System

Abstract

A system is provided.SOLUTION: A system, comprising: means for collecting work data of an employee; means for preprocessing the collected data; means for training a machine learning model based on the preprocessed data; means for evaluating performance of the employee using the trained machine learning model; and means for feeding back an evaluation result to the employee.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] In many companies today, personnel evaluations rely on subjective factors, often resulting in a sense of unfairness. This sense of unfairness leads to lower employee motivation and productivity, negatively impacting the company's overall performance. Furthermore, excessive deadlines have become the norm, and setting deadlines with ample time creates the risk of fraud. To solve these problems, it is necessary to properly evaluate each employee's performance and efforts and create a healthy work environment. [Means for solving the problem]

[0005] The present invention is a system including a means for collecting employee work data, a means for preprocessing the collected data, a means for training a machine learning model based on the preprocessed data, a means for evaluating employee performance using the trained machine learning model, and a means for providing feedback on the evaluation results to the employee. Furthermore, the evaluation means uses the employee's past performance data, current work status data, and work execution efficiency data as input, and the feedback means provides advice for performance improvement based on the evaluation results, thereby enabling an objective and fair evaluation of employee performance, eliminating feelings of unfairness, and creating a healthy work environment.

[0006] An "employee" is an individual employed by a business or organization to perform a specific job.

[0007] "Business data" refers to information related to the work performed by an employee, such as the progress of the work, the results achieved, and the time spent on the work.

[0008] "Preprocessing" refers to operations for converting raw data into an analyzable format, including standardizing data formats, filling in missing values, and correcting outliers.

[0009] A "machine learning model" is an algorithm that learns specific patterns and relationships based on past data and uses the results of that learning to evaluate and predict new data.

[0010] "Training" is the process by which a machine learning model learns from past data, adjusts parameters, and improves its evaluation and prediction capabilities.

[0011] An "evaluation tool" is a function or process for evaluating employee performance using collected work data.

[0012] "Feedback" refers to presenting the evaluation results to employees and providing advice and suggestions for improvement to improve their performance based on the results.

[0013] "Performance" refers to the results of the work activities carried out by an employee and indicates how much they have contributed to achieving the goals of the company or organization.

[0014] A "system" is an entire system that combines multiple elements (hardware, software, algorithms, etc.) designed to achieve a specific purpose. [Brief explanation of the drawings]

[0015] [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

[0016] 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.

[0017] First, the terms used in the following description will be explained.

[0018] 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).

[0019] 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.

[0020] 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.

[0021] 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.

[0022] 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."

[0023] [First embodiment]

[0024] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0025] 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.

[0026] 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).

[0027] 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.

[0028] 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.

[0029] 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.

[0030] 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.

[0031] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0032] 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.

[0033] 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.

[0034] 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.

[0035] 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."

[0036] The present invention is a system for objectively evaluating employee performance using AI technology, and is implemented through the following steps.

[0037] Data collection

[0038] The server automatically collects employee work data, including past performance data, current work status data, and work execution efficiency data, such as completed project tasks, achieved goals, login times, and task completion times. This data is obtained from various internal databases and external APIs.

[0039] Data Preprocessing

[0040] The collected data is pre-processed by the server, which includes the following operations:

[0041] 1. Unifying data formats: Converting data from different sources into a common format.

[0042] 2. Data Combination: Combining information from different data sources into a single data set.

[0043] 3. Missing Value Imputation: Properly impute missing data to create a consistent dataset.

[0044] 4. Normalization and scaling: Reduce the variability in the data and convert it into a format suitable for machine learning models.

[0045] Training the model

[0046] The server uses the preprocessed data to train machine learning models, including algorithms such as random forests, XGBoost, and neural networks, using historical performance data and corresponding evaluation results to enable the models to accurately predict employee performance.

[0047] Performance evaluation

[0048] Using the trained machine learning model, the server evaluates employee performance, using real-time updated work data as input to calculate an employee performance score, taking into account current project progress, daily work performance, and feedback from colleagues and superiors.

[0049] feedback

[0050] The evaluation results are fed back to the employee via a device. For example, an employee may receive a notification on their smartphone or computer saying, "This month's performance score is 80 points." In addition, specific advice on how to improve performance based on the evaluation results and information useful for setting goals for the next time are also provided.

[0051] Specific examples

[0052] As an example, consider the case of evaluating employee A, a salesperson.

[0053] 1. The server collects employee A's sales data, number of clients, and number of deals for the past year from the CRM system.

