System

Abstract

A system is provided.SOLUTION: A system, comprising: means for collecting transaction data and meeting logs from terminals; means for pre-processing the collected transaction data and meeting logs; means for using a generative AI model to perform qualitative and quantitative assessments; and means for outputting assessment results from the generative AI model in a report format.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] The current evaluation system has the problem that superiors do not have sufficient visibility into the work content of employees, making it difficult to evaluate fairly. In particular, evaluations of specific activities such as what kind of work an employee does or what they say in meetings tend to be subjective, which is a factor in lowering employee satisfaction. For this reason, it is necessary to make the work achievements of each employee visible and to realize a fair and convincing evaluation system. [Means for solving the problem]

[0005] To solve this problem, the present invention proposes the following configuration: It includes a means for collecting business data and meeting logs from terminals and a means for preprocessing the collected business data and meeting logs. It also includes a means for analyzing the preprocessed data and using a generative AI model to perform qualitative and quantitative evaluations. The above problem is solved by creating a system that includes a means for outputting the evaluation results from the generative AI model in report format. This system further includes a means for providing supplementary materials that allow managers to use the evaluation results to perform management by objectives (MBO) evaluations, and a means for converting voice data into text data. This makes it possible to objectively evaluate employee work content and comments, realizing a fair and convincing evaluation system.

[0006] A "terminal" is a computer or device that a user uses to perform work.

[0007] "Business data" refers to the work that users do on their devices and related information, including application usage history, file access logs, and email content.

[0008] A "conference log" is data that includes a record of a video conference or audio conference and the content of what was said there.

[0009] "Preprocessing" refers to the process of converting the raw data collected into an analyzable form, which may include morphological analysis of text data or normalization of numerical data.

[0010] A "generative AI model" is an artificial intelligence that uses machine learning and deep learning to analyze data and generate evaluation results.

[0011] Qualitative evaluation is the process of analyzing the content and quality of collected data and quantifying subjective evaluation criteria, such as the impact of a statement or the effectiveness of a proposal.

[0012] Quantitative evaluation is the process of evaluating performance and work efficiency using collected numerical data, such as time spent on a task or frequency of application use.

[0013] A "report format" is a presentation format for documents and data that organizes and integrates analysis results in an easy-to-read format.

[0014] Management by Objectives (MBO) evaluation is an evaluation method in which managers evaluate employees' goal achievement and determine the next goal setting and compensation based on the results.

[0015] A "voice recognition tool" is software or hardware that analyzes voice data and converts it into text data. [Brief explanation of the drawings]

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

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

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

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

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

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

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

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

[0024] [First embodiment]

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

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

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

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

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

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

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

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

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

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

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

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

[0037] The present invention is a system that visualizes employees' work performance by performing qualitative and quantitative evaluations using a generative AI model using business data and meeting logs collected from terminals, thereby realizing a fair and convincing evaluation system. Specific embodiments for implementing the present invention are described below.

[0038] Data collection

[0039] The device continuously collects the user's business data and meeting logs. Specifically, it acquires business data such as keystrokes, application usage history, file access logs, and email content. It also collects audio and video recordings of video and audio conferences.

[0040] Examples:

[0041] For example, the terminal of user A records a log of the applications used throughout the day and saves the contents of statements made during a meeting as audio data.

[0042] Data Preprocessing

[0043] The server converts the collected raw data into an analyzable format. First, the voice data is converted into text data using a speech recognition tool. Next, the text data is subjected to morphological analysis using natural language processing (NLP) techniques to analyze grammatical structure and meaning. Numerical data is normalized and converted into a unified format.

[0044] Examples:

[0045] For example, audio data from a meeting can be converted into text using a transcription tool, and morphological analysis can be performed to make what was said easier to understand. Numerical data such as application usage time can also be converted into a unified format.

[0046] Analysis using generative AI models

[0047] The server inputs the preprocessed data into the generative AI model and performs qualitative and quantitative evaluations. Qualitative evaluations evaluate the content of text and the quality of speech in meetings. Quantitative evaluations use log data such as work time and application usage.

[0048] Examples:

[0049] For example, a generative AI model can extract meaningful comments about project progress from User A's emails and give them a high rating. At the same time, it can quantitatively evaluate User A's spending more than 40 hours per week on a specific project based on the amount of time worked.

[0050] Output of evaluation results

[0051] The server compiles the results of the analysis by the generative AI model into a report and provides it to the user (supervisor). The report includes qualitative evaluations (e.g., quality of comments) and quantitative evaluations (e.g., work time, statistics on applications used).

[0052] Examples:

[0053] The server generates a report and provides it to the manager, which includes a qualitative assessment of User A's project management skills and problem-solving abilities, as well as quantitative data showing User A's time contribution to a particular project.

[0054] Complementing MBO evaluation

[0055] The user (supervisor) performs an MBO evaluation based on the provided report, checks the validity of the evaluation content, makes adjustments as necessary, and provides feedback to the employee.

[0056] Examples:

[0057] The supervisor uses the analysis results of the generated AI model to conduct an annual evaluation of User A. User A's leadership and focus are particularly highly evaluated, and a high MBO score is awarded.

[0058] In this way, the system of the present invention clearly shows employees' work achievements and enables convincing and fair evaluations. This system also contributes to improving employee satisfaction by providing managers with objective data to complement MBO evaluations.

[0059] The processing flow will be explained below.

[0060] Step 1: Data collection

[0061] The device collects the user's business data and meeting logs. Specifically, it periodically records the user's application usage history, file access logs, email content, keystroke data, etc. It also collects audio and video recording data of video and audio conferences.

[0062] Specific operation:

[0063] The terminal starts the log collection agent and collects business data every 15 minutes.

[0064] Acquire recording data via the video conferencing tool's API.

[0065] Step 2: Data Preprocessing

[0066] The server preprocesses the collected raw data. First, it converts the voice data into text data using a speech recognition tool, then performs morphological analysis of the text data using natural language processing (NLP) techniques. Numerical data is normalized and converted into a unified format.

[0067] Specific operation:

[0068] The server inputs the audio data into a transcription tool and converts it into text data.

[0069] The server performs morphological analysis, analyzes the text data, and extracts grammatical structure and semantic information.

[0070] The server normalizes the numerical data and converts it into a unified data format.

[0071] Step 3: Analysis by generative AI model

[0072] The server inputs the preprocessed data into the generative AI model and performs qualitative and quantitative evaluations. The qualitative evaluation involves analyzing the sentiment and importance of the comments, while the quantitative evaluation involves calculating metrics such as work time and application usage frequency.

[0073] Specific operation:

[0074] The server inputs the text data extracted from the meeting log into the generative AI model.

[0075] The server uses the generative AI model to generate sentiment analysis and content evaluation scores for the comments.

[0076] The server inputs the numerical data into the AI ​​model and generates an evaluation score based on work time and frequency of application use.

[0077] Step 4: Output of evaluation results

[0078] The server compiles the evaluation results from the generative AI model into a report and provides it to the user (supervisor). The report includes a qualitative evaluation (quality of speech content) and a quantitative evaluation (task time, statistics on applications used).

[0079] Specific operation:

[0080] The server automatically generates a report based on the evaluation results from the generated AI model.

[0081] The server periodically displays the report on the manager's dashboard or sends it by email.

[0082] Step 5: Complementing the MBO evaluation

[0083] The user (supervisor) performs an MBO evaluation based on the provided report, checks the validity of the evaluation content, makes adjustments as necessary, and provides feedback to the employee.

[0084] Specific operation:

[0085] The user checks the report and completes the MBO evaluation sheet based on the evaluation results of the generated AI model.

[0086] The user conducts a feedback interview based on the evaluation results and shares the evaluation details with the employee.

[0087] Through the above steps, the system of the present invention makes employees' work achievements visible and realizes a fair and convincing evaluation system.

[0088] Example 1

[0089] 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."

[0090] Traditional evaluation systems have the problem of making it difficult to accurately and fairly evaluate employees' work performance. In particular, there is a lack of quantitative methods for qualitative evaluations and contributions in meetings, which often leads to biased evaluations and a sense of unfairness. There is also a lack of materials for managers to conduct effective management-by-objectives (MBO) evaluations.

[0091] 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.

[0092] In this invention, the server includes means for continuously collecting business data and meeting logs from user terminals, means for preprocessing the collected business data and meeting logs using a speech recognition tool and natural language processing technology, means for inputting the preprocessed data into a generative AI model to perform qualitative and quantitative evaluations, and means for outputting the evaluation results by the generative AI model in report format. This enables employees' work performance to be evaluated objectively and fairly, and provides managers with complementary information for conducting effective MBO evaluations.

[0093] A "terminal" is an electronic device that a user uses to collect business data and meeting logs.

[0094] "Business data" refers to information related to a user's business, such as keystrokes, application usage history, file access logs, and email content.

[0095] "Meeting Log" refers to records of a meeting, including audio and video recordings of video and audio conferences, speeches, and related information.

[0096] "Speech recognition tool" refers to software or services for converting voice data into text data.

[0097] "Natural language processing technology" refers to computer science technology for analyzing text data and performing morphological analysis and grammatical structure analysis.

[0098] "Generative AI model" refers to an artificial intelligence model that uses preprocessed data to perform qualitative and quantitative evaluations.

[0099] "Qualitative evaluation" refers to the process of evaluating the content of a written piece or the quality of what is said in a meeting.

[0100] "Quantitative evaluation" refers to the process of evaluating based on numerical data such as work time and application usage.

[0101] "Report format" refers to the form of a document that summarizes the evaluation results in an easy-to-understand format, and the output includes formats such as Excel and PDF.

[0102] "Management by Objectives (MBO) evaluation" refers to a management technique in which managers evaluate employees' performance based on the degree to which they achieve goals.

[0103] The present invention provides a system that realizes a fair and convincing evaluation system by collecting users' business data and meeting logs and performing evaluations using a generative AI model. Specific embodiments for implementing the present invention are described below.

[0104] Data collection

[0105] The device continuously collects the user's business data and meeting logs. The device is equipped with a keyboard hook and a file system mini-filter driver to collect the following data:

[0106] Keystrokes

[0107] Application Usage History

[0108] File Access Log

[0109] Email content

[0110] Recordings of video and audio conferences

[0111] Examples:

[0112] For example, User A's terminal records the usage logs of all applications used during work and saves speeches made during meetings as audio data.

[0113] Data Preprocessing

[0114] The server converts the collected raw data into an analyzable format, specifically by following these steps:

[0115] The voice data is converted into text data using a voice recognition tool (e.g., Google Speech-to-Text API).

[0116] The text data is morphologically analyzed using natural language processing techniques (e.g., spaCy, NLTK) to analyze grammatical structure and meaning.

[0117] Numerical data is normalized using scaling and normalization techniques.

[0118] Examples:

[0119] The audio data from the meeting is converted into text data using a transcription tool, and morphological analysis is performed to evaluate the quality of what is being said. The numerical data of application usage time is also normalized into a unified format.

[0120] Analysis using generative AI models

[0121] The server inputs the preprocessed data into a generative AI model (e.g., OpenAI GPT-4) and performs qualitative and quantitative evaluation. The specific steps of the evaluation are as follows:

[0122] In qualitative evaluation, text data is analyzed to evaluate the quality of comments made during meetings and the importance of email content.

[0123] Quantitative evaluation involves analyzing log data and evaluating work time and application usage.

[0124] Examples:

[0125] The generative AI model receives a prompt such as "Evaluate User A's comments during the meeting and analyze how constructive they were," and the model outputs the evaluation results. The model also analyzes work time data using a prompt such as "Calculate User A's weekly work hours and evaluate their productivity."

[0126] Output of evaluation results

[0127] The server compiles the results of the analysis by the generative AI model into a report and provides it to the user (supervisor). This report includes the following:

[0128] Qualitative assessment (e.g., quality of content)

[0129] Quantitative evaluation (e.g., work time, application usage statistics)

[0130] Examples:

[0131] The server generates a report and provides it to the manager, which includes a qualitative assessment of User A's project management skills and problem-solving abilities, as well as quantitative data showing User A's time contribution to a particular project.

[0132] Complementing MBO evaluation

[0133] The user (supervisor) performs a management by objectives (MBO) evaluation based on the provided report, checks the validity of the evaluation content, makes adjustments as necessary, and provides feedback to the employee.

[0134] Examples:

[0135] The supervisor uses the analysis results of the generated AI model to conduct an annual evaluation of User A. User A's leadership and focus are particularly highly evaluated, and a high MBO score is awarded.

[0136] In this way, the system of the present invention clearly shows employees' work achievements and achieves fair evaluations. This system provides managers with objective data to complement MBO evaluations, thereby contributing to improved employee satisfaction.

[0137] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0138] Step 1: Data collection

[0139] The device continuously collects business data and meeting logs. Specifically, it collects user keystrokes, application usage history, file access logs, email content, and audio and video conference recordings. The input is user interactions and meeting participation, and the output is log data stored on the device.

[0140] Specific behavior:

[0141] The device records keystrokes using keyboard hooks, captures application usage history and file access logs using a dedicated service, and, when audio and video conferences are taking place, captures conference logs using dedicated audio and video recording software.

[0142] Step 2: Data Preprocessing

[0143] The server converts the collected raw data into an analyzable format. Specifically, it converts voice data into text data using a speech recognition tool (e.g., Google Speech-to-Text API), and then performs morphological analysis on the text data using natural language processing technology (e.g., spaCy, NLTK). Numerical data is normalized using scaling and normalization techniques. The input is the collected log data, and the output is preprocessed, analyzable data.

[0144] Specific behavior:

[0145] The server sends the audio file to a speech recognition API to obtain the transcribed text, then feeds the text data into an NLP tool for morphological analysis, and normalizes the numerical data using an appropriate scaling algorithm.

[0146] Step 3: Analysis by generative AI model

[0147] The server inputs the preprocessed data into a generative AI model (e.g., OpenAI GPT-4) to perform qualitative and quantitative evaluations. Qualitative evaluations assess the quality of comments made during meetings and the importance of email content, while quantitative evaluations assess task time and application usage. The input is the preprocessed data, and the output is the evaluation results.

[0148] Specific behavior:

[0149] The server inputs a prompt to the generative AI model, such as "Evaluate User A's comments during the meeting and analyze how constructive they were," and obtains a response from the model. It also analyzes the work time data using a prompt, such as "Calculate User A's weekly work hours and evaluate their productivity."

[0150] Step 4: Output of evaluation results

[0151] The server compiles the evaluation results analyzed by the generative AI model into a report. Specifically, it integrates qualitative evaluations (e.g., quality of speech) and quantitative evaluations (e.g., work time, statistics on applications used) to create a report. The input is the analyzed evaluation results, and the output is the report.

[0152] Specific behavior:

[0153] The server receives the output of the generative AI model and creates a report in formats such as Excel or PDF, which is visually organized using graphs and tables so that administrators can easily understand it.

[0154] Step 5: Complementing the MBO evaluation

[0155] The user (supervisor) performs a management by objectives (MBO) evaluation based on the provided report. Specifically, the user checks the validity of the evaluation content and makes adjustments as necessary. Finally, the user provides feedback to the employee. The input is the generated report, and the output is the revised evaluation and feedback.

[0156] Specific behavior:

[0157] The supervisor reviews the report and adds comments such as, "User A's leadership is evident, but we would like to see more specific examples." In addition, the supervisor discusses with the employee specific evaluation items and proposes the next steps for improvement.

[0158] (Application example 1)

[0159] 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."

[0160] At the workplace, it is important to accurately evaluate the performance and contribution of workers and to create a fair and transparent evaluation system. Factory floors, in particular, require a system that efficiently collects and analyzes work logs and robot interaction data, and evaluates them qualitatively and quantitatively.

[0161] 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.

[0162] In this invention, the server includes means for collecting business data and meeting logs from terminals, means for preprocessing the collected business data and meeting logs, means for analyzing the preprocessed data and using a generative AI model for qualitative and quantitative evaluation, means for outputting the evaluation results by the generative AI model in report format, means for collecting operation data and work logs of robots in the factory, means for collecting interaction data with workers, and means for analyzing the collected data and evaluating the performance of workers. This makes it possible to clarify the work achievements of workers and provide convincing and fair evaluations.

[0163] "Device" means a computer device used to collect and process data.

[0164] "Business data" refers to information such as applications that users operate during work, files that they access, and communication content.

[0165] "Meeting logs" are audio and video recordings related to statements made during a meeting and the agenda, as well as textual versions of such recordings.

[0166] "Preprocessing" is the process of converting collected raw data into an analyzable format.

[0167] A "generative AI model" is an artificial intelligence model that uses collected data to perform qualitative and quantitative evaluations.

[0168] "Qualitative evaluation" is the process of evaluating the content of a written piece or the quality of what is said in a meeting.

[0169] "Quantitative evaluation" is a process of evaluation based on numerical data such as log data and work time.

[0170] "Report format" refers to the format of a document summarizing the evaluation results.

[0171] "Factory robots" are automated machinery used to streamline factory work.

[0172] "Operation data" refers to records and log data relating to the robot's operations.

[0173] A "work log" is recorded data on the actions and work details of workers.

[0174] "Interaction data" is recorded data of the operations and communication that take place alternately between workers and robots.

[0175] "Evaluation results" refer to the qualitative and quantitative evaluation values ​​and evaluation content calculated by the generative AI model.

[0176] To implement this invention, a terminal first collects business data and a meeting log. Business data includes information such as the operation history of applications used by the user, files accessed, and emails sent and received. The meeting log includes audio and video data related to comments made during the meeting and the agenda.

[0177] The collected data is sent to a server where it undergoes preprocessing. During the preprocessing stage, audio data is converted into text data using a transcription tool. Next, natural language processing (NLP) techniques are used to perform morphological analysis of the text data, analyzing its grammatical structure and meaning. Numerical data is normalized and converted into a unified format.

[0178] Once preprocessed, the data is input into a generative AI model for qualitative and quantitative evaluation. In qualitative evaluation, the generative AI model evaluates the content of text and the quality of comments made in meetings. For example, it extracts meaningful comments about project progress from the contents of a user's email. In quantitative evaluation, it performs evaluation based on log data and work time. For example, it measures the total amount of time a user spends on a specific project.

[0179] The evaluation results are compiled in a report format by the server and provided to the user (manager). This report includes both qualitative and quantitative evaluations. Based on this report, the manager can obtain objective information for performing a management by objectives (MBO) evaluation. Using this information, the manager can check the validity of the evaluation content and make adjustments as necessary.

[0180] Furthermore, the system can collect operation data and work logs of robots in factories and analyze interaction data with workers, which will further improve the efficiency of factory work and enable fair evaluations.

[0181] As a concrete example, a robot in a factory collects speech data from workers, converts it into text data, performs morphological analysis, evaluates the quality of the speech, quantitatively evaluates the worker's performance using work log data, and integrates this data to provide the manager with the evaluation results in the form of a report.

[0182] An example of a prompt sentence to input to the generative AI model is, "Please summarize what was said in this meeting and extract the most important points." Using this prompt allows for efficient qualitative evaluation of meeting logs.

[0183] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0184] Step 1:

[0185] The device collects business data and meeting logs. Business data includes the operation history of applications used by the user, files accessed, and emails sent and received. Meeting logs include audio and video recordings related to comments made during meetings and the agenda. This data is obtained using software (e.g., data collection API). Input is the user's activity log and meeting audio, and output is the collected raw data.

[0186] Step 2:

[0187] The terminal sends the collected data to the server. The address of the server is specified and the data is sent in encrypted format via the required protocol (e.g. HTTP / HTTPS). The input is the raw data acquired, and the output is the data sent to the server.

[0188] Step 3:

[0189] The raw data received by the server is preprocessed. First, the voice data is converted into text data using a speech recognition tool (e.g., Python's speech_recognition library). Next, the text data is subjected to morphological analysis using natural language processing (NLP) techniques (e.g., the spacy library) to analyze grammatical structure and meaning. Numerical data is converted into a data frame and normalized. The input is raw data, and the output is preprocessed data.

[0190] Step 4:

[0191] The preprocessed data is input into a generative AI model. The generative AI model (e.g., GPT-2 from Hugging Face) analyzes the text data and performs a qualitative evaluation. An example of a prompt used here is, "Please summarize what was said in this meeting and extract the most important points." The input is the preprocessed text data and the prompt, and the output is the result of the qualitative evaluation.

[0192] Step 5:

[0193] The preprocessed numerical data is used to perform quantitative evaluation. The server analyzes the data frame and calculates statistics on work time and application usage time. For example, it calculates the time spent on a specific project based on work time. The input is the preprocessed numerical data, and the output is the results of the quantitative evaluation.

[0194] Step 6:

[0195] The server integrates the results of the qualitative and quantitative evaluations and generates a report. The report includes qualitative analysis results and quantitative statistics. It displays a detailed analysis of the content of the text, the quality of the comments, and the working time. The input is the results of the qualitative and quantitative evaluations, and the output is a report-format file.

[0196] Step 7:

[0197] The server provides the generated report to the user (administrator). The report is made accessible in dashboard or PDF format so that the administrator can easily check the evaluation. The input is the generated report, and the output is the report used by the user in the actual evaluation process.

[0198] 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.

[0199] The present invention is a system that visualizes employees' work performance by performing qualitative and quantitative evaluations using a generative AI model using business data and meeting logs collected from terminals, thereby realizing a fair and convincing evaluation system. Furthermore, the present invention improves the accuracy and comprehensiveness of evaluations by combining it with an emotion engine that recognizes the user's emotions. Specific embodiments for implementing the present invention are described below.

[0200] Data collection

[0201] The device continuously collects the user's business data and meeting logs. Specifically, it acquires business data such as the user's application usage history, file access logs, email content, and keystroke data. It also collects audio and video recordings of video and audio conferences.

[0202] Examples:

[0203] For example, the terminal of user A records a log of the applications used throughout the day and saves the contents of statements made during a meeting as audio data.

