System

Abstract

A system is provided.SOLUTION: A system comprising: means for inputting a brief comment on an evaluation target person by a person involved in daily work; means for storing the input brief comment in a database; means for generating feedback content using a generation AI model that learns the stored brief comment; and means for notifying the evaluation target person of the generated feedback content at the end of a term.SELECTED DRAWING: Figure 1

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] Current feedback systems often provide feedback based on the most recent behavior of the person being evaluated, which is insufficient to provide true feedback. Furthermore, because the number of evaluators is limited, it is possible that only a portion of the person being evaluated is visible. Furthermore, due to input time constraints, there is a risk that comments may be insufficient, resulting in inconsistent feedback quality. This creates a need to address the issues of unfair evaluations and a lack of clear direction for growth. [Means for solving the problem]

[0005] The present invention provides a system that includes a means for people involved in daily work to input short comments about the person being evaluated, a means for saving the input short comments in a database, a means for generating feedback content using a generative AI model that learns from the saved short comments, and a means for notifying the person being evaluated of the generated feedback content at the end of the period. This system allows the person being evaluated to receive feedback from all people involved, freeing the evaluator from the constraints of relying on memory and input periods and allowing them to enter detailed comments as appropriate. Furthermore, the generative AI model generates appropriate feedback content, improving the accuracy and multifaceted perspectives of the evaluation, thereby contributing to the promotion of the person being evaluated's growth.

[0006] "Daily work" refers to the work activities and actions that are carried out on a daily basis in the course of carrying out one's work.

[0007] "Involved persons" are those who have business contact with the person being evaluated and can provide information necessary for the evaluation.

[0008] A "shorthand comment" is a short sentence that expresses a brief evaluation or impression of the person being evaluated.

[0009] "Input means" refers to a system including a device and software for inputting a short comment about the person being evaluated.

[0010] The "database" is a data storage system for structuring, storing, and managing input short comments.

[0011] The "generative AI model" is an artificial intelligence model that analyzes and learns from accumulated short comments and generates appropriate feedback.

[0012] "Feedback content" refers to a comprehensive evaluation document that includes evaluation and advice created by the generative AI model based on a short comment.

[0013] "End of period" refers to the end of a particular assessment period, at which point feedback is compiled.

[0014] "Means for notifying" refers to a communication and display system for informing the assessee of the generated feedback.

[0015] An "evaluation target" is a person who receives feedback on the performance of their work.

[0016] "Memory entry" is a method in which the evaluator enters feedback comments based on his or her own memory.

[0017] "Input period restriction" refers to a restriction that limits the input of feedback comments to a specific period.

[0018] "Growth promotion" is the act or process of the person being evaluated aiming to improve their abilities and skills based on the feedback.

[0019] A "multifaceted perspective" is a method of evaluating the person being evaluated from various angles and perspectives, meaning an evaluation that is not one-sided. [Brief explanation of the drawings]

[0020] [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

[0021] 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.

[0022] First, the terms used in the following description will be explained.

[0023] 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).

[0024] 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.

[0025] 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.

[0026] 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.

[0027] 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."

[0028] [First embodiment]

[0029] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0030] 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.

[0031] 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).

[0032] 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.

[0033] 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.

[0034] 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.

[0035] 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.

[0036] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0037] 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.

[0038] 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.

[0039] 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.

[0040] 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."

[0041] In order to realize a multifaceted feedback system, the present invention collects short comments about the person being evaluated on a daily basis, analyzes them using a generative AI model, and generates and notifies appropriate feedback content at the end of the period. Specific embodiments of the system are described below.

[0042] System configuration

[0043] This system consists of three main components: a user terminal, a server, and a generative AI model.

[0044] User terminal

[0045] The user terminal is a device that allows the evaluator to input short comments about the person being evaluated based on their impressions during their daily work, and can include PCs, smartphones, tablets, etc. A dedicated feedback input application is installed on the user terminal. Using this application, the user (evaluator) can input concise comments about the person being evaluated.

[0046] server

[0047] The server has the following functions:

[0048] 1. Database function: A function that receives comment data entered by users on a daily basis and stores it in a database as structured data.

[0049] 2. Managing generative AI models: A function that allows you to feed saved comment data to generative AI models for learning.

[0050] 3. Feedback generation: A function that uses a generative AI model to generate feedback for each person being evaluated at the end of the period.

[0051] 4. Notification function: A function to notify the generated feedback to the device of the person being evaluated.

[0052] Generative AI Models

[0053] The generative AI model uses natural language processing technology to analyze the saved one-word comment data and generate appropriate feedback for each person being evaluated. This model provides comprehensive feedback by taking into account the relationship between the evaluator and the person being evaluated, the content, frequency, and emotion of the comment, etc.

[0054] Explaining the feedback generation process

[0055] The following describes how this system works, using specific examples.

[0056] Comment input

[0057] A user (e.g., colleague A) uses a terminal to input a short comment about the person being evaluated (e.g., colleague B) into a feedback input application. For example, the user might leave a comment in the form of, "Mr. B made a great proposal in today's meeting."

[0058] Saving comment data

[0059] The comment data sent from the device is sent to a server, which stores the received comment data in a database and organizes it by adding metadata to each comment, such as the ID of the person being evaluated, the comment content, and a timestamp.

[0060] Data training

[0061] The server periodically supplies the saved comment data to the AI ​​model for learning. The AI ​​model updates the feedback model for each person being evaluated, taking into account the content of the comments, the relationship between the evaluator and the person being evaluated, and the frequency of comments.

[0062] Generate feedback

[0063] At the end of the period, the server uses the generative AI model to generate feedback for each person being evaluated, such as, "Mr. B made an important proposal in this project and made a significant contribution to the team. He also showed a proactive attitude toward seeking feedback."

[0064] Feedback Notification

[0065] The generated feedback is stored in a database and sent from the server to the device of the person being evaluated. The person being evaluated can then check the generated feedback using a dedicated feedback confirmation application on their device.

[0066] This allows the person being evaluated to receive multifaceted and detailed feedback that can be used to improve themselves. It also makes it easier for the evaluator to enter comments, freeing them from having to rely on memory or input constraints, allowing them to provide high-quality feedback.

[0067] The processing flow will be explained below.

[0068] Program processing flow

[0069] Step 1:

[0070] Users can input short comments about the person being evaluated into a dedicated feedback input application during work or daily conversations. For example, colleague A might input, "Mr. B made a very useful suggestion in today's meeting."

[0071] Step 2:

[0072] The device organizes the input comment data as structured data and sends it to the server, adding metadata such as the user's ID, comment content, and timestamp.

[0073] Step 3:

[0074] The server stores the received comment data in a database. For example, the data is stored in the format of "evaluator ID: B123, comment content: 'Very useful suggestion', timestamp: 2023-10-01 10:30".

[0075] Step 4:

[0076] Each time a new comment is added, the server feeds the data to the generative AI model, which performs an initial learning process, such as cleaning the text (correcting unnecessary characters and typos) and categorizing the comment.

[0077] Step 5:

[0078] The generative AI model learns from the provided data, using natural language processing technology to analyze the content of comments, the relationship between the evaluator and the person being rated, the sentiment of the ratings, frequency, etc.

[0079] Step 6:

[0080] The server sets a schedule to automatically supply additional data to the AI ​​model at regular intervals (e.g., one week) to continuously update and train the model.

[0081] Step 7:

[0082] As the end of the term approaches, the server issues an instruction to re-feed all comment data to the generative AI model. For example, it will feed all data one week before the end of the term.

[0083] Step 8:

[0084] A generative AI model uses all stored comment data to generate appropriate feedback for each ratee, balancing positive and negative feedback.

[0085] Step 9:

[0086] The server stores the generated feedback content in a database and simultaneously sends a notification to the terminal of the person being evaluated that the feedback has been generated.

[0087] Step 10:

[0088] The device receives a notification, and the person being evaluated can check the generated feedback through a dedicated feedback confirmation application. Specifically, when the person being evaluated opens the application, they can see feedback such as, "Mr. B made important suggestions in this project and made a significant contribution to the team. He also showed an attitude of proactively seeking feedback."

[0089] This allows the person being evaluated to receive detailed feedback from all angles, which can be used to improve themselves. In addition, the evaluator can enter detailed comments when necessary, freeing them from the constraints of relying on memory or a limited input period.

[0090] Example 1

[0091] 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."

[0092] Current feedback systems have the drawback of making it difficult for evaluators to provide timely and detailed feedback on the person being evaluated. Because the feedback relies on the evaluator's subjectivity, it lacks consistency and objectivity, making it difficult to effectively support the growth of the person being evaluated. Another issue is that the cumbersome process of inputting and managing feedback places an excessive burden on the evaluator.

[0093] 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.

[0094] In this invention, the server includes: a means for a user to input a one-line comment about the person to be evaluated; a means for a terminal to transmit the input one-line comment to the server; a means for the server to save the received one-line comment data in a database; a means for the server to periodically supply the saved one-line comment data to a generative AI model for learning; a means for the server to generate feedback content for each person to be evaluated using the generative AI model at the end of the period; and a means for the server to notify the terminal of the person to be evaluated of the generated feedback content. This makes it possible to provide consistent and objective feedback to the person to be evaluated and enables the evaluator to easily input and manage comments.

[0095] A "user" is a person who uses the system to input a short comment about the person being evaluated.

[0096] A "terminal" is a device used by a user, and includes a personal computer, a smartphone, a tablet, and the like.

[0097] "Server" refers to a device that stores comment data, manages generative AI models, and generates and notifies feedback.

[0098] A "shorthand comment" is a short textual evaluation content entered by a user about the person being evaluated.

[0099] The "database" is a system that stores received short comment data and generated feedback content as structured data.

[0100] The "generative AI model" is an algorithm that learns from saved one-line comment data and generates feedback content using natural language processing technology.

[0101] "Feedback content" refers to the content of the evaluation and advice given to the person being evaluated, generated by the generative AI model.

[0102] "Notification" is the process of transmitting the generated feedback content to the terminal of the person being evaluated.

[0103] "End of period" refers to the end of the cycle for generating and providing feedback.

[0104] A "prompt sentence" is an input sentence used to have a generative AI model generate feedback content.

[0105] In order to realize a multifaceted feedback system, this invention collects short comments about the person being evaluated on a daily basis, analyzes them using a generative AI model, and generates and notifies appropriate feedback content at the end of the period.

[0106] System configuration

[0107] This system consists of three main components: the user terminal, the server, and the generative AI model.

[0108] User terminal

[0109] The user terminal is a device that allows the evaluator to input short comments about the person being evaluated based on their impressions during their daily work, and can include a personal computer, smartphone, tablet, etc. A dedicated feedback input application is installed on the user terminal. Using this application, the user (evaluator) can input concise comments about the person being evaluated.

[0110] Examples:

[0111] A user (e.g., colleague A) uses a terminal to input a short comment about the person being evaluated (e.g., colleague B) into a feedback input application. For example, the user might leave a comment in the form of, "Mr. B made a great proposal in today's meeting."

[0112] server

[0113] The server has the following functions:

[0114] 1. Database function: A function that receives comment data entered by users on a daily basis and stores it in a database as structured data.

[0115] 2. Managing generative AI models: A function that allows you to feed saved comment data to generative AI models for learning.

[0116] 3. Feedback generation: A function that uses a generative AI model to generate feedback for each person being evaluated at the end of the period.

[0117] 4. Notification function: A function to notify the generated feedback to the device of the person being evaluated.

[0118] Examples:

[0119] Comment data sent from the device is sent to a server, which stores the received comment data in a database and organizes it by adding metadata to each comment, such as the user's ID, comment content, and timestamp.

[0120] Generative AI Models

[0121] The generative AI model uses natural language processing technology to analyze the saved one-word comment data and generate appropriate feedback for each person being evaluated. This model provides comprehensive feedback by taking into account the relationship between the evaluator and the person being evaluated, the content, frequency, and emotion of the comment, etc.

[0122] Examples:

[0123] At the end of the period, the server uses the generative AI model to generate feedback for each person being evaluated, such as, "Mr. B made an important proposal in this project and made a significant contribution to the team. He also showed a proactive attitude toward seeking feedback."

[0124] Feedback Generation Process

[0125] A specific example of the feedback generation process is as follows:

[0126] Learning from comment data

[0127] The server periodically supplies the comment data stored in the database to the generative AI model, which then learns from it, taking into account the content of the comment, the relationship between the evaluator and the person being evaluated, and the frequency of the comment.

[0128] Example prompt sentence:

[0129] Generate feedback for Subject B based on the following comments:

[0130] 1. Person B asked insightful questions during team meetings, which sparked deeper discussions.

[0131] 2. Mr. B's new idea was very original and caught everyone's attention.

[0132] Feedback content:

[0133] Feedback Notification

[0134] The server stores the generated feedback in a database and notifies the evaluation recipient's device, where the evaluation recipient can check the generated feedback using a dedicated feedback checking application.

[0135] This allows the person being evaluated to receive multifaceted and detailed feedback that can be used to improve themselves. It also makes it easier for the evaluator to enter comments, freeing them from having to rely on memory or input constraints, allowing for the provision of high-quality feedback.

[0136] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0137] Step 1:

[0138] Enter a comment

[0139] The user inputs a comment about the person being evaluated.

[0140] Input: Comment text entered by the evaluator through the terminal.

[0141] Specific operation: The user launches a dedicated feedback input application and enters a comment about the person being evaluated, such as "Mr. B made a great proposal in today's meeting." The terminal has a "Send" button, which the user clicks.

[0142] Output: The comment text entered is ready to be sent.

[0143] Step 2:

[0144] Submitting comment data

[0145] The terminal transmits the input comment data to the server.

[0146] Input: The comment text that you have completed.

[0147] Specific operation: The device sends the comment text as an HTTP request to a specific API endpoint (e.g., POST / comments) on the server.

[0148] Output: The comment data sent to the server.

[0149] Step 3:

[0150] Saving comment data

[0151] The server stores the received comment data in a database.

[0152] Input: The comment text received by the server.

[0153] Specific operation: The server receives the HTTP request, adds metadata such as the rater's ID, comment content, and timestamp, and structures the data. It then generates and executes an INSERT query in a database (e.g., MySQL) to save the comment data.

[0154] Output: Comment data stored in the database.

[0155] Step 4:

[0156] Data training

[0157] The server periodically supplies the saved comment data to the generative AI model for learning.

[0158] Input: Multiple comment data stored in the database.

[0159] How it works: The server periodically retrieves comment data from the database as a scheduled job and supplies it to the API of a generative AI model (e.g., the GPT-3 model). When supplied, it converts it to JSON format and sends it to the model's endpoint. The model learns by taking into account the comment content, relationships, frequency, etc.

[0160] Output: A trained generative AI model.

[0161] Step 5:

[0162] Generate feedback

[0163] At the end of the period, the server uses the generated AI model to generate feedback content for each person being evaluated.

[0164] Input: Trained generative AI model and comment data.

[0165] Specific operation: The server generates a prompt based on the ID of the person being evaluated and sends it to the generation AI model. For example, a prompt like "Generate feedback for person B based on the following comments: 1. Person B asked a sharp question in the team meeting, which sparked a deeper discussion. 2. Person B's new idea was very novel and attracted everyone's attention. Feedback content:" is used. The feedback returned by the model is then processed and organized.

[0166] Output: The generated feedback.

[0167] Step 6:

[0168] Feedback Notification

[0169] The server notifies the generated feedback content to the terminal of the person being evaluated.

[0170] Input: The generated feedback.

[0171] Specific operation: The server saves the feedback content in a database and sends a notification to the assessee's device using a notification service (e.g., Firebase Cloud Messaging). The application on the assessee's device receives the notification and prompts the user to confirm.

[0172] Output: Feedback provided and confirmation by the assessee.

[0173] (Application example 1)

[0174] 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."

[0175] It is important to routinely evaluate the operating status and performance of automated equipment operating on-site and provide appropriate feedback. However, in many production sites, such evaluations are often dependent on a few managers, resulting in subjective and inconsistent evaluations. Furthermore, it has been difficult to quickly and effectively monitor the status of automated equipment during daily operations and detect problems early. Furthermore, delays in such evaluations and feedback can result in equipment deterioration and reduced production efficiency.

[0176] 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.

[0177] In this invention, the server includes means for inputting a short comment about the person being evaluated by a person involved in daily work, means for saving the input short comment in a database, means for generating feedback content using a generative AI model that learns the saved short comment, means for notifying the person being evaluated of the generated feedback content at the end of the period, means for inputting comments about the operating status and performance of automated operation equipment operating in a factory, and means for evaluating the performance of the automated operation equipment based on the input comments. This enables daily feedback about the operating status and performance of the automated operation equipment in a factory, makes evaluations consistent and objective, and enables efficient operation of the automated operation equipment and early problem detection.

[0178] "Evaluation target" refers to the person who is evaluated in the feedback system, and refers to automated work equipment and robots operating in a factory.

[0179] A "shorthand comment" is a short written opinion or observation about the operating status or performance of the subject of evaluation.

[0180] The "database" is a system for storing and managing input short comments as structured data.

[0181] A "generative AI model" is an artificial intelligence model that learns from input one-line comments and generates feedback content based on them.

[0182] "End of period" refers to the end of a specific time or period when feedback is communicated to the assessee.

[0183] "Notification means" refers to a method or device for notifying the generated feedback content to the person being evaluated.

[0184] "Automated work equipment" refers to machines or robots used to automatically perform specific tasks in a factory.

[0185] "Performance evaluation" refers to the task of evaluating the operating status and performance of the person being evaluated.

[0186] This system evaluates the operational status and performance of automated work equipment in a factory on a daily basis and provides appropriate feedback. This system consists of three main components: a user terminal, a server, and a generative AI model.

[0187] System configuration

[0188] User terminal

[0189] The user terminal is a device that allows the evaluator to input a brief comment about the automated operation equipment operating in the factory, and can include PCs, smartphones, tablets, etc. A dedicated feedback input application is installed on the user terminal. Using this application, the user (evaluator) can input a brief comment about the evaluation target (automated operation equipment).

[0190] server

[0191] The server has the following functions:

[0192] 1. Database function: A function that receives comment data entered by users and stores it in a database as structured data.

[0193] 2. Managing generative AI models: A function that allows you to feed saved comment data to generative AI models for learning.

[0194] 3. Feedback generation: A function that uses a generative AI model to generate feedback for each person being evaluated at the end of the period.

[0195] 4. Notification function: A function to notify the generated feedback to the device of the person being evaluated.

[0196] Generative AI Models

[0197] The generative AI model uses natural language processing technology to analyze the saved one-line comment data and generate appropriate feedback for each person being evaluated. This model provides comprehensive feedback by taking into account the content, frequency, importance, etc. of the comment.

[0198] Feedback Generation Process

[0199] The following describes how this system works, using specific examples.

[0200] Comment input

[0201] A user (e.g., Manager A) uses a terminal to input a brief comment about an automated work device (e.g., Robot A) in a factory into a feedback input application. For example, the user leaves a comment in the form of "Robot A's work speed seemed slow today."

[0202] Saving comment data

[0203] The comment data sent from the device is sent to a server, which stores the received comment data in a database and organizes it by adding metadata to each comment, such as the automated work device's ID, comment content, and timestamp.

[0204] Data training

[0205] The server periodically supplies the saved comment data to the AI ​​model for learning. The AI ​​model updates the feedback model for each person being evaluated, taking into account the content of the comments, the relationship between the evaluator and the automated work device, and the frequency of comments.

[0206] Specific examples

[0207] Examples of comment input:

[0208] Automated Work Device ID: RobotA

[0209] Comment: "You seem to be working a lot slower today."

[0210] Example prompt sentence:

[0211] Generate feedback for each robot based on the comment data below:

[0212] Robot ID: RobotA, Comment: "You seem to be working a lot slower today."

[0213] Examples of feedback generated include:

[0214] Robot A requires regular maintenance. It has recently been running slower, possibly due to wear on the belt.

[0215] In this way, users can receive comprehensive feedback about the automated work equipment in their factories, which can be used to improve maintenance and operation.

[0216] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0217] Step 1:

[0218] A user uses a terminal to input a brief comment about an automated work device in a factory into a feedback input application.

[0219] Input: ID of the automated work device, comment (e.g., "Robot A seemed to be working slowly today")

[0220] Output: Comment data is sent from the device to the server.

