finding and extracting necessary tasks from the outline of new work.

[0492] "WBS (Work Breakdown Structure)" refers to a hierarchical organization and structure of the tasks required to achieve the project's goals.

[0493] "Organizational methods" refer to techniques and tools for organizing and categorizing extracted tasks into a rational order and structure.

[0494] "Visual display means" refers to graphical interfaces and tools that present the generated WBS to users in an easy-to-understand manner.

[0495] "Feedback methods" refer to methods and tools for collecting user opinions and corrections and reflecting them in the system.

[0496] "Retraining methods" refer to methods and tools used to retrain machine learning models based on feedback collected from users, further improving their predictive accuracy.

[0497] This invention relates to a system that automatically identifies tasks required for new work by utilizing past work data and generates an efficient work breakdown structure (WBS). This system consists of three main elements: a server, terminals, and users.

[0498] First, the server collects past business data. This data includes data from project management tools, emails, message logs, documents, etc. During the collection process, APIs and data extraction scripts are used for each data source (e.g., project management tools, email servers, chat tools, cloud storage) to obtain information and manage it centrally. The server then stores the collected data in a specific directory.

[0499] Next, the server preprocesses the collected data. Specifically, it uses Python's Pandas library to read the data and remove duplicates and noise. It then uses regular expressions to filter out the noise and extract only the necessary information. To standardize the data format, it uses preprocessing modules from Numpy and Scikit-learn. This stage ensures the quality of the data and creates a dataset suitable for training subsequent machine learning models.

[0500] Using the preprocessed data, the server trains a machine learning model. TensorFlow or PyTorch is used to train a recurrent neural network (RNN) or Transformer model, which learns task patterns and dependencies from past project data. Cross-validation is used to evaluate the model, and hyperparameters are tuned as needed.

[0501] A user inputs the outline of a new task via a terminal. For example, if a user inputs "Run a marketing campaign for a new product," the data is sent from the terminal to the server in JSON format. The server analyzes this input data using natural language processing technology (e.g., BERT, GPT-3) and uses machine learning models to extract the necessary tasks.

[0502] Once tasks are extracted, the server automatically generates a WBS based on them. The generated WBS includes the task hierarchy, dependencies, and priorities. This WBS is displayed on the device using visualization libraries such as D3.js and Highcharts. Users can review the generated WBS and add, delete, or modify tasks as needed.

[0503] User feedback is sent from the device to the server, which then uses this feedback to retrain the machine learning model, improving the accuracy of task extraction and WBS generation from the next time onwards.

[0504] Specific examples

[0505] For example, when a user is running a marketing campaign for a new product, they can input the following prompt: "What are the tasks required to run a marketing campaign for a new product?" Based on this, the server analyzes data from similar past projects and generates a task list like this:

[0506] Marketing strategy development

[0507] Target Audience Research

[0508] Creating advertising materials

[0509] Advertisement delivery plan

[0510] Campaign monitoring and evaluation

[0511] As users review and modify the generated WBS, feedback is collected and the machine learning model is retrained based on this data, improving the efficiency of new work and the success rate of projects.

[0512] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0513] Step 1:

[0514] The server collects past business data from various data sources. The collected data includes project management tools, emails, message logs, documents, and more. Specifically, project data is obtained using the project management tool's API, and email data is collected using the email server's IMAP protocol. Message logs are downloaded using the chat tool's API, and document data is downloaded using the cloud storage API. The input is raw data obtained from each data source, and the output is a data file in a unified format stored on the server.

[0515] Step 2:

[0516] The server preprocesses the collected data. It uses Python's Pandas library to read the data and remove duplicates and noise. It uses regular expressions to filter out unnecessary strings and noise and extract the necessary information. To unify the data format, it uses preprocessing modules from Numpy and Scikit-learn. The input is the collected raw data, and the output is clean data after preprocessing.

[0517] Step 3:

[0518] The server uses the preprocessed data to train a machine learning model. It uses TensorFlow or PyTorch to train a recurrent neural network (RNN) or Transformer model. The model is trained by learning task patterns and dependencies based on past project data. The input is clean data that has been preprocessed, and the output is a trained machine learning model.

[0519] Step 4:

[0520] The user inputs the outline of a new business via the terminal. For example, the user inputs "Run a marketing campaign for a new product" in text format. The terminal converts this data into JSON format and sends it to the server. The input is the outline of the new business entered by the user, and the output is the JSON format data sent to the server.

[0521] Step 5:

[0522] When the server receives the input outline of the new work, it analyzes it using natural language processing technology (e.g., BERT, GPT-3). It then uses a machine learning model to extract the necessary tasks. The input is new work data in JSON format, and the output is a set of necessary tasks. Once the tasks are extracted, a WBS is automatically generated based on them. The generated WBS includes the hierarchical structure, dependencies, and priorities of the tasks.

[0523] Step 6:

[0524] The server sends the generated WBS to the terminal, which uses a visualization library such as D3.js or Highcharts to visually display the WBS to the user. The input is the task group and the generated data of the WBS, and the output is the visualized WBS that is displayed to the user.

[0525] Step 7:

[0526] The user checks the displayed WBS and adds, deletes, or modifies tasks as necessary. The results of the user's operations are sent from the terminal to the server as feedback. The input is the user's feedback operations, and the output is the feedback data.

[0527] Step 8:

[0528] The server retrains the machine learning model based on the collected feedback. The feedback data is used as new training data to update the machine learning model. This improves the accuracy of task extraction and WBS generation from the next time onwards. The input is the feedback data, and the output is the updated machine learning model.

[0529] (Application example 1)

[0530] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0531] Optimizing production processes in traditional factories relies primarily on the experience and intuition of managers, making efficient project management difficult. Furthermore, identifying appropriate tasks and creating a work breakdown structure (WBS) is particularly difficult for new work. This results in lower productivity and project success rates. Furthermore, it is not easy to monitor the progress of each process in real time and make appropriate readjustments.

[0532] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0533] In this invention, the server includes means for collecting past work data, means for preprocessing the collected data, means for training a machine learning model using the preprocessed data, means for inputting an outline of a new work, means for extracting necessary tasks based on the input outline of the new work, means for organizing the extracted tasks to generate a WBS, means for displaying the generated WBS, means for collecting feedback on the generated WBS, means for retraining the machine learning model based on the collected feedback, means for assigning tasks to robots to optimize production processes in a factory, and means for monitoring and readjusting the progress of each process in real time. This enables automatic and efficient optimization of production processes, thereby improving project success rates and productivity.

[0534] "Past business data" refers to all information related to past projects or work, including project management tools, emails, chat logs, and documents.

[0535] "Preprocessing means" refers to the process of removing redundancy and noise from collected data and converting the data into a unified format.

[0536] "Means for training machine learning models" refers to running machine learning algorithms using pre-processed historical business data to learn task patterns and dependencies.

[0537] The "means for inputting the outline of a new business" refers to an interface through which the user inputs details of a new business to be started.

[0538] "Means for extracting necessary tasks" refers to the process of automatically identifying appropriate tasks from the input outline of new work.

[0539] "Method for generating WBS" refers to the method for organizing the extracted tasks into a hierarchical structure and creating a work breakdown structure.

[0540] "Means for displaying the generated WBS" refers to an interface that visually displays the generated WBS to the user.

[0541] "Means for collecting feedback" refers to an interface for collecting comments and correction requests from users.

[0542] "Means for retraining the machine learning model" refers to the process of updating the machine learning model based on collected feedback to improve the accuracy of task extraction and WBS generation from the next time onwards.

[0543] "Means for assigning tasks to robots to optimize production processes" refers to a system that automatically distributes appropriate work instructions to robots in a factory.

[0544] "Means for monitoring and readjusting the progress of the process in real time" refers to a system for monitoring the progress of the production process in real time and readjusting it as necessary.

[0545] This invention is a system that uses past business data to automatically identify tasks required for new work and generate an efficient work breakdown structure (WBS).Application examples include a system that assigns tasks to robots to optimize production processes within a factory, and monitors and readjusts the progress of each process in real time.

[0546] The server first collects past business data, including project management tools, emails, chat logs, documents, etc. Data collection is centrally managed using APIs and data extraction scripts.

[0547] The server then preprocesses the collected data, which includes removing duplicates and noise and converting the data into a unified format, ensuring data quality and creating a dataset suitable for training machine learning models.

[0548] The server uses the preprocessed data to train a machine learning model. The model analyzes past project data and learns task patterns and dependencies. By using a large amount of past data for this training, highly accurate predictions are possible.

[0549] The user inputs an outline of the new job through a terminal. This input data is sent to the server, which then analyzes the received data using natural language processing technology and extracts the necessary tasks using a machine learning model.

[0550] The server automatically generates a WBS based on the extracted tasks. The generated WBS includes the hierarchical structure, dependencies, and priorities of the tasks. The terminal visually displays the generated WBS to the user.

[0551] Furthermore, the system has the ability to optimize production processes within the factory and assign tasks to robots. Specifically, it lists the tasks for robots in each process, monitors their progress in real time, and adjusts them as necessary.

[0552] Users can check the generated WBS and add, delete, or modify tasks as needed. This feedback is sent to the server, which then uses it to retrain the machine learning model and improve prediction accuracy for future projects.

[0553] As a concrete example, consider the case where a new plan for setting up a new product line is entered into a "production process optimization app" on a smartphone. The server generates the following task list based on similar past data:

[0554] 1. Manufacturing line design

[0555] 2. Machine settings

[0556] 3. Production Test

[0557] 4. Quality check

[0558] An example of a prompt sentence is "Please generate the optimal production process for designing a new manufacturing line for airplane parts." By using this prompt sentence, the server generates the optimal task list and WBS, realizing the efficiency of the production process.

[0559] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0560] Step 1:

[0561] The server collects historical business data using APIs and data extraction scripts. It receives input from data sources such as project management tools, emails, chat logs, and documents. It integrates this data and stores it in a central database. The output is unified historical business data.

[0562] Step 2:

[0563] The server preprocesses the collected data, specifically removing duplicate and noisy data and converting the data into a unified format. The input for this process is historical business data, and the output is a high-quality, unified dataset.

[0564] Step 3:

[0565] The server uses the preprocessed dataset to train a machine learning model by running a machine learning algorithm that learns patterns and dependencies in the data. The input to this process is the preprocessed dataset, and the output is a machine learning model that has learned the patterns and dependencies for the task.

[0566] Step 4:

[0567] The user inputs the outline of the new business through the terminal. The details of the business to be newly started are input as input to the terminal. The output is the outline data of the new business sent from the terminal.

[0568] Step 5:

[0569] The server analyzes the received outline of the new business using natural language processing technology. The input is the outline data of the new business, and the output is data extracting the necessary tasks.

[0570] Step 6:

[0571] The server generates a WBS based on the extracted task data. The input is the analyzed task data. The output is a WBS that includes the hierarchical structure, dependencies, and priorities of the tasks.

[0572] Step 7:

[0573] The terminal visually displays the generated WBS to the user. The input is the generated WBS, and the output is a representation of the WBS that the user can visually confirm.

[0574] Step 8:

[0575] The user inputs feedback for the generated WBS. The input is the feedback entered by the user via the terminal, and the output is the feedback data sent from the terminal to the server.

[0576] Step 9:

[0577] The server retrains the machine learning model based on the collected feedback: the input is the user feedback data, and the output is the updated machine learning model.

[0578] Step 10:

[0579] The server assigns tasks to robots to optimize production processes in the factory. The inputs are the generated WBS and robot work information, and the output is specific work instructions delivered to the robots.

[0580] Step 11:

[0581] The server monitors the progress of each process in real time and adjusts it as necessary. The input is real-time progress data, and the output is optimized process information.

[0582] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0583] This invention combines a system that uses past business data to automatically identify tasks required for new business and generate an efficient WBS (Work Breakdown Structure), with an emotion engine that recognizes user emotions. This system consists of four main elements: a server, a terminal, a user, and an emotion engine.

[0584] First, the server collects past business data. This data includes project management tools, emails, chat logs, documents, etc. The server uses APIs and data extraction scripts for each data source to collect information and manage it centrally.

[0585] The server then preprocesses the collected data, removing duplicates and noise and converting it into a unified format. Ensuring data quality at this stage creates a dataset suitable for training subsequent machine learning models.

[0586] Using the preprocessed data, the server trains a machine learning model. This model analyzes past project data and learns task patterns and dependencies. By using a large amount of past data to train the model, it is possible to make highly accurate predictions.

[0587] The user inputs an outline of a new job via a terminal. The terminal then sends this input data to the server. The server then analyzes the received data using natural language processing technology and uses a machine learning model to extract the necessary tasks.

[0588] Once tasks are extracted, the server automatically generates a WBS based on them. The generated WBS includes the hierarchical structure, dependencies, and priorities of the tasks. The terminal visually displays this WBS to the user. Furthermore, by incorporating an emotion engine, the system analyzes the user's emotional state and adjusts the way the WBS is displayed.

[0589] As a concrete example, consider the case of a new project to launch a new product. When a user types "launch a new product," the server generates the following task list based on data from similar past projects:

[0590] Product packaging design

[0591] Package design approval

[0592] Selection of sales channels

[0593] Sales channel contract negotiations

[0594] Creating a marketing plan

[0595] Marketing plan approval

[0596] Creating advertising materials

[0597] Formulating an advertising distribution plan

[0598] If a user reviews the generated WBS and feels, for example, nervous or anxious, the emotion engine will recognize this and determine that more support is needed, providing more detailed explanations or additional guidance.

[0599] The user then provides feedback on the generated WBS. The device collects the user's feedback and sends it to the server. The server then retrains the machine learning model based on the feedback, improving prediction accuracy for future iterations.

[0600] This emotion engine recognizes the user's emotional state and adjusts the interface and feedback based on that state, providing a user-friendly system. This further strengthens rapid response to new tasks and transfers, further improving work productivity and project success rates.

[0601] The processing flow will be explained below.

[0602] Step 1: Data collection

[0603] The server collects historical business data using APIs and scripts that retrieve data from project management tools, emails, chat logs, documents, etc.

[0604] Specific operation: Obtain information such as project names, task names, deadlines, and assignees from project management tools via API. Filter and obtain related messages from emails and chat logs, and extract text data containing related keywords from documents.

[0605] Step 2: Data Preprocessing

[0606] The server pre-processes the collected data, which is the process of cleansing the data, removing duplicates and noise data, and converting it into a unified format.

[0607] What it does: It removes duplicate entries, maps information from different data sources into one unified format, filters out noisy data, and fills in gaps and incomplete data.