[0204] Data Preprocessing

[0205] The server converts the collected raw data into an analyzable format. First, the voice data is converted into text data using a speech recognition tool. Next, the text data is subjected to morphological analysis using natural language processing (NLP) techniques to analyze grammatical structure and meaning. Numerical data is normalized and converted into a unified format. An emotion engine is also implemented at this stage.

[0206] Examples:

[0207] For example, audio data from a meeting is converted into text data using a transcription tool, and morphological analysis is performed to make what is being said easier to understand. Numerical data such as application usage time is also converted into a unified format. An emotion engine analyzes the user's emotions from the text data.

[0208] Analysis using generative AI models

[0209] The server inputs the preprocessed data into a generative AI model for qualitative and quantitative evaluation. Qualitative evaluation evaluates the content of text and the quality of speech in meetings. The emotion engine also analyzes the user's emotional state and incorporates it into the evaluation. Quantitative evaluation uses log data such as work time and application usage.

[0210] Examples:

[0211] For example, a generative AI model extracts meaningful comments about project progress from User A's email content and gives a high rating. At the same time, it quantitatively evaluates that User A is spending more than 40 hours per week on a specific project based on the amount of time worked. An emotion engine analyzes the emotions in User A's email and reflects their emotional state, such as "positive" or "negative," in the rating.

[0212] Output of evaluation results

[0213] The server compiles the results of the analysis by the generative AI model and emotion engine into a report and provides it to the user (supervisor). The report includes qualitative evaluations (quality of speech content, emotional state) and quantitative evaluations (work time, statistics on applications used).

[0214] Examples:

[0215] The server generates a report and provides it to the manager, which includes a qualitative assessment of User A's project management skills and problem-solving abilities, as well as quantitative data showing his time contribution to a specific project. The report also includes an emotional assessment based on the emotion engine.

[0216] Complementing MBO evaluation

[0217] The user (supervisor) performs an MBO evaluation based on the provided report, checks the validity of the evaluation content, makes adjustments as necessary, and provides feedback to the employee.

[0218] Examples:

[0219] The supervisor uses the analysis results of the generative AI model and the emotion engine to conduct an annual evaluation of User A. In particular, User A's leadership and focus are highly evaluated, and a high MBO score is awarded. User A's emotional state is also taken into consideration, and the evaluation content is adjusted accordingly.

[0220] In this way, the system of the present invention clearly shows the employee's work achievements and enables convincing and fair evaluations. At the same time, the introduction of an emotion engine enables comprehensive evaluations that take into account the employee's emotional state.

[0221] The processing flow will be explained below.

[0222] Step 1: Data collection

[0223] The device collects the user's business data and conference logs. Specifically, it acquires business data such as the user's application usage history, file access logs, email content, and keystroke data. It also collects audio and video recordings of video and audio conferences.

[0224] Specific operation:

[0225] The terminal starts the log collection agent and collects business data every 15 minutes.

[0226] Acquire recording data via the video conferencing tool's API.

[0227] Step 2: Data Preprocessing

[0228] The server preprocesses the collected raw data. First, the voice data is converted into text data using a voice recognition tool. Next, the text data undergoes morphological analysis using natural language processing (NLP) techniques. Numerical data is normalized and converted into a unified format. An emotion engine analyzes the user's emotions from the text data.

[0229] Specific operation:

[0230] The server inputs the audio data into a transcription tool and converts it into text data.

[0231] The server performs morphological analysis, analyzes the text data, and extracts grammatical structure and semantic information.

[0232] The server normalizes the numerical data and converts it into a unified data format.

[0233] The emotion engine analyzes the user's emotional state from the text data.

[0234] Step 3: Analysis by generative AI models and emotion engines

[0235] The server inputs the preprocessed data into the generative AI model and emotion engine to perform qualitative and quantitative evaluations. Qualitative evaluation evaluates the content of text and the quality of speech in meetings. The emotion engine also analyzes the user's emotional state and incorporates it into the evaluation. Quantitative evaluation uses log data such as work time and application usage.

[0236] Specific operation:

[0237] The server inputs the text data extracted from the meeting log into the generative AI model.

[0238] The server uses the generative AI model to generate sentiment analysis and content evaluation scores for the comments.

[0239] The server inputs the numerical data into the AI ​​model and generates an evaluation score based on work time and frequency of application use.

[0240] The emotion engine analyzes the user's text data and reflects their emotional state in the evaluation score.

[0241] Step 4: Output of evaluation results

[0242] The server compiles the evaluation results from the generative AI model and emotion engine into a report and provides it to the user (supervisor). The report includes qualitative evaluations (quality of speech content, emotional state) and quantitative evaluations (work time, statistics on applications used).

[0243] Specific operation:

[0244] The server automatically generates a report based on the evaluation results from the AI ​​model and emotion engine.

[0245] The server periodically displays the report on the manager's dashboard or sends it by email.

[0246] Step 5: Complementing the MBO evaluation

[0247] The user (supervisor) performs an MBO evaluation based on the provided report, checks the validity of the evaluation content, makes adjustments as necessary, and provides feedback to the employee.

[0248] Specific operation:

[0249] The user reviews the report and completes the MBO evaluation sheet based on the evaluation results of the generative AI model and emotion engine.

[0250] The user conducts a feedback interview based on the evaluation results and shares the evaluation details with the employee.

[0251] Through these steps, the system of the present invention visualizes employees' work performance and realizes a fair and convincing evaluation system. In addition, the introduction of an emotion engine enables comprehensive evaluation that takes into account the emotional aspects of employees.

[0252] Example 2

[0253] 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."

[0254] Conventional performance evaluation systems often evaluate employees' work performance qualitatively, making it difficult to base evaluations on concrete data. They also lack comprehensive evaluations that take emotional aspects into account, which can lead to evaluations that lack fairness and satisfaction. Furthermore, insufficient pre-processing of meeting logs and work data often reduces the accuracy of evaluation results.

[0255] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for collecting business data and meeting logs from terminals, a means for preprocessing the collected business data and meeting logs, a means for analyzing the preprocessed data and using a generative AI model for qualitative and quantitative evaluation, a means for outputting the evaluation results by the generative AI model in report format, and a means for recognizing emotions from text data using an emotion engine. This enables a fair and comprehensive evaluation of employees' work performance based on specific data and taking into account their emotional aspects.

[0256] A "terminal" is a computer or mobile device used by a user, and is a device for collecting business data and meeting logs.

[0257] "Business data" refers to all data related to a user's business activities, including application usage history, file access logs, email content, keystroke data, and the like.

[0258] A "meeting log" refers to recorded data of a meeting, including audio and video, and is data that captures the content of statements and discussions made during the meeting.

[0259] "Collection means" refers to the method or function for acquiring business data and meeting logs from a terminal, and takes the form of software or hardware.

[0260] "Preprocessing" refers to the process of converting raw data into an analyzable format, and includes transcribing audio data, morphological analysis of text data, and normalizing numerical data.

[0261] A "generative AI model" refers to an artificial intelligence model that has functions such as response generation and data analysis, and is used here to perform qualitative and quantitative evaluations.

[0262] An "emotion engine" refers to a program or tool that recognizes and analyzes a user's emotional state from text data.

[0263] "Report format" refers to a format that visually summarizes the evaluation results from the generative AI model and emotion engine in an easy-to-understand manner, and includes formats such as PDF and web dashboards.

[0264] "Manager" refers to the person or position with the authority to conduct MBO evaluations based on the evaluation results, and is usually a supervisor or human resources officer.

[0265] "MBO" is an abbreviation for Management by Objectives, a management method that evaluates employees based on the degree to which they have achieved their goals.

[0266] The present invention provides a system that visualizes employees' work performance by performing qualitative and quantitative evaluations using a generative AI model using business data and meeting logs collected from terminals, thereby realizing a fair and convincing evaluation system. The present invention also provides a system that improves the accuracy and comprehensiveness of evaluations by combining it with an emotion engine that recognizes user emotions. Specific embodiments for implementing the present invention are described below.

[0267] Data collection

[0268] The device continuously collects the user's business data and meeting logs. Specifically, it acquires business data such as the user's application usage history, file access logs, email content, and keystroke data. It also collects audio and video recordings of video and audio conferences.

[0269] Examples:

[0270] For example, the terminal of user A records a log of the applications used throughout the day and saves the contents of statements made during a meeting as audio data.

[0271] Data Transfer

[0272] The device sends the collected data to the server in real time. The data collection program in the device periodically uploads the locally stored data to the server. For security reasons, the data is sent in encrypted form.

[0273] Data Preprocessing

[0274] The server converts the received raw data into an analyzable format. First, the voice data is converted into text data using a speech recognition tool (e.g., Google Cloud Speech-to-Text). Next, the text data is subjected to morphological analysis using natural language processing (NLP) technology (e.g., Spacy) to analyze grammatical structure and meaning. Numerical data is normalized from Excel or CSV format and converted into a unified format (e.g., JSON format). Finally, the text data is analyzed for sentiment using the API of an emotion engine (e.g., Microsoft Azure Text Analytics).

[0275] Examples:

[0276] For example, audio data from a meeting is transcribed using Google Cloud Speech-to-Text, and morphological analysis is performed using Spacy to make what is being said easier to understand. Numerical data such as application usage time is converted from Excel or CSV format into a unified format. An emotion engine analyzes user emotions from the text data.

[0277] Analysis using generative AI models

[0278] The server inputs the preprocessed data into a generative AI model (e.g., OpenAI GPT-3) to perform qualitative and quantitative evaluations. Qualitative evaluations evaluate the content of text and the quality of speech in meetings. An emotion engine also analyzes the user's emotional state and incorporates it into the evaluation. Quantitative evaluations use log data such as work time and application usage.

[0279] Examples:

[0280] For example, a generative AI model extracts meaningful comments about project progress from User A's email content and gives a high rating. At the same time, it quantitatively evaluates that User A is spending more than 40 hours per week on a specific project based on the amount of time worked. An emotion engine analyzes the emotions in User A's email and reflects their emotional state, such as "positive" or "negative," in the rating.

[0281] Example prompt sentence:

[0282] "Evaluate User A's important contributions to the project progress based on his emails and meeting comments over the past month."

[0283] "Quantitatively evaluate User A's time contribution to a specific project based on his / her weekly work time data."

[0284] Sentiment analysis results integration

[0285] The server uses an emotion engine to integrate the results of the emotion analysis with the evaluation from the generative AI model. The emotion analysis results are combined with the output of the AI ​​model to reflect positive and negative comments in the evaluation score. Detailed results of the emotion analysis are also added to the report.

[0286] Generating evaluation results

[0287] The server compiles the generated evaluation results into a report. The server uses a report generation tool (e.g., JasperReports) to create a detailed report that includes both qualitative and quantitative evaluations. The report is output in PDF format or web dashboard format.

[0288] Providing reports

[0289] The server provides the completed report to the user (supervisor). The report is automatically sent to the supervisor's email address periodically (for example, at the end of the month). The report is also uploaded to the company portal so that the user can access it.

[0290] Complementing MBO evaluation

[0291] The user (supervisor) performs an MBO evaluation based on the provided report. The supervisor reviews the report and provides additional feedback or supplementary information as needed. Once the evaluation is complete, a session is set up to provide feedback to the employee being evaluated.

[0292] In this way, each processing step of the system works together in a concrete way to clarify employees' work achievements and achieve fair and convincing evaluations. The introduction of an emotion engine makes it possible to provide comprehensive evaluations that take into account employees' emotional states.

[0293] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0294] Step 1: Data collection

[0295] The terminal collects the user's business data and meeting logs.

[0296] Input: User application usage history, file access logs, email content, keystroke data, and audio and video conference recordings.

[0297] Data processing: Raw data is automatically acquired and collected 24 / 7.

[0298] Output: Locally stored business data and meeting logs.

[0299] Specific operation: The device will periodically record application logs and file access logs, and will automatically start recording when an audio conference is started.

[0300] Step 2: Data Transfer

[0301] The data collected by the device is sent to the server in real time.

[0302] Input: Business data and meeting logs stored locally on the device.

[0303] Data processing: Encrypt the data and upload it securely to the server.

[0304] Output: Aggregated business data and meeting logs on the server.

[0305] Specific operation: The data collection program in the device periodically (for example, every hour) encrypts the data and uploads it to the server.

[0306] Step 3: Data Preprocessing

[0307] The server converts the raw data it receives into a format that can be analyzed.

[0308] Input: Raw data (business data, meeting logs) aggregated on the server.

[0309] Data processing: Convert the voice data into text data using Google Cloud Speech-to-Text, perform morphological analysis on the text data using Spacy, normalize the numerical data and convert it into a unified format (e.g., JSON), and analyze emotions from the text data using an emotion engine.

[0310] Output: Data converted into an analyzable format (text data, morphological analysis results, normalized numerical data, sentiment analysis results).

[0311] How it works: The server transcribes the audio data using Google Cloud Speech-to-Text, then performs morphological analysis using Spacy. Numerical data is converted from Excel or CSV to JSON format, and sentiment analysis is performed using a sentiment analysis tool.

[0312] Step 4: Data integration and storage

[0313] The server consolidates the pre-processed data and stores it in a database.

[0314] Input: Preprocessed data (text data, morphological analysis results, normalized numerical data, sentiment analysis results).

[0315] Data processing: Integrate the data, classify it by user, and store it in a NoSQL database (e.g., MongoDB).

[0316] Output: Data integrated and stored in a database.

[0317] Specific operation: The server inserts the preprocessed data into MongoDB, classifies it by user, and creates indexes to enable efficient searches.

[0318] Step 5: Analysis by generative AI model

[0319] The server inputs the preprocessed data into a generative AI model for qualitative and quantitative evaluation.

[0320] Input: Preprocessed data extracted from the database.

[0321] Data processing: A generative AI model (e.g., OpenAI GPT-3) is invoked with a prompt sentence to perform qualitative evaluation (text content and speech quality) and quantitative evaluation (task time, application usage).

[0322] Output: Analysis results from the generative AI model.

[0323] Specific operation: The server inputs data into the OpenAI GPT-3 model using the prompt sentence and generates an appropriate evaluation score and comment.

[0324] Step 6: Integrating the results of sentiment analysis

[0325] The server uses an emotion engine to integrate the results of emotion analysis with the evaluation from the generative AI model.

[0326] Input: Analysis results from generative AI model, sentiment analysis results.

[0327] Data processing: The results of sentiment analysis are combined with the output of the AI ​​model and reflected in the evaluation score.

[0328] Output: Evaluation results integrated with sentiment analysis results.

[0329] Specific operation: The server adds the results from the emotion engine to the output of the generative AI model, and reflects positive or negative emotions in the evaluation score.

[0330] Step 7: Generate evaluation results

[0331] The server compiles the generated evaluation results in a report format.

[0332] Input: Evaluation results integrated with sentiment analysis results.

[0333] Data processing: Create reports using a report generation tool (e.g., JasperReports) and convert them into PDF or web dashboard formats.

[0334] Output: Assessment report in PDF format and web dashboard format.

[0335] What it does: The server compiles the results of the assessment into a detailed report and converts it into a visually understandable format.

[0336] Step 8: Providing the report

[0337] The server provides the completed report to the user (supervisor).

[0338] Input: Assessment report in PDF format and web dashboard format.

[0339] Data Processing: Present the report to your superior in an appropriate manner.

[0340] Output: Evaluation report provided to supervisor.

[0341] What it does: The server sends the report to the manager's email address at the end of the month and also uploads it to the company's internal portal for access.

[0342] Step 9: Complementing the MBO evaluation

[0343] The user (supervisor) performs an MBO evaluation based on the provided report.

[0344] Input: Evaluation report provided to supervisor.

[0345] Data processing: Check the evaluation content and provide any necessary feedback or supplementary information.

[0346] Output: Completed MBO evaluation and feedback.

[0347] What happens next: The manager reviews the report, notes any necessary corrections or additional information, and schedules a feedback session.

[0348] In this way, the system of the present invention evaluates employees' work performance in a concrete and data-based manner, and achieves a fair and comprehensive evaluation that also takes into account their emotional aspects.

[0349] (Application example 2)

[0350] 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."

[0351] Existing evaluation systems do not adequately perform qualitative and quantitative evaluations based on work data and meeting logs. This makes it difficult to fairly evaluate employees' work performance and the operational efficiency of production equipment. Another problem is the lack of systems that can perform comprehensive evaluations using robot operation logs and performance indicators. Furthermore, it is necessary to improve the accuracy and comprehensiveness of evaluations by combining an emotion engine.

[0352] The identification processing 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 means for collecting business data and meeting logs from terminals, means for preprocessing the collected business data and meeting logs, means for analyzing the preprocessed data and using a generative AI model for qualitative and quantitative evaluation, means for outputting the evaluation results by the generative AI model in report format, means for collecting operation log data and performance indicators of the production equipment, means for converting the collected operation log data and performance indicators into an analyzable format, and means for qualitatively evaluating the operation log data and quantitatively evaluating the performance indicators using the generative AI model and an emotion recognition engine. This not only enables fair and convincing evaluation based on the business data and meeting logs, but also enables comprehensive evaluation of the operational efficiency of the production equipment.

[0353] A "terminal" is an electronic device such as a computer or smartphone used by a user, and is used to collect business data and meeting logs.

[0354] "Business data" refers to records such as application usage history, file access logs, email content, and keystroke data that are generated when a user performs a task.

[0355] "Meeting logs" refer to audio and video recordings of video and audio conferences, including audio data converted into text data and minutes of meetings.

[0356] "Preprocessing" is the process of converting collected raw data into an analyzable format, and includes transcribing audio data, morphological analysis of text data, and normalizing numerical data.

[0357] A "generative AI model" is a model that uses machine learning algorithms to analyze collected data and perform qualitative and quantitative evaluations.

[0358] An "emotion recognition engine" is a technology for analyzing a user's emotional state from text data and voice data, and reflects "positive," "negative," and changes therein in the evaluation.

[0359] "Production equipment" refers to equipment such as robots used in factories and production lines, and is the subject of collection of operation log data and performance indicators.

[0360] "Operation log data" refers to data that includes the operation history and operation records of robots and production equipment, and mainly includes detailed records of the production process.

[0361] "Performance index" refers to the key performance metrics, such as the efficiency or error rate achieved by a robot or production equipment, expressed in specific numerical values.

[0362] "Qualitative evaluation" is a type of evaluation performed by a generative AI model, and is a qualitative evaluation that is primarily based on the content of statements and operation logs.

[0363] "Quantitative evaluation" is a type of evaluation performed by a generative AI model, and is a quantitative evaluation based primarily on numerical data such as work time and performance indicators.

[0364] This invention provides a system for fair and comprehensive evaluation of the operational efficiency of production equipment and the work performance of employees. The system includes a terminal, a server, a generative AI model, and an emotion recognition engine.

[0365] First, the terminal continuously collects operation log data and performance indicators of production equipment. For example, it acquires performance data such as operation logs, work efficiency, and error rates of robots installed in a factory. This allows the operation history and operation records of production equipment, as well as key performance metrics, to be collected.

[0366] Next, the server preprocesses the collected data. At this stage, the collected raw data is converted into an analyzable format. Specifically, the operation log data is subjected to morphological analysis, and the performance indicator data is normalized. If voice data is included, it is converted into text data using a voice recognition tool.

[0367] The preprocessed data is input into a generative AI model, which performs both qualitative and quantitative evaluations. In qualitative evaluation, meaningful utterances and behavioral patterns are extracted from the operation log data, and the emotional state is analyzed using an emotion recognition engine. In quantitative evaluation, evaluation is performed based on performance indicators such as work efficiency and error rate.

[0368] The server outputs the analysis results of the generative AI model and emotion recognition engine in the form of a report. This report includes a qualitative evaluation of the robot's operation and emotional state, as well as a quantitative evaluation of each performance indicator. This allows administrators to conduct comprehensive and fair evaluations.

[0369] As a specific example, the operation log of the robot 123 is acquired as "log data of robot 123", and the evaluation results include quantitative and qualitative evaluations as follows:

[0370] Qualitative evaluation: Extract meaningful operation patterns from the robot's operation log and give a positive evaluation using an emotion engine.

[0371] Quantitative evaluation: Since the work efficiency is 95% and the error rate is 2%, the evaluation is "Efficiency: High" and "Error rate: Low."

[0372] Example prompt sentence:

[0373] "Collect the operation log data and performance metrics of Robot 123 and evaluate it using the generative AI model and emotion engine."

[0374] This system allows for a comprehensive evaluation of production equipment and employee performance, resulting in rational and convincing evaluations.

[0375] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0376] Step 1:

[0377] The terminal collects operation log data and performance indicators of production equipment. Specifically, data is acquired using sensors and controllers that record the operation history, operation records, work efficiency, and error rates of robots installed in the factory. The input is real-time operation log data and performance indicators collected from the robots, which are stored in a database. The output is the collected operation log data and raw performance indicator data.

[0378] Step 2:

[0379] The server preprocesses the collected raw data. Specifically, it analyzes the operation log data using a morphological analysis tool (e.g., MeCab) and converts the grammatical structure and meaning into an understandable format. It also normalizes the performance indicator data and converts it into a unified format. If voice data is included, it is converted into text data using voice recognition software (e.g., Google Speech-to-Text). The input is raw data obtained from the device, and the output is preprocessed data that can be analyzed.

[0380] Step 3:

[0381] The server inputs the preprocessed data into a generative AI model to perform qualitative and quantitative evaluations. In qualitative evaluation, a generative AI model (e.g., Hugging Face Transformers) is used to analyze the contents of the operation log data and extract meaningful operation patterns and comments. An emotion recognition engine (e.g., Sentiment Analysis API) is used to evaluate the user's emotional state from the operation log. In quantitative evaluation, work efficiency and error rates are analyzed based on performance index data. The input is preprocessed data, and the output is qualitative and quantitative evaluation results.

[0382] Step 4:

[0383] The server outputs the evaluation results of the generated AI model and emotion recognition engine in report format. Specifically, the evaluation results are compiled in Excel or PDF format and provided to the administrator. The report includes the robot's operation details, emotional state, and evaluation results of each performance indicator. The input is the evaluation result data, and the output is the evaluation results in report format.