[0221] Step 2:

[0222] The comment data sent from the terminal is received by the server.

[0223] Input: Comment data sent from the device (automated work device ID, comment content, timestamp)

[0224] Output: Comment data is saved in the database.

[0225] Step 3:

[0226] The server periodically supplies the comment data stored in the database to the generative AI model, allowing it to learn.

[0227] Input: Comment data stored in the database

[0228] Output: The generative AI model learns from the comment data and updates the feedback model.

[0229] Step 4:

[0230] At the end of the term, the server uses a generative AI model to generate feedback content for each person being evaluated.

[0231] Input: Comment data trained by the generative AI model

[0232] Output: End-of-period feedback (e.g., "Robot A needs regular maintenance. It has been running slowly recently, which may be due to wear on the belt.")

[0233] Step 5:

[0234] The generated feedback content is notified from the server to the terminal of the person being evaluated.

[0235] Input: Generated feedback

[0236] Output: A feedback notification is displayed on the assessee's device.

[0237] Step 6:

[0238] The person being evaluated checks the generated feedback content using the terminal.

[0239] Input: Feedback content sent to the device of the person being evaluated

[0240] Output: The person being evaluated will view the feedback and use it as a reference for implementing any necessary improvements or measures.

[0241] 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.

[0242] This system provides multifaceted feedback on the person being evaluated. The system allows people who work with the person on a daily basis to enter short comments about the person being evaluated, which are then stored in a database. A generative AI model then learns from these comments and generates feedback content. Furthermore, by combining this with an emotion engine, the system analyzes the emotions in the comments and provides more accurate feedback.

[0243] System configuration

[0244] The system consists of four main components: a user terminal, a server, a generative AI model, and an emotion engine.

[0245] User terminal

[0246] The user terminal is a device that allows the evaluator to enter short comments about the person being evaluated, and can include PCs, smartphones, tablets, etc. A dedicated feedback input application is installed on the terminal, and a comment input screen is provided.

[0247] server

[0248] The server has the following functions:

[0249] 1. Database function: A function to save comment data received from users in a database.

[0250] 2. Sentiment analysis function: A function that uses an emotion engine to analyze the sentiment of one-line comments and classify them as positive, negative, or neutral.

[0251] 3. Managing generative AI models: A function that allows saved comment data and its sentiment classification results to be fed into generative AI models for learning.

[0252] 4. Feedback generation: A function that generates feedback for each person being evaluated at the end of the period using a generative AI model.

[0253] 5. Notification function: A function to notify the assessee of the generated feedback.

[0254] Generative AI Models

[0255] The generative AI model uses natural language processing technology to analyze the saved one-word comment data and emotion classification results, and generates appropriate feedback content for each person being evaluated. This model reflects the emotional information of the comments received from the emotion engine and provides comprehensive feedback.

[0256] Emotion Engine

[0257] The emotion engine analyzes the sentiment of short comments entered by users and categorizes them as positive, negative, or neutral. The emotion engine uses natural language processing technology to extract sentiment from the text and sends the results to the server.

[0258] Explaining the feedback generation process

[0259] The following describes how this system works, using specific examples.

[0260] Comment input

[0261] A user (e.g., colleague A) uses a terminal to input a short comment about the person being evaluated (e.g., colleague B) into a feedback input application. For example, the user might leave a comment in the form of, "Mr. B made a great proposal in today's meeting."

[0262] Saving comment data

[0263] Comment data sent from the device is sent to a server, which stores the received comment data in a database and organizes it by adding metadata to each comment, such as the ID of the person being evaluated, the comment content, and a timestamp.

[0264] Emotion Analysis

[0265] The server provides the stored comment data to the emotion engine for sentiment analysis. The emotion engine analyzes the content of each comment and classifies it as positive, negative, or neutral. For example, "Mr. B made a great suggestion" is classified as positive.

[0266] Data training

[0267] The server supplies the comment data and its emotion classification results to the generative AI model, which then learns from it. The generative AI model reflects the emotion classification results and updates the feedback model for each person being evaluated, taking into account the comment content, the relationship between the evaluator and the person being evaluated, the emotion and frequency of the evaluation, etc.

[0268] Generate feedback

[0269] At the end of the period, the server uses the generative AI model to generate feedback for each individual being evaluated. For example, it generates multifaceted, emotionally-charged feedback such as, "Mr. B made an important proposal in this project and made a significant contribution to the team. He also showed a proactive attitude toward seeking feedback."

[0270] Feedback Notification

[0271] The generated feedback is stored in a database and sent from the server to the device of the person being evaluated. The person being evaluated can then check the generated feedback using a dedicated feedback confirmation application on their device.

[0272] This allows those being evaluated to receive detailed, accurate feedback from all angles, which can be used to improve themselves. Furthermore, evaluators can enter detailed comments when needed, freeing them from the constraints of relying on memory or time constraints. Furthermore, the introduction of an emotion engine improves the balance and accuracy of feedback, providing a fair and multifaceted evaluation.

[0273] The processing flow will be explained below.

[0274] Processing flow of a feedback system that combines an emotion engine

[0275] Step 1:

[0276] Users can input short comments about the person being evaluated into a dedicated feedback input application during work or daily conversations. For example, colleague A might input, "Mr. B made a very useful suggestion in today's meeting."

[0277] Step 2:

[0278] The device organizes the input comment data as structured data and sends it to the server, adding metadata such as the user's ID, comment content, and timestamp.

[0279] Step 3:

[0280] The server saves the received comment data in the database in the following format: "Evaluator ID: B123, Comment content: 'Very useful suggestion', Timestamp: 2023-10-01 10:30".

[0281] Step 4:

[0282] The server provides the stored comment data to the emotion engine for sentiment analysis. The emotion engine analyzes the content of each comment and classifies it as positive, negative, or neutral. For example, "Mr. B made a great suggestion" would be classified as positive.

[0283] Step 5:

[0284] The server adds the emotion classification results obtained from the emotion engine to the comment data and saves it in the database. For example, "evaluator ID: B123, comment content: 'Very useful suggestion', emotion: positive" is saved.

[0285] Step 6:

[0286] The server periodically supplies all comment data and emotion classification results to the AI ​​model, allowing it to update and learn. For example, you can set it to automatically start AI training on weekends.

[0287] Step 7:

[0288] The generative AI model learns from the provided data, using natural language processing technology to analyze the content of comments, the relationship between the rater and the person being rated, the sentiment classification results, the sentiment of the ratings, frequency, etc.

[0289] Step 8:

[0290] As the end of the term approaches, the server again supplies all comment data and emotion classification results to the generative AI model and instructs it to generate feedback. For example, the server issues a data supply instruction one week before the end of the term.

[0291] Step 9:

[0292] The generative AI model generates appropriate feedback for each person being evaluated based on all stored comment data and emotion classification results. For example, it considers a balance of positive and negative comments and generates feedback such as, "Mr. B made important suggestions in this project and made a significant contribution to the team. He also showed an attitude of actively seeking feedback."

[0293] Step 10:

[0294] The server stores the generated feedback content in a database and simultaneously sends a notification to the terminal of the person being evaluated that the feedback has been generated.

[0295] Step 11:

[0296] The device receives a notification, and the person being evaluated can check the generated feedback through a dedicated feedback review application. Specifically, when the person being evaluated opens the application, a list of the feedback and its details are displayed.

[0297] This allows those being evaluated to receive detailed and accurate feedback from all angles, which can be used to improve their own performance. Furthermore, users (evaluators) can enter detailed comments whenever necessary, freeing them from the constraints of memory and time constraints. Furthermore, the introduction of an emotion engine improves the balance and accuracy of feedback, providing a fair and multifaceted evaluation.

[0298] Example 2

[0299] 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."

[0300] In conventional evaluation systems, evaluators rely on their memory to make evaluations, which can lead to problems with the accuracy and fairness of the evaluations. Furthermore, because the evaluation content is one-sided, it is difficult to grasp the overall picture of the person being evaluated's performance and behavior. Furthermore, evaluations are not based on emotions, which can lead to insufficient consideration of the evaluation content.

[0301] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0302] In this invention, the server includes: a means for people involved in daily work to input short comments about the person being evaluated; a means for saving the input short comments in a database; a means for analyzing the sentiment of the saved short comments and classifying them as positive, negative, or neutral; a means for generating feedback content using a generative AI model that learns short comments including the sentiment analysis results; and a means for notifying the person being evaluated of the generated feedback content at the end of the period. This improves the accuracy and fairness of the evaluation content and enables multifaceted evaluation of the person's performance and behavior. Furthermore, incorporating sentiment analysis allows the evaluation content to reflect emotional aspects.

[0303] A "shorthand comment" is a short piece of feedback entered by the evaluator about the person being evaluated during their daily work.

[0304] "Database" means an information storage system for storing and managing entered short comments and other related data.

[0305] "Sentiment analysis" is a technology that uses natural language processing technology to analyze the emotions contained in short comments and classify them as positive, negative, or neutral.

[0306] The "generative AI model" is an artificial intelligence model that learns from saved one-line comments and the results of their emotional analysis, and generates feedback content for each person being evaluated.

[0307] "Notification" is a system function for notifying the subject of evaluation of the generated feedback content.

[0308] The present invention is a system that provides multifaceted feedback to an individual by inputting a short comment about the individual involved in daily work, storing the comment in a database, performing sentiment analysis, generating feedback using a generative AI model, and notifying the individual. Each component of this system and its embodiment will be described in detail below.

[0309] System configuration

[0310] The system consists of four main components: a user terminal, a server, a generative AI model, and a sentiment analysis engine.

[0311] User terminal

[0312] The user terminal is a device on which the evaluator inputs a short comment about the person being evaluated. It can be a PC, smartphone, tablet, etc. A dedicated feedback input application is installed on the user terminal, and a comment input screen is provided. For example, the evaluator might input a comment in the form of "Mr. B made a great proposal in today's meeting."

[0313] server

[0314] The server has the following features:

[0315] 1. Database function: Comment data received from users is saved in a database. The database uses MySQL or PostgreSQL.

[0316] 2. Sentiment analysis function: Using an emotion engine, the sentiment of one-line comments is classified as positive, negative, or neutral. Sentiment analysis is performed using Python's VADER and TextBlob.

[0317] 3. Managing the generative AI model: The saved comment data and its sentiment classification results are fed into the generative AI model for training. The generative AI model uses a Transformer-based model such as GPT-3 or BERT.

[0318] 4. Feedback generation: At the end of the period, feedback content is generated for each person being evaluated using the generative AI model.

[0319] 5. Notification function: Notify the assessee of the generated feedback.

[0320] Generative AI Models

[0321] The generative AI model uses natural language processing technology to analyze the saved one-line comment data and sentiment analysis results, generating appropriate feedback for each individual being evaluated. The generative AI model is implemented using Python or other programming languages, utilizing machine learning frameworks such as TensorFlow and PyTorch. For example, it might generate feedback such as, "Mr. B made important suggestions in this project and made a significant contribution to the team. He also showed a proactive attitude toward seeking feedback."

[0322] Sentiment Analysis Engine

[0323] The sentiment analysis engine analyzes the sentiment of short comments entered by users and classifies them as positive, negative, or neutral. The sentiment analysis engine uses natural language processing technology to extract the sentiment of the text and sends the results to a server. Technologies used include VADER and TextBlob.

[0324] Examples of concrete examples and prompts

[0325] For example, if a user (colleague A) enters a comment such as "Mr. B made a great proposal in today's meeting," the server receives this comment and stores it in a database. It is then classified as "positive" using a sentiment analysis engine. The generative AI model generates feedback for the person being evaluated (colleague B) based on this data.

[0326] Examples:

[0327] Comment: "Mr. B made a great suggestion at today's meeting."

[0328] Sentiment Analysis: Positive

[0329] Feedback: "B made important suggestions on this project and contributed greatly to the team. He also proactively sought feedback."

[0330] Example prompt for a generative AI model:

[0331] "Generate the following feedback: 'Person B made a great suggestion in today's meeting.'"

[0332] In this way, the system according to the present invention can provide multifaceted and accurate feedback regarding the person being evaluated.

[0333] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0334] Step 1:

[0335] Comments entered by the user

[0336] Operation: The user opens a feedback application on a device (PC, smartphone, tablet, etc.). They access the comment input screen and enter a short comment about the person being evaluated. For example, they enter a comment such as, "Mr. B made a great proposal in today's meeting."

[0337] Input: Text comments about the person being assessed.

[0338] Output: The entered comment data is sent from the terminal to the server.

[0339] Step 2:

[0340] Sending and saving comment data to the server

[0341] Operation: Comment data sent from the device is sent to the server via an HTTP request. The server analyzes the received comment data, extracts the ID of the person being rated, the ID of the evaluator, the comment content, and the timestamp, and stores them in a database.

[0342] Input: Comment data sent from the device (text comment, ID of the person being evaluated, ID of the evaluator, timestamp).

[0343] Output: Comment data stored in a database.

[0344] Step 3:

[0345] Performing sentiment analysis

[0346] How it works: The server passes the comment data stored in the database to a sentiment analysis engine, which uses natural language processing libraries such as VADER or TextBlob to classify the sentiment of each comment as positive, negative, or neutral.

[0347] Input: Text comment stored in the database.

[0348] Output: Comment data with sentiment classification results (positive, negative, neutral).

[0349] Step 4:

[0350] Supplying and training generative AI models

[0351] How it works: The server supplies the sentiment-analyzed comment data to the generative AI model, which then uses natural language processing technology to update the feedback model for each subject based on the data provided.

[0352] Input: Comment data with sentiment classification results.

[0353] Output: An updated generative AI model.

[0354] Step 5:

[0355] Generate feedback

[0356] How it works: At the end of the term, the server uses the updated generative AI model to generate feedback for each person being evaluated. For example, it generates feedback in the form of, "Mr. B made important suggestions in this project and made a significant contribution to the team. He also showed a proactive attitude of seeking feedback."

[0357] Input: Updated generative AI model, ID of the person being evaluated.

[0358] Output: Generated feedback for each ratee.

[0359] Step 6:

[0360] Feedback Notification

[0361] Operation: The server notifies the subject of the generated feedback to the subject's device. The subject then checks the received feedback using a dedicated feedback checking application.

[0362] Input: Generated feedback content, ID of the person being assessed.

[0363] Output: Feedback sent to the subject's device.

[0364] Through these steps, users can provide multifaceted and accurate feedback to those being evaluated. Raters can easily enter comments, and sentiment analysis and generative AI models ensure detailed and fair feedback.

[0365] (Application example 2)

[0366] 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."

[0367] In conventional feedback systems, evaluations are often based on the subjective opinions of the evaluator or one-sided information, making it difficult to accurately grasp the overall growth and areas for improvement of the person being evaluated.Furthermore, because the emotional aspects of the comments are not taken into consideration, there are problems with biased feedback content and unfair evaluations.

[0368] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a means for a person who is involved in daily work to input a short comment about the person to be evaluated, a means for saving the input short comment in a database, a means for generating feedback content using a generative AI model that learns from the saved short comment, a means for analyzing the input comment with an emotion engine and classifying emotions, and a means for notifying the person to be evaluated of the generated feedback content at the end of the period. This enables fair feedback that is multifaceted and reflects emotions.

[0369] "People involved in daily work" refers to colleagues, superiors, subordinates, and other people who work with the person being evaluated on a daily basis.

[0370] "Evaluation target" refers to the person, employee, worker, etc. who is the subject of feedback.

[0371] A "shorthand comment" is a short feedback message about the person being appraised that describes a behavior or performance observed during work.

[0372] "Means of storing in a database" refers to the systems and technologies for systematically storing and managing the entered comment data.

[0373] A "generative AI model" refers to an artificial intelligence module that learns based on input data and generates feedback content.

[0374] "Means for generating feedback content" refers to the process of analyzing collected data using a generative AI model and creating appropriate feedback for each person being evaluated.

[0375] An "emotion engine" refers to technology that analyzes the emotional aspects of input comments and classifies them as positive, negative, neutral, etc.

[0376] "Sentiment classification" refers to the process of identifying and categorizing the emotional aspects of comments using an emotion engine.

[0377] "Means of notifying at the end of the period" refers to the mechanisms and techniques for communicating the generated feedback content to the person being evaluated at the end of the evaluation period.

[0378] The present invention is a system in which people who interact with the person being evaluated in their daily work enter short comments about the person being evaluated, save them in a database, and a generative AI model learns from them to generate feedback content. The system is configured as follows:

[0379] Hardware and Software

[0380] 1. User device: A device used by the user to enter a short comment about the person being evaluated. This can be a PC, smartphone, head-mounted display (HMD), etc. A dedicated feedback input application is installed and a comment input screen is provided.

[0381] 2. Server:

[0382] Database function: A function to save comment data received from users in a database.

[0383] Sentiment analysis function: A function that uses an emotion engine to analyze the sentiment of short comments and classify them as positive, negative, or neutral.

[0384] Managing generative AI models: A function that allows saved comment data and its sentiment classification results to be fed into generative AI models for training.

[0385] Feedback generation: A function that uses a generative AI model to generate feedback for each assessee.

[0386] Notification function: A function to notify the assessee of the generated feedback at the end of the period.

[0387] Processing flow and data calculation

[0388] 1. Comment entry:

[0389] A user (e.g., a factory worker) uses a terminal to input a short comment about the work of the person being evaluated (e.g., a robot) into a feedback input application. For example, the user might leave a comment in the form of, "Robot A carried the pallet smoothly today."

[0390] Example prompt: "In one sentence, tell us what you noticed about the robot's work."

[0391] 2. Storage of Comment Data:

[0392] Comment data sent from the device is sent to a server, which stores the received comment data in a database and organizes it by adding metadata to each comment, such as the ID of the person being evaluated, the comment content, and a timestamp.

[0393] 3. Emotion analysis:

[0394] The server provides the stored comment data to the emotion engine for emotion analysis. The emotion engine analyzes the content of each comment and classifies it as positive, negative, or neutral. For example, "Robot A handled the situation smoothly" is classified as positive.

[0395] Example prompt: "Analyze the sentiment of this comment and categorize it as positive, negative, or neutral."

[0396] 4. Data training:

[0397] The server supplies the comment data and its emotion classification results to the generative AI model, which then learns from it. The generative AI model reflects the emotion classification results and updates the feedback model for each person being evaluated, taking into account the comment content, the relationship between the evaluator and the person being evaluated, the emotion and frequency of the evaluation, etc.

[0398] 5. Feedback Generation:

[0399] At the end of the period, the server uses the generative AI model to generate feedback for each individual being evaluated, generating multifaceted and emotional feedback such as, "Robot A was recognized for its smooth movements in transporting pallets, and its efficiency has improved."

[0400] Example prompt: "Based on this sentiment analysis and comment data, generate feedback to improve the robot's performance."

[0401] 6. Feedback Notification:

[0402] The generated feedback is stored in a database and sent from the server to the device of the person being evaluated. The person being evaluated can then check the generated feedback using a dedicated feedback confirmation application on their device.

[0403] This will enable multifaceted, emotional feedback to be provided to robots in factories, which is expected to improve work efficiency and performance.

[0404] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0405] Step 1:

[0406] Comment input from user terminal:

[0407] A user (such as a factory worker) uses a device (such as a smartphone or head-mounted display) to input a comment about the work of the person being evaluated (such as a robot). The comment is sent to a feedback input application on the device. Example: "Robot A carried the pallet smoothly today."

[0408] Input: A short comment entered by the user

[0409] Output: Comment data sent from the device to the server

[0410] Step 2:

[0411] Comment data storage on server:

[0412] The submitted comment data is sent to a server and stored in a database. The server organizes the data by adding metadata to each comment, such as the user's ID, comment content, and timestamp.

[0413] Input: Comment data sent from the device

[0414] Output: Comment data stored in the database

[0415] Step 3:

[0416] On-server sentiment analysis:

[0417] The comment data stored on the server is fed to the emotion engine, which analyzes the content of the comment and classifies it as positive, negative, or neutral. For example, "Robot A handled the situation smoothly" is classified as positive.