[0608] Step 3: Data training

[0609] The server uses the preprocessed data to train machine learning models, which analyze past project data and learn task patterns and dependencies.

[0610] Specific operation: Input data into a machine learning algorithm (e.g., decision tree, neural network) to generate and learn task co-occurrence patterns and dependency graphs within the project.

[0611] Step 4: Enter an outline of the new job

[0612] The terminal provides an interface for users to input the outline of a new project or task. The user inputs basic information about the new project or task (type of task, goals, resources, etc.).

[0613] Specific operation: An interface is displayed through the terminal input form, allowing the user to enter the type of work, goals, resources, deadlines, etc., and enter the required information.

[0614] Step 5: Task extraction

[0615] The server uses a machine learning model to extract the necessary tasks based on the input summary of the new business.

[0616] Specific operation: The input data is analyzed using a natural language processing algorithm, and the necessary task list is generated based on task patterns from related past projects.

[0617] Step 6: Generate WBS

[0618] The server automatically generates a work breakdown structure (WBS) based on the extracted tasks, including the hierarchical structure, dependencies, and priorities of the tasks.

[0619] Specific operations: Organize task lists hierarchically, represent dependencies between tasks in a graph structure, and set priorities and deadlines to build a WBS.

[0620] Step 7: Emotion Recognition

[0621] The device sends the user's input data and operation history to the emotion engine, which analyzes the user's emotional state.

[0622] Specific operation: Input data (keyboard input speed and pressure, screen operation history) is input into the emotion engine, and an emotion analysis algorithm is used to recognize the user's emotional state (e.g., stress, anxiety, concentration, etc.).

[0623] Step 8: Emotion-Based Display Adjustment

[0624] The server receives feedback from the emotion engine and adjusts the display of the WBS, providing an environment that is easy for users to understand and work in.

[0625] Specific behavior: If the user is stressed, simplify the WBS display and add specific guidelines and support messages. Conversely, if the user is focused, display more detailed information and options.

[0626] Step 9: Gather feedback

[0627] The user can review the generated WBS and make adjustments as needed, such as adding, deleting, or reordering tasks.

[0628] The terminal collects user feedback.

[0629] Specific operation: Record user operations (addition, deletion, change) for each task in the WBS and send them as feedback data.

[0630] Step 10: Retrain the model

[0631] The server retrains the machine learning model based on the collected feedback, improving the accuracy of future predictions.

[0632] What it does: Analyzes collected feedback data and retrains machine learning models to learn feedback patterns.

[0633] This emotion engine recognizes the user's emotional state and adjusts the interface and feedback based on that state, providing a user-friendly system. This further strengthens the ability to respond quickly and efficiently to new tasks and transfers, further improving work productivity and project success rates.

[0634] Example 2

[0635] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0636] In today's business environment, launching and executing new projects requires significant planning and management. However, traditional project management systems are unable to effectively utilize past work data, making it difficult to efficiently identify the tasks required for new projects. Furthermore, systems are not designed to take into account the user's emotional state, resulting in a suboptimal user experience. Furthermore, even if feedback on the generated WBS is collected, there is insufficient means for applying that feedback to future projects. These issues result in reduced work efficiency and project success rates.

[0637] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for collecting past business data, a means for preprocessing the collected data, and a means for training a machine learning model using the preprocessed data. This makes it possible to effectively utilize past business data and efficiently identify tasks required for a new project. The server also includes a means for inputting an outline of a new business, a means for extracting necessary tasks based on the input outline of the new business, a means for organizing the extracted tasks to generate a WBS, a means for displaying the generated WBS, a means for collecting feedback on the generated WBS, a means for retraining a machine learning model based on the collected feedback, a means for analyzing a user's emotional state, and a means for adjusting the interface and feedback based on the analyzed emotional state. This makes it possible to improve the user experience and utilize feedback to increase the system's prediction accuracy.

[0638] "Historical business data" refers to information about previously executed tasks or projects, and is data obtained from project management tools, emails, electronic chat logs, documents, etc.

[0639] "Preprocessing" is the process of organizing collected data by removing redundancy and noise and converting it into a unified format, making it suitable for training machine learning models.

[0640] A "machine learning model" is a set of algorithms that are trained using collected and preprocessed data to learn patterns and dependencies of tasks required for new business operations.

[0641] The "new business outline" refers to detailed information about a new project or task, and is input data from the user that includes specific goals and necessary requirements.

[0642] "Task extraction" is the process of using a machine learning model to identify necessary tasks from the outline of a new business and define them as specific activity items.

[0643] "WBS" stands for "Work Breakdown Structure" and is a visual representation of the hierarchical tasks required to complete a project.

[0644] "Feedback" refers to information including opinions and improvements provided by users regarding the generated WBS, and is used as data to improve the system's prediction accuracy.

[0645] "Emotional state" refers to the user's emotional response or mood, and is the psychological state that is analyzed while using the system.

[0646] An "emotion engine" is a program or set of algorithms for detecting and analyzing a user's emotional state, with the goal of improving the user experience.

[0647] "Interface" refers to the screen and operating means that users use to interact with the system, and must be designed to be easy for users to use.

[0648] The present invention provides a system that combines four main elements: a server, a terminal, a user, and an emotion engine. A specific embodiment of this system will be described below.

[0649] First, the server collects historical business data from data sources, including project management tools, email systems, electronic chat logs, and document management systems. Specifically, it uses tools such as JIRA, Slack, mail servers, and Google Drive to execute APIs and data extraction scripts, and stores the necessary data in a centralized database.

[0650] The server then preprocesses the collected data. This includes removing duplicate data, reducing noise, and converting the data into a unified format, such as JSON, to ensure data quality. Specifically, the data is cleaned and centralized using an SQLite database and regular expressions.

[0651] The server uses the preprocessed data to train a machine learning model using a machine learning framework such as TensorFlow or PyTorch, which analyzes past project data and learns task patterns and dependencies.

[0652] Next, the user enters a summary of the new business via the terminal, for example, by entering "Implement market launch of new product" into a web form. This information is sent from the terminal to the server as an HTTP POST request.

[0653] The server analyzes the new business data it receives using natural language processing (NLP) technology and uses machine learning models to extract necessary tasks. Specifically, it uses NLP libraries such as spaCy and NLTK to tokenize and analyze the data. It then identifies tasks based on the analysis results and defines them as specific actions.

[0654] Once the tasks are extracted, the server automatically generates a work breakdown structure (WBS) based on the extracted tasks. The generated WBS includes the hierarchical structure, dependencies, and priorities of the tasks. The terminal visually displays this WBS to the user.

[0655] Furthermore, an emotion engine built into the server analyzes the user's emotional state. When a user checks the WBS, emotion analysis is performed based on their reaction. For example, if the user is feeling nervous or anxious, the emotion engine will recognize this and display additional guidance or detailed explanations. This analysis uses a cloud emotion analysis API (for example, IBM Watson).

[0656] Users provide feedback on the generated WBS. The device collects the feedback and sends it to the server, which then retrains the machine learning model based on the feedback to improve prediction accuracy in the future.

[0657] As a concrete example, when a user starts a new project to launch a new product, they can input "launch a new product," and the server will generate the following task list based on data from similar past projects:

[0658] Product packaging design

[0659] Package design approval

[0660] Selection of sales channels

[0661] Sales channel contract negotiations

[0662] Creating a marketing plan

[0663] Marketing plan approval

[0664] Creating advertising materials

[0665] Formulating an advertising distribution plan

[0666] An example of a prompt sentence to be input to the generative AI model is, "Please tell me the tasks required to launch a new product to the market." By using this prompt sentence, the AI ​​model can efficiently extract the tasks required for the new business and propose an appropriate WBS.

[0667] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0668] Program processing flow

[0669] Step 1:

[0670] The server collects past business data.

[0671] Input: API requests from project management tools, email systems, electronic chat logs, and document management systems.

[0672] Data processing / calculation: Send API requests and extract the necessary information from each data source.

[0673] Output: Save the retrieved data to the database.

[0674] Specific behavior:

[0675] Get project ticket information from the project management tool API.

[0676] Email data is retrieved using the IMAP email system.

[0677] Extract chat logs via API.

[0678] Download relevant documents from your document management system.

[0679] Step 2:

[0680] The server preprocesses the collected data.

[0681] Input: Collected business data.

[0682] Data processing / calculation: Delete duplicate data, reduce noise data, and convert to a unified format.

[0683] Output: Preprocessed data converted into a unified format.

[0684] Specific behavior:

[0685] Execute a SQL query to remove duplicate data in an SQLite database.

[0686] Use regular expressions to filter out noisy messages.

[0687] All data will be unified and centralized in JSON format.

[0688] Step 3:

[0689] The server trains the machine learning model.

[0690] Input: Preprocessed business data.

[0691] Data processing / computation: Input the data into a framework such as TensorFlow or PyTorch and train a model.

[0692] Output: A trained machine learning model.

[0693] Specific behavior:

[0694] Input the dataset into the TensorFlow model and run the training loop.

[0695] Tune hyperparameters to improve model accuracy.

[0696] Save the trained model and use it in a later step.

[0697] Step 4:

[0698] The user inputs an outline of a new job via a terminal.

[0699] Input: Enter text about the new business.

[0700] Data processing / calculation: The data entered by the user is sent as an HTTP POST request.

[0701] Output: A summary of the new job is sent to the server.

[0702] Specific behavior:

[0703] A user fills out a web form with a project description (e.g., "Implement market introduction of a new product").

[0704] The device generates and sends an HTTP POST request to send the input content to the server.

[0705] Step 5:

[0706] The server analyzes the new business data and extracts tasks.

[0707] Input: New business summary, trained machine learning model.

[0708] Data processing / computation: Tokenize the data using natural language processing and use models to extract tasks.

[0709] Output: The extracted task list.

[0710] Specific behavior:

[0711] Use spaCy to tokenize the outline of the new business and perform semantic analysis.

[0712] Input data into a trained machine learning model to extract the required tasks.

[0713] Step 6:

[0714] The server automatically generates the WBS.

[0715] Input: The extracted task list.

[0716] Data processing / calculation: Calculate the hierarchical structure and dependencies of tasks and generate a WBS.

[0717] Output: The generated WBS.

[0718] Specific behavior:

[0719] Based on the extracted tasks, an algorithm is used to generate a hierarchical structure for the WBS.

[0720] Save the generated WBS in JSON format and use it in a later step.

[0721] Step 7:

[0722] The server uses an emotion engine to analyze the user's emotional state.

[0723] Input: User operation logs and input data.

[0724] Data processing / calculation: Estimate emotional state using emotion analysis API.

[0725] Output: Parsed emotion data.

[0726] Specific behavior:

[0727] Collects logs of user keyboard input and mouse operations.

[0728] Emotional states are estimated using a cloud-based emotion analysis API.

[0729] Step 8:

[0730] The user provides feedback on the generated WBS.

[0731] Input: User feedback.

[0732] Data processing / calculation: Generate an HTTP POST request to send the feedback to the server.

[0733] Output: The feedback is sent to the server.

[0734] Specific behavior:

[0735] The user fills out the feedback form and clicks the submit button.

[0736] The device generates and sends an HTTP POST request to send the feedback content to the server.

[0737] Step 9:

[0738] The server retrains the machine learning model based on the collected feedback.

[0739] Input: Feedback data, existing trained model.

[0740] Data processing / computation: Add feedback data and retrain the model.

[0741] Output: A machine learning model with improved accuracy.

[0742] Specific behavior:

[0743] Add new feedback data to the dataset.

[0744] Retrain your machine learning model and see improvements in accuracy.

[0745] Save the improved model.

[0746] (Application example 2)

[0747] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0748] Conventional work management systems have difficulty extracting appropriate tasks from past data and generating an efficient work breakdown structure (WBS) when starting a new task. Furthermore, they do not provide an interface or support that takes into account the user's emotional state, which can cause stress for users and lead to lower productivity and a deterioration in work efficiency. Furthermore, they lack a mechanism for appropriately collecting feedback and improving the accuracy of future predictions. These issues need to be resolved.

[0749] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0750] In this invention, the server includes means for collecting past business data, means for preprocessing the collected data, means for training a machine learning model using the preprocessed data, means for inputting an outline of a new business, means for extracting necessary tasks based on the input outline of the new business, means for organizing the extracted tasks to generate a work breakdown structure (WBS), means for displaying the generated WBS, means for collecting feedback on the generated WBS, means for retraining the machine learning model based on the collected feedback, and means for adjusting the WBS display method and support content based on the user's emotional state using an emotion engine that recognizes user emotions. This makes it possible to efficiently extract tasks necessary for new business based on past data, generate and display a WBS, and provide an interface and support that responds to the user's emotions.

[0751] "Past business data" refers to information about various business operations that have been carried out in the past, including data from project management tools, emails, chat logs, documents, etc.

[0752] "Preprocessing" refers to the process of removing duplication and noise from collected data and standardizing the format.

[0753] A "machine learning model" refers to an algorithm or computational model used to analyze and predict data based on collected data.

[0754] "Outline of new business" refers to basic information such as the purpose, scope, and conditions of the new business.

[0755] "Task extraction" refers to identifying and listing the necessary work items based on the outline of the new business.

[0756] A "Work Breakdown Structure (WBS)" is a tool used to hierarchically break down the work involved in a project or task and organize dependencies and priorities.

[0757] "Display" refers to visually presenting the generated information and WBS to the user.

[0758] "Feedback" refers to information collected from users, such as opinions and ratings, that is used to improve the system and retrain the model.

[0759] An "emotion engine" refers to a system that analyzes a user's emotional state and adjusts the system's behavior and display accordingly.

[0760] An embodiment of this invention will be described in detail. This system automatically generates work schedules within a factory and efficiently manages robot work. It generates new tasks based on past work data and automatically creates a work breakdown structure (WBS). Furthermore, it recognizes the emotions of factory staff and provides support and alerts according to their emotions.

[0761] Overview of program processing

[0762] The system starts by collecting historical business data and preprocessing it. The server collects data from project management tools, emails, chat logs, documents, etc. This data is collected and preprocessed using Python. Pandas is used to shape and clean the data, and Scikit-Learn is used to train machine learning models.

[0763] When a new task summary is input from a terminal, the server analyzes it and extracts the necessary tasks. It uses natural language processing technology to analyze the new task summary and generates appropriate tasks by referencing past data. To do this, it uses TfidfVectorizer to extract text features and RandomForestClassifier to predict tasks.

[0764] Next, a WBS is automatically created based on the generated tasks. The generated WBS is structured based on the time and dependencies of each task and displayed on the terminal. This display is provided as a user-friendly interface.