[0384] Step 5:

[0385] The user evaluates production equipment and employees based on the generated report. The manager reviews the report content and uses it as a supplementary reference for management by objectives (MBO) evaluation. Specifically, the report comprehensively evaluates the robot's operating efficiency, error rate, employee work performance, etc. The input is the evaluation results in report format, and the output is the manager's final evaluation results.

[0386] 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.

[0387] 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.

[0388] 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.

[0389] [Second embodiment]

[0390] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0391] 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.

[0392] 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).

[0393] 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.

[0394] 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.

[0395] 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).

[0396] 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.

[0397] 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.

[0398] 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.

[0399] 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.

[0400] 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.

[0401] 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."

[0402] The present invention is a system that visualizes employees' work performance by performing qualitative and quantitative evaluations using a generative AI model using business data and meeting logs collected from terminals, thereby realizing a fair and convincing evaluation system. Specific embodiments for implementing the present invention are described below.

[0403] Data collection

[0404] The device continuously collects the user's business data and meeting logs. Specifically, it acquires business data such as keystrokes, application usage history, file access logs, and email content. It also collects audio and video recordings of video and audio conferences.

[0405] Examples:

[0406] For example, the terminal of user A records a log of the applications used throughout the day and saves the contents of statements made during a meeting as audio data.

[0407] Data Preprocessing

[0408] The server converts the collected raw data into an analyzable format. First, the voice data is converted into text data using a speech recognition tool. Next, the text data is subjected to morphological analysis using natural language processing (NLP) techniques to analyze grammatical structure and meaning. Numerical data is normalized and converted into a unified format.

[0409] Examples:

[0410] For example, audio data from a meeting can be converted into text using a transcription tool, and morphological analysis can be performed to make what was said easier to understand. Numerical data such as application usage time can also be converted into a unified format.

[0411] Analysis using generative AI models

[0412] The server inputs the preprocessed data into the generative AI model and performs qualitative and quantitative evaluations. Qualitative evaluations evaluate the content of text and the quality of speech in meetings. Quantitative evaluations use log data such as work time and application usage.

[0413] Examples:

[0414] For example, a generative AI model can extract meaningful comments about project progress from User A's emails and give them a high rating. At the same time, it can quantitatively evaluate User A's spending more than 40 hours per week on a specific project based on the amount of time worked.

[0415] Output of evaluation results

[0416] The server compiles the results of the analysis by the generative AI model into a report and provides it to the user (supervisor). The report includes qualitative evaluations (e.g., quality of comments) and quantitative evaluations (e.g., work time, statistics on applications used).

[0417] Examples:

[0418] The server generates a report and provides it to the manager, which includes a qualitative assessment of User A's project management skills and problem-solving abilities, as well as quantitative data showing User A's time contribution to a particular project.

[0419] Complementing MBO evaluation

[0420] The user (supervisor) performs an MBO evaluation based on the provided report, checks the validity of the evaluation content, makes adjustments as necessary, and provides feedback to the employee.

[0421] Examples:

[0422] The supervisor uses the analysis results of the generated AI model to conduct an annual evaluation of User A. User A's leadership and focus are particularly highly evaluated, and a high MBO score is awarded.

[0423] In this way, the system of the present invention clearly shows employees' work achievements and enables convincing and fair evaluations. This system also contributes to improving employee satisfaction by providing managers with objective data to complement MBO evaluations.

[0424] The processing flow will be explained below.

[0425] Step 1: Data collection

[0426] The device collects the user's business data and meeting logs. Specifically, it periodically records the user's application usage history, file access logs, email content, keystroke data, etc. It also collects audio and video recording data of video and audio conferences.

[0427] Specific operation:

[0428] The terminal starts the log collection agent and collects business data every 15 minutes.

[0429] Acquire recording data via the video conferencing tool's API.

[0430] Step 2: Data Preprocessing

[0431] The server preprocesses the collected raw data. First, it converts the voice data into text data using a speech recognition tool, then performs morphological analysis of the text data using natural language processing (NLP) techniques. Numerical data is normalized and converted into a unified format.

[0432] Specific operation:

[0433] The server inputs the audio data into a transcription tool and converts it into text data.

[0434] The server performs morphological analysis, analyzes the text data, and extracts grammatical structure and semantic information.

[0435] The server normalizes the numerical data and converts it into a unified data format.

[0436] Step 3: Analysis by generative AI model

[0437] The server inputs the preprocessed data into the generative AI model and performs qualitative and quantitative evaluations. The qualitative evaluation involves analyzing the sentiment and importance of the comments, while the quantitative evaluation involves calculating metrics such as work time and application usage frequency.

[0438] Specific operation:

[0439] The server inputs the text data extracted from the meeting log into the generative AI model.

[0440] The server uses the generative AI model to generate sentiment analysis and content evaluation scores for the comments.

[0441] The server inputs the numerical data into the AI ​​model and generates an evaluation score based on work time and frequency of application use.

[0442] Step 4: Output of evaluation results

[0443] The server compiles the evaluation results from the generative AI model into a report and provides it to the user (supervisor). The report includes a qualitative evaluation (quality of speech content) and a quantitative evaluation (task time, statistics on applications used).

[0444] Specific operation:

[0445] The server automatically generates a report based on the evaluation results from the generated AI model.

[0446] The server periodically displays the report on the manager's dashboard or sends it by email.

[0447] Step 5: Complementing the MBO evaluation

[0448] The user (supervisor) performs an MBO evaluation based on the provided report, checks the validity of the evaluation content, makes adjustments as necessary, and provides feedback to the employee.

[0449] Specific operation:

[0450] The user checks the report and completes the MBO evaluation sheet based on the evaluation results of the generated AI model.

[0451] The user conducts a feedback interview based on the evaluation results and shares the evaluation details with the employee.

[0452] Through the above steps, the system of the present invention makes employees' work achievements visible and realizes a fair and convincing evaluation system.

[0453] Example 1

[0454] 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."

[0455] Traditional evaluation systems have the problem of making it difficult to accurately and fairly evaluate employees' work performance. In particular, there is a lack of quantitative methods for qualitative evaluations and contributions in meetings, which often leads to biased evaluations and a sense of unfairness. There is also a lack of materials for managers to conduct effective management-by-objectives (MBO) evaluations.

[0456] 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.

[0457] In this invention, the server includes means for continuously collecting business data and meeting logs from user terminals, means for preprocessing the collected business data and meeting logs using a speech recognition tool and natural language processing technology, means for inputting the preprocessed data into a generative AI model to perform qualitative and quantitative evaluations, and means for outputting the evaluation results by the generative AI model in report format. This enables employees' work performance to be evaluated objectively and fairly, and provides managers with complementary information for conducting effective MBO evaluations.

[0458] A "terminal" is an electronic device that a user uses to collect business data and meeting logs.

[0459] "Business data" refers to information related to a user's business, such as keystrokes, application usage history, file access logs, and email content.

[0460] "Meeting Log" refers to records of a meeting, including audio and video recordings of video and audio conferences, speeches, and related information.

[0461] "Speech recognition tool" refers to software or services for converting voice data into text data.

[0462] "Natural language processing technology" refers to computer science technology for analyzing text data and performing morphological analysis and grammatical structure analysis.

[0463] "Generative AI model" refers to an artificial intelligence model that uses preprocessed data to perform qualitative and quantitative evaluations.

[0464] "Qualitative evaluation" refers to the process of evaluating the content of a written piece or the quality of what is said in a meeting.

[0465] "Quantitative evaluation" refers to the process of evaluating based on numerical data such as work time and application usage.

[0466] "Report format" refers to the form of a document that summarizes the evaluation results in an easy-to-understand format, and the output includes formats such as Excel and PDF.

[0467] "Management by Objectives (MBO) evaluation" refers to a management technique in which managers evaluate employees' performance based on the degree to which they achieve goals.

[0468] The present invention provides a system that realizes a fair and convincing evaluation system by collecting users' business data and meeting logs and performing evaluations using a generative AI model. Specific embodiments for implementing the present invention are described below.

[0469] Data collection

[0470] The device continuously collects the user's business data and meeting logs. The device is equipped with a keyboard hook and a file system mini-filter driver to collect the following data:

[0471] Keystrokes

[0472] Application Usage History

[0473] File Access Log

[0474] Email content

[0475] Recordings of video and audio conferences

[0476] Examples:

[0477] For example, User A's terminal records the usage logs of all applications used during work and saves speeches made during meetings as audio data.

[0478] Data Preprocessing

[0479] The server converts the collected raw data into an analyzable format, specifically by following these steps:

[0480] The voice data is converted into text data using a voice recognition tool (e.g., Google Speech-to-Text API).

[0481] The text data is morphologically analyzed using natural language processing techniques (e.g., spaCy, NLTK) to analyze grammatical structure and meaning.

[0482] Numerical data is normalized using scaling and normalization techniques.

[0483] Examples:

[0484] The audio data from the meeting is converted into text data using a transcription tool, and morphological analysis is performed to evaluate the quality of what is being said. The numerical data of application usage time is also normalized into a unified format.

[0485] Analysis using generative AI models

[0486] The server inputs the preprocessed data into a generative AI model (e.g., OpenAI GPT-4) and performs qualitative and quantitative evaluation. The specific steps of the evaluation are as follows:

[0487] In qualitative evaluation, text data is analyzed to evaluate the quality of comments made during meetings and the importance of email content.

[0488] Quantitative evaluation involves analyzing log data and evaluating work time and application usage.

[0489] Examples:

[0490] The generative AI model receives a prompt such as "Evaluate User A's comments during the meeting and analyze how constructive they were," and the model outputs the evaluation results. The model also analyzes work time data using a prompt such as "Calculate User A's weekly work hours and evaluate their productivity."

[0491] Output of evaluation results

[0492] The server compiles the results of the analysis by the generative AI model into a report and provides it to the user (supervisor). This report includes the following:

[0493] Qualitative assessment (e.g., quality of content)

[0494] Quantitative evaluation (e.g., work time, application usage statistics)

[0495] Examples:

[0496] The server generates a report and provides it to the manager, which includes a qualitative assessment of User A's project management skills and problem-solving abilities, as well as quantitative data showing User A's time contribution to a particular project.

[0497] Complementing MBO evaluation

[0498] The user (supervisor) performs a management by objectives (MBO) evaluation based on the provided report, checks the validity of the evaluation content, makes adjustments as necessary, and provides feedback to the employee.

[0499] Examples:

[0500] The supervisor uses the analysis results of the generated AI model to conduct an annual evaluation of User A. User A's leadership and focus are particularly highly evaluated, and a high MBO score is awarded.

[0501] In this way, the system of the present invention clearly shows employees' work achievements and achieves fair evaluations. This system provides managers with objective data to complement MBO evaluations, thereby contributing to improved employee satisfaction.

[0502] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0503] Step 1: Data collection

[0504] The device continuously collects business data and meeting logs. Specifically, it collects user keystrokes, application usage history, file access logs, email content, and audio and video conference recordings. The input is user interactions and meeting participation, and the output is log data stored on the device.

[0505] Specific behavior:

[0506] The device records keystrokes using keyboard hooks, captures application usage history and file access logs using a dedicated service, and, when audio and video conferences are taking place, captures conference logs using dedicated audio and video recording software.

[0507] Step 2: Data Preprocessing

[0508] The server converts the collected raw data into an analyzable format. Specifically, it converts voice data into text data using a speech recognition tool (e.g., Google Speech-to-Text API), and then performs morphological analysis on the text data using natural language processing technology (e.g., spaCy, NLTK). Numerical data is normalized using scaling and normalization techniques. The input is the collected log data, and the output is preprocessed, analyzable data.

[0509] Specific behavior:

[0510] The server sends the audio file to a speech recognition API to obtain the transcribed text, then feeds the text data into an NLP tool for morphological analysis, and normalizes the numerical data using an appropriate scaling algorithm.

[0511] Step 3: Analysis by generative AI model

[0512] The server inputs the preprocessed data into a generative AI model (e.g., OpenAI GPT-4) to perform qualitative and quantitative evaluations. Qualitative evaluations assess the quality of comments made during meetings and the importance of email content, while quantitative evaluations assess task time and application usage. The input is the preprocessed data, and the output is the evaluation results.

[0513] Specific behavior:

[0514] The server inputs a prompt to the generative AI model, such as "Evaluate User A's comments during the meeting and analyze how constructive they were," and obtains a response from the model. It also analyzes the work time data using a prompt, such as "Calculate User A's weekly work hours and evaluate their productivity."

[0515] Step 4: Output of evaluation results

[0516] The server compiles the evaluation results analyzed by the generative AI model into a report. Specifically, it integrates qualitative evaluations (e.g., quality of speech) and quantitative evaluations (e.g., work time, statistics on applications used) to create a report. The input is the analyzed evaluation results, and the output is the report.

[0517] Specific behavior:

[0518] The server receives the output of the generative AI model and creates a report in formats such as Excel or PDF, which is visually organized using graphs and tables so that administrators can easily understand it.

[0519] Step 5: Complementing the MBO evaluation

[0520] The user (supervisor) performs a management by objectives (MBO) evaluation based on the provided report. Specifically, the user checks the validity of the evaluation content and makes adjustments as necessary. Finally, the user provides feedback to the employee. The input is the generated report, and the output is the revised evaluation and feedback.

[0521] Specific behavior:

[0522] The supervisor reviews the report and adds comments such as, "User A's leadership is evident, but we would like to see more specific examples." In addition, the supervisor discusses with the employee specific evaluation items and proposes the next steps for improvement.

[0523] (Application example 1)

[0524] 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."

[0525] At the workplace, it is important to accurately evaluate the performance and contribution of workers and to create a fair and transparent evaluation system. Factory floors, in particular, require a system that efficiently collects and analyzes work logs and robot interaction data, and evaluates them qualitatively and quantitatively.

[0526] 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.

[0527] In this invention, the server includes means for collecting business data and meeting logs from terminals, means for preprocessing the collected business data and meeting logs, means for analyzing the preprocessed data and using a generative AI model for qualitative and quantitative evaluation, means for outputting the evaluation results by the generative AI model in report format, means for collecting operation data and work logs of robots in the factory, means for collecting interaction data with workers, and means for analyzing the collected data and evaluating the performance of workers. This makes it possible to clarify the work achievements of workers and provide convincing and fair evaluations.

[0528] "Device" means a computer device used to collect and process data.

[0529] "Business data" refers to information such as applications that users operate during work, files that they access, and communication content.

[0530] "Meeting logs" are audio and video recordings related to statements made during a meeting and the agenda, as well as textual versions of such recordings.

[0531] "Preprocessing" is the process of converting collected raw data into an analyzable format.

[0532] A "generative AI model" is an artificial intelligence model that uses collected data to perform qualitative and quantitative evaluations.

[0533] "Qualitative evaluation" is the process of evaluating the content of a written piece or the quality of what is said in a meeting.

[0534] "Quantitative evaluation" is a process of evaluation based on numerical data such as log data and work time.

[0535] "Report format" refers to the format of a document summarizing the evaluation results.

[0536] "Factory robots" are automated machinery used to streamline factory work.

[0537] "Operation data" refers to records and log data relating to the robot's operations.

[0538] A "work log" is recorded data on the actions and work details of workers.

[0539] "Interaction data" is recorded data of the operations and communication that take place alternately between workers and robots.

[0540] "Evaluation results" refer to the qualitative and quantitative evaluation values ​​and evaluation content calculated by the generative AI model.

[0541] To implement this invention, a terminal first collects business data and a meeting log. Business data includes information such as the operation history of applications used by the user, files accessed, and emails sent and received. The meeting log includes audio and video data related to comments made during the meeting and the agenda.

[0542] The collected data is sent to a server where it undergoes preprocessing. During the preprocessing stage, audio data is converted into text data using a transcription tool. Next, natural language processing (NLP) techniques are used to perform morphological analysis of the text data, analyzing its grammatical structure and meaning. Numerical data is normalized and converted into a unified format.

[0543] Once preprocessed, the data is input into a generative AI model for qualitative and quantitative evaluation. In qualitative evaluation, the generative AI model evaluates the content of text and the quality of comments made in meetings. For example, it extracts meaningful comments about project progress from the contents of a user's email. In quantitative evaluation, it performs evaluation based on log data and work time. For example, it measures the total amount of time a user spends on a specific project.

[0544] The evaluation results are compiled in a report format by the server and provided to the user (manager). This report includes both qualitative and quantitative evaluations. Based on this report, the manager can obtain objective information for performing a management by objectives (MBO) evaluation. Using this information, the manager can check the validity of the evaluation content and make adjustments as necessary.

[0545] Furthermore, the system can collect operation data and work logs of robots in factories and analyze interaction data with workers, which will further improve the efficiency of factory work and enable fair evaluations.

[0546] As a concrete example, a robot in a factory collects speech data from workers, converts it into text data, performs morphological analysis, evaluates the quality of the speech, quantitatively evaluates the worker's performance using work log data, and integrates this data to provide the manager with the evaluation results in the form of a report.

[0547] An example of a prompt sentence to input to the generative AI model is, "Please summarize what was said in this meeting and extract the most important points." Using this prompt allows for efficient qualitative evaluation of meeting logs.

[0548] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0549] Step 1:

[0550] The device collects business data and meeting logs. Business data includes the operation history of applications used by the user, files accessed, and emails sent and received. Meeting logs include audio and video recordings related to comments made during meetings and the agenda. This data is obtained using software (e.g., data collection API). Input is the user's activity log and meeting audio, and output is the collected raw data.

[0551] Step 2:

[0552] The terminal sends the collected data to the server. The address of the server is specified and the data is sent in encrypted format via the required protocol (e.g. HTTP / HTTPS). The input is the raw data acquired, and the output is the data sent to the server.

[0553] Step 3:

[0554] The raw data received by the server is preprocessed. First, the voice data is converted into text data using a speech recognition tool (e.g., Python's speech_recognition library). Next, the text data is subjected to morphological analysis using natural language processing (NLP) techniques (e.g., the spacy library) to analyze grammatical structure and meaning. Numerical data is converted into a data frame and normalized. The input is raw data, and the output is preprocessed data.

[0555] Step 4:

[0556] The preprocessed data is input into a generative AI model. The generative AI model (e.g., GPT-2 from Hugging Face) analyzes the text data and performs a qualitative evaluation. An example of a prompt used here is, "Please summarize what was said in this meeting and extract the most important points." The input is the preprocessed text data and the prompt, and the output is the result of the qualitative evaluation.

[0557] Step 5:

[0558] The preprocessed numerical data is used to perform quantitative evaluation. The server analyzes the data frame and calculates statistics on work time and application usage time. For example, it calculates the time spent on a specific project based on work time. The input is the preprocessed numerical data, and the output is the results of the quantitative evaluation.

[0559] Step 6:

[0560] The server integrates the results of the qualitative and quantitative evaluations and generates a report. The report includes qualitative analysis results and quantitative statistics. It displays a detailed analysis of the content of the text, the quality of the comments, and the working time. The input is the results of the qualitative and quantitative evaluations, and the output is a report-format file.

[0561] Step 7:

[0562] The server provides the generated report to the user (administrator). The report is made accessible in dashboard or PDF format so that the administrator can easily check the evaluation. The input is the generated report, and the output is the report used by the user in the actual evaluation process.

[0563] 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.

[0564] The present invention is a system that visualizes employees' work performance by performing qualitative and quantitative evaluations using a generative AI model using business data and meeting logs collected from terminals, thereby realizing a fair and convincing evaluation system. Furthermore, the present invention improves the accuracy and comprehensiveness of evaluations by combining it with an emotion engine that recognizes the user's emotions. Specific embodiments for implementing the present invention are described below.

[0565] Data collection

[0566] The device continuously collects the user's business data and meeting logs. Specifically, it acquires business data such as the user's application usage history, file access logs, email content, and keystroke data. It also collects audio and video recordings of video and audio conferences.

[0567] Examples:

[0568] For example, the terminal of user A records a log of the applications used throughout the day and saves the contents of statements made during a meeting as audio data.

[0569] Data Preprocessing

[0570] The server converts the collected raw data into an analyzable format. First, the voice data is converted into text data using a speech recognition tool. Next, the text data is subjected to morphological analysis using natural language processing (NLP) techniques to analyze grammatical structure and meaning. Numerical data is normalized and converted into a unified format. An emotion engine is also implemented at this stage.

[0571] Examples:

[0572] For example, audio data from a meeting is converted into text data using a transcription tool, and morphological analysis is performed to make what is being said easier to understand. Numerical data such as application usage time is also converted into a unified format. An emotion engine analyzes the user's emotions from the text data.

[0573] Analysis using generative AI models

[0574] The server inputs the preprocessed data into a generative AI model for qualitative and quantitative evaluation. Qualitative evaluation evaluates the content of text and the quality of speech in meetings. The emotion engine also analyzes the user's emotional state and incorporates it into the evaluation. Quantitative evaluation uses log data such as work time and application usage.

[0575] Examples:

[0576] For example, a generative AI model extracts meaningful comments about project progress from User A's email content and gives a high rating. At the same time, it quantitatively evaluates that User A is spending more than 40 hours per week on a specific project based on the amount of time worked. An emotion engine analyzes the emotions in User A's email and reflects their emotional state, such as "positive" or "negative," in the rating.

[0577] Output of evaluation results

[0578] The server compiles the results of the analysis by the generative AI model and emotion engine into a report and provides it to the user (supervisor). The report includes qualitative evaluations (quality of speech content, emotional state) and quantitative evaluations (work time, statistics on applications used).

[0579] Examples:

[0580] The server generates a report and provides it to the manager, which includes a qualitative assessment of User A's project management skills and problem-solving abilities, as well as quantitative data showing his time contribution to a specific project. The report also includes an emotional assessment based on the emotion engine.