[0418] Input: Comment data stored in the database

[0419] Output: Emotion classification result

[0420] Step 4:

[0421] Training generative AI models:

[0422] The server supplies the comment data and its emotion classification results to the generative AI model for learning. The generative AI model reflects the emotion classification results and updates the feedback model for each person being evaluated, taking into account the comment content, the relationship between the evaluator and the person being evaluated, the emotion and frequency of the evaluation, etc.

[0423] Input: Comment data and sentiment classification results

[0424] Output: Updated generative AI model

[0425] Step 5:

[0426] Generate feedback:

[0427] At the end of the period, the server uses the generative AI model to generate feedback for each assessee, such as "Robot A is recognized for its smooth movements in transporting pallets, and has improved efficiency," which reflects both multifaceted and emotional feedback.

[0428] Input: Updated generative AI model

[0429] Output: Generated feedback

[0430] Step 6:

[0431] Feedback Notification:

[0432] The generated feedback content is stored in a database and is also sent from the server to the device of the person being evaluated, who can then check the generated feedback content using a dedicated feedback checking application on their device.

[0433] Input: Generated feedback

[0434] Output: Feedback notification to the device of the person being evaluated

[0435] 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.

[0436] 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.

[0437] 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.

[0438] [Second embodiment]

[0439] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0440] 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.

[0441] 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).

[0442] 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.

[0443] 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.

[0444] 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).

[0445] 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.

[0446] 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.

[0447] 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.

[0448] 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.

[0449] 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.

[0450] 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."

[0451] In order to realize a multifaceted feedback system, the present invention collects short comments about the person being evaluated on a daily basis, analyzes them using a generative AI model, and generates and notifies appropriate feedback content at the end of the period. Specific embodiments of the system are described below.

[0452] System configuration

[0453] This system consists of three main components: a user terminal, a server, and a generative AI model.

[0454] User terminal

[0455] The user terminal is a device that allows the evaluator to input short comments about the person being evaluated based on their impressions during their daily work, and can include PCs, smartphones, tablets, etc. A dedicated feedback input application is installed on the user terminal. Using this application, the user (evaluator) can input concise comments about the person being evaluated.

[0456] server

[0457] The server has the following functions:

[0458] 1. Database function: A function that receives comment data entered by users on a daily basis and stores it in a database as structured data.

[0459] 2. Managing generative AI models: A function that allows you to feed saved comment data to generative AI models for learning.

[0460] 3. Feedback generation: A function that uses a generative AI model to generate feedback for each person being evaluated at the end of the period.

[0461] 4. Notification function: A function to notify the generated feedback to the device of the person being evaluated.

[0462] Generative AI Models

[0463] The generative AI model uses natural language processing technology to analyze the saved one-word comment data and generate appropriate feedback for each person being evaluated. This model provides comprehensive feedback by taking into account the relationship between the evaluator and the person being evaluated, the content, frequency, and emotion of the comment, etc.

[0464] Explaining the feedback generation process

[0465] The following describes how this system works, using specific examples.

[0466] Comment input

[0467] A user (e.g., colleague A) uses a terminal to input a short comment about the person being evaluated (e.g., colleague B) into a feedback input application. For example, the user might leave a comment in the form of, "Mr. B made a great proposal in today's meeting."

[0468] Saving comment data

[0469] The comment data sent from the device is sent to a server, which stores the received comment data in a database and organizes it by adding metadata to each comment, such as the ID of the person being evaluated, the comment content, and a timestamp.

[0470] Data training

[0471] The server periodically supplies the saved comment data to the AI ​​model for learning. The AI ​​model updates the feedback model for each person being evaluated, taking into account the content of the comments, the relationship between the evaluator and the person being evaluated, and the frequency of comments.

[0472] Generate feedback

[0473] At the end of the period, the server uses the generative AI model to generate feedback for each person being evaluated, such as, "Mr. B made an important proposal in this project and made a significant contribution to the team. He also showed a proactive attitude toward seeking feedback."

[0474] Feedback Notification

[0475] The generated feedback is stored in a database and sent from the server to the device of the person being evaluated. The person being evaluated can then check the generated feedback using a dedicated feedback confirmation application on their device.

[0476] This allows the person being evaluated to receive multifaceted and detailed feedback that can be used to improve themselves. It also makes it easier for the evaluator to enter comments, freeing them from having to rely on memory or input constraints, allowing them to provide high-quality feedback.

[0477] The processing flow will be explained below.

[0478] Program processing flow

[0479] Step 1:

[0480] Users can input short comments about the person being evaluated into a dedicated feedback input application during work or daily conversations. For example, colleague A might input, "Mr. B made a very useful suggestion in today's meeting."

[0481] Step 2:

[0482] The device organizes the input comment data as structured data and sends it to the server, adding metadata such as the user's ID, comment content, and timestamp.

[0483] Step 3:

[0484] The server stores the received comment data in a database. For example, the data is stored in the format of "evaluator ID: B123, comment content: 'Very useful suggestion', timestamp: 2023-10-01 10:30".

[0485] Step 4:

[0486] Each time a new comment is added, the server feeds the data to the generative AI model, which performs an initial learning process, such as cleaning the text (correcting unnecessary characters and typos) and categorizing the comment.

[0487] Step 5:

[0488] The generative AI model learns from the provided data, using natural language processing technology to analyze the content of comments, the relationship between the evaluator and the person being rated, the sentiment of the ratings, frequency, etc.

[0489] Step 6:

[0490] The server sets a schedule to automatically supply additional data to the AI ​​model at regular intervals (e.g., one week) to continuously update and train the model.

[0491] Step 7:

[0492] As the end of the term approaches, the server issues an instruction to re-feed all comment data to the generative AI model. For example, it will feed all data one week before the end of the term.

[0493] Step 8:

[0494] A generative AI model uses all stored comment data to generate appropriate feedback for each ratee, balancing positive and negative feedback.

[0495] Step 9:

[0496] The server stores the generated feedback content in a database and simultaneously sends a notification to the terminal of the person being evaluated that the feedback has been generated.

[0497] Step 10:

[0498] The device receives a notification, and the person being evaluated can check the generated feedback through a dedicated feedback confirmation application. Specifically, when the person being evaluated opens the application, they can see feedback such as, "Mr. B made important suggestions in this project and made a significant contribution to the team. He also showed an attitude of proactively seeking feedback."

[0499] This allows the person being evaluated to receive detailed feedback from all angles, which can be used to improve themselves. In addition, the evaluator can enter detailed comments when necessary, freeing them from the constraints of relying on memory or a limited input period.

[0500] Example 1

[0501] 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."

[0502] Current feedback systems have the drawback of making it difficult for evaluators to provide timely and detailed feedback on the person being evaluated. Because the feedback relies on the evaluator's subjectivity, it lacks consistency and objectivity, making it difficult to effectively support the growth of the person being evaluated. Another issue is that the cumbersome process of inputting and managing feedback places an excessive burden on the evaluator.

[0503] 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.

[0504] In this invention, the server includes: a means for a user to input a one-line comment about the person to be evaluated; a means for a terminal to transmit the input one-line comment to the server; a means for the server to save the received one-line comment data in a database; a means for the server to periodically supply the saved one-line comment data to a generative AI model for learning; a means for the server to generate feedback content for each person to be evaluated using the generative AI model at the end of the period; and a means for the server to notify the terminal of the person to be evaluated of the generated feedback content. This makes it possible to provide consistent and objective feedback to the person to be evaluated and enables the evaluator to easily input and manage comments.

[0505] A "user" is a person who uses the system to input a short comment about the person being evaluated.

[0506] A "terminal" is a device used by a user, and includes a personal computer, a smartphone, a tablet, and the like.

[0507] "Server" refers to a device that stores comment data, manages generative AI models, and generates and notifies feedback.

[0508] A "shorthand comment" is a short textual evaluation content entered by a user about the person being evaluated.

[0509] The "database" is a system that stores received short comment data and generated feedback content as structured data.

[0510] The "generative AI model" is an algorithm that learns from saved one-line comment data and generates feedback content using natural language processing technology.

[0511] "Feedback content" refers to the content of the evaluation and advice given to the person being evaluated, generated by the generative AI model.

[0512] "Notification" is the process of transmitting the generated feedback content to the terminal of the person being evaluated.

[0513] "End of period" refers to the end of the cycle for generating and providing feedback.

[0514] A "prompt sentence" is an input sentence used to have a generative AI model generate feedback content.

[0515] In order to realize a multifaceted feedback system, this invention collects short comments about the person being evaluated on a daily basis, analyzes them using a generative AI model, and generates and notifies appropriate feedback content at the end of the period.

[0516] System configuration

[0517] This system consists of three main components: the user terminal, the server, and the generative AI model.

[0518] User terminal

[0519] The user terminal is a device that allows the evaluator to input short comments about the person being evaluated based on their impressions during their daily work, and can include a personal computer, smartphone, tablet, etc. A dedicated feedback input application is installed on the user terminal. Using this application, the user (evaluator) can input concise comments about the person being evaluated.

[0520] Examples:

[0521] A user (e.g., colleague A) uses a terminal to input a short comment about the person being evaluated (e.g., colleague B) into a feedback input application. For example, the user might leave a comment in the form of, "Mr. B made a great proposal in today's meeting."

[0522] server

[0523] The server has the following functions:

[0524] 1. Database function: A function that receives comment data entered by users on a daily basis and stores it in a database as structured data.

[0525] 2. Managing generative AI models: A function that allows you to feed saved comment data to generative AI models for learning.

[0526] 3. Feedback generation: A function that uses a generative AI model to generate feedback for each person being evaluated at the end of the period.

[0527] 4. Notification function: A function to notify the generated feedback to the device of the person being evaluated.

[0528] Examples:

[0529] Comment data sent from the device is sent to a server, which stores the received comment data in a database and organizes it by adding metadata to each comment, such as the user's ID, comment content, and timestamp.

[0530] Generative AI Models

[0531] The generative AI model uses natural language processing technology to analyze the saved one-word comment data and generate appropriate feedback for each person being evaluated. This model provides comprehensive feedback by taking into account the relationship between the evaluator and the person being evaluated, the content, frequency, and emotion of the comment, etc.

[0532] Examples:

[0533] At the end of the period, the server uses the generative AI model to generate feedback for each person being evaluated, such as, "Mr. B made an important proposal in this project and made a significant contribution to the team. He also showed a proactive attitude toward seeking feedback."

[0534] Feedback Generation Process

[0535] A specific example of the feedback generation process is as follows:

[0536] Learning from comment data

[0537] The server periodically supplies the comment data stored in the database to the generative AI model, which then learns from it, taking into account the content of the comment, the relationship between the evaluator and the person being evaluated, and the frequency of the comment.

[0538] Example prompt sentence:

[0539] Generate feedback for Subject B based on the following comments:

[0540] 1. Person B asked insightful questions during team meetings, which sparked deeper discussions.

[0541] 2. Mr. B's new idea was very original and caught everyone's attention.

[0542] Feedback content:

[0543] Feedback Notification

[0544] The server stores the generated feedback in a database and notifies the evaluation recipient's device, where the evaluation recipient can check the generated feedback using a dedicated feedback checking application.

[0545] This allows the person being evaluated to receive multifaceted and detailed feedback that can be used to improve themselves. It also makes it easier for the evaluator to enter comments, freeing them from having to rely on memory or input constraints, allowing for the provision of high-quality feedback.

[0546] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0547] Step 1:

[0548] Enter a comment

[0549] The user inputs a comment about the person being evaluated.

[0550] Input: Comment text entered by the evaluator through the terminal.

[0551] Specific operation: The user launches a dedicated feedback input application and enters a comment about the person being evaluated, such as "Mr. B made a great proposal in today's meeting." The terminal has a "Send" button, which the user clicks.

[0552] Output: The comment text entered is ready to be sent.

[0553] Step 2:

[0554] Submitting comment data

[0555] The terminal transmits the input comment data to the server.

[0556] Input: The comment text that you have completed.

[0557] Specific operation: The device sends the comment text as an HTTP request to a specific API endpoint (e.g., POST / comments) on the server.

[0558] Output: The comment data sent to the server.

[0559] Step 3:

[0560] Saving comment data

[0561] The server stores the received comment data in a database.

[0562] Input: The comment text received by the server.

[0563] Specific operation: The server receives the HTTP request, adds metadata such as the rater's ID, comment content, and timestamp, and structures the data. It then generates and executes an INSERT query in a database (e.g., MySQL) to save the comment data.

[0564] Output: Comment data stored in the database.

[0565] Step 4:

[0566] Data training

[0567] The server periodically supplies the saved comment data to the generative AI model for learning.

[0568] Input: Multiple comment data stored in the database.

[0569] How it works: The server periodically retrieves comment data from the database as a scheduled job and supplies it to the API of a generative AI model (e.g., the GPT-3 model). When supplied, it converts it to JSON format and sends it to the model's endpoint. The model learns by taking into account the comment content, relationships, frequency, etc.

[0570] Output: A trained generative AI model.

[0571] Step 5:

[0572] Generate feedback

[0573] At the end of the period, the server uses the generated AI model to generate feedback content for each person being evaluated.

[0574] Input: Trained generative AI model and comment data.

[0575] Specific operation: The server generates a prompt based on the ID of the person being evaluated and sends it to the generation AI model. For example, a prompt like "Generate feedback for person B based on the following comments: 1. Person B asked a sharp question in the team meeting, which sparked a deeper discussion. 2. Person B's new idea was very novel and attracted everyone's attention. Feedback content:" is used. The feedback returned by the model is then processed and organized.

[0576] Output: The generated feedback.

[0577] Step 6:

[0578] Feedback Notification

[0579] The server notifies the generated feedback content to the terminal of the person being evaluated.

[0580] Input: The generated feedback.

[0581] Specific operation: The server saves the feedback content in a database and sends a notification to the assessee's device using a notification service (e.g., Firebase Cloud Messaging). The application on the assessee's device receives the notification and prompts the user to confirm.

[0582] Output: Feedback provided and confirmation by the assessee.

[0583] (Application example 1)

[0584] 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."

[0585] It is important to routinely evaluate the operating status and performance of automated equipment operating on-site and provide appropriate feedback. However, in many production sites, such evaluations are often dependent on a few managers, resulting in subjective and inconsistent evaluations. Furthermore, it has been difficult to quickly and effectively monitor the status of automated equipment during daily operations and detect problems early. Furthermore, delays in such evaluations and feedback can result in equipment deterioration and reduced production efficiency.

[0586] 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.

[0587] In this invention, the server includes means for inputting a short comment about the person being evaluated by a person involved in daily work, means for saving the input short comment in a database, means for generating feedback content using a generative AI model that learns the saved short comment, means for notifying the person being evaluated of the generated feedback content at the end of the period, means for inputting comments about the operating status and performance of automated operation equipment operating in a factory, and means for evaluating the performance of the automated operation equipment based on the input comments. This enables daily feedback about the operating status and performance of the automated operation equipment in a factory, makes evaluations consistent and objective, and enables efficient operation of the automated operation equipment and early problem detection.

[0588] "Evaluation target" refers to the person who is evaluated in the feedback system, and refers to automated work equipment and robots operating in a factory.

[0589] A "shorthand comment" is a short written opinion or observation about the operating status or performance of the subject of evaluation.

[0590] The "database" is a system for storing and managing input short comments as structured data.

[0591] A "generative AI model" is an artificial intelligence model that learns from input one-line comments and generates feedback content based on them.

[0592] "End of period" refers to the end of a specific time or period when feedback is communicated to the assessee.

[0593] "Notification means" refers to a method or device for notifying the generated feedback content to the person being evaluated.

[0594] "Automated work equipment" refers to machines or robots used to automatically perform specific tasks in a factory.

[0595] "Performance evaluation" refers to the task of evaluating the operating status and performance of the person being evaluated.

[0596] This system evaluates the operational status and performance of automated work equipment in a factory on a daily basis and provides appropriate feedback. This system consists of three main components: a user terminal, a server, and a generative AI model.

[0597] System configuration

[0598] User terminal

[0599] The user terminal is a device that allows the evaluator to input a brief comment about the automated operation equipment operating in the factory, and can include PCs, smartphones, tablets, etc. A dedicated feedback input application is installed on the user terminal. Using this application, the user (evaluator) can input a brief comment about the evaluation target (automated operation equipment).

[0600] server

[0601] The server has the following functions:

[0602] 1. Database function: A function that receives comment data entered by users and stores it in a database as structured data.

[0603] 2. Managing generative AI models: A function that allows you to feed saved comment data to generative AI models for learning.

[0604] 3. Feedback generation: A function that uses a generative AI model to generate feedback for each person being evaluated at the end of the period.

[0605] 4. Notification function: A function to notify the generated feedback to the device of the person being evaluated.

[0606] Generative AI Models

[0607] The generative AI model uses natural language processing technology to analyze the saved one-line comment data and generate appropriate feedback for each person being evaluated. This model provides comprehensive feedback by taking into account the content, frequency, importance, etc. of the comment.

[0608] Feedback Generation Process

[0609] The following describes how this system works, using specific examples.

[0610] Comment input

[0611] A user (e.g., Manager A) uses a terminal to input a brief comment about an automated work device (e.g., Robot A) in a factory into a feedback input application. For example, the user leaves a comment in the form of "Robot A's work speed seemed slow today."

[0612] Saving comment data

[0613] The comment data sent from the device is sent to a server, which stores the received comment data in a database and organizes it by adding metadata to each comment, such as the automated work device's ID, comment content, and timestamp.

[0614] Data training

[0615] The server periodically supplies the saved comment data to the AI ​​model for learning. The AI ​​model updates the feedback model for each person being evaluated, taking into account the content of the comments, the relationship between the evaluator and the automated work device, and the frequency of comments.

[0616] Specific examples

[0617] Examples of comment input:

[0618] Automated Work Device ID: RobotA

[0619] Comment: "You seem to be working a lot slower today."

[0620] Example prompt sentence:

[0621] Generate feedback for each robot based on the comment data below:

[0622] Robot ID: RobotA, Comment: "You seem to be working a lot slower today."

[0623] Examples of feedback generated include:

[0624] Robot A requires regular maintenance. It has recently been running slower, possibly due to wear on the belt.

[0625] In this way, users can receive comprehensive feedback about the automated work equipment in their factories, which can be used to improve maintenance and operation.

[0626] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0627] Step 1:

[0628] A user uses a terminal to input a brief comment about an automated work device in a factory into a feedback input application.

[0629] Input: ID of the automated work device, comment (e.g., "Robot A seemed to be working slowly today")

[0630] Output: Comment data is sent from the device to the server.

[0631] Step 2:

[0632] The comment data sent from the terminal is received by the server.

[0633] Input: Comment data sent from the device (automated work device ID, comment content, timestamp)

[0634] Output: Comment data is saved in the database.

[0635] Step 3:

[0636] The server periodically supplies the comment data stored in the database to the generative AI model, allowing it to learn.

[0637] Input: Comment data stored in the database

[0638] Output: The generative AI model learns from the comment data and updates the feedback model.

[0639] Step 4:

[0640] At the end of the term, the server uses a generative AI model to generate feedback content for each person being evaluated.

[0641] Input: Comment data trained by the generative AI model

[0642] Output: End-of-period feedback (e.g., "Robot A needs regular maintenance. It has been running slowly recently, which may be due to wear on the belt.")

[0643] Step 5:

[0644] The generated feedback content is notified from the server to the terminal of the person being evaluated.

[0645] Input: Generated feedback

[0646] Output: A feedback notification is displayed on the assessee's device.

[0647] Step 6:

[0648] The person being evaluated checks the generated feedback content using the terminal.

[0649] Input: Feedback content sent to the device of the person being evaluated

[0650] Output: The person being evaluated will view the feedback and use it as a reference for implementing any necessary improvements or measures.

[0651] 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.

[0652] This system provides multifaceted feedback on the person being evaluated. The system allows people who work with the person on a daily basis to enter short comments about the person being evaluated, which are then stored in a database. A generative AI model then learns from these comments and generates feedback content. Furthermore, by combining this with an emotion engine, the system analyzes the emotions in the comments and provides more accurate feedback.

[0653] System configuration

[0654] The system consists of four main components: a user terminal, a server, a generative AI model, and an emotion engine.

[0655] User terminal

[0656] The user terminal is a device that allows the evaluator to enter short comments about the person being evaluated, and can include PCs, smartphones, tablets, etc. A dedicated feedback input application is installed on the terminal, and a comment input screen is provided.