[0765] Additionally, the robot is equipped with an emotion engine that analyzes the emotional state of factory staff based on their input and behavior. The emotion engine uses the Hugging Face Transformer to tailor feedback and support content based on specific emotions. If the emotional state is likely to be negatively affected, the robot will provide special support or direct customers to a hotline.

[0766] Specific examples

[0767] For example, if a factory needs to perform a final inspection of a new part and reset the assembly line, enter "Final inspection of new part" and "Resetting the assembly line" as new tasks. The server generates the following tasks based on past similar projects:

[0768] Setting inspection standards for new parts

[0769] Inspection

[0770] Assembly line circuit settings

[0771] Test run of the reconfiguration

[0772] The WBS generated in this way is displayed visually on the terminal. If a factory worker types, "I'm feeling stressed today," the emotion engine analyzes it and provides an appropriate support message (e.g., "Attention, you need emotional support").

[0773] Prompt Sentence Examples

[0774] Example of new task input: "Final inspection of new parts"

[0775] Example of emotional state input: "I'm stressed today."

[0776] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0777] Step 1:

[0778] The server collects past business data. This data includes project management tools, emails, chat logs, and documents. Specifically, it collects information from each data source using APIs and data extraction scripts. The collected data is then stored in a database.

[0779] Inputs: Project management tools, emails, chat logs, documents

[0780] Output: Collected business data (stored in a database)

[0781] Step 2:

[0782] The server preprocesses the collected business data. Specifically, it uses Pandas to remove noise and duplicate data, and converts the data into a unified format. This step ensures the quality of the data.

[0783] Input: Collected business data

[0784] Output: Preprocessed business data

[0785] Step 3:

[0786] The server trains a machine learning model using the preprocessed data: it uses Scikit-Learn's TfidfVectorizer to extract text features for the task, and RandomForestClassifier to train a predictive model for the task.

[0787] Input: Preprocessed business data

[0788] Output: A trained machine learning model

[0789] Step 4:

[0790] The user inputs the outline of the new task via the terminal. For this, the task content is input in simple text format. When the outline of the new task is input, the terminal transmits the data to the server.

[0791] Input: New business summary (text format)

[0792] Output: Sent business summary data

[0793] Step 5:

[0794] The server uses natural language processing technology to analyze the outline of the new business and generate the necessary tasks based on past data. Specifically, it extracts text features of the new business using TfidfVectorizer and predicts tasks using a trained machine learning model.

[0795] Input: Summary data for new business

[0796] Output: Generated task list

[0797] Step 6:

[0798] The server automatically creates a work breakdown structure (WBS) based on the generated task list, including the task hierarchy, dependencies, and priorities.

[0799] Input: Generated task list

[0800] Output: Generated WBS

[0801] Step 7:

[0802] The terminal visually displays the generated WBS to the user, who can then check it and provide corrections or feedback as necessary.

[0803] Input: Generated WBS

[0804] Output: A visual representation of the WBS

[0805] Step 8:

[0806] The terminal collects user feedback on the generated WBS and transmits the information to the server.

[0807] Input: User feedback

[0808] Output: Feedback data sent

[0809] Step 9:

[0810] The server uses the collected feedback to retrain the machine learning model, which then improves its prediction accuracy in future runs.

[0811] Input: User feedback data

[0812] Output: A retrained machine learning model

[0813] Step 10:

[0814] The emotion engine analyzes the emotional state of factory staff from their input and actions, and uses the Hugging Face Transformer model to determine their emotional state, adjusting the interface and support content based on the user's emotional state.

[0815] Input: Text data expressing the emotions of factory staff

[0816] Output: Analyzed emotional state and support message based on it

[0817] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0818] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0819] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0820] [Third embodiment]

[0821] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0822] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0823] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0824] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0825] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0826] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0827] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0828] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0829] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0830] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0831] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0832] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0833] This invention relates to a system that automatically identifies tasks required for new work by utilizing past work data and generates an efficient work breakdown structure (WBS). This system consists of three main elements: a server, terminals, and users.

[0834] First, the server collects past business data, including project management tools, emails, chat logs, documents, etc. During this collection process, APIs and data extraction scripts are used for each data source to collect information and manage it centrally.

[0835] The server then preprocesses the collected data, removing duplicates and noise and converting it into a unified format. Ensuring data quality at this stage creates a dataset suitable for training subsequent machine learning models.

[0836] Using the preprocessed data, the server trains a machine learning model. This model analyzes past project data and learns task patterns and dependencies. By using a large amount of past data to train the model, it is possible to make highly accurate predictions.

[0837] The user inputs an outline of a new job via a terminal. The terminal then sends this input data to the server. The server then analyzes the received data using natural language processing technology and uses a machine learning model to extract the necessary tasks.

[0838] Once the tasks are extracted, the server automatically generates a WBS based on them. The generated WBS includes the hierarchical structure, dependencies, and priorities of the tasks. The terminal visually displays this WBS to the user.

[0839] As a concrete example, consider the case of launching a new marketing campaign. When a user types "launch a marketing campaign for a new product," the server generates the following task list based on data from similar past projects:

[0840] Marketing strategy development

[0841] Target Audience Research

[0842] Creating advertising materials

[0843] Advertisement delivery plan

[0844] Campaign monitoring and evaluation

[0845] The user can then review the generated WBS and add, delete, or modify tasks as needed. The device then sends the user's feedback to the server, which then retrains the machine learning model based on that feedback. This improves prediction accuracy for future tasks.

[0846] This system enables companies to respond quickly and efficiently to new tasks within the company or tasks that employees have no experience with due to transfers, improving work productivity and the success rate of projects.

[0847] The processing flow will be explained below.

[0848] Step 1: Data collection

[0849] The server collects historical business data using APIs and scripts that retrieve data from project management tools, emails, chat logs, documents, etc.

[0850] Specific operation: Obtain information such as project names, task names, deadlines, and assignees from project management tools via API. Filter and obtain related messages from emails and chat logs, and extract text data containing related keywords from documents.

[0851] Step 2: Data Preprocessing

[0852] The server pre-processes the collected data, which is the process of cleansing the data, removing duplicates and noise data, and converting it into a unified format.

[0853] What it does: It removes duplicate entries, maps information from different data sources into one unified format, filters out noisy data, and fills in gaps and incomplete data.

[0854] Step 3: Data training

[0855] The server uses the preprocessed data to train machine learning models, which analyze past project data and learn task patterns and dependencies.

[0856] Specific operation: Input data into a machine learning algorithm (e.g., decision tree, neural network) to generate and learn task co-occurrence patterns and dependency graphs within the project.

[0857] Step 4: Enter an outline of the new job

[0858] The terminal provides an interface for users to input the outline of a new project or task. The user inputs basic information about the new project or task (type of task, goals, resources, etc.).

[0859] Specific operation: An interface is displayed through the terminal input form, allowing the user to enter the type of work, goals, resources, deadlines, etc., and enter the required information.

[0860] Step 5: Task extraction

[0861] The server uses a machine learning model to extract the necessary tasks based on the input summary of the new business.

[0862] Specific operation: The input data is analyzed using a natural language processing algorithm, and the necessary task list is generated based on task patterns from related past projects.

[0863] Step 6: Generate WBS

[0864] The server automatically generates a work breakdown structure (WBS) based on the extracted tasks, including the hierarchical structure, dependencies, and priorities of the tasks.

[0865] Specific operations: Organize task lists hierarchically, represent dependencies between tasks in a graph structure, and set priorities and deadlines to build a WBS.

[0866] Step 7: View the WBS

[0867] The terminal displays the generated WBS to the user, allowing the user to visually check the overall task structure and dependencies.

[0868] Specific operation: Renders the WBS in Gantt chart format and hierarchical display format, and provides an interface that allows interactive operation.

[0869] Step 8: User Feedback

[0870] The user can review the generated WBS and make adjustments as needed, such as adding, deleting, or reordering tasks.

[0871] The terminal collects user feedback.

[0872] Specific operation: Record user operations (addition, deletion, change) for each task in the WBS and send them as feedback data.

[0873] Step 9: Retrain the model

[0874] The server retrains the machine learning model based on the collected feedback, improving the accuracy of future predictions.

[0875] What it does: Analyzes collected feedback data and retrains machine learning models to learn feedback patterns.

[0876] Through this processing flow, the system generates an efficient and accurate WBS, enabling companies to respond quickly to new business and transfers.

[0877] Example 1

[0878] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0879] Modern business activities require the ability to quickly and efficiently handle new tasks and unfamiliar work resulting from transfers. However, conventional methods have struggled to effectively utilize past work data to automatically identify the tasks required for new work. Furthermore, there has been a lack of feedback mechanisms to ensure the accuracy and usefulness of the generated work breakdown structure (WBS). This has led to issues such as reduced work productivity and project success rates.

[0880] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0881] In this invention, the server includes a means for collecting past business data, a means for preprocessing the collected data, a means for training a machine learning model using the preprocessed data, a means for inputting an outline of a new business, a means for extracting necessary tasks based on the input outline of the new business, a means for organizing the extracted tasks to generate a WBS, a means for visually displaying the generated WBS, a means for collecting feedback on the generated WBS, and a means for retraining the machine learning model based on the collected feedback. This makes it possible to effectively utilize past business data and respond to new business quickly and efficiently. Furthermore, the accuracy and usefulness of the generated WBS are improved, thereby improving business productivity and project success rates.

[0882] "Historical Business Data" refers to all records and information relating to previously executed projects or business operations, including project management tools, emails, message logs, documents, etc.

[0883] "Collection methods" refers to the methods and tools used to obtain necessary information from various data sources and manage it centrally.

[0884] "Preprocessing means" refers to techniques and tools for removing redundant data and noise data from collected data and converting it into a unified format.

[0885] A "machine learning model" refers to an algorithm or system that learns and predicts task patterns and dependencies based on data.

[0886] "Training methods" refers to techniques and tools used to train machine learning models using preprocessed data to improve their predictive accuracy.

[0887] The "new business overview" refers to the basic content and goals of the new business or project that the user intends to implement.

[0888] "Task" refers to the specific work or activities required to accomplish new business.

[0889] "Task extraction means" refers to methods and tools for finding and extracting necessary tasks from the outline of new work.

[0890] "WBS (Work Breakdown Structure)" refers to a hierarchical organization and structure of the tasks required to achieve the project's goals.

[0891] "Organizational methods" refer to techniques and tools for organizing and categorizing extracted tasks into a rational order and structure.

[0892] "Visual display means" refers to graphical interfaces and tools that present the generated WBS to users in an easy-to-understand manner.

[0893] "Feedback methods" refer to methods and tools for collecting user opinions and corrections and reflecting them in the system.

[0894] "Retraining methods" refer to methods and tools used to retrain machine learning models based on feedback collected from users, further improving their predictive accuracy.

[0895] This invention relates to a system that automatically identifies tasks required for new work by utilizing past work data and generates an efficient work breakdown structure (WBS). This system consists of three main elements: a server, terminals, and users.

[0896] First, the server collects past business data. This data includes data from project management tools, emails, message logs, documents, etc. During the collection process, APIs and data extraction scripts are used for each data source (e.g., project management tools, email servers, chat tools, cloud storage) to obtain information and manage it centrally. The server then stores the collected data in a specific directory.

[0897] Next, the server preprocesses the collected data. Specifically, it uses Python's Pandas library to read the data and remove duplicates and noise. It then uses regular expressions to filter out the noise and extract only the necessary information. To standardize the data format, it uses preprocessing modules from Numpy and Scikit-learn. This stage ensures the quality of the data and creates a dataset suitable for training subsequent machine learning models.

[0898] Using the preprocessed data, the server trains a machine learning model. TensorFlow or PyTorch is used to train a recurrent neural network (RNN) or Transformer model, which learns task patterns and dependencies from past project data. Cross-validation is used to evaluate the model, and hyperparameters are tuned as needed.

[0899] A user inputs the outline of a new task via a terminal. For example, if a user inputs "Run a marketing campaign for a new product," the data is sent from the terminal to the server in JSON format. The server analyzes this input data using natural language processing technology (e.g., BERT, GPT-3) and uses machine learning models to extract the necessary tasks.

[0900] Once tasks are extracted, the server automatically generates a WBS based on them. The generated WBS includes the task hierarchy, dependencies, and priorities. This WBS is displayed on the device using visualization libraries such as D3.js and Highcharts. Users can review the generated WBS and add, delete, or modify tasks as needed.

[0901] User feedback is sent from the device to the server, which then uses this feedback to retrain the machine learning model, improving the accuracy of task extraction and WBS generation from the next time onwards.

[0902] Specific examples

[0903] For example, when a user is running a marketing campaign for a new product, they can input the following prompt: "What are the tasks required to run a marketing campaign for a new product?" Based on this, the server analyzes data from similar past projects and generates a task list like this:

[0904] Marketing strategy development

[0905] Target Audience Research

[0906] Creating advertising materials

[0907] Advertisement delivery plan

[0908] Campaign monitoring and evaluation

[0909] As users review and modify the generated WBS, feedback is collected and the machine learning model is retrained based on this data, improving the efficiency of new work and the success rate of projects.

[0910] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0911] Step 1:

[0912] The server collects past business data from various data sources. The collected data includes project management tools, emails, message logs, documents, and more. Specifically, project data is obtained using the project management tool's API, and email data is collected using the email server's IMAP protocol. Message logs are downloaded using the chat tool's API, and document data is downloaded using the cloud storage API. The input is raw data obtained from each data source, and the output is a data file in a unified format stored on the server.

[0913] Step 2:

[0914] The server preprocesses the collected data. It uses Python's Pandas library to read the data and remove duplicates and noise. It uses regular expressions to filter out unnecessary strings and noise and extract the necessary information. To unify the data format, it uses preprocessing modules from Numpy and Scikit-learn. The input is the collected raw data, and the output is clean data after preprocessing.

[0915] Step 3:

[0916] The server uses the preprocessed data to train a machine learning model. It uses TensorFlow or PyTorch to train a recurrent neural network (RNN) or Transformer model. The model is trained by learning task patterns and dependencies based on past project data. The input is clean data that has been preprocessed, and the output is a trained machine learning model.

[0917] Step 4:

[0918] The user inputs the outline of a new business via the terminal. For example, the user inputs "Run a marketing campaign for a new product" in text format. The terminal converts this data into JSON format and sends it to the server. The input is the outline of the new business entered by the user, and the output is the JSON format data sent to the server.

[0919] Step 5:

[0920] When the server receives the input outline of the new work, it analyzes it using natural language processing technology (e.g., BERT, GPT-3). It then uses a machine learning model to extract the necessary tasks. The input is new work data in JSON format, and the output is a set of necessary tasks. Once the tasks are extracted, a WBS is automatically generated based on them. The generated WBS includes the hierarchical structure, dependencies, and priorities of the tasks.