[0054] 2. The server appropriately preprocesses the collected data, imputes missing values, and combines the data.

[0055] 3. The server trains a machine learning model using the preprocessed data and generates a model suitable for evaluating A.

[0056] 4. The server uses the current month's sales data and customer feedback as inputs to calculate A's performance score in real time.

[0057] 5. The device notifies Employee A of the evaluation results and displays this month's score and specific suggestions for improvement.

[0058] The above is the basic form for carrying out the invention. This system allows employee performance evaluation to be carried out objectively and fairly, which is expected to improve employee motivation and the overall productivity of the company.

[0059] The processing flow will be explained below.

[0060] Step 1:

[0061] The server collects employee work data. Specifically, it obtains data on past performance, current work status, and work execution efficiency from databases and APIs. For example, it collects task completion status from a project management system, sales data from a CRM system, and login time data from a time attendance management system.

[0062] Step 2:

[0063] The server preprocesses the collected data. First, it standardizes the data format and converts data obtained from different systems into a common format. Next, it links the data and integrates data from different sources into a single dataset. If there is incomplete data or missing values, it fills them in to create a consistent dataset. It also performs normalization and scaling to reduce data variability.

[0064] Step 3:

[0065] The server trains the machine learning model. It uses the preprocessed data to build a model using algorithms such as random forest, XGBoost, and neural networks. The training dataset includes past performance data and evaluation results. Based on this data, the model adjusts parameters to predict employee performance.

[0066] Step 4:

[0067] The server uses a trained machine learning model to evaluate employee performance. It inputs data on the progress of ongoing tasks, daily work performance data, and feedback from colleagues and superiors into the model in real time to calculate an employee performance score. This score is continuously updated to reflect the latest work situation.

[0068] Step 5:

[0069] The device then provides feedback on the evaluation results to the employee. The server then sends the generated evaluation results and feedback to the device, which then notifies the employee on their smartphone or computer. For example, the device displays a score such as "This month's performance score: 85 / 100," as well as specific areas for improvement and advice on setting goals for the next time.

[0070] Step 6:

[0071] Users (employees) check the evaluation results and feedback and use them to improve their work. Based on the evaluation results, they clarify goals and areas for improvement for the next month and create plans to improve their own work performance.

[0072] This processing flow ensures that employee performance evaluations are conducted objectively and fairly, leading to increased employee motivation and improved corporate productivity.

[0073] Example 1

[0074] 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."

[0075] In today's workplaces, employee performance evaluations often contain many subjective elements, making it difficult to provide fair and objective evaluations. Furthermore, the process from data collection to evaluation is cumbersome, making it difficult to carry out the evaluation efficiently. Furthermore, the lack of specific feedback or advice based on the evaluation results means that employees are not provided with information that will help them improve their performance or set goals.

[0076] 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.

[0077] In this invention, the server includes means for collecting employee work data, means for preprocessing the collected data, means for training a machine learning model based on the preprocessed data, means for evaluating employee performance using the trained machine learning model, and means for providing feedback on the evaluation results to the employee, thereby enabling objective and fair evaluation of employee performance and providing efficient data processing and specific feedback.

[0078] "Employee business data" is a general term for various data generated in the course of an employee's work, and specifically includes project completion tasks, goals achieved, login times, and the time required to complete tasks.

[0079] "Means of collection" refers to a mechanism for automatically or manually obtaining necessary business data from a server or database, including database queries and API requests.

[0080] "Preprocessing measures" are a series of operations to convert collected data into a form suitable for analysis and model training, including data format unification, data merging, missing value imputation, normalization, and scaling.

[0081] "Means for training a machine learning model" means the process of using pre-processed data to build a machine learning model based on a specific algorithm and then optimizing the model, including, for example, the use of random forests or neural networks.

[0082] The "means for evaluating employee performance" refers to a system that utilizes a trained machine learning model to calculate employee performance scores based on collected work data, and includes real-time data entry and score calculation using the model.

[0083] "Means of providing feedback" refers to methods for notifying employees of the evaluation results and providing information such as specific suggestions for improvement and assistance with setting goals for the next time, and includes notification APIs and notification functions on devices.

[0084] Unifying "formats" refers to the process of converting data obtained from different sources into a common format, including, for example, unifying date formats and numerical units.

[0085] "Scaling" is the process of adjusting for variability in data and transforming it so that machine learning algorithms can function effectively, and includes techniques such as standardization and normalization.