[0657] server

[0658] The server has the following functions:

[0659] 1. Database function: A function to save comment data received from users in a database.

[0660] 2. Sentiment analysis function: A function that uses an emotion engine to analyze the sentiment of one-line comments and classify them as positive, negative, or neutral.

[0661] 3. Managing generative AI models: A function that allows saved comment data and its sentiment classification results to be fed into generative AI models for learning.

[0662] 4. Feedback generation: A function that generates feedback for each person being evaluated at the end of the period using a generative AI model.

[0663] 5. Notification function: A function to notify the assessee of the generated feedback.

[0664] Generative AI Models

[0665] The generative AI model uses natural language processing technology to analyze the saved one-word comment data and emotion classification results, and generates appropriate feedback content for each person being evaluated. This model reflects the emotional information of the comments received from the emotion engine and provides comprehensive feedback.

[0666] Emotion Engine

[0667] The emotion engine analyzes the sentiment of short comments entered by users and categorizes them as positive, negative, or neutral. The emotion engine uses natural language processing technology to extract sentiment from the text and sends the results to the server.

[0668] Explaining the feedback generation process

[0669] The following describes how this system works, using specific examples.

[0670] Comment input

[0671] A user (e.g., colleague A) uses a terminal to input a short comment about the person being evaluated (e.g., colleague B) into a feedback input application. For example, the user might leave a comment in the form of, "Mr. B made a great proposal in today's meeting."

[0672] Saving comment data

[0673] Comment data sent from the device is sent to a server, which stores the received comment data in a database and organizes it by adding metadata to each comment, such as the ID of the person being evaluated, the comment content, and a timestamp.

[0674] Emotion Analysis

[0675] The server provides the stored comment data to the emotion engine for sentiment analysis. The emotion engine analyzes the content of each comment and classifies it as positive, negative, or neutral. For example, "Mr. B made a great suggestion" is classified as positive.

[0676] Data training

[0677] The server supplies the comment data and its emotion classification results to the generative AI model, which then learns from it. The generative AI model reflects the emotion classification results and updates the feedback model for each person being evaluated, taking into account the comment content, the relationship between the evaluator and the person being evaluated, the emotion and frequency of the evaluation, etc.

[0678] Generate feedback

[0679] At the end of the period, the server uses the generative AI model to generate feedback for each individual being evaluated. For example, it generates multifaceted, emotionally-charged feedback such as, "Mr. B made an important proposal in this project and made a significant contribution to the team. He also showed a proactive attitude toward seeking feedback."

[0680] Feedback Notification

[0681] The generated feedback is stored in a database and sent from the server to the device of the person being evaluated. The person being evaluated can then check the generated feedback using a dedicated feedback confirmation application on their device.

[0682] This allows those being evaluated to receive detailed, accurate feedback from all angles, which can be used to improve themselves. Furthermore, evaluators can enter detailed comments when needed, freeing them from the constraints of relying on memory or time constraints. Furthermore, the introduction of an emotion engine improves the balance and accuracy of feedback, providing a fair and multifaceted evaluation.

[0683] The processing flow will be explained below.

[0684] Processing flow of a feedback system that combines an emotion engine

[0685] Step 1:

[0686] Users can input short comments about the person being evaluated into a dedicated feedback input application during work or daily conversations. For example, colleague A might input, "Mr. B made a very useful suggestion in today's meeting."

[0687] Step 2:

[0688] The device organizes the input comment data as structured data and sends it to the server, adding metadata such as the user's ID, comment content, and timestamp.

[0689] Step 3:

[0690] The server saves the received comment data in the database in the following format: "Evaluator ID: B123, Comment content: 'Very useful suggestion', Timestamp: 2023-10-01 10:30".

[0691] Step 4:

[0692] The server provides the stored comment data to the emotion engine for sentiment analysis. The emotion engine analyzes the content of each comment and classifies it as positive, negative, or neutral. For example, "Mr. B made a great suggestion" would be classified as positive.

[0693] Step 5:

[0694] The server adds the emotion classification results obtained from the emotion engine to the comment data and saves it in the database. For example, "evaluator ID: B123, comment content: 'Very useful suggestion', emotion: positive" is saved.

[0695] Step 6:

[0696] The server periodically supplies all comment data and emotion classification results to the AI ​​model, allowing it to update and learn. For example, you can set it to automatically start AI training on weekends.

[0697] Step 7:

[0698] The generative AI model learns from the provided data, using natural language processing technology to analyze the content of comments, the relationship between the rater and the person being rated, the sentiment classification results, the sentiment of the ratings, frequency, etc.

[0699] Step 8:

[0700] As the end of the term approaches, the server again supplies all comment data and emotion classification results to the generative AI model and instructs it to generate feedback. For example, the server issues a data supply instruction one week before the end of the term.

[0701] Step 9:

[0702] The generative AI model generates appropriate feedback for each person being evaluated based on all stored comment data and emotion classification results. For example, it considers a balance of positive and negative comments and generates feedback such as, "Mr. B made important suggestions in this project and made a significant contribution to the team. He also showed an attitude of actively seeking feedback."

[0703] Step 10:

[0704] The server stores the generated feedback content in a database and simultaneously sends a notification to the terminal of the person being evaluated that the feedback has been generated.

[0705] Step 11:

[0706] The device receives a notification, and the person being evaluated can check the generated feedback through a dedicated feedback review application. Specifically, when the person being evaluated opens the application, a list of the feedback and its details are displayed.

[0707] This allows those being evaluated to receive detailed and accurate feedback from all angles, which can be used to improve their own performance. Furthermore, users (evaluators) can enter detailed comments whenever necessary, freeing them from the constraints of memory and time constraints. Furthermore, the introduction of an emotion engine improves the balance and accuracy of feedback, providing a fair and multifaceted evaluation.

[0708] Example 2

[0709] 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."

[0710] In conventional evaluation systems, evaluators rely on their memory to make evaluations, which can lead to problems with the accuracy and fairness of the evaluations. Furthermore, because the evaluation content is one-sided, it is difficult to grasp the overall picture of the person being evaluated's performance and behavior. Furthermore, evaluations are not based on emotions, which can lead to insufficient consideration of the evaluation content.

[0711] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0712] In this invention, the server includes: a means for people involved in daily work to input short comments about the person being evaluated; a means for saving the input short comments in a database; a means for analyzing the sentiment of the saved short comments and classifying them as positive, negative, or neutral; a means for generating feedback content using a generative AI model that learns short comments including the sentiment analysis results; and a means for notifying the person being evaluated of the generated feedback content at the end of the period. This improves the accuracy and fairness of the evaluation content and enables multifaceted evaluation of the person's performance and behavior. Furthermore, incorporating sentiment analysis allows the evaluation content to reflect emotional aspects.

[0713] A "shorthand comment" is a short piece of feedback entered by the evaluator about the person being evaluated during their daily work.

[0714] "Database" means an information storage system for storing and managing entered short comments and other related data.

[0715] "Sentiment analysis" is a technology that uses natural language processing technology to analyze the emotions contained in short comments and classify them as positive, negative, or neutral.

[0716] The "generative AI model" is an artificial intelligence model that learns from saved one-line comments and the results of their emotional analysis, and generates feedback content for each person being evaluated.

[0717] "Notification" is a system function for notifying the subject of evaluation of the generated feedback content.

[0718] The present invention is a system that provides multifaceted feedback to an individual by inputting a short comment about the individual involved in daily work, storing the comment in a database, performing sentiment analysis, generating feedback using a generative AI model, and notifying the individual. Each component of this system and its embodiment will be described in detail below.

[0719] System configuration

[0720] The system consists of four main components: a user terminal, a server, a generative AI model, and a sentiment analysis engine.

[0721] User terminal

[0722] The user terminal is a device on which the evaluator inputs a short comment about the person being evaluated. It can be a PC, smartphone, tablet, etc. A dedicated feedback input application is installed on the user terminal, and a comment input screen is provided. For example, the evaluator might input a comment in the form of "Mr. B made a great proposal in today's meeting."

[0723] server

[0724] The server has the following features:

[0725] 1. Database function: Comment data received from users is saved in a database. The database uses MySQL or PostgreSQL.

[0726] 2. Sentiment analysis function: Using an emotion engine, the sentiment of one-line comments is classified as positive, negative, or neutral. Sentiment analysis is performed using Python's VADER and TextBlob.

[0727] 3. Managing the generative AI model: The saved comment data and its sentiment classification results are fed into the generative AI model for training. The generative AI model uses a Transformer-based model such as GPT-3 or BERT.

[0728] 4. Feedback generation: At the end of the period, feedback content is generated for each person being evaluated using the generative AI model.

[0729] 5. Notification function: Notify the assessee of the generated feedback.

[0730] Generative AI Models

[0731] The generative AI model uses natural language processing technology to analyze the saved one-line comment data and sentiment analysis results, generating appropriate feedback for each individual being evaluated. The generative AI model is implemented using Python or other programming languages, utilizing machine learning frameworks such as TensorFlow and PyTorch. For example, it might generate feedback such as, "Mr. B made important suggestions in this project and made a significant contribution to the team. He also showed a proactive attitude toward seeking feedback."

[0732] Sentiment Analysis Engine

[0733] The sentiment analysis engine analyzes the sentiment of short comments entered by users and classifies them as positive, negative, or neutral. The sentiment analysis engine uses natural language processing technology to extract the sentiment of the text and sends the results to a server. Technologies used include VADER and TextBlob.

[0734] Examples of concrete examples and prompts

[0735] For example, if a user (colleague A) enters a comment such as "Mr. B made a great proposal in today's meeting," the server receives this comment and stores it in a database. It is then classified as "positive" using a sentiment analysis engine. The generative AI model generates feedback for the person being evaluated (colleague B) based on this data.

[0736] Examples:

[0737] Comment: "Mr. B made a great suggestion at today's meeting."

[0738] Sentiment Analysis: Positive

[0739] Feedback: "B made important suggestions on this project and contributed greatly to the team. He also proactively sought feedback."

[0740] Example prompt for a generative AI model:

[0741] "Generate the following feedback: 'Person B made a great suggestion in today's meeting.'"

[0742] In this way, the system according to the present invention can provide multifaceted and accurate feedback regarding the person being evaluated.

[0743] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0744] Step 1:

[0745] Comments entered by the user

[0746] Operation: The user opens a feedback application on a device (PC, smartphone, tablet, etc.). They access the comment input screen and enter a short comment about the person being evaluated. For example, they enter a comment such as, "Mr. B made a great proposal in today's meeting."

[0747] Input: Text comments about the person being assessed.

[0748] Output: The entered comment data is sent from the terminal to the server.

[0749] Step 2:

[0750] Sending and saving comment data to the server

[0751] Operation: Comment data sent from the device is sent to the server via an HTTP request. The server analyzes the received comment data, extracts the ID of the person being rated, the ID of the evaluator, the comment content, and the timestamp, and stores them in a database.

[0752] Input: Comment data sent from the device (text comment, ID of the person being evaluated, ID of the evaluator, timestamp).

[0753] Output: Comment data stored in a database.

[0754] Step 3:

[0755] Performing sentiment analysis

[0756] How it works: The server passes the comment data stored in the database to a sentiment analysis engine, which uses natural language processing libraries such as VADER or TextBlob to classify the sentiment of each comment as positive, negative, or neutral.

[0757] Input: Text comment stored in the database.

[0758] Output: Comment data with sentiment classification results (positive, negative, neutral).

[0759] Step 4:

[0760] Supplying and training generative AI models

[0761] How it works: The server supplies the sentiment-analyzed comment data to the generative AI model, which then uses natural language processing technology to update the feedback model for each subject based on the data provided.

[0762] Input: Comment data with sentiment classification results.

[0763] Output: An updated generative AI model.

[0764] Step 5:

[0765] Generate feedback

[0766] How it works: At the end of the term, the server uses the updated generative AI model to generate feedback for each person being evaluated. For example, it generates feedback in the form of, "Mr. B made important suggestions in this project and made a significant contribution to the team. He also showed a proactive attitude of seeking feedback."

[0767] Input: Updated generative AI model, ID of the person being evaluated.

[0768] Output: Generated feedback for each ratee.

[0769] Step 6:

[0770] Feedback Notification

[0771] Operation: The server notifies the subject of the generated feedback to the subject's device. The subject then checks the received feedback using a dedicated feedback checking application.

[0772] Input: Generated feedback content, ID of the person being assessed.

[0773] Output: Feedback sent to the subject's device.

[0774] Through these steps, users can provide multifaceted and accurate feedback to those being evaluated. Raters can easily enter comments, and sentiment analysis and generative AI models ensure detailed and fair feedback.

[0775] (Application example 2)

[0776] 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."

[0777] In conventional feedback systems, evaluations are often based on the subjective opinions of the evaluator or one-sided information, making it difficult to accurately grasp the overall growth and areas for improvement of the person being evaluated.Furthermore, because the emotional aspects of the comments are not taken into consideration, there are problems with biased feedback content and unfair evaluations.

[0778] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a means for a person who is involved in daily work to input a short comment about the person to be evaluated, a means for saving the input short comment in a database, a means for generating feedback content using a generative AI model that learns from the saved short comment, a means for analyzing the input comment with an emotion engine and classifying emotions, and a means for notifying the person to be evaluated of the generated feedback content at the end of the period. This enables fair feedback that is multifaceted and reflects emotions.

[0779] "People involved in daily work" refers to colleagues, superiors, subordinates, and other people who work with the person being evaluated on a daily basis.

[0780] "Evaluation target" refers to the person, employee, worker, etc. who is the subject of feedback.

[0781] A "shorthand comment" is a short feedback message about the person being appraised that describes a behavior or performance observed during work.

[0782] "Means of storing in a database" refers to the systems and technologies for systematically storing and managing the entered comment data.

[0783] A "generative AI model" refers to an artificial intelligence module that learns based on input data and generates feedback content.

[0784] "Means for generating feedback content" refers to the process of analyzing collected data using a generative AI model and creating appropriate feedback for each person being evaluated.

[0785] An "emotion engine" refers to technology that analyzes the emotional aspects of input comments and classifies them as positive, negative, neutral, etc.

[0786] "Sentiment classification" refers to the process of identifying and categorizing the emotional aspects of comments using an emotion engine.

[0787] "Means of notifying at the end of the period" refers to the mechanisms and techniques for communicating the generated feedback content to the person being evaluated at the end of the evaluation period.

[0788] The present invention is a system in which people who interact with the person being evaluated in their daily work enter short comments about the person being evaluated, save them in a database, and a generative AI model learns from them to generate feedback content. The system is configured as follows:

[0789] Hardware and Software

[0790] 1. User device: A device used by the user to enter a short comment about the person being evaluated. This can be a PC, smartphone, head-mounted display (HMD), etc. A dedicated feedback input application is installed and a comment input screen is provided.

[0791] 2. Server:

[0792] Database function: A function to save comment data received from users in a database.

[0793] Sentiment analysis function: A function that uses an emotion engine to analyze the sentiment of short comments and classify them as positive, negative, or neutral.

[0794] Managing generative AI models: A function that allows saved comment data and its sentiment classification results to be fed into generative AI models for training.

[0795] Feedback generation: A function that uses a generative AI model to generate feedback for each assessee.

[0796] Notification function: A function to notify the assessee of the generated feedback at the end of the period.

[0797] Processing flow and data calculation

[0798] 1. Comment entry:

[0799] A user (e.g., a factory worker) uses a terminal to input a short comment about the work of the person being evaluated (e.g., a robot) into a feedback input application. For example, the user might leave a comment in the form of, "Robot A carried the pallet smoothly today."

[0800] Example prompt: "In one sentence, tell us what you noticed about the robot's work."

[0801] 2. Storage of Comment Data:

[0802] Comment data sent from the device is sent to a server, which stores the received comment data in a database and organizes it by adding metadata to each comment, such as the ID of the person being evaluated, the comment content, and a timestamp.

[0803] 3. Emotion analysis:

[0804] The server provides the stored comment data to the emotion engine for emotion analysis. The emotion engine analyzes the content of each comment and classifies it as positive, negative, or neutral. For example, "Robot A handled the situation smoothly" is classified as positive.

[0805] Example prompt: "Analyze the sentiment of this comment and categorize it as positive, negative, or neutral."

[0806] 4. Data training:

[0807] The server supplies the comment data and its emotion classification results to the generative AI model, which then learns from it. The generative AI model reflects the emotion classification results and updates the feedback model for each person being evaluated, taking into account the comment content, the relationship between the evaluator and the person being evaluated, the emotion and frequency of the evaluation, etc.

[0808] 5. Feedback Generation:

[0809] At the end of the period, the server uses the generative AI model to generate feedback for each individual being evaluated, generating multifaceted and emotional feedback such as, "Robot A was recognized for its smooth movements in transporting pallets, and its efficiency has improved."

[0810] Example prompt: "Based on this sentiment analysis and comment data, generate feedback to improve the robot's performance."

[0811] 6. Feedback Notification:

[0812] The generated feedback is stored in a database and sent from the server to the device of the person being evaluated. The person being evaluated can then check the generated feedback using a dedicated feedback confirmation application on their device.

[0813] This will enable multifaceted, emotional feedback to be provided to robots in factories, which is expected to improve work efficiency and performance.

[0814] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0815] Step 1:

[0816] Comment input from user terminal:

[0817] A user (such as a factory worker) uses a device (such as a smartphone or head-mounted display) to input a comment about the work of the person being evaluated (such as a robot). The comment is sent to a feedback input application on the device. Example: "Robot A carried the pallet smoothly today."

[0818] Input: A short comment entered by the user

[0819] Output: Comment data sent from the device to the server

[0820] Step 2:

[0821] Comment data storage on server:

[0822] The submitted comment data is sent to a server and stored in a database. The server organizes the data by adding metadata to each comment, such as the user's ID, comment content, and timestamp.

[0823] Input: Comment data sent from the device

[0824] Output: Comment data stored in the database

[0825] Step 3:

[0826] On-server sentiment analysis:

[0827] The comment data stored on the server is fed to the emotion engine, which analyzes the content of the comment and classifies it as positive, negative, or neutral. For example, "Robot A handled the situation smoothly" is classified as positive.

[0828] Input: Comment data stored in the database

[0829] Output: Emotion classification result

[0830] Step 4:

[0831] Training generative AI models:

[0832] The server supplies the comment data and its emotion classification results to the generative AI model for learning. The generative AI model reflects the emotion classification results and updates the feedback model for each person being evaluated, taking into account the comment content, the relationship between the evaluator and the person being evaluated, the emotion and frequency of the evaluation, etc.

[0833] Input: Comment data and sentiment classification results

[0834] Output: Updated generative AI model

[0835] Step 5:

[0836] Generate feedback:

[0837] At the end of the period, the server uses the generative AI model to generate feedback for each assessee, such as "Robot A is recognized for its smooth movements in transporting pallets, and has improved efficiency," which reflects both multifaceted and emotional feedback.

[0838] Input: Updated generative AI model

[0839] Output: Generated feedback

[0840] Step 6:

[0841] Feedback Notification:

[0842] The generated feedback content is stored in a database and is also sent from the server to the device of the person being evaluated, who can then check the generated feedback content using a dedicated feedback checking application on their device.

[0843] Input: Generated feedback

[0844] Output: Feedback notification to the device of the person being evaluated

[0845] 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.

[0846] 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.

[0847] 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.

[0848] [Third embodiment]

[0849] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0850] 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.

[0851] 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).

[0852] 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.

[0853] 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.

[0854] 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).

[0855] 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.

[0856] 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.

[0857] 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.

[0858] 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.

[0859] 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.

[0860] 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."

[0861] In order to realize a multifaceted feedback system, the present invention collects short comments about the person being evaluated on a daily basis, analyzes them using a generative AI model, and generates and notifies appropriate feedback content at the end of the period. Specific embodiments of the system are described below.

[0862] System configuration

[0863] This system consists of three main components: a user terminal, a server, and a generative AI model.

[0864] User terminal

[0865] The user terminal is a device that allows the evaluator to input short comments about the person being evaluated based on their impressions during their daily work, and can include PCs, smartphones, tablets, etc. A dedicated feedback input application is installed on the user terminal. Using this application, the user (evaluator) can input concise comments about the person being evaluated.