[0921] Step 6:

[0922] The server sends the generated WBS to the terminal, which uses a visualization library such as D3.js or Highcharts to visually display the WBS to the user. The input is the task group and the generated data of the WBS, and the output is the visualized WBS that is displayed to the user.

[0923] Step 7:

[0924] The user checks the displayed WBS and adds, deletes, or modifies tasks as necessary. The results of the user's operations are sent from the terminal to the server as feedback. The input is the user's feedback operations, and the output is the feedback data.

[0925] Step 8:

[0926] The server retrains the machine learning model based on the collected feedback. The feedback data is used as new training data to update the machine learning model. This improves the accuracy of task extraction and WBS generation from the next time onwards. The input is the feedback data, and the output is the updated machine learning model.

[0927] (Application example 1)

[0928] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0929] Optimizing production processes in traditional factories relies primarily on the experience and intuition of managers, making efficient project management difficult. Furthermore, identifying appropriate tasks and creating a work breakdown structure (WBS) is particularly difficult for new work. This results in lower productivity and project success rates. Furthermore, it is not easy to monitor the progress of each process in real time and make appropriate readjustments.

[0930] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0931] In this invention, the server includes means for collecting past work data, means for preprocessing the collected data, means for training a machine learning model using the preprocessed data, means for inputting an outline of a new work, means for extracting necessary tasks based on the input outline of the new work, means for organizing the extracted tasks to generate a WBS, means for displaying the generated WBS, means for collecting feedback on the generated WBS, means for retraining the machine learning model based on the collected feedback, means for assigning tasks to robots to optimize production processes in a factory, and means for monitoring and readjusting the progress of each process in real time. This enables automatic and efficient optimization of production processes, thereby improving project success rates and productivity.

[0932] "Past business data" refers to all information related to past projects or work, including project management tools, emails, chat logs, and documents.

[0933] "Preprocessing means" refers to the process of removing redundancy and noise from collected data and converting the data into a unified format.

[0934] "Means for training machine learning models" refers to running machine learning algorithms using pre-processed historical business data to learn task patterns and dependencies.

[0935] The "means for inputting the outline of a new business" refers to an interface through which the user inputs details of a new business to be started.

[0936] "Means for extracting necessary tasks" refers to the process of automatically identifying appropriate tasks from the input outline of new work.

[0937] "Method for generating WBS" refers to the method for organizing the extracted tasks into a hierarchical structure and creating a work breakdown structure.

[0938] "Means for displaying the generated WBS" refers to an interface that visually displays the generated WBS to the user.

[0939] "Means for collecting feedback" refers to an interface for collecting comments and correction requests from users.

[0940] "Means for retraining the machine learning model" refers to the process of updating the machine learning model based on collected feedback to improve the accuracy of task extraction and WBS generation from the next time onwards.

[0941] "Means for assigning tasks to robots to optimize production processes" refers to a system that automatically distributes appropriate work instructions to robots in a factory.

[0942] "Means for monitoring and readjusting the progress of the process in real time" refers to a system for monitoring the progress of the production process in real time and readjusting it as necessary.

[0943] This invention is a system that uses past business data to automatically identify tasks required for new work and generate an efficient work breakdown structure (WBS).Application examples include a system that assigns tasks to robots to optimize production processes within a factory, and monitors and readjusts the progress of each process in real time.

[0944] The server first collects past business data, including project management tools, emails, chat logs, documents, etc. Data collection is centrally managed using APIs and data extraction scripts.

[0945] The server then preprocesses the collected data, which includes removing duplicates and noise and converting the data into a unified format, ensuring data quality and creating a dataset suitable for training machine learning models.

[0946] The server uses the preprocessed data to train a machine learning model. The model analyzes past project data and learns task patterns and dependencies. By using a large amount of past data for this training, highly accurate predictions are possible.

[0947] The user inputs an outline of the new job through a terminal. This input data is sent to the server, which then analyzes the received data using natural language processing technology and extracts the necessary tasks using a machine learning model.

[0948] The server automatically generates a WBS based on the extracted tasks. The generated WBS includes the hierarchical structure, dependencies, and priorities of the tasks. The terminal visually displays the generated WBS to the user.

[0949] Furthermore, the system has the ability to optimize production processes within the factory and assign tasks to robots. Specifically, it lists the tasks for robots in each process, monitors their progress in real time, and adjusts them as necessary.

[0950] Users can check the generated WBS and add, delete, or modify tasks as needed. This feedback is sent to the server, which then uses it to retrain the machine learning model and improve prediction accuracy for future projects.

[0951] As a concrete example, consider the case where a new plan for setting up a new product line is entered into a "production process optimization app" on a smartphone. The server generates the following task list based on similar past data:

[0952] 1. Manufacturing line design

[0953] 2. Machine settings

[0954] 3. Production Test

[0955] 4. Quality check

[0956] An example of a prompt sentence is "Please generate the optimal production process for designing a new manufacturing line for airplane parts." By using this prompt sentence, the server generates the optimal task list and WBS, realizing the efficiency of the production process.

[0957] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0958] Step 1:

[0959] The server collects historical business data using APIs and data extraction scripts. It receives input from data sources such as project management tools, emails, chat logs, and documents. It integrates this data and stores it in a central database. The output is unified historical business data.

[0960] Step 2:

[0961] The server preprocesses the collected data, specifically removing duplicate and noisy data and converting the data into a unified format. The input for this process is historical business data, and the output is a high-quality, unified dataset.

[0962] Step 3:

[0963] The server uses the preprocessed dataset to train a machine learning model by running a machine learning algorithm that learns patterns and dependencies in the data. The input to this process is the preprocessed dataset, and the output is a machine learning model that has learned the patterns and dependencies for the task.

[0964] Step 4:

[0965] The user inputs the outline of the new business through the terminal. The details of the business to be newly started are input as input to the terminal. The output is the outline data of the new business sent from the terminal.

[0966] Step 5:

[0967] The server analyzes the received outline of the new business using natural language processing technology. The input is the outline data of the new business, and the output is data extracting the necessary tasks.

[0968] Step 6:

[0969] The server generates a WBS based on the extracted task data. The input is the analyzed task data. The output is a WBS that includes the hierarchical structure, dependencies, and priorities of the tasks.

[0970] Step 7:

[0971] The terminal visually displays the generated WBS to the user. The input is the generated WBS, and the output is a representation of the WBS that the user can visually confirm.

[0972] Step 8:

[0973] The user inputs feedback for the generated WBS. The input is the feedback entered by the user via the terminal, and the output is the feedback data sent from the terminal to the server.

[0974] Step 9:

[0975] The server retrains the machine learning model based on the collected feedback: the input is the user feedback data, and the output is the updated machine learning model.

[0976] Step 10:

[0977] The server assigns tasks to robots to optimize production processes in the factory. The inputs are the generated WBS and robot work information, and the output is specific work instructions delivered to the robots.

[0978] Step 11:

[0979] The server monitors the progress of each process in real time and adjusts it as necessary. The input is real-time progress data, and the output is optimized process information.

[0980] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0981] This invention combines a system that uses past business data to automatically identify tasks required for new business and generate an efficient WBS (Work Breakdown Structure), with an emotion engine that recognizes user emotions. This system consists of four main elements: a server, a terminal, a user, and an emotion engine.

[0982] First, the server collects past business data. This data includes project management tools, emails, chat logs, documents, etc. The server uses APIs and data extraction scripts for each data source to collect information and manage it centrally.

[0983] The server then preprocesses the collected data, removing duplicates and noise and converting it into a unified format. Ensuring data quality at this stage creates a dataset suitable for training subsequent machine learning models.

[0984] Using the preprocessed data, the server trains a machine learning model. This model analyzes past project data and learns task patterns and dependencies. By using a large amount of past data to train the model, it is possible to make highly accurate predictions.

[0985] The user inputs an outline of a new job via a terminal. The terminal then sends this input data to the server. The server then analyzes the received data using natural language processing technology and uses a machine learning model to extract the necessary tasks.

[0986] Once tasks are extracted, the server automatically generates a WBS based on them. The generated WBS includes the hierarchical structure, dependencies, and priorities of the tasks. The terminal visually displays this WBS to the user. Furthermore, by incorporating an emotion engine, the system analyzes the user's emotional state and adjusts the way the WBS is displayed.

[0987] As a concrete example, consider the case of a new project to launch a new product. When a user types "launch a new product," the server generates the following task list based on data from similar past projects:

[0988] Product packaging design

[0989] Package design approval

[0990] Selection of sales channels

[0991] Sales channel contract negotiations

[0992] Creating a marketing plan

[0993] Marketing plan approval

[0994] Creating advertising materials

[0995] Formulating an advertising distribution plan

[0996] If a user reviews the generated WBS and feels, for example, nervous or anxious, the emotion engine will recognize this and determine that more support is needed, providing more detailed explanations or additional guidance.

[0997] The user then provides feedback on the generated WBS. The device collects the user's feedback and sends it to the server. The server then retrains the machine learning model based on the feedback, improving prediction accuracy for future iterations.

[0998] This emotion engine recognizes the user's emotional state and adjusts the interface and feedback based on that state, providing a user-friendly system. This further strengthens rapid response to new tasks and transfers, further improving work productivity and project success rates.

[0999] The processing flow will be explained below.

[1000] Step 1: Data collection

[1001] The server collects historical business data using APIs and scripts that retrieve data from project management tools, emails, chat logs, documents, etc.

[1002] Specific operation: Obtain information such as project names, task names, deadlines, and assignees from project management tools via API. Filter and obtain related messages from emails and chat logs, and extract text data containing related keywords from documents.

[1003] Step 2: Data Preprocessing

[1004] The server pre-processes the collected data, which is the process of cleansing the data, removing duplicates and noise data, and converting it into a unified format.

[1005] What it does: It removes duplicate entries, maps information from different data sources into one unified format, filters out noisy data, and fills in gaps and incomplete data.

[1006] Step 3: Data training

[1007] The server uses the preprocessed data to train machine learning models, which analyze past project data and learn task patterns and dependencies.

[1008] Specific operation: Input data into a machine learning algorithm (e.g., decision tree, neural network) to generate and learn task co-occurrence patterns and dependency graphs within the project.

[1009] Step 4: Enter an outline of the new job

[1010] The terminal provides an interface for users to input the outline of a new project or task. The user inputs basic information about the new project or task (type of task, goals, resources, etc.).

[1011] Specific operation: An interface is displayed through the terminal input form, allowing the user to enter the type of work, goals, resources, deadlines, etc., and enter the required information.

[1012] Step 5: Task extraction

[1013] The server uses a machine learning model to extract the necessary tasks based on the input summary of the new business.

[1014] Specific operation: The input data is analyzed using a natural language processing algorithm, and the necessary task list is generated based on task patterns from related past projects.

[1015] Step 6: Generate WBS

[1016] The server automatically generates a work breakdown structure (WBS) based on the extracted tasks, including the hierarchical structure, dependencies, and priorities of the tasks.

[1017] Specific operations: Organize task lists hierarchically, represent dependencies between tasks in a graph structure, and set priorities and deadlines to build a WBS.

[1018] Step 7: Emotion Recognition

[1019] The device sends the user's input data and operation history to the emotion engine, which analyzes the user's emotional state.

[1020] Specific operation: Input data (keyboard input speed and pressure, screen operation history) is input into the emotion engine, and an emotion analysis algorithm is used to recognize the user's emotional state (e.g., stress, anxiety, concentration, etc.).

[1021] Step 8: Emotion-Based Display Adjustment

[1022] The server receives feedback from the emotion engine and adjusts the display of the WBS, providing an environment that is easy for users to understand and work in.

[1023] Specific behavior: If the user is stressed, simplify the WBS display and add specific guidelines and support messages. Conversely, if the user is focused, display more detailed information and options.

[1024] Step 9: Gather feedback

[1025] The user can review the generated WBS and make adjustments as needed, such as adding, deleting, or reordering tasks.

[1026] The terminal collects user feedback.

[1027] Specific operation: Record user operations (addition, deletion, change) for each task in the WBS and send them as feedback data.

[1028] Step 10: Retrain the model

[1029] The server retrains the machine learning model based on the collected feedback, improving the accuracy of future predictions.

[1030] What it does: Analyzes collected feedback data and retrains machine learning models to learn feedback patterns.

[1031] This emotion engine recognizes the user's emotional state and adjusts the interface and feedback based on that state, providing a user-friendly system. This further strengthens the ability to respond quickly and efficiently to new tasks and transfers, further improving work productivity and project success rates.

[1032] Example 2

[1033] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1034] In today's business environment, launching and executing new projects requires significant planning and management. However, traditional project management systems are unable to effectively utilize past work data, making it difficult to efficiently identify the tasks required for new projects. Furthermore, systems are not designed to take into account the user's emotional state, resulting in a suboptimal user experience. Furthermore, even if feedback on the generated WBS is collected, there is insufficient means for applying that feedback to future projects. These issues result in reduced work efficiency and project success rates.

[1035] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for collecting past business data, a means for preprocessing the collected data, and a means for training a machine learning model using the preprocessed data. This makes it possible to effectively utilize past business data and efficiently identify tasks required for a new project. The server also includes a means for inputting an outline of a new business, a means for extracting necessary tasks based on the input outline of the new business, a means for organizing the extracted tasks to generate a WBS, a means for displaying the generated WBS, a means for collecting feedback on the generated WBS, a means for retraining a machine learning model based on the collected feedback, a means for analyzing a user's emotional state, and a means for adjusting the interface and feedback based on the analyzed emotional state. This makes it possible to improve the user experience and utilize feedback to increase the system's prediction accuracy.

[1036] "Historical business data" refers to information about previously executed tasks or projects, and is data obtained from project management tools, emails, electronic chat logs, documents, etc.

[1037] "Preprocessing" is the process of organizing collected data by removing redundancy and noise and converting it into a unified format, making it suitable for training machine learning models.

[1038] A "machine learning model" is a set of algorithms that are trained using collected and preprocessed data to learn patterns and dependencies of tasks required for new business operations.

[1039] The "new business outline" refers to detailed information about a new project or task, and is input data from the user that includes specific goals and necessary requirements.

[1040] "Task extraction" is the process of using a machine learning model to identify necessary tasks from the outline of a new business and define them as specific activity items.

[1041] "WBS" stands for "Work Breakdown Structure" and is a visual representation of the hierarchical tasks required to complete a project.