[0086] A "machine learning model" is an algorithm that learns specific patterns and features from data and uses them to make predictions and classifications from future data, and includes random forests, XGBoost, neural networks, and more.

[0087] A "performance score" is an indicator used by machine learning models to quantify or evaluate employee performance, and serves as the basis for performance evaluation and feedback.

[0088] The present invention is a system for objectively and fairly evaluating employee performance, which includes a number of processing steps. The system includes a server, a terminal, and a user.

[0089] Server Roles

[0090] Data collection

[0091] The server is responsible for automatically collecting employee business data, including information such as:

[0092] Completed tasks in a project

[0093] Goals achieved

[0094] Login Time

[0095] Time required to complete the task

[0096] The server retrieves this data from internal databases or external APIs, accessing the data using SQL queries or RESTful APIs.

[0097] Data Preprocessing

[0098] The collected data is pre-processed by the server in the following ways:

[0099] 1. Data format unification: Converting data from different sources into a common format. For example, converting different date formats into the standard ISO 8601 format.

[0100] 2. Data Combining: Combining information from different data sources into a single dataset. A specific example is "Merging Data Frames Using the pandas Library."

[0101] 3. Missing value imputation: Properly impute missing data to create a consistent dataset. For example, imputing missing values ​​with the mean or median, or imputing using k-nearest neighbors, are some of the methods.

[0102] 4. Normalization and scaling: Transform the data into a format suitable for machine learning models. Specifically, "Standardize the data using scikit-learn's StandardScaler."

[0103] Training the model

[0104] During the training process, the server uses the preprocessed data to train machine learning models. The algorithms used in this system include Random Forest, XGBoost, and neural networks. Training is performed using past performance data and corresponding evaluation results.

[0105] The specific steps are:

[0106] We build a model using scikit-learn's RandomForestClassifier and perform 10-fold cross-validation.

[0107] Use GridSearchCV to tune hyperparameters and select the best model.

[0108] Performance evaluation

[0109] Using the trained machine learning model, the server evaluates employee performance in real time using the following data:

[0110] Current month's sales data

[0111] Customer Feedback

[0112] Daily business performance

[0113] A specific example is the action "predict a performance score using the predict method."

[0114] Device Role

[0115] feedback

[0116] The terminal is responsible for notifying the user of the evaluation results. Notification is done via the employee's smartphone or computer. Specifically, this involves the process of calling a notification API and sending the evaluation score.

[0117] Based on the evaluation results, the system will also display specific suggestions for improvement and information that will help employees set their next goal. For example, for an employee who received a score of 80, the system will display advice such as "Set your goal for next month at 90 and take an online course to improve specific skills."

[0118] Specific examples

[0119] For example, to evaluate Employee A in sales:

[0120] 1. The server collects Employee A's sales data, number of clients, and number of deals for the past year from the CRM system. Specifically, it executes the SQL query "SELECT FROM SalesData WHERE EmployeeID = 'A'".

[0121] 2. The server reads the collected data as a data frame using the pandas library, imputes missing values, and combines the data.

[0122] 3. The server uses the preprocessed data to train a model using scikit-learn's RandomForestClassifier and optimizes the model through 10-fold cross-validation.

[0123] 4. The server uses the model to calculate a performance score based on the current month's sales data and customer feedback.

[0124] 5. The device notifies Employee A of the evaluation results and displays the score and specific improvement suggestions.

[0125] Prompt Sentence Examples

[0126] Example prompts to input to the generative AI model:

[0127] "Evaluate Employee A's performance based on sales data, number of clients, and number of deals closed over the past year. Also consider the current month's sales data and customer feedback."

[0128] The above is the basic form for carrying out the present invention. This system allows employee performance evaluations to be carried out objectively and fairly, which is expected to improve employee motivation and corporate productivity.

[0129] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0130] Step 1: Data collection

[0131] The server automatically collects employee work data. Specifically, the server periodically retrieves data from a database or external API. For example, the server uses SQL queries to retrieve "project completed tasks," "goals achieved," "login time," and "time required to complete tasks" from the database. The server also calls RESTful APIs to retrieve customer feedback and industry-standard data from external services. This data serves as input data for performance evaluation. The collected work data is generated as output.

[0132] Step 2: Data Preprocessing

[0133] The server pre-processes the collected data in the following ways:

[0134] 1. Unifying data formats: If input data has different formats, the server converts them into a common format. For example, it converts all different date formats into ISO 8601 format.