[0866] server

[0867] The server has the following functions:

[0868] 1. Database function: A function that receives comment data entered by users on a daily basis and stores it in a database as structured data.

[0869] 2. Managing generative AI models: A function that allows you to feed saved comment data to generative AI models for learning.

[0870] 3. Feedback generation: A function that uses a generative AI model to generate feedback for each person being evaluated at the end of the period.

[0871] 4. Notification function: A function to notify the generated feedback to the device of the person being evaluated.

[0872] Generative AI Models

[0873] The generative AI model uses natural language processing technology to analyze the saved one-word comment data and generate appropriate feedback for each person being evaluated. This model provides comprehensive feedback by taking into account the relationship between the evaluator and the person being evaluated, the content, frequency, and emotion of the comment, etc.

[0874] Explaining the feedback generation process

[0875] The following describes how this system works, using specific examples.

[0876] Comment input

[0877] A user (e.g., colleague A) uses a terminal to input a short comment about the person being evaluated (e.g., colleague B) into a feedback input application. For example, the user might leave a comment in the form of, "Mr. B made a great proposal in today's meeting."

[0878] Saving comment data

[0879] The comment data sent from the device is sent to a server, which stores the received comment data in a database and organizes it by adding metadata to each comment, such as the ID of the person being evaluated, the comment content, and a timestamp.

[0880] Data training

[0881] The server periodically supplies the saved comment data to the AI ​​model for learning. The AI ​​model updates the feedback model for each person being evaluated, taking into account the content of the comments, the relationship between the evaluator and the person being evaluated, and the frequency of comments.

[0882] Generate feedback

[0883] At the end of the period, the server uses the generative AI model to generate feedback for each person being evaluated, such as, "Mr. B made an important proposal in this project and made a significant contribution to the team. He also showed a proactive attitude toward seeking feedback."

[0884] Feedback Notification

[0885] The generated feedback is stored in a database and sent from the server to the device of the person being evaluated. The person being evaluated can then check the generated feedback using a dedicated feedback confirmation application on their device.

[0886] This allows the person being evaluated to receive multifaceted and detailed feedback that can be used to improve themselves. It also makes it easier for the evaluator to enter comments, freeing them from having to rely on memory or input constraints, allowing them to provide high-quality feedback.

[0887] The processing flow will be explained below.

[0888] Program processing flow

[0889] Step 1:

[0890] Users can input short comments about the person being evaluated into a dedicated feedback input application during work or daily conversations. For example, colleague A might input, "Mr. B made a very useful suggestion in today's meeting."

[0891] Step 2:

[0892] The device organizes the input comment data as structured data and sends it to the server, adding metadata such as the user's ID, comment content, and timestamp.

[0893] Step 3:

[0894] The server stores the received comment data in a database. For example, the data is stored in the format of "evaluator ID: B123, comment content: 'Very useful suggestion', timestamp: 2023-10-01 10:30".

[0895] Step 4:

[0896] Each time a new comment is added, the server feeds the data to the generative AI model, which performs an initial learning process, such as cleaning the text (correcting unnecessary characters and typos) and categorizing the comment.

[0897] Step 5:

[0898] The generative AI model learns from the provided data, using natural language processing technology to analyze the content of comments, the relationship between the evaluator and the person being rated, the sentiment of the ratings, frequency, etc.

[0899] Step 6:

[0900] The server sets a schedule to automatically supply additional data to the AI ​​model at regular intervals (e.g., one week) to continuously update and train the model.

[0901] Step 7:

[0902] As the end of the term approaches, the server issues an instruction to re-feed all comment data to the generative AI model. For example, it will feed all data one week before the end of the term.

[0903] Step 8:

[0904] A generative AI model uses all stored comment data to generate appropriate feedback for each ratee, balancing positive and negative feedback.

[0905] Step 9:

[0906] The server stores the generated feedback content in a database and simultaneously sends a notification to the terminal of the person being evaluated that the feedback has been generated.

[0907] Step 10:

[0908] The device receives a notification, and the person being evaluated can check the generated feedback through a dedicated feedback confirmation application. Specifically, when the person being evaluated opens the application, they can see feedback such as, "Mr. B made important suggestions in this project and made a significant contribution to the team. He also showed an attitude of proactively seeking feedback."

[0909] This allows the person being evaluated to receive detailed feedback from all angles, which can be used to improve themselves. In addition, the evaluator can enter detailed comments when necessary, freeing them from the constraints of relying on memory or a limited input period.

[0910] Example 1

[0911] 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."

[0912] Current feedback systems have the drawback of making it difficult for evaluators to provide timely and detailed feedback on the person being evaluated. Because the feedback relies on the evaluator's subjectivity, it lacks consistency and objectivity, making it difficult to effectively support the growth of the person being evaluated. Another issue is that the cumbersome process of inputting and managing feedback places an excessive burden on the evaluator.

[0913] 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.

[0914] In this invention, the server includes: a means for a user to input a one-line comment about the person to be evaluated; a means for a terminal to transmit the input one-line comment to the server; a means for the server to save the received one-line comment data in a database; a means for the server to periodically supply the saved one-line comment data to a generative AI model for learning; a means for the server to generate feedback content for each person to be evaluated using the generative AI model at the end of the period; and a means for the server to notify the terminal of the person to be evaluated of the generated feedback content. This makes it possible to provide consistent and objective feedback to the person to be evaluated and enables the evaluator to easily input and manage comments.

[0915] A "user" is a person who uses the system to input a short comment about the person being evaluated.

[0916] A "terminal" is a device used by a user, and includes a personal computer, a smartphone, a tablet, and the like.

[0917] "Server" refers to a device that stores comment data, manages generative AI models, and generates and notifies feedback.

[0918] A "shorthand comment" is a short textual evaluation content entered by a user about the person being evaluated.

[0919] The "database" is a system that stores received short comment data and generated feedback content as structured data.

[0920] The "generative AI model" is an algorithm that learns from saved one-line comment data and generates feedback content using natural language processing technology.

[0921] "Feedback content" refers to the content of the evaluation and advice given to the person being evaluated, generated by the generative AI model.

[0922] "Notification" is the process of transmitting the generated feedback content to the terminal of the person being evaluated.

[0923] "End of period" refers to the end of the cycle for generating and providing feedback.

[0924] A "prompt sentence" is an input sentence used to have a generative AI model generate feedback content.

[0925] In order to realize a multifaceted feedback system, this invention collects short comments about the person being evaluated on a daily basis, analyzes them using a generative AI model, and generates and notifies appropriate feedback content at the end of the period.

[0926] System configuration

[0927] This system consists of three main components: the user terminal, the server, and the generative AI model.

[0928] User terminal

[0929] The user terminal is a device that allows the evaluator to input short comments about the person being evaluated based on their impressions during their daily work, and can include a personal computer, smartphone, tablet, etc. A dedicated feedback input application is installed on the user terminal. Using this application, the user (evaluator) can input concise comments about the person being evaluated.

[0930] Examples:

[0931] A user (e.g., colleague A) uses a terminal to input a short comment about the person being evaluated (e.g., colleague B) into a feedback input application. For example, the user might leave a comment in the form of, "Mr. B made a great proposal in today's meeting."

[0932] server

[0933] The server has the following functions:

[0934] 1. Database function: A function that receives comment data entered by users on a daily basis and stores it in a database as structured data.

[0935] 2. Managing generative AI models: A function that allows you to feed saved comment data to generative AI models for learning.

[0936] 3. Feedback generation: A function that uses a generative AI model to generate feedback for each person being evaluated at the end of the period.

[0937] 4. Notification function: A function to notify the generated feedback to the device of the person being evaluated.

[0938] Examples:

[0939] Comment data sent from the device is sent to a server, which stores the received comment data in a database and organizes it by adding metadata to each comment, such as the user's ID, comment content, and timestamp.

[0940] Generative AI Models

[0941] The generative AI model uses natural language processing technology to analyze the saved one-word comment data and generate appropriate feedback for each person being evaluated. This model provides comprehensive feedback by taking into account the relationship between the evaluator and the person being evaluated, the content, frequency, and emotion of the comment, etc.

[0942] Examples:

[0943] At the end of the period, the server uses the generative AI model to generate feedback for each person being evaluated, such as, "Mr. B made an important proposal in this project and made a significant contribution to the team. He also showed a proactive attitude toward seeking feedback."

[0944] Feedback Generation Process

[0945] A specific example of the feedback generation process is as follows:

[0946] Learning from comment data

[0947] The server periodically supplies the comment data stored in the database to the generative AI model, which then learns from it, taking into account the content of the comment, the relationship between the evaluator and the person being evaluated, and the frequency of the comment.

[0948] Example prompt sentence:

[0949] Generate feedback for Subject B based on the following comments:

[0950] 1. Person B asked insightful questions during team meetings, which sparked deeper discussions.

[0951] 2. Mr. B's new idea was very original and caught everyone's attention.

[0952] Feedback content:

[0953] Feedback Notification

[0954] The server stores the generated feedback in a database and notifies the evaluation recipient's device, where the evaluation recipient can check the generated feedback using a dedicated feedback checking application.

[0955] This allows the person being evaluated to receive multifaceted and detailed feedback that can be used to improve themselves. It also makes it easier for the evaluator to enter comments, freeing them from having to rely on memory or input constraints, allowing for the provision of high-quality feedback.

[0956] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0957] Step 1:

[0958] Enter a comment

[0959] The user inputs a comment about the person being evaluated.

[0960] Input: Comment text entered by the evaluator through the terminal.

[0961] Specific operation: The user launches a dedicated feedback input application and enters a comment about the person being evaluated, such as "Mr. B made a great proposal in today's meeting." The terminal has a "Send" button, which the user clicks.

[0962] Output: The comment text entered is ready to be sent.

[0963] Step 2:

[0964] Submitting comment data

[0965] The terminal transmits the input comment data to the server.

[0966] Input: The comment text that you have completed.

[0967] Specific operation: The device sends the comment text as an HTTP request to a specific API endpoint (e.g., POST / comments) on the server.

[0968] Output: The comment data sent to the server.

[0969] Step 3:

[0970] Saving comment data

[0971] The server stores the received comment data in a database.

[0972] Input: The comment text received by the server.

[0973] Specific operation: The server receives the HTTP request, adds metadata such as the rater's ID, comment content, and timestamp, and structures the data. It then generates and executes an INSERT query in a database (e.g., MySQL) to save the comment data.

[0974] Output: Comment data stored in the database.

[0975] Step 4:

[0976] Data training

[0977] The server periodically supplies the saved comment data to the generative AI model for learning.

[0978] Input: Multiple comment data stored in the database.

[0979] How it works: The server periodically retrieves comment data from the database as a scheduled job and supplies it to the API of a generative AI model (e.g., the GPT-3 model). When supplied, it converts it to JSON format and sends it to the model's endpoint. The model learns by taking into account the comment content, relationships, frequency, etc.

[0980] Output: A trained generative AI model.

[0981] Step 5:

[0982] Generate feedback

[0983] At the end of the period, the server uses the generated AI model to generate feedback content for each person being evaluated.

[0984] Input: Trained generative AI model and comment data.

[0985] Specific operation: The server generates a prompt based on the ID of the person being evaluated and sends it to the generation AI model. For example, a prompt like "Generate feedback for person B based on the following comments: 1. Person B asked a sharp question in the team meeting, which sparked a deeper discussion. 2. Person B's new idea was very novel and attracted everyone's attention. Feedback content:" is used. The feedback returned by the model is then processed and organized.

[0986] Output: The generated feedback.

[0987] Step 6:

[0988] Feedback Notification

[0989] The server notifies the generated feedback content to the terminal of the person being evaluated.

[0990] Input: The generated feedback.

[0991] Specific operation: The server saves the feedback content in a database and sends a notification to the assessee's device using a notification service (e.g., Firebase Cloud Messaging). The application on the assessee's device receives the notification and prompts the user to confirm.

[0992] Output: Feedback provided and confirmation by the assessee.

[0993] (Application example 1)

[0994] 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."

[0995] It is important to routinely evaluate the operating status and performance of automated equipment operating on-site and provide appropriate feedback. However, in many production sites, such evaluations are often dependent on a few managers, resulting in subjective and inconsistent evaluations. Furthermore, it has been difficult to quickly and effectively monitor the status of automated equipment during daily operations and detect problems early. Furthermore, delays in such evaluations and feedback can result in equipment deterioration and reduced production efficiency.

[0996] 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.

[0997] In this invention, the server includes means for inputting a short comment about the person being evaluated by a person involved in daily work, means for saving the input short comment in a database, means for generating feedback content using a generative AI model that learns the saved short comment, means for notifying the person being evaluated of the generated feedback content at the end of the period, means for inputting comments about the operating status and performance of automated operation equipment operating in a factory, and means for evaluating the performance of the automated operation equipment based on the input comments. This enables daily feedback about the operating status and performance of the automated operation equipment in a factory, makes evaluations consistent and objective, and enables efficient operation of the automated operation equipment and early problem detection.

[0998] "Evaluation target" refers to the person who is evaluated in the feedback system, and refers to automated work equipment and robots operating in a factory.

[0999] A "shorthand comment" is a short written opinion or observation about the operating status or performance of the subject of evaluation.

[1000] The "database" is a system for storing and managing input short comments as structured data.

[1001] A "generative AI model" is an artificial intelligence model that learns from input one-line comments and generates feedback content based on them.

[1002] "End of period" refers to the end of a specific time or period when feedback is communicated to the assessee.

[1003] "Notification means" refers to a method or device for notifying the generated feedback content to the person being evaluated.

[1004] "Automated work equipment" refers to machines or robots used to automatically perform specific tasks in a factory.

[1005] "Performance evaluation" refers to the task of evaluating the operating status and performance of the person being evaluated.

[1006] This system evaluates the operational status and performance of automated work equipment in a factory on a daily basis and provides appropriate feedback. This system consists of three main components: a user terminal, a server, and a generative AI model.

[1007] System configuration

[1008] User terminal

[1009] The user terminal is a device that allows the evaluator to input a brief comment about the automated operation equipment operating in the factory, and can include PCs, smartphones, tablets, etc. A dedicated feedback input application is installed on the user terminal. Using this application, the user (evaluator) can input a brief comment about the evaluation target (automated operation equipment).

[1010] server

[1011] The server has the following functions:

[1012] 1. Database function: A function that receives comment data entered by users and stores it in a database as structured data.

[1013] 2. Managing generative AI models: A function that allows you to feed saved comment data to generative AI models for learning.

[1014] 3. Feedback generation: A function that uses a generative AI model to generate feedback for each person being evaluated at the end of the period.

[1015] 4. Notification function: A function to notify the generated feedback to the device of the person being evaluated.

[1016] Generative AI Models

[1017] The generative AI model uses natural language processing technology to analyze the saved one-line comment data and generate appropriate feedback for each person being evaluated. This model provides comprehensive feedback by taking into account the content, frequency, importance, etc. of the comment.

[1018] Feedback Generation Process

[1019] The following describes how this system works, using specific examples.

[1020] Comment input

[1021] A user (e.g., Manager A) uses a terminal to input a brief comment about an automated work device (e.g., Robot A) in a factory into a feedback input application. For example, the user leaves a comment in the form of "Robot A's work speed seemed slow today."

[1022] Saving comment data

[1023] The comment data sent from the device is sent to a server, which stores the received comment data in a database and organizes it by adding metadata to each comment, such as the automated work device's ID, comment content, and timestamp.

[1024] Data training

[1025] The server periodically supplies the saved comment data to the AI ​​model for learning. The AI ​​model updates the feedback model for each person being evaluated, taking into account the content of the comments, the relationship between the evaluator and the automated work device, and the frequency of comments.

[1026] Specific examples

[1027] Examples of comment input:

[1028] Automated Work Device ID: RobotA

[1029] Comment: "You seem to be working a lot slower today."

[1030] Example prompt sentence:

[1031] Generate feedback for each robot based on the comment data below:

[1032] Robot ID: RobotA, Comment: "You seem to be working a lot slower today."

[1033] Examples of feedback generated include:

[1034] Robot A requires regular maintenance. It has recently been running slower, possibly due to wear on the belt.

[1035] In this way, users can receive comprehensive feedback about the automated work equipment in their factories, which can be used to improve maintenance and operation.

[1036] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1037] Step 1:

[1038] A user uses a terminal to input a brief comment about an automated work device in a factory into a feedback input application.

[1039] Input: ID of the automated work device, comment (e.g., "Robot A seemed to be working slowly today")

[1040] Output: Comment data is sent from the device to the server.

[1041] Step 2:

[1042] The comment data sent from the terminal is received by the server.

[1043] Input: Comment data sent from the device (automated work device ID, comment content, timestamp)

[1044] Output: Comment data is saved in the database.

[1045] Step 3:

[1046] The server periodically supplies the comment data stored in the database to the generative AI model, allowing it to learn.

[1047] Input: Comment data stored in the database

[1048] Output: The generative AI model learns from the comment data and updates the feedback model.

[1049] Step 4:

[1050] At the end of the term, the server uses a generative AI model to generate feedback content for each person being evaluated.

[1051] Input: Comment data trained by the generative AI model

[1052] Output: End-of-period feedback (e.g., "Robot A needs regular maintenance. It has been running slowly recently, which may be due to wear on the belt.")

[1053] Step 5:

[1054] The generated feedback content is notified from the server to the terminal of the person being evaluated.

[1055] Input: Generated feedback

[1056] Output: A feedback notification is displayed on the assessee's device.

[1057] Step 6:

[1058] The person being evaluated checks the generated feedback content using the terminal.

[1059] Input: Feedback content sent to the device of the person being evaluated

[1060] Output: The person being evaluated will view the feedback and use it as a reference for implementing any necessary improvements or measures.

[1061] 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.

[1062] This system provides multifaceted feedback on the person being evaluated. The system allows people who work with the person on a daily basis to enter short comments about the person being evaluated, which are then stored in a database. A generative AI model then learns from these comments and generates feedback content. Furthermore, by combining this with an emotion engine, the system analyzes the emotions in the comments and provides more accurate feedback.

[1063] System configuration

[1064] The system consists of four main components: a user terminal, a server, a generative AI model, and an emotion engine.

[1065] User terminal

[1066] The user terminal is a device that allows the evaluator to enter short comments about the person being evaluated, and can include PCs, smartphones, tablets, etc. A dedicated feedback input application is installed on the terminal, and a comment input screen is provided.

[1067] server

[1068] The server has the following functions:

[1069] 1. Database function: A function to save comment data received from users in a database.

[1070] 2. Sentiment analysis function: A function that uses an emotion engine to analyze the sentiment of one-line comments and classify them as positive, negative, or neutral.

[1071] 3. Managing generative AI models: A function that allows saved comment data and its sentiment classification results to be fed into generative AI models for learning.

[1072] 4. Feedback generation: A function that generates feedback for each person being evaluated at the end of the period using a generative AI model.

[1073] 5. Notification function: A function to notify the assessee of the generated feedback.

[1074] Generative AI Models

[1075] The generative AI model uses natural language processing technology to analyze the saved one-word comment data and emotion classification results, and generates appropriate feedback content for each person being evaluated. This model reflects the emotional information of the comments received from the emotion engine and provides comprehensive feedback.

[1076] Emotion Engine

[1077] The emotion engine analyzes the sentiment of short comments entered by users and categorizes them as positive, negative, or neutral. The emotion engine uses natural language processing technology to extract sentiment from the text and sends the results to the server.

[1078] Explaining the feedback generation process

[1079] The following describes how this system works, using specific examples.

[1080] Comment input

[1081] A user (e.g., colleague A) uses a terminal to input a short comment about the person being evaluated (e.g., colleague B) into a feedback input application. For example, the user might leave a comment in the form of, "Mr. B made a great proposal in today's meeting."

[1082] Saving comment data

[1083] Comment data sent from the device is sent to a server, which stores the received comment data in a database and organizes it by adding metadata to each comment, such as the ID of the person being evaluated, the comment content, and a timestamp.

[1084] Emotion Analysis

[1085] The server provides the stored comment data to the emotion engine for sentiment analysis. The emotion engine analyzes the content of each comment and classifies it as positive, negative, or neutral. For example, "Mr. B made a great suggestion" is classified as positive.