[1042] "Feedback" refers to information including opinions and improvements provided by users regarding the generated WBS, and is used as data to improve the system's prediction accuracy.

[1043] "Emotional state" refers to the user's emotional response or mood, and is the psychological state that is analyzed while using the system.

[1044] An "emotion engine" is a program or set of algorithms for detecting and analyzing a user's emotional state, with the goal of improving the user experience.

[1045] "Interface" refers to the screen and operating means that users use to interact with the system, and must be designed to be easy for users to use.

[1046] The present invention provides a system that combines four main elements: a server, a terminal, a user, and an emotion engine. A specific embodiment of this system will be described below.

[1047] First, the server collects historical business data from data sources, including project management tools, email systems, electronic chat logs, and document management systems. Specifically, it uses tools such as JIRA, Slack, mail servers, and Google Drive to execute APIs and data extraction scripts, and stores the necessary data in a centralized database.

[1048] The server then preprocesses the collected data. This includes removing duplicate data, reducing noise, and converting the data into a unified format, such as JSON, to ensure data quality. Specifically, the data is cleaned and centralized using an SQLite database and regular expressions.

[1049] The server uses the preprocessed data to train a machine learning model using a machine learning framework such as TensorFlow or PyTorch, which analyzes past project data and learns task patterns and dependencies.

[1050] Next, the user enters a summary of the new business via the terminal, for example, by entering "Implement market launch of new product" into a web form. This information is sent from the terminal to the server as an HTTP POST request.

[1051] The server analyzes the new business data it receives using natural language processing (NLP) technology and uses machine learning models to extract necessary tasks. Specifically, it uses NLP libraries such as spaCy and NLTK to tokenize and analyze the data. It then identifies tasks based on the analysis results and defines them as specific actions.

[1052] Once the tasks are extracted, the server automatically generates a work breakdown structure (WBS) based on the extracted tasks. The generated WBS includes the hierarchical structure, dependencies, and priorities of the tasks. The terminal visually displays this WBS to the user.

[1053] Furthermore, an emotion engine built into the server analyzes the user's emotional state. When a user checks the WBS, emotion analysis is performed based on their reaction. For example, if the user is feeling nervous or anxious, the emotion engine will recognize this and display additional guidance or detailed explanations. This analysis uses a cloud emotion analysis API (for example, IBM Watson).

[1054] Users provide feedback on the generated WBS. The device collects the feedback and sends it to the server, which then retrains the machine learning model based on the feedback to improve prediction accuracy in the future.

[1055] As a concrete example, when a user starts a new project to launch a new product, they can input "launch a new product," and the server will generate the following task list based on data from similar past projects:

[1056] Product packaging design

[1057] Package design approval

[1058] Selection of sales channels

[1059] Sales channel contract negotiations

[1060] Creating a marketing plan

[1061] Marketing plan approval

[1062] Creating advertising materials

[1063] Formulating an advertising distribution plan

[1064] An example of a prompt sentence to be input to the generative AI model is, "Please tell me the tasks required to launch a new product to the market." By using this prompt sentence, the AI ​​model can efficiently extract the tasks required for the new business and propose an appropriate WBS.

[1065] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1066] Program processing flow

[1067] Step 1:

[1068] The server collects past business data.

[1069] Input: API requests from project management tools, email systems, electronic chat logs, and document management systems.

[1070] Data processing / calculation: Send API requests and extract the necessary information from each data source.

[1071] Output: Save the retrieved data to the database.

[1072] Specific behavior:

[1073] Get project ticket information from the project management tool API.

[1074] Email data is retrieved using the IMAP email system.

[1075] Extract chat logs via API.

[1076] Download relevant documents from your document management system.

[1077] Step 2:

[1078] The server preprocesses the collected data.

[1079] Input: Collected business data.

[1080] Data processing / calculation: Delete duplicate data, reduce noise data, and convert to a unified format.

[1081] Output: Preprocessed data converted into a unified format.

[1082] Specific behavior:

[1083] Execute a SQL query to remove duplicate data in an SQLite database.

[1084] Use regular expressions to filter out noisy messages.

[1085] All data will be unified and centralized in JSON format.

[1086] Step 3:

[1087] The server trains the machine learning model.

[1088] Input: Preprocessed business data.

[1089] Data processing / computation: Input the data into a framework such as TensorFlow or PyTorch and train a model.

[1090] Output: A trained machine learning model.

[1091] Specific behavior:

[1092] Input the dataset into the TensorFlow model and run the training loop.

[1093] Tune hyperparameters to improve model accuracy.

[1094] Save the trained model and use it in a later step.

[1095] Step 4:

[1096] The user inputs an outline of a new job via a terminal.

[1097] Input: Enter text about the new business.

[1098] Data processing / calculation: The data entered by the user is sent as an HTTP POST request.

[1099] Output: A summary of the new job is sent to the server.

[1100] Specific behavior:

[1101] A user fills out a web form with a project description (e.g., "Implement market introduction of a new product").

[1102] The device generates and sends an HTTP POST request to send the input content to the server.

[1103] Step 5:

[1104] The server analyzes the new business data and extracts tasks.

[1105] Input: New business summary, trained machine learning model.

[1106] Data processing / computation: Tokenize the data using natural language processing and use models to extract tasks.

[1107] Output: The extracted task list.

[1108] Specific behavior:

[1109] Use spaCy to tokenize the outline of the new business and perform semantic analysis.

[1110] Input data into a trained machine learning model to extract the required tasks.

[1111] Step 6:

[1112] The server automatically generates the WBS.

[1113] Input: The extracted task list.

[1114] Data processing / calculation: Calculate the hierarchical structure and dependencies of tasks and generate a WBS.

[1115] Output: The generated WBS.

[1116] Specific behavior:

[1117] Based on the extracted tasks, an algorithm is used to generate a hierarchical structure for the WBS.

[1118] Save the generated WBS in JSON format and use it in a later step.

[1119] Step 7:

[1120] The server uses an emotion engine to analyze the user's emotional state.

[1121] Input: User operation logs and input data.

[1122] Data processing / calculation: Estimate emotional state using emotion analysis API.

[1123] Output: Parsed emotion data.

[1124] Specific behavior:

[1125] Collects logs of user keyboard input and mouse operations.

[1126] Emotional states are estimated using a cloud-based emotion analysis API.

[1127] Step 8:

[1128] The user provides feedback on the generated WBS.

[1129] Input: User feedback.

[1130] Data processing / calculation: Generate an HTTP POST request to send the feedback to the server.

[1131] Output: The feedback is sent to the server.

[1132] Specific behavior:

[1133] The user fills out the feedback form and clicks the submit button.

[1134] The device generates and sends an HTTP POST request to send the feedback content to the server.

[1135] Step 9:

[1136] The server retrains the machine learning model based on the collected feedback.

[1137] Input: Feedback data, existing trained model.

[1138] Data processing / computation: Add feedback data and retrain the model.

[1139] Output: A machine learning model with improved accuracy.

[1140] Specific behavior:

[1141] Add new feedback data to the dataset.

[1142] Retrain your machine learning model and see improvements in accuracy.

[1143] Save the improved model.

[1144] (Application example 2)

[1145] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1146] Conventional work management systems have difficulty extracting appropriate tasks from past data and generating an efficient work breakdown structure (WBS) when starting a new task. Furthermore, they do not provide an interface or support that takes into account the user's emotional state, which can cause stress for users and lead to lower productivity and a deterioration in work efficiency. Furthermore, they lack a mechanism for appropriately collecting feedback and improving the accuracy of future predictions. These issues need to be resolved.

[1147] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1148] In this invention, the server includes means for collecting past business data, means for preprocessing the collected data, means for training a machine learning model using the preprocessed data, means for inputting an outline of a new business, means for extracting necessary tasks based on the input outline of the new business, means for organizing the extracted tasks to generate a work breakdown structure (WBS), means for displaying the generated WBS, means for collecting feedback on the generated WBS, means for retraining the machine learning model based on the collected feedback, and means for adjusting the WBS display method and support content based on the user's emotional state using an emotion engine that recognizes user emotions. This makes it possible to efficiently extract tasks necessary for new business based on past data, generate and display a WBS, and provide an interface and support that responds to the user's emotions.

[1149] "Past business data" refers to information about various business operations that have been carried out in the past, including data from project management tools, emails, chat logs, documents, etc.

[1150] "Preprocessing" refers to the process of removing duplication and noise from collected data and standardizing the format.

[1151] A "machine learning model" refers to an algorithm or computational model used to analyze and predict data based on collected data.

[1152] "Outline of new business" refers to basic information such as the purpose, scope, and conditions of the new business.

[1153] "Task extraction" refers to identifying and listing the necessary work items based on the outline of the new business.

[1154] A "Work Breakdown Structure (WBS)" is a tool used to hierarchically break down the work involved in a project or task and organize dependencies and priorities.

[1155] "Display" refers to visually presenting the generated information and WBS to the user.

[1156] "Feedback" refers to information collected from users, such as opinions and ratings, that is used to improve the system and retrain the model.

[1157] An "emotion engine" refers to a system that analyzes a user's emotional state and adjusts the system's behavior and display accordingly.

[1158] An embodiment of this invention will be described in detail. This system automatically generates work schedules within a factory and efficiently manages robot work. It generates new tasks based on past work data and automatically creates a work breakdown structure (WBS). Furthermore, it recognizes the emotions of factory staff and provides support and alerts according to their emotions.

[1159] Overview of program processing

[1160] The system starts by collecting historical business data and preprocessing it. The server collects data from project management tools, emails, chat logs, documents, etc. This data is collected and preprocessed using Python. Pandas is used to shape and clean the data, and Scikit-Learn is used to train machine learning models.

[1161] When a new task summary is input from a terminal, the server analyzes it and extracts the necessary tasks. It uses natural language processing technology to analyze the new task summary and generates appropriate tasks by referencing past data. To do this, it uses TfidfVectorizer to extract text features and RandomForestClassifier to predict tasks.

[1162] Next, a WBS is automatically created based on the generated tasks. The generated WBS is structured based on the time and dependencies of each task and displayed on the terminal. This display is provided as a user-friendly interface.

[1163] Additionally, the robot is equipped with an emotion engine that analyzes the emotional state of factory staff based on their input and behavior. The emotion engine uses the Hugging Face Transformer to tailor feedback and support content based on specific emotions. If the emotional state is likely to be negatively affected, the robot will provide special support or direct customers to a hotline.

[1164] Specific examples

[1165] For example, if a factory needs to perform a final inspection of a new part and reset the assembly line, enter "Final inspection of new part" and "Resetting the assembly line" as new tasks. The server generates the following tasks based on past similar projects:

[1166] Setting inspection standards for new parts

[1167] Inspection

[1168] Assembly line circuit settings

[1169] Test run of the reconfiguration

[1170] The WBS generated in this way is displayed visually on the terminal. If a factory worker types, "I'm feeling stressed today," the emotion engine analyzes it and provides an appropriate support message (e.g., "Attention, you need emotional support").

[1171] Prompt Sentence Examples

[1172] Example of new task input: "Final inspection of new parts"

[1173] Example of emotional state input: "I'm stressed today."

[1174] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1175] Step 1:

[1176] The server collects past business data. This data includes project management tools, emails, chat logs, and documents. Specifically, it collects information from each data source using APIs and data extraction scripts. The collected data is then stored in a database.

[1177] Inputs: Project management tools, emails, chat logs, documents

[1178] Output: Collected business data (stored in a database)

[1179] Step 2:

[1180] The server preprocesses the collected business data. Specifically, it uses Pandas to remove noise and duplicate data, and converts the data into a unified format. This step ensures the quality of the data.

[1181] Input: Collected business data

[1182] Output: Preprocessed business data

[1183] Step 3:

[1184] The server trains a machine learning model using the preprocessed data: it uses Scikit-Learn's TfidfVectorizer to extract text features for the task, and RandomForestClassifier to train a predictive model for the task.

[1185] Input: Preprocessed business data

[1186] Output: A trained machine learning model

[1187] Step 4:

[1188] The user inputs the outline of the new task via the terminal. For this, the task content is input in simple text format. When the outline of the new task is input, the terminal transmits the data to the server.

[1189] Input: New business summary (text format)

[1190] Output: Sent business summary data

[1191] Step 5:

[1192] The server uses natural language processing technology to analyze the outline of the new business and generate the necessary tasks based on past data. Specifically, it extracts text features of the new business using TfidfVectorizer and predicts tasks using a trained machine learning model.

[1193] Input: Summary data for new business

[1194] Output: Generated task list

[1195] Step 6:

[1196] The server automatically creates a work breakdown structure (WBS) based on the generated task list, including the task hierarchy, dependencies, and priorities.

[1197] Input: Generated task list

[1198] Output: Generated WBS

[1199] Step 7:

[1200] The terminal visually displays the generated WBS to the user, who can then check it and provide corrections or feedback as necessary.

[1201] Input: Generated WBS

[1202] Output: A visual representation of the WBS

[1203] Step 8:

[1204] The terminal collects user feedback on the generated WBS and transmits the information to the server.

[1205] Input: User feedback

[1206] Output: Feedback data sent

[1207] Step 9:

[1208] The server uses the collected feedback to retrain the machine learning model, which then improves its prediction accuracy in future runs.

[1209] Input: User feedback data

[1210] Output: A retrained machine learning model

[1211] Step 10:

[1212] The emotion engine analyzes the emotional state of factory staff from their input and actions, and uses the Hugging Face Transformer model to determine their emotional state, adjusting the interface and support content based on the user's emotional state.

[1213] Input: Text data expressing the emotions of factory staff

[1214] Output: Analyzed emotional state and support message based on it

[1215] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1216] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1217] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1218] [Fourth embodiment]

[1219] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1220] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1221] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1222] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1223] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1224] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1225] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1226] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1227] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1228] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1229] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1230] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1231] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1232] This invention relates to a system that automatically identifies tasks required for new work by utilizing past work data and generates an efficient work breakdown structure (WBS). This system consists of three main elements: a server, terminals, and users.

[1233] First, the server collects past business data, including project management tools, emails, chat logs, documents, etc. During this collection process, APIs and data extraction scripts are used for each data source to collect information and manage it centrally.

[1234] The server then preprocesses the collected data, removing duplicates and noise and converting it into a unified format. Ensuring data quality at this stage creates a dataset suitable for training subsequent machine learning models.

[1235] Using the preprocessed data, the server trains a machine learning model. This model analyzes past project data and learns task patterns and dependencies. By using a large amount of past data to train the model, it is possible to make highly accurate predictions.