[0135] 2. Data Combining: The server combines the information obtained from different data sources into a single dataset. Here, we use the pandas library to perform operations to merge multiple data frames.

[0136] 3. Missing Value Imputation: The server imputes missing data using the mean, median, or k-nearest neighbors.

[0137] 4. Normalization and scaling: The server scales the collected data to reduce variability and make it easier for machine learning algorithms to work. For example, it standardizes the data using scikit-learn's StandardScaler.

[0138] This outputs a preprocessed, consistent dataset.

[0139] Step 3: Train the machine learning model

[0140] The server trains a machine learning model based on the preprocessed data. It uses the preprocessed performance data and the corresponding evaluation results as input data. The server executes the following specific algorithms:

[0141] Random Forest: We build a model using scikit-learn's RandomForestClassifier. Specifically, we create multiple decision trees using the training data and combine them to generate the final model.

[0142] Cross-validation: The server performs 10-fold cross-validation during the training process to evaluate the accuracy of the model, and also uses GridSearchCV to optimize the hyperparameters.

[0143] This outputs a trained machine learning model.

[0144] Step 4: Performance evaluation

[0145] The server evaluates employee performance using a trained machine learning model. As input data, it uses current job data and the individual performance data of the employee being evaluated. The server performs the following specific tasks:

[0146] Evaluation Data Entry: Collect current month sales data, customer feedback, and daily operational performance data.

[0147] Input data into the model: The server inputs these data into the model and uses the predict method to calculate a performance score.

[0148] This outputs a real-time performance evaluation score for each employee.

[0149] Step 5: Feedback

[0150] The terminal feeds back the evaluation results from the server to the employee. It uses the performance evaluation score and related feedback information as input data. Specifically, it performs the following processes:

[0151] Notification sending: The device calls the notification API and sends the evaluation results to the employee's smartphone or computer. For example, an employee with an evaluation score of 80 will be notified that "Your performance score this month is 80 points."

[0152] Providing improvement suggestions: Based on the evaluation results, the device will display specific improvement suggestions and support for setting goals for the next time. For example, it may suggest, "Set your goal for next month at 90 points and take an online course to improve a specific skill."

[0153] This allows employees to receive performance evaluations and specific feedback.

[0154] Prompt Sentence Examples

[0155] Example prompts to input to the generative AI model:

[0156] "Evaluate Employee A's performance based on sales data, number of clients, and number of deals closed over the past year. Also consider the current month's sales data and customer feedback."

[0157] (Application example 1)

[0158] 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."

[0159] Employee performance evaluations tend to be subjective and often lack transparency and fairness. Furthermore, the lack of a method for properly evaluating the operating status of work equipment makes efficient management difficult. This has led to a growing need for a system that comprehensively evaluates the performance of employees and work equipment and provides feedback to managers.

[0160] 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.

[0161] In this invention, the server includes means for collecting employee work data, means for preprocessing the collected data, means for training a machine learning model based on the preprocessed data, means for evaluating employee performance using the trained machine learning model, means for providing feedback on the evaluation results to the employee, means for collecting labor equipment operation data, means for preprocessing the collected labor equipment operation data, means for training a machine learning model based on the preprocessed labor equipment operation data, means for evaluating the labor equipment performance using the trained machine learning model, and means for providing feedback on the evaluation results to a manager. This makes it possible to objectively and fairly evaluate the performance of employees and labor equipment and provide appropriate feedback to the manager.

[0162] An "employee" is an individual employed by an organization or company to perform work.

[0163] "Business data" refers to various information generated by employees in the course of performing their work, including completed project tasks, goals achieved, login times, and the time required to complete tasks.

[0164] "Labor equipment" refers to equipment used in factories and businesses that performs physical work, such as machinery and robots.

[0165] "Operation data" refers to information generated when labor equipment is operating, and specifically includes operating time, number of completed tasks, maintenance logs, error occurrence status, etc.

[0166] "Preprocessing" refers to the process of converting collected data into a format suitable for analysis and machine learning, and includes unifying data formats, combining data, imputing missing values, and normalizing and scaling data.

[0167] A "machine learning model" is an algorithm that is trained using collected and preprocessed data to identify specific patterns and relationships and use them to make predictions and evaluations.

[0168] "Training" refers to the process of training an algorithm with data to build a machine learning model.

[0169] "Feedback" refers to the act of notifying employees and managers of evaluation results and providing advice on how to improve and enhance performance based on those results.

[0170] "Evaluation" refers to calculating the performance and accomplishments of employees and work equipment based on collected data and trained machine learning models, and judging their performance.