[1086] Data training

[1087] The server supplies the comment data and its emotion classification results to the generative AI model, which then learns from it. The generative AI model reflects the emotion classification results and updates the feedback model for each person being evaluated, taking into account the comment content, the relationship between the evaluator and the person being evaluated, the emotion and frequency of the evaluation, etc.

[1088] Generate feedback

[1089] At the end of the period, the server uses the generative AI model to generate feedback for each individual being evaluated. For example, it generates multifaceted, emotionally-charged feedback such as, "Mr. B made an important proposal in this project and made a significant contribution to the team. He also showed a proactive attitude toward seeking feedback."

[1090] Feedback Notification

[1091] The generated feedback is stored in a database and sent from the server to the device of the person being evaluated. The person being evaluated can then check the generated feedback using a dedicated feedback confirmation application on their device.

[1092] This allows those being evaluated to receive detailed, accurate feedback from all angles, which can be used to improve themselves. Furthermore, evaluators can enter detailed comments when needed, freeing them from the constraints of relying on memory or time constraints. Furthermore, the introduction of an emotion engine improves the balance and accuracy of feedback, providing a fair and multifaceted evaluation.

[1093] The processing flow will be explained below.

[1094] Processing flow of a feedback system that combines an emotion engine

[1095] Step 1:

[1096] Users can input short comments about the person being evaluated into a dedicated feedback input application during work or daily conversations. For example, colleague A might input, "Mr. B made a very useful suggestion in today's meeting."

[1097] Step 2:

[1098] The device organizes the input comment data as structured data and sends it to the server, adding metadata such as the user's ID, comment content, and timestamp.

[1099] Step 3:

[1100] The server saves the received comment data in the database in the following format: "Evaluator ID: B123, Comment content: 'Very useful suggestion', Timestamp: 2023-10-01 10:30".

[1101] Step 4:

[1102] The server provides the stored comment data to the emotion engine for sentiment analysis. The emotion engine analyzes the content of each comment and classifies it as positive, negative, or neutral. For example, "Mr. B made a great suggestion" would be classified as positive.

[1103] Step 5:

[1104] The server adds the emotion classification results obtained from the emotion engine to the comment data and saves it in the database. For example, "evaluator ID: B123, comment content: 'Very useful suggestion', emotion: positive" is saved.

[1105] Step 6:

[1106] The server periodically supplies all comment data and emotion classification results to the AI ​​model, allowing it to update and learn. For example, you can set it to automatically start AI training on weekends.

[1107] Step 7:

[1108] The generative AI model learns from the provided data, using natural language processing technology to analyze the content of comments, the relationship between the rater and the person being rated, the sentiment classification results, the sentiment of the ratings, frequency, etc.

[1109] Step 8:

[1110] As the end of the term approaches, the server again supplies all comment data and emotion classification results to the generative AI model and instructs it to generate feedback. For example, the server issues a data supply instruction one week before the end of the term.

[1111] Step 9:

[1112] The generative AI model generates appropriate feedback for each person being evaluated based on all stored comment data and emotion classification results. For example, it considers a balance of positive and negative comments and generates feedback such as, "Mr. B made important suggestions in this project and made a significant contribution to the team. He also showed an attitude of actively seeking feedback."

[1113] Step 10:

[1114] The server stores the generated feedback content in a database and simultaneously sends a notification to the terminal of the person being evaluated that the feedback has been generated.

[1115] Step 11:

[1116] The device receives a notification, and the person being evaluated can check the generated feedback through a dedicated feedback review application. Specifically, when the person being evaluated opens the application, a list of the feedback and its details are displayed.

[1117] This allows those being evaluated to receive detailed and accurate feedback from all angles, which can be used to improve their own performance. Furthermore, users (evaluators) can enter detailed comments whenever necessary, freeing them from the constraints of memory and time constraints. Furthermore, the introduction of an emotion engine improves the balance and accuracy of feedback, providing a fair and multifaceted evaluation.

[1118] Example 2

[1119] 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."

[1120] In conventional evaluation systems, evaluators rely on their memory to make evaluations, which can lead to problems with the accuracy and fairness of the evaluations. Furthermore, because the evaluation content is one-sided, it is difficult to grasp the overall picture of the person being evaluated's performance and behavior. Furthermore, evaluations are not based on emotions, which can lead to insufficient consideration of the evaluation content.

[1121] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1122] In this invention, the server includes: a means for people involved in daily work to input short comments about the person being evaluated; a means for saving the input short comments in a database; a means for analyzing the sentiment of the saved short comments and classifying them as positive, negative, or neutral; a means for generating feedback content using a generative AI model that learns short comments including the sentiment analysis results; and a means for notifying the person being evaluated of the generated feedback content at the end of the period. This improves the accuracy and fairness of the evaluation content and enables multifaceted evaluation of the person's performance and behavior. Furthermore, incorporating sentiment analysis allows the evaluation content to reflect emotional aspects.

[1123] A "shorthand comment" is a short piece of feedback entered by the evaluator about the person being evaluated during their daily work.

[1124] "Database" means an information storage system for storing and managing entered short comments and other related data.

[1125] "Sentiment analysis" is a technology that uses natural language processing technology to analyze the emotions contained in short comments and classify them as positive, negative, or neutral.

[1126] The "generative AI model" is an artificial intelligence model that learns from saved one-line comments and the results of their emotional analysis, and generates feedback content for each person being evaluated.

[1127] "Notification" is a system function for notifying the subject of evaluation of the generated feedback content.

[1128] The present invention is a system that provides multifaceted feedback to an individual by inputting a short comment about the individual involved in daily work, storing the comment in a database, performing sentiment analysis, generating feedback using a generative AI model, and notifying the individual. Each component of this system and its embodiment will be described in detail below.

[1129] System configuration

[1130] The system consists of four main components: a user terminal, a server, a generative AI model, and a sentiment analysis engine.

[1131] User terminal

[1132] The user terminal is a device on which the evaluator inputs a short comment about the person being evaluated. It can be a PC, smartphone, tablet, etc. A dedicated feedback input application is installed on the user terminal, and a comment input screen is provided. For example, the evaluator might input a comment in the form of "Mr. B made a great proposal in today's meeting."

[1133] server

[1134] The server has the following features:

[1135] 1. Database function: Comment data received from users is saved in a database. The database uses MySQL or PostgreSQL.

[1136] 2. Sentiment analysis function: Using an emotion engine, the sentiment of one-line comments is classified as positive, negative, or neutral. Sentiment analysis is performed using Python's VADER and TextBlob.

[1137] 3. Managing the generative AI model: The saved comment data and its sentiment classification results are fed into the generative AI model for training. The generative AI model uses a Transformer-based model such as GPT-3 or BERT.

[1138] 4. Feedback generation: At the end of the period, feedback content is generated for each person being evaluated using the generative AI model.

[1139] 5. Notification function: Notify the assessee of the generated feedback.

[1140] Generative AI Models

[1141] The generative AI model uses natural language processing technology to analyze the saved one-line comment data and sentiment analysis results, generating appropriate feedback for each individual being evaluated. The generative AI model is implemented using Python or other programming languages, utilizing machine learning frameworks such as TensorFlow and PyTorch. For example, it might generate feedback such as, "Mr. B made important suggestions in this project and made a significant contribution to the team. He also showed a proactive attitude toward seeking feedback."

[1142] Sentiment Analysis Engine

[1143] The sentiment analysis engine analyzes the sentiment of short comments entered by users and classifies them as positive, negative, or neutral. The sentiment analysis engine uses natural language processing technology to extract the sentiment of the text and sends the results to a server. Technologies used include VADER and TextBlob.

[1144] Examples of concrete examples and prompts

[1145] For example, if a user (colleague A) enters a comment such as "Mr. B made a great proposal in today's meeting," the server receives this comment and stores it in a database. It is then classified as "positive" using a sentiment analysis engine. The generative AI model generates feedback for the person being evaluated (colleague B) based on this data.

[1146] Examples:

[1147] Comment: "Mr. B made a great suggestion at today's meeting."

[1148] Sentiment Analysis: Positive

[1149] Feedback: "B made important suggestions on this project and contributed greatly to the team. He also proactively sought feedback."

[1150] Example prompt for a generative AI model:

[1151] "Generate the following feedback: 'Person B made a great suggestion in today's meeting.'"

[1152] In this way, the system according to the present invention can provide multifaceted and accurate feedback regarding the person being evaluated.

[1153] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1154] Step 1:

[1155] Comments entered by the user

[1156] Operation: The user opens a feedback application on a device (PC, smartphone, tablet, etc.). They access the comment input screen and enter a short comment about the person being evaluated. For example, they enter a comment such as, "Mr. B made a great proposal in today's meeting."

[1157] Input: Text comments about the person being assessed.

[1158] Output: The entered comment data is sent from the terminal to the server.

[1159] Step 2:

[1160] Sending and saving comment data to the server

[1161] Operation: Comment data sent from the device is sent to the server via an HTTP request. The server analyzes the received comment data, extracts the ID of the person being rated, the ID of the evaluator, the comment content, and the timestamp, and stores them in a database.

[1162] Input: Comment data sent from the device (text comment, ID of the person being evaluated, ID of the evaluator, timestamp).

[1163] Output: Comment data stored in a database.

[1164] Step 3:

[1165] Performing sentiment analysis

[1166] How it works: The server passes the comment data stored in the database to a sentiment analysis engine, which uses natural language processing libraries such as VADER or TextBlob to classify the sentiment of each comment as positive, negative, or neutral.

[1167] Input: Text comment stored in the database.

[1168] Output: Comment data with sentiment classification results (positive, negative, neutral).

[1169] Step 4:

[1170] Supplying and training generative AI models

[1171] How it works: The server supplies the sentiment-analyzed comment data to the generative AI model, which then uses natural language processing technology to update the feedback model for each subject based on the data provided.

[1172] Input: Comment data with sentiment classification results.

[1173] Output: An updated generative AI model.

[1174] Step 5:

[1175] Generate feedback

[1176] How it works: At the end of the term, the server uses the updated generative AI model to generate feedback for each person being evaluated. For example, it generates feedback in the form of, "Mr. B made important suggestions in this project and made a significant contribution to the team. He also showed a proactive attitude of seeking feedback."

[1177] Input: Updated generative AI model, ID of the person being evaluated.

[1178] Output: Generated feedback for each ratee.

[1179] Step 6:

[1180] Feedback Notification

[1181] Operation: The server notifies the subject of the generated feedback to the subject's device. The subject then checks the received feedback using a dedicated feedback checking application.

[1182] Input: Generated feedback content, ID of the person being assessed.

[1183] Output: Feedback sent to the subject's device.

[1184] Through these steps, users can provide multifaceted and accurate feedback to those being evaluated. Raters can easily enter comments, and sentiment analysis and generative AI models ensure detailed and fair feedback.

[1185] (Application example 2)

[1186] 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."

[1187] In conventional feedback systems, evaluations are often based on the subjective opinions of the evaluator or one-sided information, making it difficult to accurately grasp the overall growth and areas for improvement of the person being evaluated.Furthermore, because the emotional aspects of the comments are not taken into consideration, there are problems with biased feedback content and unfair evaluations.

[1188] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a means for a person who is involved in daily work to input a short comment about the person to be evaluated, a means for saving the input short comment in a database, a means for generating feedback content using a generative AI model that learns from the saved short comment, a means for analyzing the input comment with an emotion engine and classifying emotions, and a means for notifying the person to be evaluated of the generated feedback content at the end of the period. This enables fair feedback that is multifaceted and reflects emotions.

[1189] "People involved in daily work" refers to colleagues, superiors, subordinates, and other people who work with the person being evaluated on a daily basis.

[1190] "Evaluation target" refers to the person, employee, worker, etc. who is the subject of feedback.

[1191] A "shorthand comment" is a short feedback message about the person being appraised that describes a behavior or performance observed during work.

[1192] "Means of storing in a database" refers to the systems and technologies for systematically storing and managing the entered comment data.

[1193] A "generative AI model" refers to an artificial intelligence module that learns based on input data and generates feedback content.

[1194] "Means for generating feedback content" refers to the process of analyzing collected data using a generative AI model and creating appropriate feedback for each person being evaluated.

[1195] An "emotion engine" refers to technology that analyzes the emotional aspects of input comments and classifies them as positive, negative, neutral, etc.

[1196] "Sentiment classification" refers to the process of identifying and categorizing the emotional aspects of comments using an emotion engine.

[1197] "Means of notifying at the end of the period" refers to the mechanisms and techniques for communicating the generated feedback content to the person being evaluated at the end of the evaluation period.

[1198] The present invention is a system in which people who interact with the person being evaluated in their daily work enter short comments about the person being evaluated, save them in a database, and a generative AI model learns from them to generate feedback content. The system is configured as follows:

[1199] Hardware and Software

[1200] 1. User device: A device used by the user to enter a short comment about the person being evaluated. This can be a PC, smartphone, head-mounted display (HMD), etc. A dedicated feedback input application is installed and a comment input screen is provided.

[1201] 2. Server:

[1202] Database function: A function to save comment data received from users in a database.

[1203] Sentiment analysis function: A function that uses an emotion engine to analyze the sentiment of short comments and classify them as positive, negative, or neutral.

[1204] Managing generative AI models: A function that allows saved comment data and its sentiment classification results to be fed into generative AI models for training.

[1205] Feedback generation: A function that uses a generative AI model to generate feedback for each assessee.

[1206] Notification function: A function to notify the assessee of the generated feedback at the end of the period.

[1207] Processing flow and data calculation

[1208] 1. Comment entry:

[1209] A user (e.g., a factory worker) uses a terminal to input a short comment about the work of the person being evaluated (e.g., a robot) into a feedback input application. For example, the user might leave a comment in the form of, "Robot A carried the pallet smoothly today."

[1210] Example prompt: "In one sentence, tell us what you noticed about the robot's work."

[1211] 2. Storage of Comment Data:

[1212] Comment data sent from the device is sent to a server, which stores the received comment data in a database and organizes it by adding metadata to each comment, such as the ID of the person being evaluated, the comment content, and a timestamp.

[1213] 3. Emotion analysis:

[1214] The server provides the stored comment data to the emotion engine for emotion analysis. The emotion engine analyzes the content of each comment and classifies it as positive, negative, or neutral. For example, "Robot A handled the situation smoothly" is classified as positive.

[1215] Example prompt: "Analyze the sentiment of this comment and categorize it as positive, negative, or neutral."

[1216] 4. Data training:

[1217] The server supplies the comment data and its emotion classification results to the generative AI model, which then learns from it. The generative AI model reflects the emotion classification results and updates the feedback model for each person being evaluated, taking into account the comment content, the relationship between the evaluator and the person being evaluated, the emotion and frequency of the evaluation, etc.

[1218] 5. Feedback Generation:

[1219] At the end of the period, the server uses the generative AI model to generate feedback for each individual being evaluated, generating multifaceted and emotional feedback such as, "Robot A was recognized for its smooth movements in transporting pallets, and its efficiency has improved."

[1220] Example prompt: "Based on this sentiment analysis and comment data, generate feedback to improve the robot's performance."

[1221] 6. Feedback Notification:

[1222] The generated feedback is stored in a database and sent from the server to the device of the person being evaluated. The person being evaluated can then check the generated feedback using a dedicated feedback confirmation application on their device.

[1223] This will enable multifaceted, emotional feedback to be provided to robots in factories, which is expected to improve work efficiency and performance.

[1224] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1225] Step 1:

[1226] Comment input from user terminal:

[1227] A user (such as a factory worker) uses a device (such as a smartphone or head-mounted display) to input a comment about the work of the person being evaluated (such as a robot). The comment is sent to a feedback input application on the device. Example: "Robot A carried the pallet smoothly today."

[1228] Input: A short comment entered by the user

[1229] Output: Comment data sent from the device to the server

[1230] Step 2:

[1231] Comment data storage on server:

[1232] The submitted comment data is sent to a server and stored in a database. The server organizes the data by adding metadata to each comment, such as the user's ID, comment content, and timestamp.

[1233] Input: Comment data sent from the device

[1234] Output: Comment data stored in the database

[1235] Step 3:

[1236] On-server sentiment analysis:

[1237] The comment data stored on the server is fed to the emotion engine, which analyzes the content of the comment and classifies it as positive, negative, or neutral. For example, "Robot A handled the situation smoothly" is classified as positive.

[1238] Input: Comment data stored in the database

[1239] Output: Emotion classification result

[1240] Step 4:

[1241] Training generative AI models:

[1242] The server supplies the comment data and its emotion classification results to the generative AI model for learning. The generative AI model reflects the emotion classification results and updates the feedback model for each person being evaluated, taking into account the comment content, the relationship between the evaluator and the person being evaluated, the emotion and frequency of the evaluation, etc.

[1243] Input: Comment data and sentiment classification results

[1244] Output: Updated generative AI model

[1245] Step 5:

[1246] Generate feedback:

[1247] At the end of the period, the server uses the generative AI model to generate feedback for each assessee, such as "Robot A is recognized for its smooth movements in transporting pallets, and has improved efficiency," which reflects both multifaceted and emotional feedback.

[1248] Input: Updated generative AI model

[1249] Output: Generated feedback

[1250] Step 6:

[1251] Feedback Notification:

[1252] The generated feedback content is stored in a database and is also sent from the server to the device of the person being evaluated, who can then check the generated feedback content using a dedicated feedback checking application on their device.

[1253] Input: Generated feedback

[1254] Output: Feedback notification to the device of the person being evaluated

[1255] 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.

[1256] 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.

[1257] 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.

[1258] [Fourth embodiment]

[1259] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1260] 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.

[1261] 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).

[1262] 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.

[1263] 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.

[1264] 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).

[1265] 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.

[1266] 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.

[1267] 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.

[1268] 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.

[1269] 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.

[1270] 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.

[1271] 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."

[1272] In order to realize a multifaceted feedback system, the present invention collects short comments about the person being evaluated on a daily basis, analyzes them using a generative AI model, and generates and notifies appropriate feedback content at the end of the period. Specific embodiments of the system are described below.

[1273] System configuration

[1274] This system consists of three main components: a user terminal, a server, and a generative AI model.

[1275] User terminal

[1276] The user terminal is a device that allows the evaluator to input short comments about the person being evaluated based on their impressions during their daily work, and can include PCs, smartphones, tablets, etc. A dedicated feedback input application is installed on the user terminal. Using this application, the user (evaluator) can input concise comments about the person being evaluated.

[1277] server

[1278] The server has the following functions:

[1279] 1. Database function: A function that receives comment data entered by users on a daily basis and stores it in a database as structured data.

[1280] 2. Managing generative AI models: A function that allows you to feed saved comment data to generative AI models for learning.

[1281] 3. Feedback generation: A function that uses a generative AI model to generate feedback for each person being evaluated at the end of the period.

[1282] 4. Notification function: A function to notify the generated feedback to the device of the person being evaluated.

[1283] Generative AI Models

[1284] The generative AI model uses natural language processing technology to analyze the saved one-word comment data and generate appropriate feedback for each person being evaluated. This model provides comprehensive feedback by taking into account the relationship between the evaluator and the person being evaluated, the content, frequency, and emotion of the comment, etc.

[1285] Explaining the feedback generation process

[1286] The following describes how this system works, using specific examples.

[1287] Comment input

[1288] A user (e.g., colleague A) uses a terminal to input a short comment about the person being evaluated (e.g., colleague B) into a feedback input application. For example, the user might leave a comment in the form of, "Mr. B made a great proposal in today's meeting."

[1289] Saving comment data

[1290] The comment data sent from the device is sent to a server, which stores the received comment data in a database and organizes it by adding metadata to each comment, such as the ID of the person being evaluated, the comment content, and a timestamp.

[1291] Data training

[1292] The server periodically supplies the saved comment data to the AI ​​model for learning. The AI ​​model updates the feedback model for each person being evaluated, taking into account the content of the comments, the relationship between the evaluator and the person being evaluated, and the frequency of comments.

[1293] Generate feedback

[1294] At the end of the period, the server uses the generative AI model to generate feedback for each person being evaluated, such as, "Mr. B made an important proposal in this project and made a significant contribution to the team. He also showed a proactive attitude toward seeking feedback."

[1295] Feedback Notification

[1296] The generated feedback is stored in a database and sent from the server to the device of the person being evaluated. The person being evaluated can then check the generated feedback using a dedicated feedback confirmation application on their device.