[0581] Complementing MBO evaluation

[0582] The user (supervisor) performs an MBO evaluation based on the provided report, checks the validity of the evaluation content, makes adjustments as necessary, and provides feedback to the employee.

[0583] Examples:

[0584] The supervisor uses the analysis results of the generative AI model and the emotion engine to conduct an annual evaluation of User A. In particular, User A's leadership and focus are highly evaluated, and a high MBO score is awarded. User A's emotional state is also taken into consideration, and the evaluation content is adjusted accordingly.

[0585] In this way, the system of the present invention clearly shows the employee's work achievements and enables convincing and fair evaluations. At the same time, the introduction of an emotion engine enables comprehensive evaluations that take into account the employee's emotional state.

[0586] The processing flow will be explained below.

[0587] Step 1: Data collection

[0588] The device collects the user's business data and conference logs. Specifically, it acquires business data such as the user's application usage history, file access logs, email content, and keystroke data. It also collects audio and video recordings of video and audio conferences.

[0589] Specific operation:

[0590] The terminal starts the log collection agent and collects business data every 15 minutes.

[0591] Acquire recording data via the video conferencing tool's API.

[0592] Step 2: Data Preprocessing

[0593] The server preprocesses the collected raw data. First, the voice data is converted into text data using a voice recognition tool. Next, the text data undergoes morphological analysis using natural language processing (NLP) techniques. Numerical data is normalized and converted into a unified format. An emotion engine analyzes the user's emotions from the text data.

[0594] Specific operation:

[0595] The server inputs the audio data into a transcription tool and converts it into text data.

[0596] The server performs morphological analysis, analyzes the text data, and extracts grammatical structure and semantic information.

[0597] The server normalizes the numerical data and converts it into a unified data format.

[0598] The emotion engine analyzes the user's emotional state from the text data.

[0599] Step 3: Analysis by generative AI models and emotion engines

[0600] The server inputs the preprocessed data into the generative AI model and emotion engine to perform qualitative and quantitative evaluations. Qualitative evaluation evaluates the content of text and the quality of speech in meetings. The emotion engine also analyzes the user's emotional state and incorporates it into the evaluation. Quantitative evaluation uses log data such as work time and application usage.

[0601] Specific operation:

[0602] The server inputs the text data extracted from the meeting log into the generative AI model.

[0603] The server uses the generative AI model to generate sentiment analysis and content evaluation scores for the comments.

[0604] The server inputs the numerical data into the AI ​​model and generates an evaluation score based on work time and frequency of application use.

[0605] The emotion engine analyzes the user's text data and reflects their emotional state in the evaluation score.

[0606] Step 4: Output of evaluation results

[0607] The server compiles the evaluation results from the generative AI model and emotion engine into a report and provides it to the user (supervisor). The report includes qualitative evaluations (quality of speech content, emotional state) and quantitative evaluations (work time, statistics on applications used).

[0608] Specific operation:

[0609] The server automatically generates a report based on the evaluation results from the AI ​​model and emotion engine.

[0610] The server periodically displays the report on the manager's dashboard or sends it by email.

[0611] Step 5: Complementing the MBO evaluation

[0612] The user (supervisor) performs an MBO evaluation based on the provided report, checks the validity of the evaluation content, makes adjustments as necessary, and provides feedback to the employee.

[0613] Specific operation:

[0614] The user reviews the report and completes the MBO evaluation sheet based on the evaluation results of the generative AI model and emotion engine.

[0615] The user conducts a feedback interview based on the evaluation results and shares the evaluation details with the employee.

[0616] Through these steps, the system of the present invention visualizes employees' work performance and realizes a fair and convincing evaluation system. In addition, the introduction of an emotion engine enables comprehensive evaluation that takes into account the emotional aspects of employees.

[0617] Example 2

[0618] 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."

[0619] Conventional performance evaluation systems often evaluate employees' work performance qualitatively, making it difficult to base evaluations on concrete data. They also lack comprehensive evaluations that take emotional aspects into account, which can lead to evaluations that lack fairness and satisfaction. Furthermore, insufficient pre-processing of meeting logs and work data often reduces the accuracy of evaluation results.

[0620] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for collecting business data and meeting logs from terminals, a means for preprocessing the collected business data and meeting logs, a means for analyzing the preprocessed data and using a generative AI model for qualitative and quantitative evaluation, a means for outputting the evaluation results by the generative AI model in report format, and a means for recognizing emotions from text data using an emotion engine. This enables a fair and comprehensive evaluation of employees' work performance based on specific data and taking into account their emotional aspects.

[0621] A "terminal" is a computer or mobile device used by a user, and is a device for collecting business data and meeting logs.

[0622] "Business data" refers to all data related to a user's business activities, including application usage history, file access logs, email content, keystroke data, and the like.

[0623] A "meeting log" refers to recorded data of a meeting, including audio and video, and is data that captures the content of statements and discussions made during the meeting.

[0624] "Collection means" refers to the method or function for acquiring business data and meeting logs from a terminal, and takes the form of software or hardware.

[0625] "Preprocessing" refers to the process of converting raw data into an analyzable format, and includes transcribing audio data, morphological analysis of text data, and normalizing numerical data.

[0626] A "generative AI model" refers to an artificial intelligence model that has functions such as response generation and data analysis, and is used here to perform qualitative and quantitative evaluations.

[0627] An "emotion engine" refers to a program or tool that recognizes and analyzes a user's emotional state from text data.

[0628] "Report format" refers to a format that visually summarizes the evaluation results from the generative AI model and emotion engine in an easy-to-understand manner, and includes formats such as PDF and web dashboards.

[0629] "Manager" refers to the person or position with the authority to conduct MBO evaluations based on the evaluation results, and is usually a supervisor or human resources officer.

[0630] "MBO" is an abbreviation for Management by Objectives, a management method that evaluates employees based on the degree to which they have achieved their goals.

[0631] The present invention provides a system that visualizes employees' work performance by performing qualitative and quantitative evaluations using a generative AI model using business data and meeting logs collected from terminals, thereby realizing a fair and convincing evaluation system. The present invention also provides a system that improves the accuracy and comprehensiveness of evaluations by combining it with an emotion engine that recognizes user emotions. Specific embodiments for implementing the present invention are described below.

[0632] Data collection

[0633] The device continuously collects the user's business data and meeting logs. Specifically, it acquires business data such as the user's application usage history, file access logs, email content, and keystroke data. It also collects audio and video recordings of video and audio conferences.

[0634] Examples:

[0635] For example, the terminal of user A records a log of the applications used throughout the day and saves the contents of statements made during a meeting as audio data.

[0636] Data Transfer

[0637] The device sends the collected data to the server in real time. The data collection program in the device periodically uploads the locally stored data to the server. For security reasons, the data is sent in encrypted form.

[0638] Data Preprocessing

[0639] The server converts the received raw data into an analyzable format. First, the voice data is converted into text data using a speech recognition tool (e.g., Google Cloud Speech-to-Text). Next, the text data is subjected to morphological analysis using natural language processing (NLP) technology (e.g., Spacy) to analyze grammatical structure and meaning. Numerical data is normalized from Excel or CSV format and converted into a unified format (e.g., JSON format). Finally, the text data is analyzed for sentiment using the API of an emotion engine (e.g., Microsoft Azure Text Analytics).

[0640] Examples:

[0641] For example, audio data from a meeting is transcribed using Google Cloud Speech-to-Text, and morphological analysis is performed using Spacy to make what is being said easier to understand. Numerical data such as application usage time is converted from Excel or CSV format into a unified format. An emotion engine analyzes user emotions from the text data.

[0642] Analysis using generative AI models

[0643] The server inputs the preprocessed data into a generative AI model (e.g., OpenAI GPT-3) to perform qualitative and quantitative evaluations. Qualitative evaluations evaluate the content of text and the quality of speech in meetings. An emotion engine also analyzes the user's emotional state and incorporates it into the evaluation. Quantitative evaluations use log data such as work time and application usage.

[0644] Examples:

[0645] For example, a generative AI model extracts meaningful comments about project progress from User A's email content and gives a high rating. At the same time, it quantitatively evaluates that User A is spending more than 40 hours per week on a specific project based on the amount of time worked. An emotion engine analyzes the emotions in User A's email and reflects their emotional state, such as "positive" or "negative," in the rating.

[0646] Example prompt sentence:

[0647] "Evaluate User A's important contributions to the project progress based on his emails and meeting comments over the past month."

[0648] "Quantitatively evaluate User A's time contribution to a specific project based on his / her weekly work time data."

[0649] Sentiment analysis results integration

[0650] The server uses an emotion engine to integrate the results of the emotion analysis with the evaluation from the generative AI model. The emotion analysis results are combined with the output of the AI ​​model to reflect positive and negative comments in the evaluation score. Detailed results of the emotion analysis are also added to the report.

[0651] Generating evaluation results

[0652] The server compiles the generated evaluation results into a report. The server uses a report generation tool (e.g., JasperReports) to create a detailed report that includes both qualitative and quantitative evaluations. The report is output in PDF format or web dashboard format.

[0653] Providing reports

[0654] The server provides the completed report to the user (supervisor). The report is automatically sent to the supervisor's email address periodically (for example, at the end of the month). The report is also uploaded to the company portal so that the user can access it.

[0655] Complementing MBO evaluation

[0656] The user (supervisor) performs an MBO evaluation based on the provided report. The supervisor reviews the report and provides additional feedback or supplementary information as needed. Once the evaluation is complete, a session is set up to provide feedback to the employee being evaluated.

[0657] In this way, each processing step of the system works together in a concrete way to clarify employees' work achievements and achieve fair and convincing evaluations. The introduction of an emotion engine makes it possible to provide comprehensive evaluations that take into account employees' emotional states.

[0658] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0659] Step 1: Data collection

[0660] The terminal collects the user's business data and meeting logs.

[0661] Input: User application usage history, file access logs, email content, keystroke data, and audio and video conference recordings.

[0662] Data processing: Raw data is automatically acquired and collected 24 / 7.

[0663] Output: Locally stored business data and meeting logs.

[0664] Specific operation: The device will periodically record application logs and file access logs, and will automatically start recording when an audio conference is started.

[0665] Step 2: Data Transfer

[0666] The data collected by the device is sent to the server in real time.

[0667] Input: Business data and meeting logs stored locally on the device.

[0668] Data processing: Encrypt the data and upload it securely to the server.

[0669] Output: Aggregated business data and meeting logs on the server.

[0670] Specific operation: The data collection program in the device periodically (for example, every hour) encrypts the data and uploads it to the server.

[0671] Step 3: Data Preprocessing

[0672] The server converts the raw data it receives into a format that can be analyzed.

[0673] Input: Raw data (business data, meeting logs) aggregated on the server.

[0674] Data processing: Convert the voice data into text data using Google Cloud Speech-to-Text, perform morphological analysis on the text data using Spacy, normalize the numerical data and convert it into a unified format (e.g., JSON), and analyze emotions from the text data using an emotion engine.

[0675] Output: Data converted into an analyzable format (text data, morphological analysis results, normalized numerical data, sentiment analysis results).

[0676] How it works: The server transcribes the audio data using Google Cloud Speech-to-Text, then performs morphological analysis using Spacy. Numerical data is converted from Excel or CSV to JSON format, and sentiment analysis is performed using a sentiment analysis tool.

[0677] Step 4: Data integration and storage

[0678] The server consolidates the pre-processed data and stores it in a database.

[0679] Input: Preprocessed data (text data, morphological analysis results, normalized numerical data, sentiment analysis results).

[0680] Data processing: Integrate the data, classify it by user, and store it in a NoSQL database (e.g., MongoDB).

[0681] Output: Data integrated and stored in a database.

[0682] Specific operation: The server inserts the preprocessed data into MongoDB, classifies it by user, and creates indexes to enable efficient searches.

[0683] Step 5: Analysis by generative AI model

[0684] The server inputs the preprocessed data into a generative AI model for qualitative and quantitative evaluation.

[0685] Input: Preprocessed data extracted from the database.

[0686] Data processing: A generative AI model (e.g., OpenAI GPT-3) is invoked with a prompt sentence to perform qualitative evaluation (text content and speech quality) and quantitative evaluation (task time, application usage).

[0687] Output: Analysis results from the generative AI model.

[0688] Specific operation: The server inputs data into the OpenAI GPT-3 model using the prompt sentence and generates an appropriate evaluation score and comment.

[0689] Step 6: Integrating the results of sentiment analysis

[0690] The server uses an emotion engine to integrate the results of emotion analysis with the evaluation from the generative AI model.

[0691] Input: Analysis results from generative AI model, sentiment analysis results.

[0692] Data processing: The results of sentiment analysis are combined with the output of the AI ​​model and reflected in the evaluation score.

[0693] Output: Evaluation results integrated with sentiment analysis results.

[0694] Specific operation: The server adds the results from the emotion engine to the output of the generative AI model, and reflects positive or negative emotions in the evaluation score.

[0695] Step 7: Generate evaluation results

[0696] The server compiles the generated evaluation results in a report format.

[0697] Input: Evaluation results integrated with sentiment analysis results.

[0698] Data processing: Create reports using a report generation tool (e.g., JasperReports) and convert them into PDF or web dashboard formats.

[0699] Output: Assessment report in PDF format and web dashboard format.

[0700] What it does: The server compiles the results of the assessment into a detailed report and converts it into a visually understandable format.

[0701] Step 8: Providing the report

[0702] The server provides the completed report to the user (supervisor).

[0703] Input: Assessment report in PDF format and web dashboard format.

[0704] Data Processing: Present the report to your superior in an appropriate manner.

[0705] Output: Evaluation report provided to supervisor.

[0706] What it does: The server sends the report to the manager's email address at the end of the month and also uploads it to the company's internal portal for access.

[0707] Step 9: Complementing the MBO evaluation

[0708] The user (supervisor) performs an MBO evaluation based on the provided report.

[0709] Input: Evaluation report provided to supervisor.

[0710] Data processing: Check the evaluation content and provide any necessary feedback or supplementary information.

[0711] Output: Completed MBO evaluation and feedback.

[0712] What happens next: The manager reviews the report, notes any necessary corrections or additional information, and schedules a feedback session.

[0713] In this way, the system of the present invention evaluates employees' work performance in a concrete and data-based manner, and achieves a fair and comprehensive evaluation that also takes into account their emotional aspects.

[0714] (Application example 2)

[0715] 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."

[0716] Existing evaluation systems do not adequately perform qualitative and quantitative evaluations based on work data and meeting logs. This makes it difficult to fairly evaluate employees' work performance and the operational efficiency of production equipment. Another problem is the lack of systems that can perform comprehensive evaluations using robot operation logs and performance indicators. Furthermore, it is necessary to improve the accuracy and comprehensiveness of evaluations by combining an emotion engine.

[0717] The identification processing 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 means for collecting business data and meeting logs from terminals, means for preprocessing the collected business data and meeting logs, means for analyzing the preprocessed data and using a generative AI model for qualitative and quantitative evaluation, means for outputting the evaluation results by the generative AI model in report format, means for collecting operation log data and performance indicators of the production equipment, means for converting the collected operation log data and performance indicators into an analyzable format, and means for qualitatively evaluating the operation log data and quantitatively evaluating the performance indicators using the generative AI model and an emotion recognition engine. This not only enables fair and convincing evaluation based on the business data and meeting logs, but also enables comprehensive evaluation of the operational efficiency of the production equipment.

[0718] A "terminal" is an electronic device such as a computer or smartphone used by a user, and is used to collect business data and meeting logs.

[0719] "Business data" refers to records such as application usage history, file access logs, email content, and keystroke data that are generated when a user performs a task.

[0720] "Meeting logs" refer to audio and video recordings of video and audio conferences, including audio data converted into text data and minutes of meetings.

[0721] "Preprocessing" is the process of converting collected raw data into an analyzable format, and includes transcribing audio data, morphological analysis of text data, and normalizing numerical data.

[0722] A "generative AI model" is a model that uses machine learning algorithms to analyze collected data and perform qualitative and quantitative evaluations.

[0723] An "emotion recognition engine" is a technology for analyzing a user's emotional state from text data and voice data, and reflects "positive," "negative," and changes therein in the evaluation.

[0724] "Production equipment" refers to equipment such as robots used in factories and production lines, and is the subject of collection of operation log data and performance indicators.

[0725] "Operation log data" refers to data that includes the operation history and operation records of robots and production equipment, and mainly includes detailed records of the production process.

[0726] "Performance index" refers to the key performance metrics, such as the efficiency or error rate achieved by a robot or production equipment, expressed in specific numerical values.

[0727] "Qualitative evaluation" is a type of evaluation performed by a generative AI model, and is a qualitative evaluation that is primarily based on the content of statements and operation logs.

[0728] "Quantitative evaluation" is a type of evaluation performed by a generative AI model, and is a quantitative evaluation based primarily on numerical data such as work time and performance indicators.

[0729] This invention provides a system for fair and comprehensive evaluation of the operational efficiency of production equipment and the work performance of employees. The system includes a terminal, a server, a generative AI model, and an emotion recognition engine.

[0730] First, the terminal continuously collects operation log data and performance indicators of production equipment. For example, it acquires performance data such as operation logs, work efficiency, and error rates of robots installed in a factory. This allows the operation history and operation records of production equipment, as well as key performance metrics, to be collected.

[0731] Next, the server preprocesses the collected data. At this stage, the collected raw data is converted into an analyzable format. Specifically, the operation log data is subjected to morphological analysis, and the performance indicator data is normalized. If voice data is included, it is converted into text data using a voice recognition tool.

[0732] The preprocessed data is input into a generative AI model, which performs both qualitative and quantitative evaluations. In qualitative evaluation, meaningful utterances and behavioral patterns are extracted from the operation log data, and the emotional state is analyzed using an emotion recognition engine. In quantitative evaluation, evaluation is performed based on performance indicators such as work efficiency and error rate.

[0733] The server outputs the analysis results of the generative AI model and emotion recognition engine in the form of a report. This report includes a qualitative evaluation of the robot's operation and emotional state, as well as a quantitative evaluation of each performance indicator. This allows administrators to conduct comprehensive and fair evaluations.

[0734] As a specific example, the operation log of the robot 123 is acquired as "log data of robot 123", and the evaluation results include quantitative and qualitative evaluations as follows:

[0735] Qualitative evaluation: Extract meaningful operation patterns from the robot's operation log and give a positive evaluation using an emotion engine.

[0736] Quantitative evaluation: Since the work efficiency is 95% and the error rate is 2%, the evaluation is "Efficiency: High" and "Error rate: Low."

[0737] Example prompt sentence:

[0738] "Collect the operation log data and performance metrics of Robot 123 and evaluate it using the generative AI model and emotion engine."

[0739] This system allows for a comprehensive evaluation of production equipment and employee performance, resulting in rational and convincing evaluations.

[0740] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0741] Step 1:

[0742] The terminal collects operation log data and performance indicators of production equipment. Specifically, data is acquired using sensors and controllers that record the operation history, operation records, work efficiency, and error rates of robots installed in the factory. The input is real-time operation log data and performance indicators collected from the robots, which are stored in a database. The output is the collected operation log data and raw performance indicator data.

[0743] Step 2:

[0744] The server preprocesses the collected raw data. Specifically, it analyzes the operation log data using a morphological analysis tool (e.g., MeCab) and converts the grammatical structure and meaning into an understandable format. It also normalizes the performance indicator data and converts it into a unified format. If voice data is included, it is converted into text data using voice recognition software (e.g., Google Speech-to-Text). The input is raw data obtained from the device, and the output is preprocessed data that can be analyzed.

[0745] Step 3:

[0746] The server inputs the preprocessed data into a generative AI model to perform qualitative and quantitative evaluations. In qualitative evaluation, a generative AI model (e.g., Hugging Face Transformers) is used to analyze the contents of the operation log data and extract meaningful operation patterns and comments. An emotion recognition engine (e.g., Sentiment Analysis API) is used to evaluate the user's emotional state from the operation log. In quantitative evaluation, work efficiency and error rates are analyzed based on performance index data. The input is preprocessed data, and the output is qualitative and quantitative evaluation results.

[0747] Step 4:

[0748] The server outputs the evaluation results of the generated AI model and emotion recognition engine in report format. Specifically, the evaluation results are compiled in Excel or PDF format and provided to the administrator. The report includes the robot's operation details, emotional state, and evaluation results of each performance indicator. The input is the evaluation result data, and the output is the evaluation results in report format.

[0749] Step 5:

[0750] The user evaluates production equipment and employees based on the generated report. The manager reviews the report content and uses it as a supplementary reference for management by objectives (MBO) evaluation. Specifically, the report comprehensively evaluates the robot's operating efficiency, error rate, employee work performance, etc. The input is the evaluation results in report format, and the output is the manager's final evaluation results.

[0751] 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.

[0752] 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.

[0753] 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.

[0754] [Third embodiment]

[0755] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0756] 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.

[0757] 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).

[0758] 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.

[0759] 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.

[0760] 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).

[0761] 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.

[0762] 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.

[0763] 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.

[0764] 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.

[0765] 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.

[0766] 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."

[0767] The present invention is a system that visualizes employees' work performance by performing qualitative and quantitative evaluations using a generative AI model using business data and meeting logs collected from terminals, thereby realizing a fair and convincing evaluation system. Specific embodiments for implementing the present invention are described below.

[0768] Data collection

[0769] The device continuously collects the user's business data and meeting logs. Specifically, it acquires business data such as keystrokes, application usage history, file access logs, and email content. It also collects audio and video recordings of video and audio conferences.

[0770] Examples:

[0771] For example, the terminal of user A records a log of the applications used throughout the day and saves the contents of statements made during a meeting as audio data.

[0772] Data Preprocessing

[0773] The server converts the collected raw data into an analyzable format. First, the voice data is converted into text data using a speech recognition tool. Next, the text data is subjected to morphological analysis using natural language processing (NLP) techniques to analyze grammatical structure and meaning. Numerical data is normalized and converted into a unified format.