[1236] The user inputs an outline of a new job via a terminal. The terminal then sends this input data to the server. The server then analyzes the received data using natural language processing technology and uses a machine learning model to extract the necessary tasks.

[1237] Once the tasks are extracted, the server automatically generates a WBS based on them. The generated WBS includes the hierarchical structure, dependencies, and priorities of the tasks. The terminal visually displays this WBS to the user.

[1238] As a concrete example, consider the case of launching a new marketing campaign. When a user types "launch a marketing campaign for a new product," the server generates the following task list based on data from similar past projects:

[1239] Marketing strategy development

[1240] Target Audience Research

[1241] Creating advertising materials

[1242] Advertisement delivery plan

[1243] Campaign monitoring and evaluation

[1244] The user can then review the generated WBS and add, delete, or modify tasks as needed. The device then sends the user's feedback to the server, which then retrains the machine learning model based on that feedback. This improves prediction accuracy for future tasks.

[1245] This system enables companies to respond quickly and efficiently to new tasks within the company or tasks that employees have no experience with due to transfers, improving work productivity and the success rate of projects.

[1246] The processing flow will be explained below.

[1247] Step 1: Data collection

[1248] The server collects historical business data using APIs and scripts that retrieve data from project management tools, emails, chat logs, documents, etc.

[1249] Specific operation: Obtain information such as project names, task names, deadlines, and assignees from project management tools via API. Filter and obtain related messages from emails and chat logs, and extract text data containing related keywords from documents.

[1250] Step 2: Data Preprocessing

[1251] The server pre-processes the collected data, which is the process of cleansing the data, removing duplicates and noise data, and converting it into a unified format.

[1252] What it does: It removes duplicate entries, maps information from different data sources into one unified format, filters out noisy data, and fills in gaps and incomplete data.

[1253] Step 3: Data training

[1254] The server uses the preprocessed data to train machine learning models, which analyze past project data and learn task patterns and dependencies.

[1255] Specific operation: Input data into a machine learning algorithm (e.g., decision tree, neural network) to generate and learn task co-occurrence patterns and dependency graphs within the project.

[1256] Step 4: Enter an outline of the new job

[1257] The terminal provides an interface for users to input the outline of a new project or task. The user inputs basic information about the new project or task (type of task, goals, resources, etc.).

[1258] Specific operation: An interface is displayed through the terminal input form, allowing the user to enter the type of work, goals, resources, deadlines, etc., and enter the required information.

[1259] Step 5: Task extraction

[1260] The server uses a machine learning model to extract the necessary tasks based on the input summary of the new business.

[1261] Specific operation: The input data is analyzed using a natural language processing algorithm, and the necessary task list is generated based on task patterns from related past projects.

[1262] Step 6: Generate WBS

[1263] The server automatically generates a work breakdown structure (WBS) based on the extracted tasks, including the hierarchical structure, dependencies, and priorities of the tasks.

[1264] Specific operations: Organize task lists hierarchically, represent dependencies between tasks in a graph structure, and set priorities and deadlines to build a WBS.

[1265] Step 7: View the WBS

[1266] The terminal displays the generated WBS to the user, allowing the user to visually check the overall task structure and dependencies.

[1267] Specific operation: Renders the WBS in Gantt chart format and hierarchical display format, and provides an interface that allows interactive operation.

[1268] Step 8: User Feedback

[1269] The user can review the generated WBS and make adjustments as needed, such as adding, deleting, or reordering tasks.

[1270] The terminal collects user feedback.

[1271] Specific operation: Record user operations (addition, deletion, change) for each task in the WBS and send them as feedback data.

[1272] Step 9: Retrain the model

[1273] The server retrains the machine learning model based on the collected feedback, improving the accuracy of future predictions.

[1274] What it does: Analyzes collected feedback data and retrains machine learning models to learn feedback patterns.

[1275] Through this processing flow, the system generates an efficient and accurate WBS, enabling companies to respond quickly to new business and transfers.

[1276] Example 1

[1277] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1278] Modern business activities require the ability to quickly and efficiently handle new tasks and unfamiliar work resulting from transfers. However, conventional methods have struggled to effectively utilize past work data to automatically identify the tasks required for new work. Furthermore, there has been a lack of feedback mechanisms to ensure the accuracy and usefulness of the generated work breakdown structure (WBS). This has led to issues such as reduced work productivity and project success rates.

[1279] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1280] In this invention, the server includes a means for collecting past business data, a means for preprocessing the collected data, a means for training a machine learning model using the preprocessed data, a means for inputting an outline of a new business, a means for extracting necessary tasks based on the input outline of the new business, a means for organizing the extracted tasks to generate a WBS, a means for visually displaying the generated WBS, a means for collecting feedback on the generated WBS, and a means for retraining the machine learning model based on the collected feedback. This makes it possible to effectively utilize past business data and respond to new business quickly and efficiently. Furthermore, the accuracy and usefulness of the generated WBS are improved, thereby improving business productivity and project success rates.

[1281] "Historical Business Data" refers to all records and information relating to previously executed projects or business operations, including project management tools, emails, message logs, documents, etc.

[1282] "Collection methods" refers to the methods and tools used to obtain necessary information from various data sources and manage it centrally.

[1283] "Preprocessing means" refers to techniques and tools for removing redundant data and noise data from collected data and converting it into a unified format.

[1284] A "machine learning model" refers to an algorithm or system that learns and predicts task patterns and dependencies based on data.

[1285] "Training methods" refers to techniques and tools used to train machine learning models using preprocessed data to improve their predictive accuracy.

[1286] The "new business overview" refers to the basic content and goals of the new business or project that the user intends to implement.

[1287] "Task" refers to the specific work or activities required to accomplish new business.

[1288] "Task extraction means" refers to methods and tools for finding and extracting necessary tasks from the outline of new work.

[1289] "WBS (Work Breakdown Structure)" refers to a hierarchical organization and structure of the tasks required to achieve the project's goals.

[1290] "Organizational methods" refer to techniques and tools for organizing and categorizing extracted tasks into a rational order and structure.

[1291] "Visual display means" refers to graphical interfaces and tools that present the generated WBS to users in an easy-to-understand manner.

[1292] "Feedback methods" refer to methods and tools for collecting user opinions and corrections and reflecting them in the system.

[1293] "Retraining methods" refer to methods and tools used to retrain machine learning models based on feedback collected from users, further improving their predictive accuracy.

[1294] This invention relates to a system that automatically identifies tasks required for new work by utilizing past work data and generates an efficient work breakdown structure (WBS). This system consists of three main elements: a server, terminals, and users.

[1295] First, the server collects past business data. This data includes data from project management tools, emails, message logs, documents, etc. During the collection process, APIs and data extraction scripts are used for each data source (e.g., project management tools, email servers, chat tools, cloud storage) to obtain information and manage it centrally. The server then stores the collected data in a specific directory.

[1296] Next, the server preprocesses the collected data. Specifically, it uses Python's Pandas library to read the data and remove duplicates and noise. It then uses regular expressions to filter out the noise and extract only the necessary information. To standardize the data format, it uses preprocessing modules from Numpy and Scikit-learn. This stage ensures the quality of the data and creates a dataset suitable for training subsequent machine learning models.

[1297] Using the preprocessed data, the server trains a machine learning model. TensorFlow or PyTorch is used to train a recurrent neural network (RNN) or Transformer model, which learns task patterns and dependencies from past project data. Cross-validation is used to evaluate the model, and hyperparameters are tuned as needed.

[1298] A user inputs the outline of a new task via a terminal. For example, if a user inputs "Run a marketing campaign for a new product," the data is sent from the terminal to the server in JSON format. The server analyzes this input data using natural language processing technology (e.g., BERT, GPT-3) and uses machine learning models to extract the necessary tasks.

[1299] Once tasks are extracted, the server automatically generates a WBS based on them. The generated WBS includes the task hierarchy, dependencies, and priorities. This WBS is displayed on the device using visualization libraries such as D3.js and Highcharts. Users can review the generated WBS and add, delete, or modify tasks as needed.

[1300] User feedback is sent from the device to the server, which then uses this feedback to retrain the machine learning model, improving the accuracy of task extraction and WBS generation from the next time onwards.

[1301] Specific examples

[1302] For example, when a user is running a marketing campaign for a new product, they can input the following prompt: "What are the tasks required to run a marketing campaign for a new product?" Based on this, the server analyzes data from similar past projects and generates a task list like this:

[1303] Marketing strategy development

[1304] Target Audience Research

[1305] Creating advertising materials

[1306] Advertisement delivery plan

[1307] Campaign monitoring and evaluation

[1308] As users review and modify the generated WBS, feedback is collected and the machine learning model is retrained based on this data, improving the efficiency of new work and the success rate of projects.

[1309] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1310] Step 1:

[1311] The server collects past business data from various data sources. The collected data includes project management tools, emails, message logs, documents, and more. Specifically, project data is obtained using the project management tool's API, and email data is collected using the email server's IMAP protocol. Message logs are downloaded using the chat tool's API, and document data is downloaded using the cloud storage API. The input is raw data obtained from each data source, and the output is a data file in a unified format stored on the server.

[1312] Step 2:

[1313] The server preprocesses the collected data. It uses Python's Pandas library to read the data and remove duplicates and noise. It uses regular expressions to filter out unnecessary strings and noise and extract the necessary information. To unify the data format, it uses preprocessing modules from Numpy and Scikit-learn. The input is the collected raw data, and the output is clean data after preprocessing.

[1314] Step 3:

[1315] The server uses the preprocessed data to train a machine learning model. It uses TensorFlow or PyTorch to train a recurrent neural network (RNN) or Transformer model. The model is trained by learning task patterns and dependencies based on past project data. The input is clean data that has been preprocessed, and the output is a trained machine learning model.

[1316] Step 4:

[1317] The user inputs the outline of a new business via the terminal. For example, the user inputs "Run a marketing campaign for a new product" in text format. The terminal converts this data into JSON format and sends it to the server. The input is the outline of the new business entered by the user, and the output is the JSON format data sent to the server.

[1318] Step 5:

[1319] When the server receives the input outline of the new work, it analyzes it using natural language processing technology (e.g., BERT, GPT-3). It then uses a machine learning model to extract the necessary tasks. The input is new work data in JSON format, and the output is a set of necessary tasks. Once the tasks are extracted, a WBS is automatically generated based on them. The generated WBS includes the hierarchical structure, dependencies, and priorities of the tasks.

[1320] Step 6:

[1321] The server sends the generated WBS to the terminal, which uses a visualization library such as D3.js or Highcharts to visually display the WBS to the user. The input is the task group and the generated data of the WBS, and the output is the visualized WBS that is displayed to the user.

[1322] Step 7:

[1323] The user checks the displayed WBS and adds, deletes, or modifies tasks as necessary. The results of the user's operations are sent from the terminal to the server as feedback. The input is the user's feedback operations, and the output is the feedback data.

[1324] Step 8:

[1325] The server retrains the machine learning model based on the collected feedback. The feedback data is used as new training data to update the machine learning model. This improves the accuracy of task extraction and WBS generation from the next time onwards. The input is the feedback data, and the output is the updated machine learning model.

[1326] (Application example 1)

[1327] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1328] Optimizing production processes in traditional factories relies primarily on the experience and intuition of managers, making efficient project management difficult. Furthermore, identifying appropriate tasks and creating a work breakdown structure (WBS) is particularly difficult for new work. This results in lower productivity and project success rates. Furthermore, it is not easy to monitor the progress of each process in real time and make appropriate readjustments.

[1329] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1330] In this invention, the server includes means for collecting past work data, means for preprocessing the collected data, means for training a machine learning model using the preprocessed data, means for inputting an outline of a new work, means for extracting necessary tasks based on the input outline of the new work, means for organizing the extracted tasks to generate a WBS, means for displaying the generated WBS, means for collecting feedback on the generated WBS, means for retraining the machine learning model based on the collected feedback, means for assigning tasks to robots to optimize production processes in a factory, and means for monitoring and readjusting the progress of each process in real time. This enables automatic and efficient optimization of production processes, thereby improving project success rates and productivity.

[1331] "Past business data" refers to all information related to past projects or work, including project management tools, emails, chat logs, and documents.

[1332] "Preprocessing means" refers to the process of removing redundancy and noise from collected data and converting the data into a unified format.

[1333] "Means for training machine learning models" refers to running machine learning algorithms using pre-processed historical business data to learn task patterns and dependencies.

[1334] The "means for inputting the outline of a new business" refers to an interface through which the user inputs details of a new business to be started.

[1335] "Means for extracting necessary tasks" refers to the process of automatically identifying appropriate tasks from the input outline of new work.

[1336] "Method for generating WBS" refers to the method for organizing the extracted tasks into a hierarchical structure and creating a work breakdown structure.

[1337] "Means for displaying the generated WBS" refers to an interface that visually displays the generated WBS to the user.

[1338] "Means for collecting feedback" refers to an interface for collecting comments and correction requests from users.

[1339] "Means for retraining the machine learning model" refers to the process of updating the machine learning model based on collected feedback to improve the accuracy of task extraction and WBS generation from the next time onwards.

[1340] "Means for assigning tasks to robots to optimize production processes" refers to a system that automatically distributes appropriate work instructions to robots in a factory.

[1341] "Means for monitoring and readjusting the progress of the process in real time" refers to a system for monitoring the progress of the production process in real time and readjusting it as necessary.

[1342] This invention is a system that uses past business data to automatically identify tasks required for new work and generate an efficient work breakdown structure (WBS).Application examples include a system that assigns tasks to robots to optimize production processes within a factory, and monitors and readjusts the progress of each process in real time.

[1343] The server first collects past business data, including project management tools, emails, chat logs, documents, etc. Data collection is centrally managed using APIs and data extraction scripts.

[1344] The server then preprocesses the collected data, which includes removing duplicates and noise and converting the data into a unified format, ensuring data quality and creating a dataset suitable for training machine learning models.

[1345] The server uses the preprocessed data to train a machine learning model. The model analyzes past project data and learns task patterns and dependencies. By using a large amount of past data for this training, highly accurate predictions are possible.

[1346] The user inputs an outline of the new job through a terminal. This input data is sent to the server, which then analyzes the received data using natural language processing technology and extracts the necessary tasks using a machine learning model.

[1347] The server automatically generates a WBS based on the extracted tasks. The generated WBS includes the hierarchical structure, dependencies, and priorities of the tasks. The terminal visually displays the generated WBS to the user.

[1348] Furthermore, the system has the ability to optimize production processes within the factory and assign tasks to robots. Specifically, it lists the tasks for robots in each process, monitors their progress in real time, and adjusts them as necessary.