[1297] This allows the person being evaluated to receive multifaceted and detailed feedback that can be used to improve themselves. It also makes it easier for the evaluator to enter comments, freeing them from having to rely on memory or input constraints, allowing them to provide high-quality feedback.

[1298] The processing flow will be explained below.

[1299] Program processing flow

[1300] Step 1:

[1301] Users can input short comments about the person being evaluated into a dedicated feedback input application during work or daily conversations. For example, colleague A might input, "Mr. B made a very useful suggestion in today's meeting."

[1302] Step 2:

[1303] The device organizes the input comment data as structured data and sends it to the server, adding metadata such as the user's ID, comment content, and timestamp.

[1304] Step 3:

[1305] The server stores the received comment data in a database. For example, the data is stored in the format of "evaluator ID: B123, comment content: 'Very useful suggestion', timestamp: 2023-10-01 10:30".

[1306] Step 4:

[1307] Each time a new comment is added, the server feeds the data to the generative AI model, which performs an initial learning process, such as cleaning the text (correcting unnecessary characters and typos) and categorizing the comment.

[1308] Step 5:

[1309] The generative AI model learns from the provided data, using natural language processing technology to analyze the content of comments, the relationship between the evaluator and the person being rated, the sentiment of the ratings, frequency, etc.

[1310] Step 6:

[1311] The server sets a schedule to automatically supply additional data to the AI ​​model at regular intervals (e.g., one week) to continuously update and train the model.

[1312] Step 7:

[1313] As the end of the term approaches, the server issues an instruction to re-feed all comment data to the generative AI model. For example, it will feed all data one week before the end of the term.

[1314] Step 8:

[1315] A generative AI model uses all stored comment data to generate appropriate feedback for each ratee, balancing positive and negative feedback.

[1316] Step 9:

[1317] The server stores the generated feedback content in a database and simultaneously sends a notification to the terminal of the person being evaluated that the feedback has been generated.

[1318] Step 10:

[1319] The device receives a notification, and the person being evaluated can check the generated feedback through a dedicated feedback confirmation application. Specifically, when the person being evaluated opens the application, they can see feedback such as, "Mr. B made important suggestions in this project and made a significant contribution to the team. He also showed an attitude of proactively seeking feedback."

[1320] This allows the person being evaluated to receive detailed feedback from all angles, which can be used to improve themselves. In addition, the evaluator can enter detailed comments when necessary, freeing them from the constraints of relying on memory or a limited input period.

[1321] Example 1

[1322] 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."

[1323] Current feedback systems have the drawback of making it difficult for evaluators to provide timely and detailed feedback on the person being evaluated. Because the feedback relies on the evaluator's subjectivity, it lacks consistency and objectivity, making it difficult to effectively support the growth of the person being evaluated. Another issue is that the cumbersome process of inputting and managing feedback places an excessive burden on the evaluator.

[1324] 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.

[1325] In this invention, the server includes: a means for a user to input a one-line comment about the person to be evaluated; a means for a terminal to transmit the input one-line comment to the server; a means for the server to save the received one-line comment data in a database; a means for the server to periodically supply the saved one-line comment data to a generative AI model for learning; a means for the server to generate feedback content for each person to be evaluated using the generative AI model at the end of the period; and a means for the server to notify the terminal of the person to be evaluated of the generated feedback content. This makes it possible to provide consistent and objective feedback to the person to be evaluated and enables the evaluator to easily input and manage comments.

[1326] A "user" is a person who uses the system to input a short comment about the person being evaluated.

[1327] A "terminal" is a device used by a user, and includes a personal computer, a smartphone, a tablet, and the like.

[1328] "Server" refers to a device that stores comment data, manages generative AI models, and generates and notifies feedback.

[1329] A "shorthand comment" is a short textual evaluation content entered by a user about the person being evaluated.

[1330] The "database" is a system that stores received short comment data and generated feedback content as structured data.

[1331] The "generative AI model" is an algorithm that learns from saved one-line comment data and generates feedback content using natural language processing technology.

[1332] "Feedback content" refers to the content of the evaluation and advice given to the person being evaluated, generated by the generative AI model.

[1333] "Notification" is the process of transmitting the generated feedback content to the terminal of the person being evaluated.

[1334] "End of period" refers to the end of the cycle for generating and providing feedback.

[1335] A "prompt sentence" is an input sentence used to have a generative AI model generate feedback content.

[1336] In order to realize a multifaceted feedback system, this invention collects short comments about the person being evaluated on a daily basis, analyzes them using a generative AI model, and generates and notifies appropriate feedback content at the end of the period.

[1337] System configuration

[1338] This system consists of three main components: the user terminal, the server, and the generative AI model.

[1339] User terminal

[1340] The user terminal is a device that allows the evaluator to input short comments about the person being evaluated based on their impressions during their daily work, and can include a personal computer, smartphone, tablet, etc. A dedicated feedback input application is installed on the user terminal. Using this application, the user (evaluator) can input concise comments about the person being evaluated.

[1341] Examples:

[1342] A user (e.g., colleague A) uses a terminal to input a short comment about the person being evaluated (e.g., colleague B) into a feedback input application. For example, the user might leave a comment in the form of, "Mr. B made a great proposal in today's meeting."

[1343] server

[1344] The server has the following functions:

[1345] 1. Database function: A function that receives comment data entered by users on a daily basis and stores it in a database as structured data.

[1346] 2. Managing generative AI models: A function that allows you to feed saved comment data to generative AI models for learning.

[1347] 3. Feedback generation: A function that uses a generative AI model to generate feedback for each person being evaluated at the end of the period.

[1348] 4. Notification function: A function to notify the generated feedback to the device of the person being evaluated.

[1349] Examples:

[1350] Comment data sent from the device is sent to a server, which stores the received comment data in a database and organizes it by adding metadata to each comment, such as the user's ID, comment content, and timestamp.

[1351] Generative AI Models

[1352] The generative AI model uses natural language processing technology to analyze the saved one-word comment data and generate appropriate feedback for each person being evaluated. This model provides comprehensive feedback by taking into account the relationship between the evaluator and the person being evaluated, the content, frequency, and emotion of the comment, etc.

[1353] Examples:

[1354] At the end of the period, the server uses the generative AI model to generate feedback for each person being evaluated, such as, "Mr. B made an important proposal in this project and made a significant contribution to the team. He also showed a proactive attitude toward seeking feedback."

[1355] Feedback Generation Process

[1356] A specific example of the feedback generation process is as follows:

[1357] Learning from comment data

[1358] The server periodically supplies the comment data stored in the database to the generative AI model, which then learns from it, taking into account the content of the comment, the relationship between the evaluator and the person being evaluated, and the frequency of the comment.

[1359] Example prompt sentence:

[1360] Generate feedback for Subject B based on the following comments:

[1361] 1. Person B asked insightful questions during team meetings, which sparked deeper discussions.

[1362] 2. Mr. B's new idea was very original and caught everyone's attention.

[1363] Feedback content:

[1364] Feedback Notification

[1365] The server stores the generated feedback in a database and notifies the evaluation recipient's device, where the evaluation recipient can check the generated feedback using a dedicated feedback checking application.

[1366] This allows the person being evaluated to receive multifaceted and detailed feedback that can be used to improve themselves. It also makes it easier for the evaluator to enter comments, freeing them from having to rely on memory or input constraints, allowing for the provision of high-quality feedback.

[1367] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1368] Step 1:

[1369] Enter a comment

[1370] The user inputs a comment about the person being evaluated.

[1371] Input: Comment text entered by the evaluator through the terminal.

[1372] Specific operation: The user launches a dedicated feedback input application and enters a comment about the person being evaluated, such as "Mr. B made a great proposal in today's meeting." The terminal has a "Send" button, which the user clicks.

[1373] Output: The comment text entered is ready to be sent.

[1374] Step 2:

[1375] Submitting comment data

[1376] The terminal transmits the input comment data to the server.

[1377] Input: The comment text that you have completed.

[1378] Specific operation: The device sends the comment text as an HTTP request to a specific API endpoint (e.g., POST / comments) on the server.

[1379] Output: The comment data sent to the server.

[1380] Step 3:

[1381] Saving comment data

[1382] The server stores the received comment data in a database.

[1383] Input: The comment text received by the server.

[1384] Specific operation: The server receives the HTTP request, adds metadata such as the rater's ID, comment content, and timestamp, and structures the data. It then generates and executes an INSERT query in a database (e.g., MySQL) to save the comment data.

[1385] Output: Comment data stored in the database.

[1386] Step 4:

[1387] Data training

[1388] The server periodically supplies the saved comment data to the generative AI model for learning.

[1389] Input: Multiple comment data stored in the database.

[1390] How it works: The server periodically retrieves comment data from the database as a scheduled job and supplies it to the API of a generative AI model (e.g., the GPT-3 model). When supplied, it converts it to JSON format and sends it to the model's endpoint. The model learns by taking into account the comment content, relationships, frequency, etc.

[1391] Output: A trained generative AI model.

[1392] Step 5:

[1393] Generate feedback

[1394] At the end of the period, the server uses the generated AI model to generate feedback content for each person being evaluated.

[1395] Input: Trained generative AI model and comment data.

[1396] Specific operation: The server generates a prompt based on the ID of the person being evaluated and sends it to the generation AI model. For example, a prompt like "Generate feedback for person B based on the following comments: 1. Person B asked a sharp question in the team meeting, which sparked a deeper discussion. 2. Person B's new idea was very novel and attracted everyone's attention. Feedback content:" is used. The feedback returned by the model is then processed and organized.

[1397] Output: The generated feedback.

[1398] Step 6:

[1399] Feedback Notification

[1400] The server notifies the generated feedback content to the terminal of the person being evaluated.

[1401] Input: The generated feedback.

[1402] Specific operation: The server saves the feedback content in a database and sends a notification to the assessee's device using a notification service (e.g., Firebase Cloud Messaging). The application on the assessee's device receives the notification and prompts the user to confirm.

[1403] Output: Feedback provided and confirmation by the assessee.

[1404] (Application example 1)

[1405] 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."

[1406] It is important to routinely evaluate the operating status and performance of automated equipment operating on-site and provide appropriate feedback. However, in many production sites, such evaluations are often dependent on a few managers, resulting in subjective and inconsistent evaluations. Furthermore, it has been difficult to quickly and effectively monitor the status of automated equipment during daily operations and detect problems early. Furthermore, delays in such evaluations and feedback can result in equipment deterioration and reduced production efficiency.

[1407] 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.

[1408] In this invention, the server includes means for inputting a short comment about the person being evaluated by a person involved in daily work, means for saving the input short comment in a database, means for generating feedback content using a generative AI model that learns the saved short comment, means for notifying the person being evaluated of the generated feedback content at the end of the period, means for inputting comments about the operating status and performance of automated operation equipment operating in a factory, and means for evaluating the performance of the automated operation equipment based on the input comments. This enables daily feedback about the operating status and performance of the automated operation equipment in a factory, makes evaluations consistent and objective, and enables efficient operation of the automated operation equipment and early problem detection.

[1409] "Evaluation target" refers to the person who is evaluated in the feedback system, and refers to automated work equipment and robots operating in a factory.

[1410] A "shorthand comment" is a short written opinion or observation about the operating status or performance of the subject of evaluation.

[1411] The "database" is a system for storing and managing input short comments as structured data.

[1412] A "generative AI model" is an artificial intelligence model that learns from input one-line comments and generates feedback content based on them.

[1413] "End of period" refers to the end of a specific time or period when feedback is communicated to the assessee.

[1414] "Notification means" refers to a method or device for notifying the generated feedback content to the person being evaluated.

[1415] "Automated work equipment" refers to machines or robots used to automatically perform specific tasks in a factory.

[1416] "Performance evaluation" refers to the task of evaluating the operating status and performance of the person being evaluated.

[1417] This system evaluates the operational status and performance of automated work equipment in a factory on a daily basis and provides appropriate feedback. This system consists of three main components: a user terminal, a server, and a generative AI model.

[1418] System configuration

[1419] User terminal

[1420] The user terminal is a device that allows the evaluator to input a brief comment about the automated operation equipment operating in the factory, and can include PCs, smartphones, tablets, etc. A dedicated feedback input application is installed on the user terminal. Using this application, the user (evaluator) can input a brief comment about the evaluation target (automated operation equipment).

[1421] server

[1422] The server has the following functions:

[1423] 1. Database function: A function that receives comment data entered by users and stores it in a database as structured data.

[1424] 2. Managing generative AI models: A function that allows you to feed saved comment data to generative AI models for learning.

[1425] 3. Feedback generation: A function that uses a generative AI model to generate feedback for each person being evaluated at the end of the period.

[1426] 4. Notification function: A function to notify the generated feedback to the device of the person being evaluated.

[1427] Generative AI Models

[1428] The generative AI model uses natural language processing technology to analyze the saved one-line comment data and generate appropriate feedback for each person being evaluated. This model provides comprehensive feedback by taking into account the content, frequency, importance, etc. of the comment.

[1429] Feedback Generation Process

[1430] The following describes how this system works, using specific examples.

[1431] Comment input

[1432] A user (e.g., Manager A) uses a terminal to input a brief comment about an automated work device (e.g., Robot A) in a factory into a feedback input application. For example, the user leaves a comment in the form of "Robot A's work speed seemed slow today."

[1433] Saving comment data

[1434] The comment data sent from the device is sent to a server, which stores the received comment data in a database and organizes it by adding metadata to each comment, such as the automated work device's ID, comment content, and timestamp.

[1435] Data training

[1436] The server periodically supplies the saved comment data to the AI ​​model for learning. The AI ​​model updates the feedback model for each person being evaluated, taking into account the content of the comments, the relationship between the evaluator and the automated work device, and the frequency of comments.

[1437] Specific examples

[1438] Examples of comment input:

[1439] Automated Work Device ID: RobotA

[1440] Comment: "You seem to be working a lot slower today."

[1441] Example prompt sentence:

[1442] Generate feedback for each robot based on the comment data below:

[1443] Robot ID: RobotA, Comment: "You seem to be working a lot slower today."

[1444] Examples of feedback generated include:

[1445] Robot A requires regular maintenance. It has recently been running slower, possibly due to wear on the belt.

[1446] In this way, users can receive comprehensive feedback about the automated work equipment in their factories, which can be used to improve maintenance and operation.

[1447] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1448] Step 1:

[1449] A user uses a terminal to input a brief comment about an automated work device in a factory into a feedback input application.

[1450] Input: ID of the automated work device, comment (e.g., "Robot A seemed to be working slowly today")

[1451] Output: Comment data is sent from the device to the server.

[1452] Step 2:

[1453] The comment data sent from the terminal is received by the server.

[1454] Input: Comment data sent from the device (automated work device ID, comment content, timestamp)

[1455] Output: Comment data is saved in the database.

[1456] Step 3:

[1457] The server periodically supplies the comment data stored in the database to the generative AI model, allowing it to learn.

[1458] Input: Comment data stored in the database

[1459] Output: The generative AI model learns from the comment data and updates the feedback model.

[1460] Step 4:

[1461] At the end of the term, the server uses a generative AI model to generate feedback content for each person being evaluated.

[1462] Input: Comment data trained by the generative AI model

[1463] Output: End-of-period feedback (e.g., "Robot A needs regular maintenance. It has been running slowly recently, which may be due to wear on the belt.")

[1464] Step 5:

[1465] The generated feedback content is notified from the server to the terminal of the person being evaluated.

[1466] Input: Generated feedback

[1467] Output: A feedback notification is displayed on the assessee's device.

[1468] Step 6:

[1469] The person being evaluated checks the generated feedback content using the terminal.

[1470] Input: Feedback content sent to the device of the person being evaluated

[1471] Output: The person being evaluated will view the feedback and use it as a reference for implementing any necessary improvements or measures.

[1472] 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.

[1473] This system provides multifaceted feedback on the person being evaluated. The system allows people who work with the person on a daily basis to enter short comments about the person being evaluated, which are then stored in a database. A generative AI model then learns from these comments and generates feedback content. Furthermore, by combining this with an emotion engine, the system analyzes the emotions in the comments and provides more accurate feedback.

[1474] System configuration

[1475] The system consists of four main components: a user terminal, a server, a generative AI model, and an emotion engine.

[1476] User terminal

[1477] The user terminal is a device that allows the evaluator to enter short comments about the person being evaluated, and can include PCs, smartphones, tablets, etc. A dedicated feedback input application is installed on the terminal, and a comment input screen is provided.

[1478] server

[1479] The server has the following functions:

[1480] 1. Database function: A function to save comment data received from users in a database.

[1481] 2. Sentiment analysis function: A function that uses an emotion engine to analyze the sentiment of one-line comments and classify them as positive, negative, or neutral.

[1482] 3. Managing generative AI models: A function that allows saved comment data and its sentiment classification results to be fed into generative AI models for learning.

[1483] 4. Feedback generation: A function that generates feedback for each person being evaluated at the end of the period using a generative AI model.

[1484] 5. Notification function: A function to notify the assessee of the generated feedback.

[1485] Generative AI Models

[1486] The generative AI model uses natural language processing technology to analyze the saved one-word comment data and emotion classification results, and generates appropriate feedback content for each person being evaluated. This model reflects the emotional information of the comments received from the emotion engine and provides comprehensive feedback.

[1487] Emotion Engine

[1488] The emotion engine analyzes the sentiment of short comments entered by users and categorizes them as positive, negative, or neutral. The emotion engine uses natural language processing technology to extract sentiment from the text and sends the results to the server.

[1489] Explaining the feedback generation process

[1490] The following describes how this system works, using specific examples.

[1491] Comment input

[1492] A user (e.g., colleague A) uses a terminal to input a short comment about the person being evaluated (e.g., colleague B) into a feedback input application. For example, the user might leave a comment in the form of, "Mr. B made a great proposal in today's meeting."

[1493] Saving comment data

[1494] Comment data sent from the device is sent to a server, which stores the received comment data in a database and organizes it by adding metadata to each comment, such as the ID of the person being evaluated, the comment content, and a timestamp.

[1495] Emotion Analysis

[1496] The server provides the stored comment data to the emotion engine for sentiment analysis. The emotion engine analyzes the content of each comment and classifies it as positive, negative, or neutral. For example, "Mr. B made a great suggestion" is classified as positive.

[1497] Data training

[1498] The server supplies the comment data and its emotion classification results to the generative AI model, which then learns from it. The generative AI model reflects the emotion classification results and updates the feedback model for each person being evaluated, taking into account the comment content, the relationship between the evaluator and the person being evaluated, the emotion and frequency of the evaluation, etc.

[1499] Generate feedback

[1500] At the end of the period, the server uses the generative AI model to generate feedback for each individual being evaluated. For example, it generates multifaceted, emotionally-charged feedback such as, "Mr. B made an important proposal in this project and made a significant contribution to the team. He also showed a proactive attitude toward seeking feedback."

[1501] Feedback Notification

[1502] The generated feedback is stored in a database and sent from the server to the device of the person being evaluated. The person being evaluated can then check the generated feedback using a dedicated feedback confirmation application on their device.

[1503] This allows those being evaluated to receive detailed, accurate feedback from all angles, which can be used to improve themselves. Furthermore, evaluators can enter detailed comments when needed, freeing them from the constraints of relying on memory or time constraints. Furthermore, the introduction of an emotion engine improves the balance and accuracy of feedback, providing a fair and multifaceted evaluation.

[1504] The processing flow will be explained below.

[1505] Processing flow of a feedback system that combines an emotion engine

[1506] Step 1:

[1507] Users can input short comments about the person being evaluated into a dedicated feedback input application during work or daily conversations. For example, colleague A might input, "Mr. B made a very useful suggestion in today's meeting."

[1508] Step 2:

[1509] The device organizes the input comment data as structured data and sends it to the server, adding metadata such as the user's ID, comment content, and timestamp.

[1510] Step 3:

[1511] The server saves the received comment data in the database in the following format: "Evaluator ID: B123, Comment content: 'Very useful suggestion', Timestamp: 2023-10-01 10:30".

[1512] Step 4:

[1513] The server provides the stored comment data to the emotion engine for sentiment analysis. The emotion engine analyzes the content of each comment and classifies it as positive, negative, or neutral. For example, "Mr. B made a great suggestion" would be classified as positive.