[0774] Examples:

[0775] For example, audio data from a meeting can be converted into text using a transcription tool, and morphological analysis can be performed to make what was said easier to understand. Numerical data such as application usage time can also be converted into a unified format.

[0776] Analysis using generative AI models

[0777] The server inputs the preprocessed data into the generative AI model and performs qualitative and quantitative evaluations. Qualitative evaluations evaluate the content of text and the quality of speech in meetings. Quantitative evaluations use log data such as work time and application usage.

[0778] Examples:

[0779] For example, a generative AI model can extract meaningful comments about project progress from User A's emails and give them a high rating. At the same time, it can quantitatively evaluate User A's spending more than 40 hours per week on a specific project based on the amount of time worked.

[0780] Output of evaluation results

[0781] The server compiles the results of the analysis by the generative AI model into a report and provides it to the user (supervisor). The report includes qualitative evaluations (e.g., quality of comments) and quantitative evaluations (e.g., work time, statistics on applications used).

[0782] Examples:

[0783] The server generates a report and provides it to the manager, which includes a qualitative assessment of User A's project management skills and problem-solving abilities, as well as quantitative data showing User A's time contribution to a particular project.

[0784] Complementing MBO evaluation

[0785] The user (supervisor) performs an MBO evaluation based on the provided report, checks the validity of the evaluation content, makes adjustments as necessary, and provides feedback to the employee.

[0786] Examples:

[0787] The supervisor uses the analysis results of the generated AI model to conduct an annual evaluation of User A. User A's leadership and focus are particularly highly evaluated, and a high MBO score is awarded.

[0788] In this way, the system of the present invention clearly shows employees' work achievements and enables convincing and fair evaluations. This system also contributes to improving employee satisfaction by providing managers with objective data to complement MBO evaluations.

[0789] The processing flow will be explained below.

[0790] Step 1: Data collection

[0791] The device collects the user's business data and meeting logs. Specifically, it periodically records the user's application usage history, file access logs, email content, keystroke data, etc. It also collects audio and video recording data of video and audio conferences.

[0792] Specific operation:

[0793] The terminal starts the log collection agent and collects business data every 15 minutes.

[0794] Acquire recording data via the video conferencing tool's API.

[0795] Step 2: Data Preprocessing

[0796] The server preprocesses the collected raw data. First, it converts the voice data into text data using a speech recognition tool, then performs morphological analysis of the text data using natural language processing (NLP) techniques. Numerical data is normalized and converted into a unified format.

[0797] Specific operation:

[0798] The server inputs the audio data into a transcription tool and converts it into text data.

[0799] The server performs morphological analysis, analyzes the text data, and extracts grammatical structure and semantic information.

[0800] The server normalizes the numerical data and converts it into a unified data format.

[0801] Step 3: Analysis by generative AI model

[0802] The server inputs the preprocessed data into the generative AI model and performs qualitative and quantitative evaluations. The qualitative evaluation involves analyzing the sentiment and importance of the comments, while the quantitative evaluation involves calculating metrics such as work time and application usage frequency.

[0803] Specific operation:

[0804] The server inputs the text data extracted from the meeting log into the generative AI model.

[0805] The server uses the generative AI model to generate sentiment analysis and content evaluation scores for the comments.

[0806] The server inputs the numerical data into the AI ​​model and generates an evaluation score based on work time and frequency of application use.

[0807] Step 4: Output of evaluation results

[0808] The server compiles the evaluation results from the generative AI model into a report and provides it to the user (supervisor). The report includes a qualitative evaluation (quality of speech content) and a quantitative evaluation (task time, statistics on applications used).

[0809] Specific operation:

[0810] The server automatically generates a report based on the evaluation results from the generated AI model.

[0811] The server periodically displays the report on the manager's dashboard or sends it by email.

[0812] Step 5: Complementing the MBO evaluation

[0813] The user (supervisor) performs an MBO evaluation based on the provided report, checks the validity of the evaluation content, makes adjustments as necessary, and provides feedback to the employee.

[0814] Specific operation:

[0815] The user checks the report and completes the MBO evaluation sheet based on the evaluation results of the generated AI model.

[0816] The user conducts a feedback interview based on the evaluation results and shares the evaluation details with the employee.

[0817] Through the above steps, the system of the present invention makes employees' work achievements visible and realizes a fair and convincing evaluation system.

[0818] Example 1

[0819] 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."

[0820] Traditional evaluation systems have the problem of making it difficult to accurately and fairly evaluate employees' work performance. In particular, there is a lack of quantitative methods for qualitative evaluations and contributions in meetings, which often leads to biased evaluations and a sense of unfairness. There is also a lack of materials for managers to conduct effective management-by-objectives (MBO) evaluations.

[0821] 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.

[0822] In this invention, the server includes means for continuously collecting business data and meeting logs from user terminals, means for preprocessing the collected business data and meeting logs using a speech recognition tool and natural language processing technology, means for inputting the preprocessed data into a generative AI model to perform qualitative and quantitative evaluations, and means for outputting the evaluation results by the generative AI model in report format. This enables employees' work performance to be evaluated objectively and fairly, and provides managers with complementary information for conducting effective MBO evaluations.

[0823] A "terminal" is an electronic device that a user uses to collect business data and meeting logs.

[0824] "Business data" refers to information related to a user's business, such as keystrokes, application usage history, file access logs, and email content.

[0825] "Meeting Log" refers to records of a meeting, including audio and video recordings of video and audio conferences, speeches, and related information.

[0826] "Speech recognition tool" refers to software or services for converting voice data into text data.

[0827] "Natural language processing technology" refers to computer science technology for analyzing text data and performing morphological analysis and grammatical structure analysis.

[0828] "Generative AI model" refers to an artificial intelligence model that uses preprocessed data to perform qualitative and quantitative evaluations.

[0829] "Qualitative evaluation" refers to the process of evaluating the content of a written piece or the quality of what is said in a meeting.

[0830] "Quantitative evaluation" refers to the process of evaluating based on numerical data such as work time and application usage.

[0831] "Report format" refers to the form of a document that summarizes the evaluation results in an easy-to-understand format, and the output includes formats such as Excel and PDF.

[0832] "Management by Objectives (MBO) evaluation" refers to a management technique in which managers evaluate employees' performance based on the degree to which they achieve goals.

[0833] The present invention provides a system that realizes a fair and convincing evaluation system by collecting users' business data and meeting logs and performing evaluations using a generative AI model. Specific embodiments for implementing the present invention are described below.

[0834] Data collection

[0835] The device continuously collects the user's business data and meeting logs. The device is equipped with a keyboard hook and a file system mini-filter driver to collect the following data:

[0836] Keystrokes

[0837] Application Usage History

[0838] File Access Log

[0839] Email content

[0840] Recordings of video and audio conferences

[0841] Examples:

[0842] For example, User A's terminal records the usage logs of all applications used during work and saves speeches made during meetings as audio data.

[0843] Data Preprocessing

[0844] The server converts the collected raw data into an analyzable format, specifically by following these steps:

[0845] The voice data is converted into text data using a voice recognition tool (e.g., Google Speech-to-Text API).

[0846] The text data is morphologically analyzed using natural language processing techniques (e.g., spaCy, NLTK) to analyze grammatical structure and meaning.

[0847] Numerical data is normalized using scaling and normalization techniques.

[0848] Examples:

[0849] The audio data from the meeting is converted into text data using a transcription tool, and morphological analysis is performed to evaluate the quality of what is being said. The numerical data of application usage time is also normalized into a unified format.

[0850] Analysis using generative AI models

[0851] The server inputs the preprocessed data into a generative AI model (e.g., OpenAI GPT-4) and performs qualitative and quantitative evaluation. The specific steps of the evaluation are as follows:

[0852] In qualitative evaluation, text data is analyzed to evaluate the quality of comments made during meetings and the importance of email content.

[0853] Quantitative evaluation involves analyzing log data and evaluating work time and application usage.

[0854] Examples:

[0855] The generative AI model receives a prompt such as "Evaluate User A's comments during the meeting and analyze how constructive they were," and the model outputs the evaluation results. The model also analyzes work time data using a prompt such as "Calculate User A's weekly work hours and evaluate their productivity."

[0856] Output of evaluation results

[0857] The server compiles the results of the analysis by the generative AI model into a report and provides it to the user (supervisor). This report includes the following:

[0858] Qualitative assessment (e.g., quality of content)

[0859] Quantitative evaluation (e.g., work time, application usage statistics)

[0860] Examples:

[0861] The server generates a report and provides it to the manager, which includes a qualitative assessment of User A's project management skills and problem-solving abilities, as well as quantitative data showing User A's time contribution to a particular project.

[0862] Complementing MBO evaluation

[0863] The user (supervisor) performs a management by objectives (MBO) evaluation based on the provided report, checks the validity of the evaluation content, makes adjustments as necessary, and provides feedback to the employee.

[0864] Examples:

[0865] The supervisor uses the analysis results of the generated AI model to conduct an annual evaluation of User A. User A's leadership and focus are particularly highly evaluated, and a high MBO score is awarded.

[0866] In this way, the system of the present invention clearly shows employees' work achievements and achieves fair evaluations. This system provides managers with objective data to complement MBO evaluations, thereby contributing to improved employee satisfaction.

[0867] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0868] Step 1: Data collection

[0869] The device continuously collects business data and meeting logs. Specifically, it collects user keystrokes, application usage history, file access logs, email content, and audio and video conference recordings. The input is user interactions and meeting participation, and the output is log data stored on the device.

[0870] Specific behavior:

[0871] The device records keystrokes using keyboard hooks, captures application usage history and file access logs using a dedicated service, and, when audio and video conferences are taking place, captures conference logs using dedicated audio and video recording software.

[0872] Step 2: Data Preprocessing

[0873] The server converts the collected raw data into an analyzable format. Specifically, it converts voice data into text data using a speech recognition tool (e.g., Google Speech-to-Text API), and then performs morphological analysis on the text data using natural language processing technology (e.g., spaCy, NLTK). Numerical data is normalized using scaling and normalization techniques. The input is the collected log data, and the output is preprocessed, analyzable data.

[0874] Specific behavior:

[0875] The server sends the audio file to a speech recognition API to obtain the transcribed text, then feeds the text data into an NLP tool for morphological analysis, and normalizes the numerical data using an appropriate scaling algorithm.

[0876] Step 3: Analysis by generative AI model

[0877] The server inputs the preprocessed data into a generative AI model (e.g., OpenAI GPT-4) to perform qualitative and quantitative evaluations. Qualitative evaluations assess the quality of comments made during meetings and the importance of email content, while quantitative evaluations assess task time and application usage. The input is the preprocessed data, and the output is the evaluation results.

[0878] Specific behavior:

[0879] The server inputs a prompt to the generative AI model, such as "Evaluate User A's comments during the meeting and analyze how constructive they were," and obtains a response from the model. It also analyzes the work time data using a prompt, such as "Calculate User A's weekly work hours and evaluate their productivity."

[0880] Step 4: Output of evaluation results

[0881] The server compiles the evaluation results analyzed by the generative AI model into a report. Specifically, it integrates qualitative evaluations (e.g., quality of speech) and quantitative evaluations (e.g., work time, statistics on applications used) to create a report. The input is the analyzed evaluation results, and the output is the report.

[0882] Specific behavior:

[0883] The server receives the output of the generative AI model and creates a report in formats such as Excel or PDF, which is visually organized using graphs and tables so that administrators can easily understand it.

[0884] Step 5: Complementing the MBO evaluation

[0885] The user (supervisor) performs a management by objectives (MBO) evaluation based on the provided report. Specifically, the user checks the validity of the evaluation content and makes adjustments as necessary. Finally, the user provides feedback to the employee. The input is the generated report, and the output is the revised evaluation and feedback.

[0886] Specific behavior:

[0887] The supervisor reviews the report and adds comments such as, "User A's leadership is evident, but we would like to see more specific examples." In addition, the supervisor discusses with the employee specific evaluation items and proposes the next steps for improvement.

[0888] (Application example 1)

[0889] 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."

[0890] At the workplace, it is important to accurately evaluate the performance and contribution of workers and to create a fair and transparent evaluation system. Factory floors, in particular, require a system that efficiently collects and analyzes work logs and robot interaction data, and evaluates them qualitatively and quantitatively.

[0891] 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.

[0892] In this invention, the server includes means for collecting business data and meeting logs from terminals, means for preprocessing the collected business data and meeting logs, means for analyzing the preprocessed data and using a generative AI model for qualitative and quantitative evaluation, means for outputting the evaluation results by the generative AI model in report format, means for collecting operation data and work logs of robots in the factory, means for collecting interaction data with workers, and means for analyzing the collected data and evaluating the performance of workers. This makes it possible to clarify the work achievements of workers and provide convincing and fair evaluations.

[0893] "Device" means a computer device used to collect and process data.

[0894] "Business data" refers to information such as applications that users operate during work, files that they access, and communication content.

[0895] "Meeting logs" are audio and video recordings related to statements made during a meeting and the agenda, as well as textual versions of such recordings.

[0896] "Preprocessing" is the process of converting collected raw data into an analyzable format.

[0897] A "generative AI model" is an artificial intelligence model that uses collected data to perform qualitative and quantitative evaluations.

[0898] "Qualitative evaluation" is the process of evaluating the content of a written piece or the quality of what is said in a meeting.

[0899] "Quantitative evaluation" is a process of evaluation based on numerical data such as log data and work time.

[0900] "Report format" refers to the format of a document summarizing the evaluation results.

[0901] "Factory robots" are automated machinery used to streamline factory work.

[0902] "Operation data" refers to records and log data relating to the robot's operations.

[0903] A "work log" is recorded data on the actions and work details of workers.

[0904] "Interaction data" is recorded data of the operations and communication that take place alternately between workers and robots.

[0905] "Evaluation results" refer to the qualitative and quantitative evaluation values ​​and evaluation content calculated by the generative AI model.

[0906] To implement this invention, a terminal first collects business data and a meeting log. Business data includes information such as the operation history of applications used by the user, files accessed, and emails sent and received. The meeting log includes audio and video data related to comments made during the meeting and the agenda.

[0907] The collected data is sent to a server where it undergoes preprocessing. During the preprocessing stage, audio data is converted into text data using a transcription tool. Next, natural language processing (NLP) techniques are used to perform morphological analysis of the text data, analyzing its grammatical structure and meaning. Numerical data is normalized and converted into a unified format.

[0908] Once preprocessed, the data is input into a generative AI model for qualitative and quantitative evaluation. In qualitative evaluation, the generative AI model evaluates the content of text and the quality of comments made in meetings. For example, it extracts meaningful comments about project progress from the contents of a user's email. In quantitative evaluation, it performs evaluation based on log data and work time. For example, it measures the total amount of time a user spends on a specific project.

[0909] The evaluation results are compiled in a report format by the server and provided to the user (manager). This report includes both qualitative and quantitative evaluations. Based on this report, the manager can obtain objective information for performing a management by objectives (MBO) evaluation. Using this information, the manager can check the validity of the evaluation content and make adjustments as necessary.

[0910] Furthermore, the system can collect operation data and work logs of robots in factories and analyze interaction data with workers, which will further improve the efficiency of factory work and enable fair evaluations.

[0911] As a concrete example, a robot in a factory collects speech data from workers, converts it into text data, performs morphological analysis, evaluates the quality of the speech, quantitatively evaluates the worker's performance using work log data, and integrates this data to provide the manager with the evaluation results in the form of a report.

[0912] An example of a prompt sentence to input to the generative AI model is, "Please summarize what was said in this meeting and extract the most important points." Using this prompt allows for efficient qualitative evaluation of meeting logs.

[0913] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0914] Step 1:

[0915] The device collects business data and meeting logs. Business data includes the operation history of applications used by the user, files accessed, and emails sent and received. Meeting logs include audio and video recordings related to comments made during meetings and the agenda. This data is obtained using software (e.g., data collection API). Input is the user's activity log and meeting audio, and output is the collected raw data.

[0916] Step 2:

[0917] The terminal sends the collected data to the server. The address of the server is specified and the data is sent in encrypted format via the required protocol (e.g. HTTP / HTTPS). The input is the raw data acquired, and the output is the data sent to the server.

[0918] Step 3:

[0919] The raw data received by the server is preprocessed. First, the voice data is converted into text data using a speech recognition tool (e.g., Python's speech_recognition library). Next, the text data is subjected to morphological analysis using natural language processing (NLP) techniques (e.g., the spacy library) to analyze grammatical structure and meaning. Numerical data is converted into a data frame and normalized. The input is raw data, and the output is preprocessed data.

[0920] Step 4:

[0921] The preprocessed data is input into a generative AI model. The generative AI model (e.g., GPT-2 from Hugging Face) analyzes the text data and performs a qualitative evaluation. An example of a prompt used here is, "Please summarize what was said in this meeting and extract the most important points." The input is the preprocessed text data and the prompt, and the output is the result of the qualitative evaluation.

[0922] Step 5:

[0923] The preprocessed numerical data is used to perform quantitative evaluation. The server analyzes the data frame and calculates statistics on work time and application usage time. For example, it calculates the time spent on a specific project based on work time. The input is the preprocessed numerical data, and the output is the results of the quantitative evaluation.

[0924] Step 6:

[0925] The server integrates the results of the qualitative and quantitative evaluations and generates a report. The report includes qualitative analysis results and quantitative statistics. It displays a detailed analysis of the content of the text, the quality of the comments, and the working time. The input is the results of the qualitative and quantitative evaluations, and the output is a report-format file.

[0926] Step 7:

[0927] The server provides the generated report to the user (administrator). The report is made accessible in dashboard or PDF format so that the administrator can easily check the evaluation. The input is the generated report, and the output is the report used by the user in the actual evaluation process.

[0928] 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.

[0929] The present invention is a system that visualizes employees' work performance by performing qualitative and quantitative evaluations using a generative AI model using business data and meeting logs collected from terminals, thereby realizing a fair and convincing evaluation system. Furthermore, the present invention improves the accuracy and comprehensiveness of evaluations by combining it with an emotion engine that recognizes the user's emotions. Specific embodiments for implementing the present invention are described below.

[0930] Data collection

[0931] The device continuously collects the user's business data and meeting logs. Specifically, it acquires business data such as the user's application usage history, file access logs, email content, and keystroke data. It also collects audio and video recordings of video and audio conferences.

[0932] Examples:

[0933] For example, the terminal of user A records a log of the applications used throughout the day and saves the contents of statements made during a meeting as audio data.

[0934] Data Preprocessing

[0935] The server converts the collected raw data into an analyzable format. First, the voice data is converted into text data using a speech recognition tool. Next, the text data is subjected to morphological analysis using natural language processing (NLP) techniques to analyze grammatical structure and meaning. Numerical data is normalized and converted into a unified format. An emotion engine is also implemented at this stage.

[0936] Examples:

[0937] For example, audio data from a meeting is converted into text data using a transcription tool, and morphological analysis is performed to make what is being said easier to understand. Numerical data such as application usage time is also converted into a unified format. An emotion engine analyzes the user's emotions from the text data.

[0938] Analysis using generative AI models

[0939] The server inputs the preprocessed data into a generative AI model for qualitative and quantitative evaluation. Qualitative evaluation evaluates the content of text and the quality of speech in meetings. The emotion engine also analyzes the user's emotional state and incorporates it into the evaluation. Quantitative evaluation uses log data such as work time and application usage.

[0940] Examples:

[0941] For example, a generative AI model extracts meaningful comments about project progress from User A's email content and gives a high rating. At the same time, it quantitatively evaluates that User A is spending more than 40 hours per week on a specific project based on the amount of time worked. An emotion engine analyzes the emotions in User A's email and reflects their emotional state, such as "positive" or "negative," in the rating.

[0942] Output of evaluation results

[0943] The server compiles the results of the analysis by the generative AI model and emotion engine into a report and provides it to the user (supervisor). The report includes qualitative evaluations (quality of speech content, emotional state) and quantitative evaluations (work time, statistics on applications used).

[0944] Examples:

[0945] The server generates a report and provides it to the manager, which includes a qualitative assessment of User A's project management skills and problem-solving abilities, as well as quantitative data showing his time contribution to a specific project. The report also includes an emotional assessment based on the emotion engine.

[0946] Complementing MBO evaluation

[0947] The user (supervisor) performs an MBO evaluation based on the provided report, checks the validity of the evaluation content, makes adjustments as necessary, and provides feedback to the employee.

[0948] Examples:

[0949] The supervisor uses the analysis results of the generative AI model and the emotion engine to conduct an annual evaluation of User A. In particular, User A's leadership and focus are highly evaluated, and a high MBO score is awarded. User A's emotional state is also taken into consideration, and the evaluation content is adjusted accordingly.

[0950] In this way, the system of the present invention clearly shows the employee's work achievements and enables convincing and fair evaluations. At the same time, the introduction of an emotion engine enables comprehensive evaluations that take into account the employee's emotional state.

[0951] The processing flow will be explained below.

[0952] Step 1: Data collection

[0953] The device collects the user's business data and conference logs. Specifically, it acquires business data such as the user's application usage history, file access logs, email content, and keystroke data. It also collects audio and video recordings of video and audio conferences.

[0954] Specific operation:

[0955] The terminal starts the log collection agent and collects business data every 15 minutes.

[0956] Acquire recording data via the video conferencing tool's API.

[0957] Step 2: Data Preprocessing

[0958] The server preprocesses the collected raw data. First, the voice data is converted into text data using a voice recognition tool. Next, the text data undergoes morphological analysis using natural language processing (NLP) techniques. Numerical data is normalized and converted into a unified format. An emotion engine analyzes the user's emotions from the text data.

[0959] Specific operation:

[0960] The server inputs the audio data into a transcription tool and converts it into text data.

[0961] The server performs morphological analysis, analyzes the text data, and extracts grammatical structure and semantic information.

[0962] The server normalizes the numerical data and converts it into a unified data format.

[0963] The emotion engine analyzes the user's emotional state from the text data.