[1349] Users can check the generated WBS and add, delete, or modify tasks as needed. This feedback is sent to the server, which then uses it to retrain the machine learning model and improve prediction accuracy for future projects.

[1350] As a concrete example, consider the case where a new plan for setting up a new product line is entered into a "production process optimization app" on a smartphone. The server generates the following task list based on similar past data:

[1351] 1. Manufacturing line design

[1352] 2. Machine settings

[1353] 3. Production Test

[1354] 4. Quality check

[1355] An example of a prompt sentence is "Please generate the optimal production process for designing a new manufacturing line for airplane parts." By using this prompt sentence, the server generates the optimal task list and WBS, realizing the efficiency of the production process.

[1356] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1357] Step 1:

[1358] The server collects historical business data using APIs and data extraction scripts. It receives input from data sources such as project management tools, emails, chat logs, and documents. It integrates this data and stores it in a central database. The output is unified historical business data.

[1359] Step 2:

[1360] The server preprocesses the collected data, specifically removing duplicate and noisy data and converting the data into a unified format. The input for this process is historical business data, and the output is a high-quality, unified dataset.

[1361] Step 3:

[1362] The server uses the preprocessed dataset to train a machine learning model by running a machine learning algorithm that learns patterns and dependencies in the data. The input to this process is the preprocessed dataset, and the output is a machine learning model that has learned the patterns and dependencies for the task.

[1363] Step 4:

[1364] The user inputs the outline of the new business through the terminal. The details of the business to be newly started are input as input to the terminal. The output is the outline data of the new business sent from the terminal.

[1365] Step 5:

[1366] The server analyzes the received outline of the new business using natural language processing technology. The input is the outline data of the new business, and the output is data extracting the necessary tasks.

[1367] Step 6:

[1368] The server generates a WBS based on the extracted task data. The input is the analyzed task data. The output is a WBS that includes the hierarchical structure, dependencies, and priorities of the tasks.

[1369] Step 7:

[1370] The terminal visually displays the generated WBS to the user. The input is the generated WBS, and the output is a representation of the WBS that the user can visually confirm.

[1371] Step 8:

[1372] The user inputs feedback for the generated WBS. The input is the feedback entered by the user via the terminal, and the output is the feedback data sent from the terminal to the server.

[1373] Step 9:

[1374] The server retrains the machine learning model based on the collected feedback: the input is the user feedback data, and the output is the updated machine learning model.

[1375] Step 10:

[1376] The server assigns tasks to robots to optimize production processes in the factory. The inputs are the generated WBS and robot work information, and the output is specific work instructions delivered to the robots.

[1377] Step 11:

[1378] The server monitors the progress of each process in real time and adjusts it as necessary. The input is real-time progress data, and the output is optimized process information.

[1379] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1380] This invention combines a system that uses past business data to automatically identify tasks required for new business and generate an efficient WBS (Work Breakdown Structure), with an emotion engine that recognizes user emotions. This system consists of four main elements: a server, a terminal, a user, and an emotion engine.

[1381] First, the server collects past business data. This data includes project management tools, emails, chat logs, documents, etc. The server uses APIs and data extraction scripts for each data source to collect information and manage it centrally.

[1382] The server then preprocesses the collected data, removing duplicates and noise and converting it into a unified format. Ensuring data quality at this stage creates a dataset suitable for training subsequent machine learning models.

[1383] Using the preprocessed data, the server trains a machine learning model. This model analyzes past project data and learns task patterns and dependencies. By using a large amount of past data to train the model, it is possible to make highly accurate predictions.

[1384] The user inputs an outline of a new job via a terminal. The terminal then sends this input data to the server. The server then analyzes the received data using natural language processing technology and uses a machine learning model to extract the necessary tasks.

[1385] Once tasks are extracted, the server automatically generates a WBS based on them. The generated WBS includes the hierarchical structure, dependencies, and priorities of the tasks. The terminal visually displays this WBS to the user. Furthermore, by incorporating an emotion engine, the system analyzes the user's emotional state and adjusts the way the WBS is displayed.

[1386] As a concrete example, consider the case of a new project to launch a new product. When a user types "launch a new product," the server generates the following task list based on data from similar past projects:

[1387] Product packaging design

[1388] Package design approval

[1389] Selection of sales channels

[1390] Sales channel contract negotiations

[1391] Creating a marketing plan

[1392] Marketing plan approval

[1393] Creating advertising materials

[1394] Formulating an advertising distribution plan

[1395] If a user reviews the generated WBS and feels, for example, nervous or anxious, the emotion engine will recognize this and determine that more support is needed, providing more detailed explanations or additional guidance.

[1396] The user then provides feedback on the generated WBS. The device collects the user's feedback and sends it to the server. The server then retrains the machine learning model based on the feedback, improving prediction accuracy for future iterations.

[1397] This emotion engine recognizes the user's emotional state and adjusts the interface and feedback based on that state, providing a user-friendly system. This further strengthens rapid response to new tasks and transfers, further improving work productivity and project success rates.

[1398] The processing flow will be explained below.

[1399] Step 1: Data collection

[1400] The server collects historical business data using APIs and scripts that retrieve data from project management tools, emails, chat logs, documents, etc.

[1401] Specific operation: Obtain information such as project names, task names, deadlines, and assignees from project management tools via API. Filter and obtain related messages from emails and chat logs, and extract text data containing related keywords from documents.

[1402] Step 2: Data Preprocessing

[1403] The server pre-processes the collected data, which is the process of cleansing the data, removing duplicates and noise data, and converting it into a unified format.

[1404] What it does: It removes duplicate entries, maps information from different data sources into one unified format, filters out noisy data, and fills in gaps and incomplete data.

[1405] Step 3: Data training

[1406] The server uses the preprocessed data to train machine learning models, which analyze past project data and learn task patterns and dependencies.

[1407] Specific operation: Input data into a machine learning algorithm (e.g., decision tree, neural network) to generate and learn task co-occurrence patterns and dependency graphs within the project.

[1408] Step 4: Enter an outline of the new job

[1409] The terminal provides an interface for users to input the outline of a new project or task. The user inputs basic information about the new project or task (type of task, goals, resources, etc.).

[1410] Specific operation: An interface is displayed through the terminal input form, allowing the user to enter the type of work, goals, resources, deadlines, etc., and enter the required information.

[1411] Step 5: Task extraction

[1412] The server uses a machine learning model to extract the necessary tasks based on the input summary of the new business.

[1413] Specific operation: The input data is analyzed using a natural language processing algorithm, and the necessary task list is generated based on task patterns from related past projects.

[1414] Step 6: Generate WBS

[1415] The server automatically generates a work breakdown structure (WBS) based on the extracted tasks, including the hierarchical structure, dependencies, and priorities of the tasks.

[1416] Specific operations: Organize task lists hierarchically, represent dependencies between tasks in a graph structure, and set priorities and deadlines to build a WBS.

[1417] Step 7: Emotion Recognition

[1418] The device sends the user's input data and operation history to the emotion engine, which analyzes the user's emotional state.

[1419] Specific operation: Input data (keyboard input speed and pressure, screen operation history) is input into the emotion engine, and an emotion analysis algorithm is used to recognize the user's emotional state (e.g., stress, anxiety, concentration, etc.).

[1420] Step 8: Emotion-Based Display Adjustment

[1421] The server receives feedback from the emotion engine and adjusts the display of the WBS, providing an environment that is easy for users to understand and work in.

[1422] Specific behavior: If the user is stressed, simplify the WBS display and add specific guidelines and support messages. Conversely, if the user is focused, display more detailed information and options.

[1423] Step 9: Gather feedback

[1424] The user can review the generated WBS and make adjustments as needed, such as adding, deleting, or reordering tasks.

[1425] The terminal collects user feedback.

[1426] Specific operation: Record user operations (addition, deletion, change) for each task in the WBS and send them as feedback data.

[1427] Step 10: Retrain the model

[1428] The server retrains the machine learning model based on the collected feedback, improving the accuracy of future predictions.

[1429] What it does: Analyzes collected feedback data and retrains machine learning models to learn feedback patterns.

[1430] This emotion engine recognizes the user's emotional state and adjusts the interface and feedback based on that state, providing a user-friendly system. This further strengthens the ability to respond quickly and efficiently to new tasks and transfers, further improving work productivity and project success rates.

[1431] Example 2

[1432] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1433] In today's business environment, launching and executing new projects requires significant planning and management. However, traditional project management systems are unable to effectively utilize past work data, making it difficult to efficiently identify the tasks required for new projects. Furthermore, systems are not designed to take into account the user's emotional state, resulting in a suboptimal user experience. Furthermore, even if feedback on the generated WBS is collected, there is insufficient means for applying that feedback to future projects. These issues result in reduced work efficiency and project success rates.

[1434] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for collecting past business data, a means for preprocessing the collected data, and a means for training a machine learning model using the preprocessed data. This makes it possible to effectively utilize past business data and efficiently identify tasks required for a new project. The server also includes a means for inputting an outline of a new business, a means for extracting necessary tasks based on the input outline of the new business, a means for organizing the extracted tasks to generate a WBS, a means for displaying the generated WBS, a means for collecting feedback on the generated WBS, a means for retraining a machine learning model based on the collected feedback, a means for analyzing a user's emotional state, and a means for adjusting the interface and feedback based on the analyzed emotional state. This makes it possible to improve the user experience and utilize feedback to increase the system's prediction accuracy.

[1435] "Historical business data" refers to information about previously executed tasks or projects, and is data obtained from project management tools, emails, electronic chat logs, documents, etc.

[1436] "Preprocessing" is the process of organizing collected data by removing redundancy and noise and converting it into a unified format, making it suitable for training machine learning models.

[1437] A "machine learning model" is a set of algorithms that are trained using collected and preprocessed data to learn patterns and dependencies of tasks required for new business operations.

[1438] The "new business outline" refers to detailed information about a new project or task, and is input data from the user that includes specific goals and necessary requirements.

[1439] "Task extraction" is the process of using a machine learning model to identify necessary tasks from the outline of a new business and define them as specific activity items.

[1440] "WBS" stands for "Work Breakdown Structure" and is a visual representation of the hierarchical tasks required to complete a project.

[1441] "Feedback" refers to information including opinions and improvements provided by users regarding the generated WBS, and is used as data to improve the system's prediction accuracy.

[1442] "Emotional state" refers to the user's emotional response or mood, and is the psychological state that is analyzed while using the system.

[1443] An "emotion engine" is a program or set of algorithms for detecting and analyzing a user's emotional state, with the goal of improving the user experience.

[1444] "Interface" refers to the screen and operating means that users use to interact with the system, and must be designed to be easy for users to use.

[1445] The present invention provides a system that combines four main elements: a server, a terminal, a user, and an emotion engine. A specific embodiment of this system will be described below.

[1446] First, the server collects historical business data from data sources, including project management tools, email systems, electronic chat logs, and document management systems. Specifically, it uses tools such as JIRA, Slack, mail servers, and Google Drive to execute APIs and data extraction scripts, and stores the necessary data in a centralized database.

[1447] The server then preprocesses the collected data. This includes removing duplicate data, reducing noise, and converting the data into a unified format, such as JSON, to ensure data quality. Specifically, the data is cleaned and centralized using an SQLite database and regular expressions.

[1448] The server uses the preprocessed data to train a machine learning model using a machine learning framework such as TensorFlow or PyTorch, which analyzes past project data and learns task patterns and dependencies.

[1449] Next, the user enters a summary of the new business via the terminal, for example, by entering "Implement market launch of new product" into a web form. This information is sent from the terminal to the server as an HTTP POST request.

[1450] The server analyzes the new business data it receives using natural language processing (NLP) technology and uses machine learning models to extract necessary tasks. Specifically, it uses NLP libraries such as spaCy and NLTK to tokenize and analyze the data. It then identifies tasks based on the analysis results and defines them as specific actions.

[1451] Once the tasks are extracted, the server automatically generates a work breakdown structure (WBS) based on the extracted tasks. The generated WBS includes the hierarchical structure, dependencies, and priorities of the tasks. The terminal visually displays this WBS to the user.

[1452] Furthermore, an emotion engine built into the server analyzes the user's emotional state. When a user checks the WBS, emotion analysis is performed based on their reaction. For example, if the user is feeling nervous or anxious, the emotion engine will recognize this and display additional guidance or detailed explanations. This analysis uses a cloud emotion analysis API (for example, IBM Watson).

[1453] Users provide feedback on the generated WBS. The device collects the feedback and sends it to the server, which then retrains the machine learning model based on the feedback to improve prediction accuracy in the future.

[1454] As a concrete example, when a user starts a new project to launch a new product, they can input "launch a new product," and the server will generate the following task list based on data from similar past projects:

[1455] Product packaging design

[1456] Package design approval

[1457] Selection of sales channels

[1458] Sales channel contract negotiations

[1459] Creating a marketing plan

[1460] Marketing plan approval

[1461] Creating advertising materials

[1462] Formulating an advertising distribution plan

[1463] An example of a prompt sentence to be input to the generative AI model is, "Please tell me the tasks required to launch a new product to the market." By using this prompt sentence, the AI ​​model can efficiently extract the tasks required for the new business and propose an appropriate WBS.

[1464] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1465] Program processing flow

[1466] Step 1:

[1467] The server collects past business data.

[1468] Input: API requests from project management tools, email systems, electronic chat logs, and document management systems.

[1469] Data processing / calculation: Send API requests and extract the necessary information from each data source.

[1470] Output: Save the retrieved data to the database.

[1471] Specific behavior:

[1472] Get project ticket information from the project management tool API.

[1473] Email data is retrieved using the IMAP email system.

[1474] Extract chat logs via API.

[1475] Download relevant documents from your document management system.

[1476] Step 2:

[1477] The server preprocesses the collected data.

[1478] Input: Collected business data.

[1479] Data processing / calculation: Delete duplicate data, reduce noise data, and convert to a unified format.

[1480] Output: Preprocessed data converted into a unified format.

[1481] Specific behavior:

[1482] Execute a SQL query to remove duplicate data in an SQLite database.

[1483] Use regular expressions to filter out noisy messages.

[1484] All data will be unified and centralized in JSON format.

[1485] Step 3:

[1486] The server trains the machine learning model.

[1487] Input: Preprocessed business data.

[1488] Data processing / computation: Input the data into a framework such as TensorFlow or PyTorch and train a model.

[1489] Output: A trained machine learning model.

[1490] Specific behavior:

[1491] Input the dataset into the TensorFlow model and run the training loop.

[1492] Tune hyperparameters to improve model accuracy.

[1493] Save the trained model and use it in a later step.

[1494] Step 4:

[1495] The user inputs an outline of a new job via a terminal.