[1514] Step 5:

[1515] The server adds the emotion classification results obtained from the emotion engine to the comment data and saves it in the database. For example, "evaluator ID: B123, comment content: 'Very useful suggestion', emotion: positive" is saved.

[1516] Step 6:

[1517] The server periodically supplies all comment data and emotion classification results to the AI ​​model, allowing it to update and learn. For example, you can set it to automatically start AI training on weekends.

[1518] Step 7:

[1519] The generative AI model learns from the provided data, using natural language processing technology to analyze the content of comments, the relationship between the rater and the person being rated, the sentiment classification results, the sentiment of the ratings, frequency, etc.

[1520] Step 8:

[1521] As the end of the term approaches, the server again supplies all comment data and emotion classification results to the generative AI model and instructs it to generate feedback. For example, the server issues a data supply instruction one week before the end of the term.

[1522] Step 9:

[1523] The generative AI model generates appropriate feedback for each person being evaluated based on all stored comment data and emotion classification results. For example, it considers a balance of positive and negative comments and generates feedback such as, "Mr. B made important suggestions in this project and made a significant contribution to the team. He also showed an attitude of actively seeking feedback."

[1524] Step 10:

[1525] The server stores the generated feedback content in a database and simultaneously sends a notification to the terminal of the person being evaluated that the feedback has been generated.

[1526] Step 11:

[1527] The device receives a notification, and the person being evaluated can check the generated feedback through a dedicated feedback review application. Specifically, when the person being evaluated opens the application, a list of the feedback and its details are displayed.

[1528] This allows those being evaluated to receive detailed and accurate feedback from all angles, which can be used to improve their own performance. Furthermore, users (evaluators) can enter detailed comments whenever necessary, freeing them from the constraints of memory and time constraints. Furthermore, the introduction of an emotion engine improves the balance and accuracy of feedback, providing a fair and multifaceted evaluation.

[1529] Example 2

[1530] 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."

[1531] In conventional evaluation systems, evaluators rely on their memory to make evaluations, which can lead to problems with the accuracy and fairness of the evaluations. Furthermore, because the evaluation content is one-sided, it is difficult to grasp the overall picture of the person being evaluated's performance and behavior. Furthermore, evaluations are not based on emotions, which can lead to insufficient consideration of the evaluation content.

[1532] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1533] In this invention, the server includes: a means for people involved in daily work to input short comments about the person being evaluated; a means for saving the input short comments in a database; a means for analyzing the sentiment of the saved short comments and classifying them as positive, negative, or neutral; a means for generating feedback content using a generative AI model that learns short comments including the sentiment analysis results; and a means for notifying the person being evaluated of the generated feedback content at the end of the period. This improves the accuracy and fairness of the evaluation content and enables multifaceted evaluation of the person's performance and behavior. Furthermore, incorporating sentiment analysis allows the evaluation content to reflect emotional aspects.

[1534] A "shorthand comment" is a short piece of feedback entered by the evaluator about the person being evaluated during their daily work.

[1535] "Database" means an information storage system for storing and managing entered short comments and other related data.

[1536] "Sentiment analysis" is a technology that uses natural language processing technology to analyze the emotions contained in short comments and classify them as positive, negative, or neutral.

[1537] The "generative AI model" is an artificial intelligence model that learns from saved one-line comments and the results of their emotional analysis, and generates feedback content for each person being evaluated.

[1538] "Notification" is a system function for notifying the subject of evaluation of the generated feedback content.

[1539] The present invention is a system that provides multifaceted feedback to an individual by inputting a short comment about the individual involved in daily work, storing the comment in a database, performing sentiment analysis, generating feedback using a generative AI model, and notifying the individual. Each component of this system and its embodiment will be described in detail below.

[1540] System configuration

[1541] The system consists of four main components: a user terminal, a server, a generative AI model, and a sentiment analysis engine.

[1542] User terminal

[1543] The user terminal is a device on which the evaluator inputs a short comment about the person being evaluated. It can be a PC, smartphone, tablet, etc. A dedicated feedback input application is installed on the user terminal, and a comment input screen is provided. For example, the evaluator might input a comment in the form of "Mr. B made a great proposal in today's meeting."

[1544] server

[1545] The server has the following features:

[1546] 1. Database function: Comment data received from users is saved in a database. The database uses MySQL or PostgreSQL.

[1547] 2. Sentiment analysis function: Using an emotion engine, the sentiment of one-line comments is classified as positive, negative, or neutral. Sentiment analysis is performed using Python's VADER and TextBlob.

[1548] 3. Managing the generative AI model: The saved comment data and its sentiment classification results are fed into the generative AI model for training. The generative AI model uses a Transformer-based model such as GPT-3 or BERT.

[1549] 4. Feedback generation: At the end of the period, feedback content is generated for each person being evaluated using the generative AI model.

[1550] 5. Notification function: Notify the assessee of the generated feedback.

[1551] Generative AI Models

[1552] The generative AI model uses natural language processing technology to analyze the saved one-line comment data and sentiment analysis results, generating appropriate feedback for each individual being evaluated. The generative AI model is implemented using Python or other programming languages, utilizing machine learning frameworks such as TensorFlow and PyTorch. For example, it might generate feedback such as, "Mr. B made important suggestions in this project and made a significant contribution to the team. He also showed a proactive attitude toward seeking feedback."

[1553] Sentiment Analysis Engine

[1554] The sentiment analysis engine analyzes the sentiment of short comments entered by users and classifies them as positive, negative, or neutral. The sentiment analysis engine uses natural language processing technology to extract the sentiment of the text and sends the results to a server. Technologies used include VADER and TextBlob.

[1555] Examples of concrete examples and prompts

[1556] For example, if a user (colleague A) enters a comment such as "Mr. B made a great proposal in today's meeting," the server receives this comment and stores it in a database. It is then classified as "positive" using a sentiment analysis engine. The generative AI model generates feedback for the person being evaluated (colleague B) based on this data.

[1557] Examples:

[1558] Comment: "Mr. B made a great suggestion at today's meeting."

[1559] Sentiment Analysis: Positive

[1560] Feedback: "B made important suggestions on this project and contributed greatly to the team. He also proactively sought feedback."

[1561] Example prompt for a generative AI model:

[1562] "Generate the following feedback: 'Person B made a great suggestion in today's meeting.'"

[1563] In this way, the system according to the present invention can provide multifaceted and accurate feedback regarding the person being evaluated.

[1564] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1565] Step 1:

[1566] Comments entered by the user

[1567] Operation: The user opens a feedback application on a device (PC, smartphone, tablet, etc.). They access the comment input screen and enter a short comment about the person being evaluated. For example, they enter a comment such as, "Mr. B made a great proposal in today's meeting."

[1568] Input: Text comments about the person being assessed.

[1569] Output: The entered comment data is sent from the terminal to the server.

[1570] Step 2:

[1571] Sending and saving comment data to the server

[1572] Operation: Comment data sent from the device is sent to the server via an HTTP request. The server analyzes the received comment data, extracts the ID of the person being rated, the ID of the evaluator, the comment content, and the timestamp, and stores them in a database.

[1573] Input: Comment data sent from the device (text comment, ID of the person being evaluated, ID of the evaluator, timestamp).

[1574] Output: Comment data stored in a database.

[1575] Step 3:

[1576] Performing sentiment analysis

[1577] How it works: The server passes the comment data stored in the database to a sentiment analysis engine, which uses natural language processing libraries such as VADER or TextBlob to classify the sentiment of each comment as positive, negative, or neutral.

[1578] Input: Text comment stored in the database.

[1579] Output: Comment data with sentiment classification results (positive, negative, neutral).

[1580] Step 4:

[1581] Supplying and training generative AI models

[1582] How it works: The server supplies the sentiment-analyzed comment data to the generative AI model, which then uses natural language processing technology to update the feedback model for each subject based on the data provided.

[1583] Input: Comment data with sentiment classification results.

[1584] Output: An updated generative AI model.

[1585] Step 5:

[1586] Generate feedback

[1587] How it works: At the end of the term, the server uses the updated generative AI model to generate feedback for each person being evaluated. For example, it generates feedback in the form of, "Mr. B made important suggestions in this project and made a significant contribution to the team. He also showed a proactive attitude of seeking feedback."

[1588] Input: Updated generative AI model, ID of the person being evaluated.

[1589] Output: Generated feedback for each ratee.

[1590] Step 6:

[1591] Feedback Notification

[1592] Operation: The server notifies the subject of the generated feedback to the subject's device. The subject then checks the received feedback using a dedicated feedback checking application.

[1593] Input: Generated feedback content, ID of the person being assessed.

[1594] Output: Feedback sent to the subject's device.

[1595] Through these steps, users can provide multifaceted and accurate feedback to those being evaluated. Raters can easily enter comments, and sentiment analysis and generative AI models ensure detailed and fair feedback.

[1596] (Application example 2)

[1597] 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."

[1598] In conventional feedback systems, evaluations are often based on the subjective opinions of the evaluator or one-sided information, making it difficult to accurately grasp the overall growth and areas for improvement of the person being evaluated.Furthermore, because the emotional aspects of the comments are not taken into consideration, there are problems with biased feedback content and unfair evaluations.

[1599] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a means for a person who is involved in daily work to input a short comment about the person to be evaluated, a means for saving the input short comment in a database, a means for generating feedback content using a generative AI model that learns from the saved short comment, a means for analyzing the input comment with an emotion engine and classifying emotions, and a means for notifying the person to be evaluated of the generated feedback content at the end of the period. This enables fair feedback that is multifaceted and reflects emotions.

[1600] "People involved in daily work" refers to colleagues, superiors, subordinates, and other people who work with the person being evaluated on a daily basis.

[1601] "Evaluation target" refers to the person, employee, worker, etc. who is the subject of feedback.

[1602] A "shorthand comment" is a short feedback message about the person being appraised that describes a behavior or performance observed during work.

[1603] "Means of storing in a database" refers to the systems and technologies for systematically storing and managing the entered comment data.

[1604] A "generative AI model" refers to an artificial intelligence module that learns based on input data and generates feedback content.

[1605] "Means for generating feedback content" refers to the process of analyzing collected data using a generative AI model and creating appropriate feedback for each person being evaluated.

[1606] An "emotion engine" refers to technology that analyzes the emotional aspects of input comments and classifies them as positive, negative, neutral, etc.

[1607] "Sentiment classification" refers to the process of identifying and categorizing the emotional aspects of comments using an emotion engine.

[1608] "Means of notifying at the end of the period" refers to the mechanisms and techniques for communicating the generated feedback content to the person being evaluated at the end of the evaluation period.

[1609] The present invention is a system in which people who interact with the person being evaluated in their daily work enter short comments about the person being evaluated, save them in a database, and a generative AI model learns from them to generate feedback content. The system is configured as follows:

[1610] Hardware and Software

[1611] 1. User device: A device used by the user to enter a short comment about the person being evaluated. This can be a PC, smartphone, head-mounted display (HMD), etc. A dedicated feedback input application is installed and a comment input screen is provided.

[1612] 2. Server:

[1613] Database function: A function to save comment data received from users in a database.

[1614] Sentiment analysis function: A function that uses an emotion engine to analyze the sentiment of short comments and classify them as positive, negative, or neutral.

[1615] Managing generative AI models: A function that allows saved comment data and its sentiment classification results to be fed into generative AI models for training.

[1616] Feedback generation: A function that uses a generative AI model to generate feedback for each assessee.

[1617] Notification function: A function to notify the assessee of the generated feedback at the end of the period.

[1618] Processing flow and data calculation

[1619] 1. Comment entry:

[1620] A user (e.g., a factory worker) uses a terminal to input a short comment about the work of the person being evaluated (e.g., a robot) into a feedback input application. For example, the user might leave a comment in the form of, "Robot A carried the pallet smoothly today."

[1621] Example prompt: "In one sentence, tell us what you noticed about the robot's work."

[1622] 2. Storage of Comment Data:

[1623] Comment data sent from the device is sent to a server, which stores the received comment data in a database and organizes it by adding metadata to each comment, such as the ID of the person being evaluated, the comment content, and a timestamp.

[1624] 3. Emotion analysis:

[1625] The server provides the stored comment data to the emotion engine for emotion analysis. The emotion engine analyzes the content of each comment and classifies it as positive, negative, or neutral. For example, "Robot A handled the situation smoothly" is classified as positive.

[1626] Example prompt: "Analyze the sentiment of this comment and categorize it as positive, negative, or neutral."

[1627] 4. Data training:

[1628] The server supplies the comment data and its emotion classification results to the generative AI model, which then learns from it. The generative AI model reflects the emotion classification results and updates the feedback model for each person being evaluated, taking into account the comment content, the relationship between the evaluator and the person being evaluated, the emotion and frequency of the evaluation, etc.

[1629] 5. Feedback Generation:

[1630] At the end of the period, the server uses the generative AI model to generate feedback for each individual being evaluated, generating multifaceted and emotional feedback such as, "Robot A was recognized for its smooth movements in transporting pallets, and its efficiency has improved."

[1631] Example prompt: "Based on this sentiment analysis and comment data, generate feedback to improve the robot's performance."

[1632] 6. Feedback Notification:

[1633] The generated feedback is stored in a database and sent from the server to the device of the person being evaluated. The person being evaluated can then check the generated feedback using a dedicated feedback confirmation application on their device.

[1634] This will enable multifaceted, emotional feedback to be provided to robots in factories, which is expected to improve work efficiency and performance.

[1635] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1636] Step 1:

[1637] Comment input from user terminal:

[1638] A user (such as a factory worker) uses a device (such as a smartphone or head-mounted display) to input a comment about the work of the person being evaluated (such as a robot). The comment is sent to a feedback input application on the device. Example: "Robot A carried the pallet smoothly today."

[1639] Input: A short comment entered by the user

[1640] Output: Comment data sent from the device to the server

[1641] Step 2:

[1642] Comment data storage on server:

[1643] The submitted comment data is sent to a server and stored in a database. The server organizes the data by adding metadata to each comment, such as the user's ID, comment content, and timestamp.

[1644] Input: Comment data sent from the device

[1645] Output: Comment data stored in the database

[1646] Step 3:

[1647] On-server sentiment analysis:

[1648] The comment data stored on the server is fed to the emotion engine, which analyzes the content of the comment and classifies it as positive, negative, or neutral. For example, "Robot A handled the situation smoothly" is classified as positive.

[1649] Input: Comment data stored in the database

[1650] Output: Emotion classification result

[1651] Step 4:

[1652] Training generative AI models:

[1653] The server supplies the comment data and its emotion classification results to the generative AI model for learning. The generative AI model reflects the emotion classification results and updates the feedback model for each person being evaluated, taking into account the comment content, the relationship between the evaluator and the person being evaluated, the emotion and frequency of the evaluation, etc.

[1654] Input: Comment data and sentiment classification results

[1655] Output: Updated generative AI model

[1656] Step 5:

[1657] Generate feedback:

[1658] At the end of the period, the server uses the generative AI model to generate feedback for each assessee, such as "Robot A is recognized for its smooth movements in transporting pallets, and has improved efficiency," which reflects both multifaceted and emotional feedback.

[1659] Input: Updated generative AI model

[1660] Output: Generated feedback

[1661] Step 6:

[1662] Feedback Notification:

[1663] The generated feedback content is stored in a database and is also sent from the server to the device of the person being evaluated, who can then check the generated feedback content using a dedicated feedback checking application on their device.

[1664] Input: Generated feedback

[1665] Output: Feedback notification to the device of the person being evaluated

[1666] 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.

[1667] 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.

[1668] 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.

[1669] 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.

[1670] FIG. 9 illustrates 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 behaviors 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.

[1671] 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.

[1672] 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).

[1673] 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.

[1674] 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."

[1675] 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.

[1676] 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).

[1677] 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.

[1678] 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.

[1679] 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.

[1680] 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.

[1681] 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.

[1682] 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.

[1683] 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.

[1684] 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.

[1685] 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.

[1686] 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.

[1687] The following is further disclosed regarding the above embodiment.

[1688] (Claim 1)

[1689] A means for people involved in daily work to enter a short comment about the person being evaluated,

[1690] A means for saving the entered short comment in a database;

[1691] A means for generating feedback content using a generative AI model that learns from the saved short comments;

[1692] A means for notifying the assessee of the generated feedback at the end of the period;

[1693] A system including:

[1694] (Claim 2)

[1695] 2. The system of claim 1, wherein the generated feedback content includes individual feedback content for each person being evaluated.

[1696] (Claim 3)

[1697] 2. The system according to claim 1, wherein when generating feedback content for a person to be evaluated, weighting is performed based on the strength of involvement and frequency of comments.

[1698] "Example 1"

[1699] (Claim 1)

[1700] A means for a user to input a short comment about the person to be evaluated;

[1701] A means for transmitting the input comment from the terminal to a server;

[1702] A means for storing the received comment data in a database;

[1703] A means for the server to periodically supply the saved comment data to the AI ​​model for learning;

[1704] A means for the server to generate feedback content for each person being evaluated using the generated AI model at the end of the term;

[1705] a means for notifying the terminal of the person being evaluated of the generated feedback content;

[1706] A system including:

[1707] (Claim 2)

[1708] 2. The system of claim 1, wherein the generated feedback content includes individual feedback content for each person being evaluated.

[1709] (Claim 3)

[1710] 2. The system according to claim 1, wherein when generating feedback content for a person to be evaluated, weighting is performed based on the strength of involvement and frequency of comments.

[1711] "Application Example 1"

[1712] (Claim 1)

[1713] A means for people involved in daily work to enter a short comment about the person being evaluated,

[1714] A means for saving the entered short comment in a database;

[1715] A means for generating feedback content using a generative AI model that learns from the saved short comments;

[1716] A means for notifying the assessee of the generated feedback at the end of the period;

[1717] A means for inputting comments about the operating status and performance of an automated work device operating in a factory;

[1718] a means for evaluating the performance of the automated work device based on the input comments;

[1719] A system including:

[1720] (Claim 2)

[1721] 2. The system of claim 1, wherein the generated feedback content includes individual feedback content for each person being evaluated.

[1722] (Claim 3)

[1723] 2. The system according to claim 1, wherein when generating feedback content for a person to be evaluated, weighting is performed based on the strength of involvement and frequency of comments.

[1724] "Example 2: Combining Emotion Engines"

[1725] (Claim 1)

[1726] A means for people involved in daily work to enter a short comment about the person being evaluated,

[1727] A means for saving the entered short comment in a database;

[1728] A means of analyzing the sentiment of the saved short comments and classifying them as positive, negative, or neutral;

[1729] A means for generating feedback content using a generative AI model that learns one-line comments including the results of sentiment analysis;

[1730] A means for notifying the assessee of the generated feedback at the end of the period;

[1731] A system including:

[1732] (Claim 2)

[1733] 2. The system of claim 1, wherein the generated feedback content includes individual feedback content for each person being evaluated.

[1734] (Claim 3)

[1735] 2. The system according to claim 1, wherein when generating feedback content for a person to be evaluated, weighting is performed based on the strength of involvement and frequency of comments.

[1736] "Application example 2 when combining emotion engines"

[1737] (Claim 1)

[1738] A means for people involved in daily work to enter a short comment about the person being evaluated,

[1739] A means for saving the entered short comment in a database;

[1740] A means for generating feedback content using a generative AI model that learns from the saved short comments;

[1741] A means of analyzing the input comments using an emotion engine and classifying the emotions;

[1742] A means for notifying the assessee of the generated feedback at the end of the period;

[1743] A system including:

[1744] (Claim 2)

[1745] 2. The system of claim 1, wherein the generated feedback content includes individual feedback content for each person being evaluated.

[1746] (Claim 3)

[1747] 2. The system according to claim 1, wherein when generating feedback content for a person to be evaluated, weighting is performed based on the strength of involvement and frequency of comments. [Explanation of symbols]

[1748] 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 for people involved in daily work to enter a short comment about the person being evaluated, A means for saving the entered short comment in a database; A means for generating feedback content using a generative AI model that learns from the saved short comments; A means for notifying the assessee of the generated feedback at the end of the period; A system including:

2. The system according to claim 1 , wherein the generated feedback content includes individual feedback content for each person to be evaluated.

3. 2. The system according to claim 1, wherein when generating feedback content for a person to be evaluated, weighting is performed based on the strength of involvement and the frequency of comments.

Images