[0964] Step 3: Analysis by generative AI models and emotion engines

[0965] The server inputs the preprocessed data into the generative AI model and emotion engine to perform qualitative and quantitative evaluations. Qualitative evaluation evaluates the content of text and the quality of speech in meetings. The emotion engine also analyzes the user's emotional state and incorporates it into the evaluation. Quantitative evaluation uses log data such as work time and application usage.

[0966] Specific operation:

[0967] The server inputs the text data extracted from the meeting log into the generative AI model.

[0968] The server uses the generative AI model to generate sentiment analysis and content evaluation scores for the comments.

[0969] The server inputs the numerical data into the AI ​​model and generates an evaluation score based on work time and frequency of application use.

[0970] The emotion engine analyzes the user's text data and reflects their emotional state in the evaluation score.

[0971] Step 4: Output of evaluation results

[0972] The server compiles the evaluation results from the generative AI model and emotion engine into a report and provides it to the user (supervisor). The report includes qualitative evaluations (quality of speech content, emotional state) and quantitative evaluations (work time, statistics on applications used).

[0973] Specific operation:

[0974] The server automatically generates a report based on the evaluation results from the AI ​​model and emotion engine.

[0975] The server periodically displays the report on the manager's dashboard or sends it by email.

[0976] Step 5: Complementing the MBO evaluation

[0977] The user (supervisor) performs an MBO evaluation based on the provided report, checks the validity of the evaluation content, makes adjustments as necessary, and provides feedback to the employee.

[0978] Specific operation:

[0979] The user reviews the report and completes the MBO evaluation sheet based on the evaluation results of the generative AI model and emotion engine.

[0980] The user conducts a feedback interview based on the evaluation results and shares the evaluation details with the employee.

[0981] Through these steps, the system of the present invention visualizes employees' work performance and realizes a fair and convincing evaluation system. In addition, the introduction of an emotion engine enables comprehensive evaluation that takes into account the emotional aspects of employees.

[0982] Example 2

[0983] 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."

[0984] Conventional performance evaluation systems often evaluate employees' work performance qualitatively, making it difficult to base evaluations on concrete data. They also lack comprehensive evaluations that take emotional aspects into account, which can lead to evaluations that lack fairness and satisfaction. Furthermore, insufficient pre-processing of meeting logs and work data often reduces the accuracy of evaluation results.

[0985] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for collecting business data and meeting logs from terminals, a means for preprocessing the collected business data and meeting logs, a means for analyzing the preprocessed data and using a generative AI model for qualitative and quantitative evaluation, a means for outputting the evaluation results by the generative AI model in report format, and a means for recognizing emotions from text data using an emotion engine. This enables a fair and comprehensive evaluation of employees' work performance based on specific data and taking into account their emotional aspects.

[0986] A "terminal" is a computer or mobile device used by a user, and is a device for collecting business data and meeting logs.

[0987] "Business data" refers to all data related to a user's business activities, including application usage history, file access logs, email content, keystroke data, and the like.

[0988] A "meeting log" refers to recorded data of a meeting, including audio and video, and is data that captures the content of statements and discussions made during the meeting.

[0989] "Collection means" refers to the method or function for acquiring business data and meeting logs from a terminal, and takes the form of software or hardware.

[0990] "Preprocessing" refers to the process of converting raw data into an analyzable format, and includes transcribing audio data, morphological analysis of text data, and normalizing numerical data.

[0991] A "generative AI model" refers to an artificial intelligence model that has functions such as response generation and data analysis, and is used here to perform qualitative and quantitative evaluations.

[0992] An "emotion engine" refers to a program or tool that recognizes and analyzes a user's emotional state from text data.

[0993] "Report format" refers to a format that visually summarizes the evaluation results from the generative AI model and emotion engine in an easy-to-understand manner, and includes formats such as PDF and web dashboards.

[0994] "Manager" refers to the person or position with the authority to conduct MBO evaluations based on the evaluation results, and is usually a supervisor or human resources officer.

[0995] "MBO" is an abbreviation for Management by Objectives, a management method that evaluates employees based on the degree to which they have achieved their goals.

[0996] The present invention provides a system that visualizes employees' work performance by performing qualitative and quantitative evaluations using a generative AI model using business data and meeting logs collected from terminals, thereby realizing a fair and convincing evaluation system. The present invention also provides a system that improves the accuracy and comprehensiveness of evaluations by combining it with an emotion engine that recognizes user emotions. Specific embodiments for implementing the present invention are described below.

[0997] Data collection

[0998] The device continuously collects the user's business data and meeting logs. Specifically, it acquires business data such as the user's application usage history, file access logs, email content, and keystroke data. It also collects audio and video recordings of video and audio conferences.

[0999] Examples:

[1000] For example, the terminal of user A records a log of the applications used throughout the day and saves the contents of statements made during a meeting as audio data.

[1001] Data Transfer

[1002] The device sends the collected data to the server in real time. The data collection program in the device periodically uploads the locally stored data to the server. For security reasons, the data is sent in encrypted form.

[1003] Data Preprocessing

[1004] The server converts the received raw data into an analyzable format. First, the voice data is converted into text data using a speech recognition tool (e.g., Google Cloud Speech-to-Text). Next, the text data is subjected to morphological analysis using natural language processing (NLP) technology (e.g., Spacy) to analyze grammatical structure and meaning. Numerical data is normalized from Excel or CSV format and converted into a unified format (e.g., JSON format). Finally, the text data is analyzed for sentiment using the API of an emotion engine (e.g., Microsoft Azure Text Analytics).

[1005] Examples:

[1006] For example, audio data from a meeting is transcribed using Google Cloud Speech-to-Text, and morphological analysis is performed using Spacy to make what is being said easier to understand. Numerical data such as application usage time is converted from Excel or CSV format into a unified format. An emotion engine analyzes user emotions from the text data.

[1007] Analysis using generative AI models

[1008] The server inputs the preprocessed data into a generative AI model (e.g., OpenAI GPT-3) to perform qualitative and quantitative evaluations. Qualitative evaluations evaluate the content of text and the quality of speech in meetings. An emotion engine also analyzes the user's emotional state and incorporates it into the evaluation. Quantitative evaluations use log data such as work time and application usage.

[1009] Examples:

[1010] For example, a generative AI model extracts meaningful comments about project progress from User A's email content and gives a high rating. At the same time, it quantitatively evaluates that User A is spending more than 40 hours per week on a specific project based on the amount of time worked. An emotion engine analyzes the emotions in User A's email and reflects their emotional state, such as "positive" or "negative," in the rating.

[1011] Example prompt sentence:

[1012] "Evaluate User A's important contributions to the project progress based on his emails and meeting comments over the past month."

[1013] "Quantitatively evaluate User A's time contribution to a specific project based on his / her weekly work time data."

[1014] Sentiment analysis results integration

[1015] The server uses an emotion engine to integrate the results of the emotion analysis with the evaluation from the generative AI model. The emotion analysis results are combined with the output of the AI ​​model to reflect positive and negative comments in the evaluation score. Detailed results of the emotion analysis are also added to the report.

[1016] Generating evaluation results

[1017] The server compiles the generated evaluation results into a report. The server uses a report generation tool (e.g., JasperReports) to create a detailed report that includes both qualitative and quantitative evaluations. The report is output in PDF format or web dashboard format.

[1018] Providing reports

[1019] The server provides the completed report to the user (supervisor). The report is automatically sent to the supervisor's email address periodically (for example, at the end of the month). The report is also uploaded to the company portal so that the user can access it.

[1020] Complementing MBO evaluation

[1021] The user (supervisor) performs an MBO evaluation based on the provided report. The supervisor reviews the report and provides additional feedback or supplementary information as needed. Once the evaluation is complete, a session is set up to provide feedback to the employee being evaluated.

[1022] In this way, each processing step of the system works together in a concrete way to clarify employees' work achievements and achieve fair and convincing evaluations. The introduction of an emotion engine makes it possible to provide comprehensive evaluations that take into account employees' emotional states.

[1023] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1024] Step 1: Data collection

[1025] The terminal collects the user's business data and meeting logs.

[1026] Input: User application usage history, file access logs, email content, keystroke data, and audio and video conference recordings.

[1027] Data processing: Raw data is automatically acquired and collected 24 / 7.

[1028] Output: Locally stored business data and meeting logs.

[1029] Specific operation: The device will periodically record application logs and file access logs, and will automatically start recording when an audio conference is started.

[1030] Step 2: Data Transfer

[1031] The data collected by the device is sent to the server in real time.

[1032] Input: Business data and meeting logs stored locally on the device.

[1033] Data processing: Encrypt the data and upload it securely to the server.

[1034] Output: Aggregated business data and meeting logs on the server.

[1035] Specific operation: The data collection program in the device periodically (for example, every hour) encrypts the data and uploads it to the server.

[1036] Step 3: Data Preprocessing

[1037] The server converts the raw data it receives into a format that can be analyzed.

[1038] Input: Raw data (business data, meeting logs) aggregated on the server.

[1039] Data processing: Convert the voice data into text data using Google Cloud Speech-to-Text, perform morphological analysis on the text data using Spacy, normalize the numerical data and convert it into a unified format (e.g., JSON), and analyze emotions from the text data using an emotion engine.

[1040] Output: Data converted into an analyzable format (text data, morphological analysis results, normalized numerical data, sentiment analysis results).

[1041] How it works: The server transcribes the audio data using Google Cloud Speech-to-Text, then performs morphological analysis using Spacy. Numerical data is converted from Excel or CSV to JSON format, and sentiment analysis is performed using a sentiment analysis tool.

[1042] Step 4: Data integration and storage

[1043] The server consolidates the pre-processed data and stores it in a database.

[1044] Input: Preprocessed data (text data, morphological analysis results, normalized numerical data, sentiment analysis results).

[1045] Data processing: Integrate the data, classify it by user, and store it in a NoSQL database (e.g., MongoDB).

[1046] Output: Data integrated and stored in a database.

[1047] Specific operation: The server inserts the preprocessed data into MongoDB, classifies it by user, and creates indexes to enable efficient searches.

[1048] Step 5: Analysis by generative AI model

[1049] The server inputs the preprocessed data into a generative AI model for qualitative and quantitative evaluation.

[1050] Input: Preprocessed data extracted from the database.

[1051] Data processing: A generative AI model (e.g., OpenAI GPT-3) is invoked with a prompt sentence to perform qualitative evaluation (text content and speech quality) and quantitative evaluation (task time, application usage).

[1052] Output: Analysis results from the generative AI model.

[1053] Specific operation: The server inputs data into the OpenAI GPT-3 model using the prompt sentence and generates an appropriate evaluation score and comment.

[1054] Step 6: Integrating the results of sentiment analysis

[1055] The server uses an emotion engine to integrate the results of emotion analysis with the evaluation from the generative AI model.

[1056] Input: Analysis results from generative AI model, sentiment analysis results.

[1057] Data processing: The results of sentiment analysis are combined with the output of the AI ​​model and reflected in the evaluation score.

[1058] Output: Evaluation results integrated with sentiment analysis results.

[1059] Specific operation: The server adds the results from the emotion engine to the output of the generative AI model, and reflects positive or negative emotions in the evaluation score.

[1060] Step 7: Generate evaluation results

[1061] The server compiles the generated evaluation results in a report format.

[1062] Input: Evaluation results integrated with sentiment analysis results.

[1063] Data processing: Create reports using a report generation tool (e.g., JasperReports) and convert them into PDF or web dashboard formats.

[1064] Output: Assessment report in PDF format and web dashboard format.

[1065] What it does: The server compiles the results of the assessment into a detailed report and converts it into a visually understandable format.

[1066] Step 8: Providing the report

[1067] The server provides the completed report to the user (supervisor).

[1068] Input: Assessment report in PDF format and web dashboard format.

[1069] Data Processing: Present the report to your superior in an appropriate manner.

[1070] Output: Evaluation report provided to supervisor.

[1071] What it does: The server sends the report to the manager's email address at the end of the month and also uploads it to the company's internal portal for access.

[1072] Step 9: Complementing the MBO evaluation

[1073] The user (supervisor) performs an MBO evaluation based on the provided report.

[1074] Input: Evaluation report provided to supervisor.

[1075] Data processing: Check the evaluation content and provide any necessary feedback or supplementary information.

[1076] Output: Completed MBO evaluation and feedback.

[1077] What happens next: The manager reviews the report, notes any necessary corrections or additional information, and schedules a feedback session.

[1078] In this way, the system of the present invention evaluates employees' work performance in a concrete and data-based manner, and achieves a fair and comprehensive evaluation that also takes into account their emotional aspects.

[1079] (Application example 2)

[1080] 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."

[1081] Existing evaluation systems do not adequately perform qualitative and quantitative evaluations based on work data and meeting logs. This makes it difficult to fairly evaluate employees' work performance and the operational efficiency of production equipment. Another problem is the lack of systems that can perform comprehensive evaluations using robot operation logs and performance indicators. Furthermore, it is necessary to improve the accuracy and comprehensiveness of evaluations by combining an emotion engine.

[1082] The identification processing 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 means for collecting business data and meeting logs from terminals, means for preprocessing the collected business data and meeting logs, means for analyzing the preprocessed data and using a generative AI model for qualitative and quantitative evaluation, means for outputting the evaluation results by the generative AI model in report format, means for collecting operation log data and performance indicators of the production equipment, means for converting the collected operation log data and performance indicators into an analyzable format, and means for qualitatively evaluating the operation log data and quantitatively evaluating the performance indicators using the generative AI model and an emotion recognition engine. This not only enables fair and convincing evaluation based on the business data and meeting logs, but also enables comprehensive evaluation of the operational efficiency of the production equipment.

[1083] A "terminal" is an electronic device such as a computer or smartphone used by a user, and is used to collect business data and meeting logs.

[1084] "Business data" refers to records such as application usage history, file access logs, email content, and keystroke data that are generated when a user performs a task.

[1085] "Meeting logs" refer to audio and video recordings of video and audio conferences, including audio data converted into text data and minutes of meetings.

[1086] "Preprocessing" is the process of converting collected raw data into an analyzable format, and includes transcribing audio data, morphological analysis of text data, and normalizing numerical data.

[1087] A "generative AI model" is a model that uses machine learning algorithms to analyze collected data and perform qualitative and quantitative evaluations.

[1088] An "emotion recognition engine" is a technology for analyzing a user's emotional state from text data and voice data, and reflects "positive," "negative," and changes therein in the evaluation.

[1089] "Production equipment" refers to equipment such as robots used in factories and production lines, and is the subject of collection of operation log data and performance indicators.

[1090] "Operation log data" refers to data that includes the operation history and operation records of robots and production equipment, and mainly includes detailed records of the production process.

[1091] "Performance index" refers to the key performance metrics, such as the efficiency or error rate achieved by a robot or production equipment, expressed in specific numerical values.

[1092] "Qualitative evaluation" is a type of evaluation performed by a generative AI model, and is a qualitative evaluation that is primarily based on the content of statements and operation logs.

[1093] "Quantitative evaluation" is a type of evaluation performed by a generative AI model, and is a quantitative evaluation based primarily on numerical data such as work time and performance indicators.

[1094] This invention provides a system for fair and comprehensive evaluation of the operational efficiency of production equipment and the work performance of employees. The system includes a terminal, a server, a generative AI model, and an emotion recognition engine.

[1095] First, the terminal continuously collects operation log data and performance indicators of production equipment. For example, it acquires performance data such as operation logs, work efficiency, and error rates of robots installed in a factory. This allows the operation history and operation records of production equipment, as well as key performance metrics, to be collected.

[1096] Next, the server preprocesses the collected data. At this stage, the collected raw data is converted into an analyzable format. Specifically, the operation log data is subjected to morphological analysis, and the performance indicator data is normalized. If voice data is included, it is converted into text data using a voice recognition tool.

[1097] The preprocessed data is input into a generative AI model, which performs both qualitative and quantitative evaluations. In qualitative evaluation, meaningful utterances and behavioral patterns are extracted from the operation log data, and the emotional state is analyzed using an emotion recognition engine. In quantitative evaluation, evaluation is performed based on performance indicators such as work efficiency and error rate.

[1098] The server outputs the analysis results of the generative AI model and emotion recognition engine in the form of a report. This report includes a qualitative evaluation of the robot's operation and emotional state, as well as a quantitative evaluation of each performance indicator. This allows administrators to conduct comprehensive and fair evaluations.

[1099] As a specific example, the operation log of the robot 123 is acquired as "log data of robot 123", and the evaluation results include quantitative and qualitative evaluations as follows:

[1100] Qualitative evaluation: Extract meaningful operation patterns from the robot's operation log and give a positive evaluation using an emotion engine.

[1101] Quantitative evaluation: Since the work efficiency is 95% and the error rate is 2%, the evaluation is "Efficiency: High" and "Error rate: Low."

[1102] Example prompt sentence:

[1103] "Collect the operation log data and performance metrics of Robot 123 and evaluate it using the generative AI model and emotion engine."

[1104] This system allows for a comprehensive evaluation of production equipment and employee performance, resulting in rational and convincing evaluations.

[1105] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1106] Step 1:

[1107] The terminal collects operation log data and performance indicators of production equipment. Specifically, data is acquired using sensors and controllers that record the operation history, operation records, work efficiency, and error rates of robots installed in the factory. The input is real-time operation log data and performance indicators collected from the robots, which are stored in a database. The output is the collected operation log data and raw performance indicator data.

[1108] Step 2:

[1109] The server preprocesses the collected raw data. Specifically, it analyzes the operation log data using a morphological analysis tool (e.g., MeCab) and converts the grammatical structure and meaning into an understandable format. It also normalizes the performance indicator data and converts it into a unified format. If voice data is included, it is converted into text data using voice recognition software (e.g., Google Speech-to-Text). The input is raw data obtained from the device, and the output is preprocessed data that can be analyzed.

[1110] Step 3:

[1111] The server inputs the preprocessed data into a generative AI model to perform qualitative and quantitative evaluations. In qualitative evaluation, a generative AI model (e.g., Hugging Face Transformers) is used to analyze the contents of the operation log data and extract meaningful operation patterns and comments. An emotion recognition engine (e.g., Sentiment Analysis API) is used to evaluate the user's emotional state from the operation log. In quantitative evaluation, work efficiency and error rates are analyzed based on performance index data. The input is preprocessed data, and the output is qualitative and quantitative evaluation results.

[1112] Step 4:

[1113] The server outputs the evaluation results of the generated AI model and emotion recognition engine in report format. Specifically, the evaluation results are compiled in Excel or PDF format and provided to the administrator. The report includes the robot's operation details, emotional state, and evaluation results of each performance indicator. The input is the evaluation result data, and the output is the evaluation results in report format.

[1114] Step 5:

[1115] The user evaluates production equipment and employees based on the generated report. The manager reviews the report content and uses it as a supplementary reference for management by objectives (MBO) evaluation. Specifically, the report comprehensively evaluates the robot's operating efficiency, error rate, employee work performance, etc. The input is the evaluation results in report format, and the output is the manager's final evaluation results.

[1116] 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.

[1117] 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.

[1118] 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.

[1119] [Fourth embodiment]

[1120] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1121] 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.

[1122] 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).

[1123] 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.

[1124] 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.

[1125] 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).

[1126] 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.

[1127] 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.

[1128] 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.

[1129] 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.

[1130] 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.

[1131] 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.

[1132] 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."

[1133] The present invention is a system that visualizes employees' work performance by performing qualitative and quantitative evaluations using a generative AI model using business data and meeting logs collected from terminals, thereby realizing a fair and convincing evaluation system. Specific embodiments for implementing the present invention are described below.

[1134] Data collection

[1135] The device continuously collects the user's business data and meeting logs. Specifically, it acquires business data such as keystrokes, application usage history, file access logs, and email content. It also collects audio and video recordings of video and audio conferences.

[1136] Examples:

[1137] For example, the terminal of user A records a log of the applications used throughout the day and saves the contents of statements made during a meeting as audio data.

[1138] Data Preprocessing

[1139] The server converts the collected raw data into an analyzable format. First, the voice data is converted into text data using a speech recognition tool. Next, the text data is subjected to morphological analysis using natural language processing (NLP) techniques to analyze grammatical structure and meaning. Numerical data is normalized and converted into a unified format.

[1140] Examples:

[1141] For example, audio data from a meeting can be converted into text using a transcription tool, and morphological analysis can be performed to make what was said easier to understand. Numerical data such as application usage time can also be converted into a unified format.

[1142] Analysis using generative AI models

[1143] The server inputs the preprocessed data into the generative AI model and performs qualitative and quantitative evaluations. Qualitative evaluations evaluate the content of text and the quality of speech in meetings. Quantitative evaluations use log data such as work time and application usage.

[1144] Examples:

[1145] For example, a generative AI model can extract meaningful comments about project progress from User A's emails and give them a high rating. At the same time, it can quantitatively evaluate User A's spending more than 40 hours per week on a specific project based on the amount of time worked.

[1146] Output of evaluation results

[1147] The server compiles the results of the analysis by the generative AI model into a report and provides it to the user (supervisor). The report includes qualitative evaluations (e.g., quality of comments) and quantitative evaluations (e.g., work time, statistics on applications used).

[1148] Examples:

[1149] The server generates a report and provides it to the manager, which includes a qualitative assessment of User A's project management skills and problem-solving abilities, as well as quantitative data showing User A's time contribution to a particular project.

[1150] Complementing MBO evaluation

[1151] The user (supervisor) performs an MBO evaluation based on the provided report, checks the validity of the evaluation content, makes adjustments as necessary, and provides feedback to the employee.

[1152] Examples:

[1153] The supervisor uses the analysis results of the generated AI model to conduct an annual evaluation of User A. User A's leadership and focus are particularly highly evaluated, and a high MBO score is awarded.

[1154] In this way, the system of the present invention clearly shows employees' work achievements and enables convincing and fair evaluations. This system also contributes to improving employee satisfaction by providing managers with objective data to complement MBO evaluations.

[1155] The processing flow will be explained below.