[1496] Input: Enter text about the new business.

[1497] Data processing / calculation: The data entered by the user is sent as an HTTP POST request.

[1498] Output: A summary of the new job is sent to the server.

[1499] Specific behavior:

[1500] A user fills out a web form with a project description (e.g., "Implement market introduction of a new product").

[1501] The device generates and sends an HTTP POST request to send the input content to the server.

[1502] Step 5:

[1503] The server analyzes the new business data and extracts tasks.

[1504] Input: New business summary, trained machine learning model.

[1505] Data processing / computation: Tokenize the data using natural language processing and use models to extract tasks.

[1506] Output: The extracted task list.

[1507] Specific behavior:

[1508] Use spaCy to tokenize the outline of the new business and perform semantic analysis.

[1509] Input data into a trained machine learning model to extract the required tasks.

[1510] Step 6:

[1511] The server automatically generates the WBS.

[1512] Input: The extracted task list.

[1513] Data processing / calculation: Calculate the hierarchical structure and dependencies of tasks and generate a WBS.

[1514] Output: The generated WBS.

[1515] Specific behavior:

[1516] Based on the extracted tasks, an algorithm is used to generate a hierarchical structure for the WBS.

[1517] Save the generated WBS in JSON format and use it in a later step.

[1518] Step 7:

[1519] The server uses an emotion engine to analyze the user's emotional state.

[1520] Input: User operation logs and input data.

[1521] Data processing / calculation: Estimate emotional state using emotion analysis API.

[1522] Output: Parsed emotion data.

[1523] Specific behavior:

[1524] Collects logs of user keyboard input and mouse operations.

[1525] Emotional states are estimated using a cloud-based emotion analysis API.

[1526] Step 8:

[1527] The user provides feedback on the generated WBS.

[1528] Input: User feedback.

[1529] Data processing / calculation: Generate an HTTP POST request to send the feedback to the server.

[1530] Output: The feedback is sent to the server.

[1531] Specific behavior:

[1532] The user fills out the feedback form and clicks the submit button.

[1533] The device generates and sends an HTTP POST request to send the feedback content to the server.

[1534] Step 9:

[1535] The server retrains the machine learning model based on the collected feedback.

[1536] Input: Feedback data, existing trained model.

[1537] Data processing / computation: Add feedback data and retrain the model.

[1538] Output: A machine learning model with improved accuracy.

[1539] Specific behavior:

[1540] Add new feedback data to the dataset.

[1541] Retrain your machine learning model and see improvements in accuracy.

[1542] Save the improved model.

[1543] (Application example 2)

[1544] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1545] Conventional work management systems have difficulty extracting appropriate tasks from past data and generating an efficient work breakdown structure (WBS) when starting a new task. Furthermore, they do not provide an interface or support that takes into account the user's emotional state, which can cause stress for users and lead to lower productivity and a deterioration in work efficiency. Furthermore, they lack a mechanism for appropriately collecting feedback and improving the accuracy of future predictions. These issues need to be resolved.

[1546] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1547] In this invention, the server includes means for collecting past business data, means for preprocessing the collected data, means for training a machine learning model using the preprocessed data, means for inputting an outline of a new business, means for extracting necessary tasks based on the input outline of the new business, means for organizing the extracted tasks to generate a work breakdown structure (WBS), means for displaying the generated WBS, means for collecting feedback on the generated WBS, means for retraining the machine learning model based on the collected feedback, and means for adjusting the WBS display method and support content based on the user's emotional state using an emotion engine that recognizes user emotions. This makes it possible to efficiently extract tasks necessary for new business based on past data, generate and display a WBS, and provide an interface and support that responds to the user's emotions.

[1548] "Past business data" refers to information about various business operations that have been carried out in the past, including data from project management tools, emails, chat logs, documents, etc.

[1549] "Preprocessing" refers to the process of removing duplication and noise from collected data and standardizing the format.

[1550] A "machine learning model" refers to an algorithm or computational model used to analyze and predict data based on collected data.

[1551] "Outline of new business" refers to basic information such as the purpose, scope, and conditions of the new business.

[1552] "Task extraction" refers to identifying and listing the necessary work items based on the outline of the new business.

[1553] A "Work Breakdown Structure (WBS)" is a tool used to hierarchically break down the work involved in a project or task and organize dependencies and priorities.

[1554] "Display" refers to visually presenting the generated information and WBS to the user.

[1555] "Feedback" refers to information collected from users, such as opinions and ratings, that is used to improve the system and retrain the model.

[1556] An "emotion engine" refers to a system that analyzes a user's emotional state and adjusts the system's behavior and display accordingly.

[1557] An embodiment of this invention will be described in detail. This system automatically generates work schedules within a factory and efficiently manages robot work. It generates new tasks based on past work data and automatically creates a work breakdown structure (WBS). Furthermore, it recognizes the emotions of factory staff and provides support and alerts according to their emotions.

[1558] Overview of program processing

[1559] The system starts by collecting historical business data and preprocessing it. The server collects data from project management tools, emails, chat logs, documents, etc. This data is collected and preprocessed using Python. Pandas is used to shape and clean the data, and Scikit-Learn is used to train machine learning models.

[1560] When a new task summary is input from a terminal, the server analyzes it and extracts the necessary tasks. It uses natural language processing technology to analyze the new task summary and generates appropriate tasks by referencing past data. To do this, it uses TfidfVectorizer to extract text features and RandomForestClassifier to predict tasks.

[1561] Next, a WBS is automatically created based on the generated tasks. The generated WBS is structured based on the time and dependencies of each task and displayed on the terminal. This display is provided as a user-friendly interface.

[1562] Additionally, the robot is equipped with an emotion engine that analyzes the emotional state of factory staff based on their input and behavior. The emotion engine uses the Hugging Face Transformer to tailor feedback and support content based on specific emotions. If the emotional state is likely to be negatively affected, the robot will provide special support or direct customers to a hotline.

[1563] Specific examples

[1564] For example, if a factory needs to perform a final inspection of a new part and reset the assembly line, enter "Final inspection of new part" and "Resetting the assembly line" as new tasks. The server generates the following tasks based on past similar projects:

[1565] Setting inspection standards for new parts

[1566] Inspection

[1567] Assembly line circuit settings

[1568] Test run of the reconfiguration

[1569] The WBS generated in this way is displayed visually on the terminal. If a factory worker types, "I'm feeling stressed today," the emotion engine analyzes it and provides an appropriate support message (e.g., "Attention, you need emotional support").

[1570] Prompt Sentence Examples

[1571] Example of new task input: "Final inspection of new parts"

[1572] Example of emotional state input: "I'm stressed today."

[1573] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1574] Step 1:

[1575] The server collects past business data. This data includes project management tools, emails, chat logs, and documents. Specifically, it collects information from each data source using APIs and data extraction scripts. The collected data is then stored in a database.

[1576] Inputs: Project management tools, emails, chat logs, documents

[1577] Output: Collected business data (stored in a database)

[1578] Step 2:

[1579] The server preprocesses the collected business data. Specifically, it uses Pandas to remove noise and duplicate data, and converts the data into a unified format. This step ensures the quality of the data.

[1580] Input: Collected business data

[1581] Output: Preprocessed business data

[1582] Step 3:

[1583] The server trains a machine learning model using the preprocessed data: it uses Scikit-Learn's TfidfVectorizer to extract text features for the task, and RandomForestClassifier to train a predictive model for the task.

[1584] Input: Preprocessed business data

[1585] Output: A trained machine learning model

[1586] Step 4:

[1587] The user inputs the outline of the new task via the terminal. For this, the task content is input in simple text format. When the outline of the new task is input, the terminal transmits the data to the server.

[1588] Input: New business summary (text format)

[1589] Output: Sent business summary data

[1590] Step 5:

[1591] The server uses natural language processing technology to analyze the outline of the new business and generate the necessary tasks based on past data. Specifically, it extracts text features of the new business using TfidfVectorizer and predicts tasks using a trained machine learning model.

[1592] Input: Summary data for new business

[1593] Output: Generated task list

[1594] Step 6:

[1595] The server automatically creates a work breakdown structure (WBS) based on the generated task list, including the task hierarchy, dependencies, and priorities.

[1596] Input: Generated task list

[1597] Output: Generated WBS

[1598] Step 7:

[1599] The terminal visually displays the generated WBS to the user, who can then check it and provide corrections or feedback as necessary.

[1600] Input: Generated WBS

[1601] Output: A visual representation of the WBS

[1602] Step 8:

[1603] The terminal collects user feedback on the generated WBS and transmits the information to the server.

[1604] Input: User feedback

[1605] Output: Feedback data sent

[1606] Step 9:

[1607] The server uses the collected feedback to retrain the machine learning model, which then improves its prediction accuracy in future runs.

[1608] Input: User feedback data

[1609] Output: A retrained machine learning model

[1610] Step 10:

[1611] The emotion engine analyzes the emotional state of factory staff from their input and actions, and uses the Hugging Face Transformer model to determine their emotional state, adjusting the interface and support content based on the user's emotional state.

[1612] Input: Text data expressing the emotions of factory staff

[1613] Output: Analyzed emotional state and support message based on it

[1614] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1615] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1616] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1617] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1618] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1619] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1620] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1621] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1622] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1623] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1624] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1625] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1626] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1627] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1628] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1629] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1630] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1631] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1632] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1633] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1634] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1635] The following is further disclosed regarding the above embodiment.

[1636] (Claim 1)

[1637] A means of collecting historical business data;

[1638] means for pre-processing the collected data;

[1639] means for training a machine learning model using the preprocessed data;

[1640] a means for inputting a summary of a new business;

[1641] A means for extracting necessary tasks based on the input outline of new work;

[1642] A means for organizing the extracted tasks to generate a WBS;

[1643] A means to display the generated WBS;

[1644] A means of gathering feedback on the generated WBS;

[1645] A means to retrain the machine learning model based on the collected feedback; and

[1646] A system including:

[1647] (Claim 2)

[1648] 2. The system of claim 1, wherein the historical business data includes project management tools, emails, chat logs, and documents.

[1649] (Claim 3)

[1650] 2. The system according to claim 1, further comprising means for representing task dependencies in a graph structure and for learning the dependency graph.

[1651] "Example 1"

[1652] (Claim 1)

[1653] A means of collecting historical business data;

[1654] means for pre-processing the collected data;

[1655] means for training a machine learning model using the preprocessed data;

[1656] a means for inputting a summary of a new business;

[1657] A means for extracting necessary tasks based on the input outline of new work;

[1658] A means for organizing the extracted tasks to generate a WBS;

[1659] A means to visually display the generated WBS;

[1660] A means of gathering feedback on the generated WBS;

[1661] A means to retrain the machine learning model based on the collected feedback; and

[1662] A system including:

[1663] (Claim 2)

[1664] 2. The system of claim 1, wherein the historical business data includes project management tools, emails, message logs, and documents.

[1665] (Claim 3)

[1666] 2. The system according to claim 1, further comprising means for representing task dependencies in a hierarchical structure and for learning the task dependency structure.

[1667] "Application Example 1"

[1668] (Claim 1)

[1669] A means of collecting historical business data;

[1670] means for pre-processing the collected data;

[1671] means for training a machine learning model using the preprocessed data;

[1672] a means for inputting a summary of a new business;

[1673] A means for extracting necessary tasks based on the input outline of new work;

[1674] A means for organizing the extracted tasks to generate a WBS;

[1675] A means to display the generated WBS;

[1676] A means of gathering feedback on the generated WBS;

[1677] A means to retrain the machine learning model based on the collected feedback; and

[1678] A means for assigning tasks to robots to optimize production processes in a factory;

[1679] A means to monitor and readjust the progress of each process in real time,

[1680] A system including:

[1681] (Claim 2)

[1682] 2. The system of claim 1, wherein the historical business data includes project management tools, emails, chat logs, and documents.

[1683] (Claim 3)

[1684] 2. The system according to claim 1, further comprising means for representing task dependencies in a graph structure and for learning the dependency graph.

[1685] "Example 2: Combining Emotion Engines"

[1686] (Claim 1)

[1687] A means of collecting historical business data;

[1688] means for pre-processing the collected data;

[1689] means for training a machine learning model using the preprocessed data;

[1690] a means for inputting a summary of a new business;

[1691] A means for extracting necessary tasks based on the input outline of new work;

[1692] A means for organizing the extracted tasks to generate a WBS;

[1693] A means to display the generated WBS;

[1694] A means of gathering feedback on the generated WBS;

[1695] A means to retrain the machine learning model based on the collected feedback; and

[1696] means for analyzing the emotional state of a user;

[1697] a means for adjusting the interface or feedback based on the analyzed emotional state;

[1698] A system including:

[1699] (Claim 2)

[1700] 2. The system of claim 1, wherein the historical business data includes project management tools, emails, electronic chat logs, and documents.

[1701] (Claim 3)

[1702] 2. The system according to claim 1, further comprising means for representing task dependencies in a graph structure and for learning the dependency graph.

[1703] "Application example 2 when combining emotion engines"

[1704] (Claim 1)

[1705] A means of collecting historical business data;

[1706] means for pre-processing the collected data;

[1707] means for training a machine learning model using the preprocessed data;

[1708] a means for inputting a summary of a new business;

[1709] A means for extracting necessary tasks based on the input outline of new work;

[1710] A means for organizing the extracted tasks to generate a work breakdown structure (WBS);

[1711] A means to display the generated WBS;

[1712] A means of gathering feedback on the generated WBS;

[1713] A means to retrain the machine learning model based on the collected feedback; and

[1714] A means for adjusting the display method of the WBS and the support content based on the user's emotional state using an emotion engine that recognizes the user's emotions;

[1715] A system including:

[1716] (Claim 2)

[1717] 2. The system of claim 1, wherein the historical business data includes project management tools, emails, chat logs, and documents.

[1718] (Claim 3)

[1719] 2. The system according to claim 1, further comprising means for representing task dependencies in a graph structure and for learning the dependency graph. [Explanation of symbols]

[1720] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of collecting historical business data; means for pre-processing the collected data; means for training a machine learning model using the preprocessed data; a means for inputting a summary of a new business; A means for extracting necessary tasks based on the input outline of new work; A means for organizing the extracted tasks to generate a WBS; A means to display the generated WBS; A means of gathering feedback on the generated WBS; A means to retrain the machine learning model based on the collected feedback; and A system including:

2. The system according to claim 1 , wherein the past business data includes project management tools, emails, chat logs, and documents.

3. 2. The system according to claim 1, further comprising means for expressing task dependencies in a graph structure and for learning the dependency graph.

Images