[1156] Step 1: Data collection

[1157] The device collects the user's business data and meeting logs. Specifically, it periodically records the user's application usage history, file access logs, email content, keystroke data, etc. It also collects audio and video recording data of video and audio conferences.

[1158] Specific operation:

[1159] The terminal starts the log collection agent and collects business data every 15 minutes.

[1160] Acquire recording data via the video conferencing tool's API.

[1161] Step 2: Data Preprocessing

[1162] The server preprocesses the collected raw data. First, it converts the voice data into text data using a speech recognition tool, then performs morphological analysis of the text data using natural language processing (NLP) techniques. Numerical data is normalized and converted into a unified format.

[1163] Specific operation:

[1164] The server inputs the audio data into a transcription tool and converts it into text data.

[1165] The server performs morphological analysis, analyzes the text data, and extracts grammatical structure and semantic information.

[1166] The server normalizes the numerical data and converts it into a unified data format.

[1167] Step 3: Analysis by generative AI model

[1168] The server inputs the preprocessed data into the generative AI model and performs qualitative and quantitative evaluations. The qualitative evaluation involves analyzing the sentiment and importance of the comments, while the quantitative evaluation involves calculating metrics such as work time and application usage frequency.

[1169] Specific operation:

[1170] The server inputs the text data extracted from the meeting log into the generative AI model.

[1171] The server uses the generative AI model to generate sentiment analysis and content evaluation scores for the comments.

[1172] The server inputs the numerical data into the AI ​​model and generates an evaluation score based on work time and frequency of application use.

[1173] Step 4: Output of evaluation results

[1174] The server compiles the evaluation results from the generative AI model into a report and provides it to the user (supervisor). The report includes a qualitative evaluation (quality of speech content) and a quantitative evaluation (task time, statistics on applications used).

[1175] Specific operation:

[1176] The server automatically generates a report based on the evaluation results from the generated AI model.

[1177] The server periodically displays the report on the manager's dashboard or sends it by email.

[1178] Step 5: Complementing the MBO evaluation

[1179] The user (supervisor) performs an MBO evaluation based on the provided report, checks the validity of the evaluation content, makes adjustments as necessary, and provides feedback to the employee.

[1180] Specific operation:

[1181] The user checks the report and completes the MBO evaluation sheet based on the evaluation results of the generated AI model.

[1182] The user conducts a feedback interview based on the evaluation results and shares the evaluation details with the employee.

[1183] Through the above steps, the system of the present invention makes employees' work achievements visible and realizes a fair and convincing evaluation system.

[1184] Example 1

[1185] 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."

[1186] Traditional evaluation systems have the problem of making it difficult to accurately and fairly evaluate employees' work performance. In particular, there is a lack of quantitative methods for qualitative evaluations and contributions in meetings, which often leads to biased evaluations and a sense of unfairness. There is also a lack of materials for managers to conduct effective management-by-objectives (MBO) evaluations.

[1187] 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.

[1188] In this invention, the server includes means for continuously collecting business data and meeting logs from user terminals, means for preprocessing the collected business data and meeting logs using a speech recognition tool and natural language processing technology, means for inputting the preprocessed data into a generative AI model to perform qualitative and quantitative evaluations, and means for outputting the evaluation results by the generative AI model in report format. This enables employees' work performance to be evaluated objectively and fairly, and provides managers with complementary information for conducting effective MBO evaluations.

[1189] A "terminal" is an electronic device that a user uses to collect business data and meeting logs.

[1190] "Business data" refers to information related to a user's business, such as keystrokes, application usage history, file access logs, and email content.

[1191] "Meeting Log" refers to records of a meeting, including audio and video recordings of video and audio conferences, speeches, and related information.

[1192] "Speech recognition tool" refers to software or services for converting voice data into text data.

[1193] "Natural language processing technology" refers to computer science technology for analyzing text data and performing morphological analysis and grammatical structure analysis.

[1194] "Generative AI model" refers to an artificial intelligence model that uses preprocessed data to perform qualitative and quantitative evaluations.

[1195] "Qualitative evaluation" refers to the process of evaluating the content of a written piece or the quality of what is said in a meeting.

[1196] "Quantitative evaluation" refers to the process of evaluating based on numerical data such as work time and application usage.

[1197] "Report format" refers to the form of a document that summarizes the evaluation results in an easy-to-understand format, and the output includes formats such as Excel and PDF.

[1198] "Management by Objectives (MBO) evaluation" refers to a management technique in which managers evaluate employees' performance based on the degree to which they achieve goals.

[1199] The present invention provides a system that realizes a fair and convincing evaluation system by collecting users' business data and meeting logs and performing evaluations using a generative AI model. Specific embodiments for implementing the present invention are described below.

[1200] Data collection

[1201] The device continuously collects the user's business data and meeting logs. The device is equipped with a keyboard hook and a file system mini-filter driver to collect the following data:

[1202] Keystrokes

[1203] Application Usage History

[1204] File Access Log

[1205] Email content

[1206] Recordings of video and audio conferences

[1207] Examples:

[1208] For example, User A's terminal records the usage logs of all applications used during work and saves speeches made during meetings as audio data.

[1209] Data Preprocessing

[1210] The server converts the collected raw data into an analyzable format, specifically by following these steps:

[1211] The voice data is converted into text data using a voice recognition tool (e.g., Google Speech-to-Text API).

[1212] The text data is morphologically analyzed using natural language processing techniques (e.g., spaCy, NLTK) to analyze grammatical structure and meaning.

[1213] Numerical data is normalized using scaling and normalization techniques.

[1214] Examples:

[1215] The audio data from the meeting is converted into text data using a transcription tool, and morphological analysis is performed to evaluate the quality of what is being said. The numerical data of application usage time is also normalized into a unified format.

[1216] Analysis using generative AI models

[1217] The server inputs the preprocessed data into a generative AI model (e.g., OpenAI GPT-4) and performs qualitative and quantitative evaluation. The specific steps of the evaluation are as follows:

[1218] In qualitative evaluation, text data is analyzed to evaluate the quality of comments made during meetings and the importance of email content.

[1219] Quantitative evaluation involves analyzing log data and evaluating work time and application usage.

[1220] Examples:

[1221] The generative AI model receives a prompt such as "Evaluate User A's comments during the meeting and analyze how constructive they were," and the model outputs the evaluation results. The model also analyzes work time data using a prompt such as "Calculate User A's weekly work hours and evaluate their productivity."

[1222] Output of evaluation results

[1223] The server compiles the results of the analysis by the generative AI model into a report and provides it to the user (supervisor). This report includes the following:

[1224] Qualitative assessment (e.g., quality of content)

[1225] Quantitative evaluation (e.g., work time, application usage statistics)

[1226] Examples:

[1227] The server generates a report and provides it to the manager, which includes a qualitative assessment of User A's project management skills and problem-solving abilities, as well as quantitative data showing User A's time contribution to a particular project.

[1228] Complementing MBO evaluation

[1229] The user (supervisor) performs a management by objectives (MBO) evaluation based on the provided report, checks the validity of the evaluation content, makes adjustments as necessary, and provides feedback to the employee.

[1230] Examples:

[1231] The supervisor uses the analysis results of the generated AI model to conduct an annual evaluation of User A. User A's leadership and focus are particularly highly evaluated, and a high MBO score is awarded.

[1232] In this way, the system of the present invention clearly shows employees' work achievements and achieves fair evaluations. This system provides managers with objective data to complement MBO evaluations, thereby contributing to improved employee satisfaction.

[1233] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1234] Step 1: Data collection

[1235] The device continuously collects business data and meeting logs. Specifically, it collects user keystrokes, application usage history, file access logs, email content, and audio and video conference recordings. The input is user interactions and meeting participation, and the output is log data stored on the device.

[1236] Specific behavior:

[1237] The device records keystrokes using keyboard hooks, captures application usage history and file access logs using a dedicated service, and, when audio and video conferences are taking place, captures conference logs using dedicated audio and video recording software.

[1238] Step 2: Data Preprocessing

[1239] The server converts the collected raw data into an analyzable format. Specifically, it converts voice data into text data using a speech recognition tool (e.g., Google Speech-to-Text API), and then performs morphological analysis on the text data using natural language processing technology (e.g., spaCy, NLTK). Numerical data is normalized using scaling and normalization techniques. The input is the collected log data, and the output is preprocessed, analyzable data.

[1240] Specific behavior:

[1241] The server sends the audio file to a speech recognition API to obtain the transcribed text, then feeds the text data into an NLP tool for morphological analysis, and normalizes the numerical data using an appropriate scaling algorithm.

[1242] Step 3: Analysis by generative AI model

[1243] The server inputs the preprocessed data into a generative AI model (e.g., OpenAI GPT-4) to perform qualitative and quantitative evaluations. Qualitative evaluations assess the quality of comments made during meetings and the importance of email content, while quantitative evaluations assess task time and application usage. The input is the preprocessed data, and the output is the evaluation results.

[1244] Specific behavior:

[1245] The server inputs a prompt to the generative AI model, such as "Evaluate User A's comments during the meeting and analyze how constructive they were," and obtains a response from the model. It also analyzes the work time data using a prompt, such as "Calculate User A's weekly work hours and evaluate their productivity."

[1246] Step 4: Output of evaluation results

[1247] The server compiles the evaluation results analyzed by the generative AI model into a report. Specifically, it integrates qualitative evaluations (e.g., quality of speech) and quantitative evaluations (e.g., work time, statistics on applications used) to create a report. The input is the analyzed evaluation results, and the output is the report.

[1248] Specific behavior:

[1249] The server receives the output of the generative AI model and creates a report in formats such as Excel or PDF, which is visually organized using graphs and tables so that administrators can easily understand it.

[1250] Step 5: Complementing the MBO evaluation

[1251] The user (supervisor) performs a management by objectives (MBO) evaluation based on the provided report. Specifically, the user checks the validity of the evaluation content and makes adjustments as necessary. Finally, the user provides feedback to the employee. The input is the generated report, and the output is the revised evaluation and feedback.

[1252] Specific behavior:

[1253] The supervisor reviews the report and adds comments such as, "User A's leadership is evident, but we would like to see more specific examples." In addition, the supervisor discusses with the employee specific evaluation items and proposes the next steps for improvement.

[1254] (Application example 1)

[1255] 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."

[1256] At the workplace, it is important to accurately evaluate the performance and contribution of workers and to create a fair and transparent evaluation system. Factory floors, in particular, require a system that efficiently collects and analyzes work logs and robot interaction data, and evaluates them qualitatively and quantitatively.

[1257] 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.

[1258] In this invention, the server includes means for collecting business data and meeting logs from terminals, means for preprocessing the collected business data and meeting logs, means for analyzing the preprocessed data and using a generative AI model for qualitative and quantitative evaluation, means for outputting the evaluation results by the generative AI model in report format, means for collecting operation data and work logs of robots in the factory, means for collecting interaction data with workers, and means for analyzing the collected data and evaluating the performance of workers. This makes it possible to clarify the work achievements of workers and provide convincing and fair evaluations.

[1259] "Device" means a computer device used to collect and process data.

[1260] "Business data" refers to information such as applications that users operate during work, files that they access, and communication content.

[1261] "Meeting logs" are audio and video recordings related to statements made during a meeting and the agenda, as well as textual versions of such recordings.

[1262] "Preprocessing" is the process of converting collected raw data into an analyzable format.

[1263] A "generative AI model" is an artificial intelligence model that uses collected data to perform qualitative and quantitative evaluations.

[1264] "Qualitative evaluation" is the process of evaluating the content of a written piece or the quality of what is said in a meeting.

[1265] "Quantitative evaluation" is a process of evaluation based on numerical data such as log data and work time.

[1266] "Report format" refers to the format of a document summarizing the evaluation results.

[1267] "Factory robots" are automated machinery used to streamline factory work.

[1268] "Operation data" refers to records and log data relating to the robot's operations.

[1269] A "work log" is recorded data on the actions and work details of workers.

[1270] "Interaction data" is recorded data of the operations and communication that take place alternately between workers and robots.

[1271] "Evaluation results" refer to the qualitative and quantitative evaluation values ​​and evaluation content calculated by the generative AI model.

[1272] To implement this invention, a terminal first collects business data and a meeting log. Business data includes information such as the operation history of applications used by the user, files accessed, and emails sent and received. The meeting log includes audio and video data related to comments made during the meeting and the agenda.

[1273] The collected data is sent to a server where it undergoes preprocessing. During the preprocessing stage, audio data is converted into text data using a transcription tool. Next, natural language processing (NLP) techniques are used to perform morphological analysis of the text data, analyzing its grammatical structure and meaning. Numerical data is normalized and converted into a unified format.

[1274] Once preprocessed, the data is input into a generative AI model for qualitative and quantitative evaluation. In qualitative evaluation, the generative AI model evaluates the content of text and the quality of comments made in meetings. For example, it extracts meaningful comments about project progress from the contents of a user's email. In quantitative evaluation, it performs evaluation based on log data and work time. For example, it measures the total amount of time a user spends on a specific project.

[1275] The evaluation results are compiled in a report format by the server and provided to the user (manager). This report includes both qualitative and quantitative evaluations. Based on this report, the manager can obtain objective information for performing a management by objectives (MBO) evaluation. Using this information, the manager can check the validity of the evaluation content and make adjustments as necessary.

[1276] Furthermore, the system can collect operation data and work logs of robots in factories and analyze interaction data with workers, which will further improve the efficiency of factory work and enable fair evaluations.

[1277] As a concrete example, a robot in a factory collects speech data from workers, converts it into text data, performs morphological analysis, evaluates the quality of the speech, quantitatively evaluates the worker's performance using work log data, and integrates this data to provide the manager with the evaluation results in the form of a report.

[1278] An example of a prompt sentence to input to the generative AI model is, "Please summarize what was said in this meeting and extract the most important points." Using this prompt allows for efficient qualitative evaluation of meeting logs.

[1279] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1280] Step 1:

[1281] The device collects business data and meeting logs. Business data includes the operation history of applications used by the user, files accessed, and emails sent and received. Meeting logs include audio and video recordings related to comments made during meetings and the agenda. This data is obtained using software (e.g., data collection API). Input is the user's activity log and meeting audio, and output is the collected raw data.

[1282] Step 2:

[1283] The terminal sends the collected data to the server. The address of the server is specified and the data is sent in encrypted format via the required protocol (e.g. HTTP / HTTPS). The input is the raw data acquired, and the output is the data sent to the server.

[1284] Step 3:

[1285] The raw data received by the server is preprocessed. First, the voice data is converted into text data using a speech recognition tool (e.g., Python's speech_recognition library). Next, the text data is subjected to morphological analysis using natural language processing (NLP) techniques (e.g., the spacy library) to analyze grammatical structure and meaning. Numerical data is converted into a data frame and normalized. The input is raw data, and the output is preprocessed data.

[1286] Step 4:

[1287] The preprocessed data is input into a generative AI model. The generative AI model (e.g., GPT-2 from Hugging Face) analyzes the text data and performs a qualitative evaluation. An example of a prompt used here is, "Please summarize what was said in this meeting and extract the most important points." The input is the preprocessed text data and the prompt, and the output is the result of the qualitative evaluation.

[1288] Step 5:

[1289] The preprocessed numerical data is used to perform quantitative evaluation. The server analyzes the data frame and calculates statistics on work time and application usage time. For example, it calculates the time spent on a specific project based on work time. The input is the preprocessed numerical data, and the output is the results of the quantitative evaluation.

[1290] Step 6:

[1291] The server integrates the results of the qualitative and quantitative evaluations and generates a report. The report includes qualitative analysis results and quantitative statistics. It displays a detailed analysis of the content of the text, the quality of the comments, and the working time. The input is the results of the qualitative and quantitative evaluations, and the output is a report-format file.

[1292] Step 7:

[1293] The server provides the generated report to the user (administrator). The report is made accessible in dashboard or PDF format so that the administrator can easily check the evaluation. The input is the generated report, and the output is the report used by the user in the actual evaluation process.

[1294] 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.

[1295] The present invention is a system that visualizes employees' work performance by performing qualitative and quantitative evaluations using a generative AI model using business data and meeting logs collected from terminals, thereby realizing a fair and convincing evaluation system. Furthermore, the present invention improves the accuracy and comprehensiveness of evaluations by combining it with an emotion engine that recognizes the user's emotions. Specific embodiments for implementing the present invention are described below.

[1296] Data collection

[1297] The device continuously collects the user's business data and meeting logs. Specifically, it acquires business data such as the user's application usage history, file access logs, email content, and keystroke data. It also collects audio and video recordings of video and audio conferences.

[1298] Examples:

[1299] For example, the terminal of user A records a log of the applications used throughout the day and saves the contents of statements made during a meeting as audio data.

[1300] Data Preprocessing

[1301] The server converts the collected raw data into an analyzable format. First, the voice data is converted into text data using a speech recognition tool. Next, the text data is subjected to morphological analysis using natural language processing (NLP) techniques to analyze grammatical structure and meaning. Numerical data is normalized and converted into a unified format. An emotion engine is also implemented at this stage.

[1302] Examples:

[1303] For example, audio data from a meeting is converted into text data using a transcription tool, and morphological analysis is performed to make what is being said easier to understand. Numerical data such as application usage time is also converted into a unified format. An emotion engine analyzes the user's emotions from the text data.

[1304] Analysis using generative AI models

[1305] The server inputs the preprocessed data into a generative AI model for qualitative and quantitative evaluation. Qualitative evaluation evaluates the content of text and the quality of speech in meetings. The emotion engine also analyzes the user's emotional state and incorporates it into the evaluation. Quantitative evaluation uses log data such as work time and application usage.

[1306] Examples:

[1307] For example, a generative AI model extracts meaningful comments about project progress from User A's email content and gives a high rating. At the same time, it quantitatively evaluates that User A is spending more than 40 hours per week on a specific project based on the amount of time worked. An emotion engine analyzes the emotions in User A's email and reflects their emotional state, such as "positive" or "negative," in the rating.

[1308] Output of evaluation results

[1309] The server compiles the results of the analysis by the generative AI model and emotion engine into a report and provides it to the user (supervisor). The report includes qualitative evaluations (quality of speech content, emotional state) and quantitative evaluations (work time, statistics on applications used).

[1310] Examples:

[1311] The server generates a report and provides it to the manager, which includes a qualitative assessment of User A's project management skills and problem-solving abilities, as well as quantitative data showing his time contribution to a specific project. The report also includes an emotional assessment based on the emotion engine.

[1312] Complementing MBO evaluation

[1313] The user (supervisor) performs an MBO evaluation based on the provided report, checks the validity of the evaluation content, makes adjustments as necessary, and provides feedback to the employee.

[1314] Examples:

[1315] The supervisor uses the analysis results of the generative AI model and the emotion engine to conduct an annual evaluation of User A. In particular, User A's leadership and focus are highly evaluated, and a high MBO score is awarded. User A's emotional state is also taken into consideration, and the evaluation content is adjusted accordingly.

[1316] In this way, the system of the present invention clearly shows the employee's work achievements and enables convincing and fair evaluations. At the same time, the introduction of an emotion engine enables comprehensive evaluations that take into account the employee's emotional state.

[1317] The processing flow will be explained below.

[1318] Step 1: Data collection

[1319] The device collects the user's business data and conference logs. Specifically, it acquires business data such as the user's application usage history, file access logs, email content, and keystroke data. It also collects audio and video recordings of video and audio conferences.

[1320] Specific operation:

[1321] The terminal starts the log collection agent and collects business data every 15 minutes.

[1322] Acquire recording data via the video conferencing tool's API.

[1323] Step 2: Data Preprocessing

[1324] The server preprocesses the collected raw data. First, the voice data is converted into text data using a voice recognition tool. Next, the text data undergoes morphological analysis using natural language processing (NLP) techniques. Numerical data is normalized and converted into a unified format. An emotion engine analyzes the user's emotions from the text data.

[1325] Specific operation:

[1326] The server inputs the audio data into a transcription tool and converts it into text data.

[1327] The server performs morphological analysis, analyzes the text data, and extracts grammatical structure and semantic information.

[1328] The server normalizes the numerical data and converts it into a unified data format.

[1329] The emotion engine analyzes the user's emotional state from the text data.

[1330] Step 3: Analysis by generative AI models and emotion engines

[1331] The server inputs the preprocessed data into the generative AI model and emotion engine to perform qualitative and quantitative evaluations. Qualitative evaluation evaluates the content of text and the quality of speech in meetings. The emotion engine also analyzes the user's emotional state and incorporates it into the evaluation. Quantitative evaluation uses log data such as work time and application usage.

[1332] Specific operation:

[1333] The server inputs the text data extracted from the meeting log into the generative AI model.

[1334] The server uses the generative AI model to generate sentiment analysis and content evaluation scores for the comments.

[1335] The server inputs the numerical data into the AI ​​model and generates an evaluation score based on work time and frequency of application use.

[1336] The emotion engine analyzes the user's text data and reflects their emotional state in the evaluation score.

[1337] Step 4: Output of evaluation results

[1338] The server compiles the evaluation results from the generative AI model and emotion engine into a report and provides it to the user (supervisor). The report includes qualitative evaluations (quality of speech content, emotional state) and quantitative evaluations (work time, statistics on applications used).

[1339] Specific operation:

[1340] The server automatically generates a report based on the evaluation results from the AI ​​model and emotion engine.

[1341] The server periodically displays the report on the manager's dashboard or sends it by email.

[1342] Step 5: Complementing the MBO evaluation

[1343] The user (supervisor) performs an MBO evaluation based on the provided report, checks the validity of the evaluation content, makes adjustments as necessary, and provides feedback to the employee.

[1344] Specific operation:

[1345] The user reviews the report and completes the MBO evaluation sheet based on the evaluation results of the generative AI model and emotion engine.

[1346] The user conducts a feedback interview based on the evaluation results and shares the evaluation details with the employee.

[1347] Through these steps, the system of the present invention visualizes employees' work performance and realizes a fair and convincing evaluation system. In addition, the introduction of an emotion engine enables comprehensive evaluation that takes into account the emotional aspects of employees.

[1348] Example 2

[1349] 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."

[1350] Conventional performance evaluation systems often evaluate employees' work performance qualitatively, making it difficult to base evaluations on concrete data. They also lack comprehensive evaluations that take emotional aspects into account, which can lead to evaluations that lack fairness and satisfaction. Furthermore, insufficient pre-processing of meeting logs and work data often reduces the accuracy of evaluation results.

[1351] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for collecting business data and meeting logs from terminals, a means for preprocessing the collected business data and meeting logs, a means for analyzing the preprocessed data and using a generative AI model for qualitative and quantitative evaluation, a means for outputting the evaluation results by the generative AI model in report format, and a means for recognizing emotions from text data using an emotion engine. This enables a fair and comprehensive evaluation of employees' work performance based on specific data and taking into account their emotional aspects.

[1352] A "terminal" is a computer or mobile device used by a user, and is a device for collecting business data and meeting logs.

[1353] "Business data" refers to all data related to a user's business activities, including application usage history, file access logs, email content, keystroke data, and the like.

[1354] A "meeting log" refers to recorded data of a meeting, including audio and video, and is data that captures the content of statements and discussions made during the meeting.

[1355] "Collection means" refers to the method or function for acquiring business data and meeting logs from a terminal, and takes the form of software or hardware.

[1356] "Preprocessing" refers to the process of converting raw data into an analyzable format, and includes transcribing audio data, morphological analysis of text data, and normalizing numerical data.

[1357] A "generative AI model" refers to an artificial intelligence model that has functions such as response generation and data analysis, and is used here to perform qualitative and quantitative evaluations.

[1358] An "emotion engine" refers to a program or tool that recognizes and analyzes a user's emotional state from text data.

[1359] "Report format" refers to a format that visually summarizes the evaluation results from the generative AI model and emotion engine in an easy-to-understand manner, and includes formats such as PDF and web dashboards.

[1360] "Manager" refers to the person or position with the authority to conduct MBO evaluations based on the evaluation results, and is usually a supervisor or human resources officer.

[1361] "MBO" is an abbreviation for Management by Objectives, a management method that evaluates employees based on the degree to which they have achieved their goals.

[1362] The present invention provides a system that visualizes employees' work performance by performing qualitative and quantitative evaluations using a generative AI model using business data and meeting logs collected from terminals, thereby realizing a fair and convincing evaluation system. The present invention also provides a system that improves the accuracy and comprehensiveness of evaluations by combining it with an emotion engine that recognizes user emotions. Specific embodiments for implementing the present invention are described below.

[1363] Data collection

[1364] The device continuously collects the user's business data and meeting logs. Specifically, it acquires business data such as the user's application usage history, file access logs, email content, and keystroke data. It also collects audio and video recordings of video and audio conferences.

[1365] Examples:

[1366] For example, the terminal of user A records a log of the applications used throughout the day and saves the contents of statements made during a meeting as audio data.

[1367] Data Transfer

[1368] The device sends the collected data to the server in real time. The data collection program in the device periodically uploads the locally stored data to the server. For security reasons, the data is sent in encrypted form.

[1369] Data Preprocessing

[1370] The server converts the received raw data into an analyzable format. First, the voice data is converted into text data using a speech recognition tool (e.g., Google Cloud Speech-to-Text). Next, the text data is subjected to morphological analysis using natural language processing (NLP) technology (e.g., Spacy) to analyze grammatical structure and meaning. Numerical data is normalized from Excel or CSV format and converted into a unified format (e.g., JSON format). Finally, the text data is analyzed for sentiment using the API of an emotion engine (e.g., Microsoft Azure Text Analytics).

[1371] Examples:

[1372] For example, audio data from a meeting is transcribed using Google Cloud Speech-to-Text, and morphological analysis is performed using Spacy to make what is being said easier to understand. Numerical data such as application usage time is converted from Excel or CSV format into a unified format. An emotion engine analyzes user emotions from the text data.

[1373] Analysis using generative AI models

[1374] The server inputs the preprocessed data into a generative AI model (e.g., OpenAI GPT-3) to perform qualitative and quantitative evaluations. Qualitative evaluations evaluate the content of text and the quality of speech in meetings. An emotion engine also analyzes the user's emotional state and incorporates it into the evaluation. Quantitative evaluations use log data such as work time and application usage.

[1375] Examples:

[1376] For example, a generative AI model extracts meaningful comments about project progress from User A's email content and gives a high rating. At the same time, it quantitatively evaluates that User A is spending more than 40 hours per week on a specific project based on the amount of time worked. An emotion engine analyzes the emotions in User A's email and reflects their emotional state, such as "positive" or "negative," in the rating.

[1377] Example prompt sentence:

[1378] "Evaluate User A's important contributions to the project progress based on his emails and meeting comments over the past month."

[1379] "Quantitatively evaluate User A's time contribution to a specific project based on his / her weekly work time data."

[1380] Sentiment analysis results integration

[1381] The server uses an emotion engine to integrate the results of the emotion analysis with the evaluation from the generative AI model. The emotion analysis results are combined with the output of the AI ​​model to reflect positive and negative comments in the evaluation score. Detailed results of the emotion analysis are also added to the report.

[1382] Generating evaluation results

[1383] The server compiles the generated evaluation results into a report. The server uses a report generation tool (e.g., JasperReports) to create a detailed report that includes both qualitative and quantitative evaluations. The report is output in PDF format or web dashboard format.

[1384] Providing reports

[1385] The server provides the completed report to the user (supervisor). The report is automatically sent to the supervisor's email address periodically (for example, at the end of the month). The report is also uploaded to the company portal so that the user can access it.

[1386] Complementing MBO evaluation

[1387] The user (supervisor) performs an MBO evaluation based on the provided report. The supervisor reviews the report and provides additional feedback or supplementary information as needed. Once the evaluation is complete, a session is set up to provide feedback to the employee being evaluated.

[1388] In this way, each processing step of the system works together in a concrete way to clarify employees' work achievements and achieve fair and convincing evaluations. The introduction of an emotion engine makes it possible to provide comprehensive evaluations that take into account employees' emotional states.

[1389] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1390] Step 1: Data collection

[1391] The terminal collects the user's business data and meeting logs.

[1392] Input: User application usage history, file access logs, email content, keystroke data, and audio and video conference recordings.

[1393] Data processing: Raw data is automatically acquired and collected 24 / 7.

[1394] Output: Locally stored business data and meeting logs.

[1395] Specific operation: The device will periodically record application logs and file access logs, and will automatically start recording when an audio conference is started.

[1396] Step 2: Data Transfer

[1397] The data collected by the device is sent to the server in real time.

[1398] Input: Business data and meeting logs stored locally on the device.

[1399] Data processing: Encrypt the data and upload it securely to the server.

[1400] Output: Aggregated business data and meeting logs on the server.

[1401] Specific operation: The data collection program in the device periodically (for example, every hour) encrypts the data and uploads it to the server.

[1402] Step 3: Data Preprocessing

[1403] The server converts the raw data it receives into a format that can be analyzed.

[1404] Input: Raw data (business data, meeting logs) aggregated on the server.

[1405] Data processing: Convert the voice data into text data using Google Cloud Speech-to-Text, perform morphological analysis on the text data using Spacy, normalize the numerical data and convert it into a unified format (e.g., JSON), and analyze emotions from the text data using an emotion engine.

[1406] Output: Data converted into an analyzable format (text data, morphological analysis results, normalized numerical data, sentiment analysis results).

[1407] How it works: The server transcribes the audio data using Google Cloud Speech-to-Text, then performs morphological analysis using Spacy. Numerical data is converted from Excel or CSV to JSON format, and sentiment analysis is performed using a sentiment analysis tool.

[1408] Step 4: Data integration and storage

[1409] The server consolidates the pre-processed data and stores it in a database.

[1410] Input: Preprocessed data (text data, morphological analysis results, normalized numerical data, sentiment analysis results).

[1411] Data processing: Integrate the data, classify it by user, and store it in a NoSQL database (e.g., MongoDB).

[1412] Output: Data integrated and stored in a database.

[1413] Specific operation: The server inserts the preprocessed data into MongoDB, classifies it by user, and creates indexes to enable efficient searches.

[1414] Step 5: Analysis by generative AI model

[1415] The server inputs the preprocessed data into a generative AI model for qualitative and quantitative evaluation.

[1416] Input: Preprocessed data extracted from the database.

[1417] Data processing: A generative AI model (e.g., OpenAI GPT-3) is invoked with a prompt sentence to perform qualitative evaluation (text content and speech quality) and quantitative evaluation (task time, application usage).

[1418] Output: Analysis results from the generative AI model.

[1419] Specific operation: The server inputs data into the OpenAI GPT-3 model using the prompt sentence and generates an appropriate evaluation score and comment.

[1420] Step 6: Integrating the results of sentiment analysis

[1421] The server uses an emotion engine to integrate the results of emotion analysis with the evaluation from the generative AI model.

[1422] Input: Analysis results from generative AI model, sentiment analysis results.

[1423] Data processing: The results of sentiment analysis are combined with the output of the AI ​​model and reflected in the evaluation score.

[1424] Output: Evaluation results integrated with sentiment analysis results.

[1425] Specific operation: The server adds the results from the emotion engine to the output of the generative AI model, and reflects positive or negative emotions in the evaluation score.

[1426] Step 7: Generate evaluation results

[1427] The server compiles the generated evaluation results in a report format.

[1428] Input: Evaluation results integrated with sentiment analysis results.

[1429] Data processing: Create reports using a report generation tool (e.g., JasperReports) and convert them into PDF or web dashboard formats.

[1430] Output: Assessment report in PDF format and web dashboard format.

[1431] What it does: The server compiles the results of the assessment into a detailed report and converts it into a visually understandable format.

[1432] Step 8: Providing the report

[1433] The server provides the completed report to the user (supervisor).

[1434] Input: Assessment report in PDF format and web dashboard format.

[1435] Data Processing: Present the report to your superior in an appropriate manner.

[1436] Output: Evaluation report provided to supervisor.

[1437] What it does: The server sends the report to the manager's email address at the end of the month and also uploads it to the company's internal portal for access.

[1438] Step 9: Complementing the MBO evaluation

[1439] The user (supervisor) performs an MBO evaluation based on the provided report.

[1440] Input: Evaluation report provided to supervisor.

[1441] Data processing: Check the evaluation content and provide any necessary feedback or supplementary information.

[1442] Output: Completed MBO evaluation and feedback.

[1443] What happens next: The manager reviews the report, notes any necessary corrections or additional information, and schedules a feedback session.

[1444] In this way, the system of the present invention evaluates employees' work performance in a concrete and data-based manner, and achieves a fair and comprehensive evaluation that also takes into account their emotional aspects.

[1445] (Application example 2)

[1446] 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."

[1447] Existing evaluation systems do not adequately perform qualitative and quantitative evaluations based on work data and meeting logs. This makes it difficult to fairly evaluate employees' work performance and the operational efficiency of production equipment. Another problem is the lack of systems that can perform comprehensive evaluations using robot operation logs and performance indicators. Furthermore, it is necessary to improve the accuracy and comprehensiveness of evaluations by combining an emotion engine.

[1448] The identification processing 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 means for collecting business data and meeting logs from terminals, means for preprocessing the collected business data and meeting logs, means for analyzing the preprocessed data and using a generative AI model for qualitative and quantitative evaluation, means for outputting the evaluation results by the generative AI model in report format, means for collecting operation log data and performance indicators of the production equipment, means for converting the collected operation log data and performance indicators into an analyzable format, and means for qualitatively evaluating the operation log data and quantitatively evaluating the performance indicators using the generative AI model and an emotion recognition engine. This not only enables fair and convincing evaluation based on the business data and meeting logs, but also enables comprehensive evaluation of the operational efficiency of the production equipment.

[1449] A "terminal" is an electronic device such as a computer or smartphone used by a user, and is used to collect business data and meeting logs.

[1450] "Business data" refers to records such as application usage history, file access logs, email content, and keystroke data that are generated when a user performs a task.

[1451] "Meeting logs" refer to audio and video recordings of video and audio conferences, including audio data converted into text data and minutes of meetings.

[1452] "Preprocessing" is the process of converting collected raw data into an analyzable format, and includes transcribing audio data, morphological analysis of text data, and normalizing numerical data.

[1453] A "generative AI model" is a model that uses machine learning algorithms to analyze collected data and perform qualitative and quantitative evaluations.

[1454] An "emotion recognition engine" is a technology for analyzing a user's emotional state from text data and voice data, and reflects "positive," "negative," and changes therein in the evaluation.

[1455] "Production equipment" refers to equipment such as robots used in factories and production lines, and is the subject of collection of operation log data and performance indicators.

[1456] "Operation log data" refers to data that includes the operation history and operation records of robots and production equipment, and mainly includes detailed records of the production process.

[1457] "Performance index" refers to the key performance metrics, such as the efficiency or error rate achieved by a robot or production equipment, expressed in specific numerical values.

[1458] "Qualitative evaluation" is a type of evaluation performed by a generative AI model, and is a qualitative evaluation that is primarily based on the content of statements and operation logs.

[1459] "Quantitative evaluation" is a type of evaluation performed by a generative AI model, and is a quantitative evaluation based primarily on numerical data such as work time and performance indicators.

[1460] This invention provides a system for fair and comprehensive evaluation of the operational efficiency of production equipment and the work performance of employees. The system includes a terminal, a server, a generative AI model, and an emotion recognition engine.

[1461] First, the terminal continuously collects operation log data and performance indicators of production equipment. For example, it acquires performance data such as operation logs, work efficiency, and error rates of robots installed in a factory. This allows the operation history and operation records of production equipment, as well as key performance metrics, to be collected.

[1462] Next, the server preprocesses the collected data. At this stage, the collected raw data is converted into an analyzable format. Specifically, the operation log data is subjected to morphological analysis, and the performance indicator data is normalized. If voice data is included, it is converted into text data using a voice recognition tool.

[1463] The preprocessed data is input into a generative AI model, which performs both qualitative and quantitative evaluations. In qualitative evaluation, meaningful utterances and behavioral patterns are extracted from the operation log data, and the emotional state is analyzed using an emotion recognition engine. In quantitative evaluation, evaluation is performed based on performance indicators such as work efficiency and error rate.

[1464] The server outputs the analysis results of the generative AI model and emotion recognition engine in the form of a report. This report includes a qualitative evaluation of the robot's operation and emotional state, as well as a quantitative evaluation of each performance indicator. This allows administrators to conduct comprehensive and fair evaluations.

[1465] As a specific example, the operation log of the robot 123 is acquired as "log data of robot 123", and the evaluation results include quantitative and qualitative evaluations as follows:

[1466] Qualitative evaluation: Extract meaningful operation patterns from the robot's operation log and give a positive evaluation using an emotion engine.

[1467] Quantitative evaluation: Since the work efficiency is 95% and the error rate is 2%, the evaluation is "Efficiency: High" and "Error rate: Low."

[1468] Example prompt sentence:

[1469] "Collect the operation log data and performance metrics of Robot 123 and evaluate it using the generative AI model and emotion engine."

[1470] This system allows for a comprehensive evaluation of production equipment and employee performance, resulting in rational and convincing evaluations.

[1471] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1472] Step 1:

[1473] The terminal collects operation log data and performance indicators of production equipment. Specifically, data is acquired using sensors and controllers that record the operation history, operation records, work efficiency, and error rates of robots installed in the factory. The input is real-time operation log data and performance indicators collected from the robots, which are stored in a database. The output is the collected operation log data and raw performance indicator data.

[1474] Step 2:

[1475] The server preprocesses the collected raw data. Specifically, it analyzes the operation log data using a morphological analysis tool (e.g., MeCab) and converts the grammatical structure and meaning into an understandable format. It also normalizes the performance indicator data and converts it into a unified format. If voice data is included, it is converted into text data using voice recognition software (e.g., Google Speech-to-Text). The input is raw data obtained from the device, and the output is preprocessed data that can be analyzed.

[1476] Step 3:

[1477] The server inputs the preprocessed data into a generative AI model to perform qualitative and quantitative evaluations. In qualitative evaluation, a generative AI model (e.g., Hugging Face Transformers) is used to analyze the contents of the operation log data and extract meaningful operation patterns and comments. An emotion recognition engine (e.g., Sentiment Analysis API) is used to evaluate the user's emotional state from the operation log. In quantitative evaluation, work efficiency and error rates are analyzed based on performance index data. The input is preprocessed data, and the output is qualitative and quantitative evaluation results.

[1478] Step 4:

[1479] The server outputs the evaluation results of the generated AI model and emotion recognition engine in report format. Specifically, the evaluation results are compiled in Excel or PDF format and provided to the administrator. The report includes the robot's operation details, emotional state, and evaluation results of each performance indicator. The input is the evaluation result data, and the output is the evaluation results in report format.

[1480] Step 5:

[1481] The user evaluates production equipment and employees based on the generated report. The manager reviews the report content and uses it as a supplementary reference for management by objectives (MBO) evaluation. Specifically, the report comprehensively evaluates the robot's operating efficiency, error rate, employee work performance, etc. The input is the evaluation results in report format, and the output is the manager's final evaluation results.

[1482] 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.

[1483] 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.

[1484] 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.

[1485] 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.

[1486] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1487] 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.

[1488] 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).

[1489] 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.

[1490] 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."

[1491] 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.

[1492] 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).

[1493] 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.

[1494] 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.

[1495] 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.

[1496] 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.

[1497] 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.

[1498] 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.

[1499] 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.

[1500] 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.

[1501] 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.

[1502] 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.

[1503] The following is further disclosed regarding the above embodiment.

[1504] (Claim 1)

[1505] A means for collecting business data and meeting logs from the terminal;

[1506] A means for pre-processing the collected business data and meeting logs;

[1507] A means for analyzing the pre-processed data and using a generative AI model to perform qualitative and quantitative evaluations;

[1508] A means for outputting the evaluation results of the generative AI model in report format;

[1509] A system including:

[1510] (Claim 2)

[1511] 10. The system of claim 1, further comprising means for providing supplemental material for a manager to conduct a management by objectives (MBO) evaluation using the evaluation results.

[1512] (Claim 3)

[1513] 10. The system of claim 1, further comprising means for converting audio data to text data.

[1514] "Example 1"

[1515] (Claim 1)

[1516] A means for continuously collecting business data and meeting logs from user terminals;

[1517] A means for preprocessing the collected business data and meeting logs using a speech recognition tool and natural language processing technology;

[1518] A means for inputting the preprocessed data into a generative AI model and performing qualitative and quantitative evaluation;

[1519] A means for outputting the evaluation results of the generative AI model in report format;

[1520] A system including:

[1521] (Claim 2)

[1522] 10. The system of claim 1, further comprising means for providing supplemental material for a manager to conduct a management by objectives (MBO) evaluation using the evaluation results.

[1523] (Claim 3)

[1524] 10. The system of claim 1, further comprising means for using a speech recognition tool to convert the audio data into text data.

[1525] "Application Example 1"

[1526] (Claim 1)

[1527] A means for collecting business data and meeting logs from the terminal;

[1528] A means for pre-processing the collected business data and meeting logs;

[1529] A means for analyzing the pre-processed data and using a generative AI model to perform qualitative and quantitative evaluations;

[1530] A means for outputting the evaluation results of the generative AI model in report format;

[1531] A means for collecting operation data and work logs of robots in a factory;

[1532] a means for collecting worker interaction data;

[1533] a means of analyzing the collected data and evaluating worker performance;

[1534] A system including:

[1535] (Claim 2)

[1536] 10. The system of claim 1, further comprising means for providing supplemental material for a manager to conduct a management by objectives (MBO) evaluation using the evaluation results.

[1537] (Claim 3)

[1538] 10. The system of claim 1, further comprising means for converting audio data to text data.

[1539] "Example 2: Combining Emotion Engines"

[1540] (Claim 1)

[1541] A means for collecting business data and meeting logs from the terminal;

[1542] A means for pre-processing the collected business data and meeting logs;

[1543] A means for analyzing the pre-processed data and using a generative AI model to perform qualitative and quantitative evaluations;

[1544] A means for outputting the evaluation results of the generative AI model in report format;

[1545] means for recognizing emotions from text data using an emotion engine;

[1546] A system including:

[1547] (Claim 2)

[1548] 10. The system of claim 1, further comprising means for providing supplemental material for a manager to use the evaluation results to conduct a management by objectives (MBO) evaluation.

[1549] (Claim 3)

[1550] 10. The system of claim 1, further comprising means for converting audio data into text data.

[1551] "Application example 2 when combining emotion engines"

[1552] (Claim 1)

[1553] A means for collecting business data and meeting logs from the terminal;

[1554] A means for pre-processing the collected business data and meeting logs;

[1555] A means for analyzing the pre-processed data and using a generative AI model to perform qualitative and quantitative evaluations;

[1556] A means for outputting the evaluation results of the generative AI model in report format;

[1557] A means for collecting operation log data and performance indicators of the production equipment;

[1558] A means for converting the collected operation log data and performance indicators into an analyzable format;

[1559] A means for qualitatively evaluating operation log data and quantitatively evaluating performance indicators using a generative AI model and an emotion recognition engine;

[1560] A system including:

[1561] (Claim 2)

[1562] 10. The system of claim 1, further comprising means for providing supplemental material for a manager to conduct a management by objectives (MBO) evaluation using the evaluation results.

[1563] (Claim 3)

[1564] 10. The system of claim 1, further comprising means for converting audio data to text data. [Explanation of symbols]

[1565] 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 collecting business data and meeting logs from the terminal; A means for pre-processing the collected business data and meeting logs; A means for analyzing the pre-processed data and using a generative AI model to perform qualitative and quantitative evaluations; A means for outputting the evaluation results of the generative AI model in report format; A system including:

2. 10. The system of claim 1, further comprising means for providing supplemental material for a manager to conduct a performance management evaluation using the evaluation results.

3. 10. The system of claim 1, further comprising means for converting audio data to text data.

Images