[0171] "Supervisor" refers to the individual responsible for the operation and supervision of employees and work equipment, and who receives evaluations and feedback.

[0172] 1. Data Collection

[0173] The server automatically collects work data from employees and work equipment. Collected data includes the number of tasks completed by employees, goals achieved, login times, and the time required to complete tasks. It also includes operational data for work equipment, such as operating hours, number of completed tasks, maintenance logs, and error occurrence status. This data is obtained from various internal databases and external APIs.

[0174] 2. Data Preprocessing

[0175] The server pre-processes the collected data, which includes the following operations:

[0176] Data format unification: Converting data from different sources into a common format.

[0177] Data Combination: Combining information from different data sources into a single data set.

[0178] Missing Value Imputation: Properly impute missing data to create a consistent dataset.

[0179] Normalization and scaling: Reduce the variability in the data and convert it into a format suitable for machine learning models.

[0180] 3. Training the model

[0181] The server uses the preprocessed data to train machine learning models, including algorithms such as random forests, XGBoost, and neural networks. Training uses historical performance data and corresponding evaluation results to enable the models to accurately predict employee and equipment performance.

[0182] 4. Performance Evaluation

[0183] Using the trained machine learning model, the server evaluates the performance of employees and work equipment. This evaluation uses real-time updated work data as input and calculates a performance score for employees and work equipment, taking into account, for example, the progress of current projects, daily work performance, feedback from colleagues and superiors, and the operation status of work equipment.

[0184] 5. Feedback

[0185] The evaluation results are fed back to employees and managers via terminals. Employees are notified on their smartphones or computers with a message such as, "This month's performance score is 80 points." Managers are provided with performance evaluations and improvement suggestions for work equipment. For example, the message may read, "Robot X's operating efficiency this month is 90%."

[0186] For example, let's assume that you are evaluating the performance of a welding robot, an example of a robot used in a factory. Past operation data (operating time, number of completed tasks, error log) for robot ID "robot1" is collected and a model is trained. Based on the training results, the current operation data is evaluated and feedback is sent to the manager. An example of a specific prompt sentence would be to evaluate performance based on the operation data "Robot ID: robot1, operating time: 120 hours, number of completed tasks: 200, error log: 5". The performance score is calculated as 85%, and feedback is sent to the manager.

[0187] The above is the basic form for carrying out the invention. This system allows for objective and fair evaluation of employee and work equipment performance, which is expected to improve employee motivation and overall company productivity.

[0188] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0189] Step 1:

[0190] The server collects work data from employees and work equipment. Specifically, it obtains the number of tasks completed by employees, the goals achieved, login time, and the time required to complete a task using an internal database or external API. Similarly, it collects the operating time of work equipment, the number of tasks completed, maintenance logs, and error occurrence status. This allows input data to be acquired.

[0191] Input: Employee work data and work equipment operation data

[0192] Output: Collected business and operational data

[0193] Step 2:

[0194] The server preprocesses the collected data. First, it converts the data formats obtained from different sources into a common format. Next, it integrates the information obtained from the different data sources into a single dataset and imputes missing values ​​appropriately. Finally, it normalizes and scales the data to reduce variability. This prepares the data in a format suitable for machine learning models.

[0195] Input: Collected business and operational data

[0196] Output: Preprocessed dataset

[0197] Step 3:

[0198] The server uses the preprocessed data to train machine learning models, specifically algorithms such as random forests, XGBoost, and neural networks, to accurately predict employee and equipment performance using past performance data and corresponding evaluation results.

[0199] Input: Preprocessed dataset

[0200] Output: A trained machine learning model

[0201] Step 4:

[0202] The server uses trained machine learning models to evaluate the performance of employees and work equipment. It inputs current work and operational data and calculates a performance score based on the model. For example, it takes into account the employee's current project progress, daily work performance, feedback from colleagues and superiors, and the operation status of work equipment.

[0203] Input: Current business and operational data

[0204] Output: Performance score

[0205] Step 5:

[0206] The terminal provides feedback on the evaluation results to employees and managers. Employees are notified via smartphone or computer in the form of, for example, "This month's performance score is 80 points." Managers are provided with the performance evaluation results of work equipment and suggestions for improvement. For example, the display might say, "Robot X's operating efficiency this month is 90%."

[0207] Input: Performance Score

[0208] Output: Feedback notification (provided to employee and manager)

[0209] These are the processing steps of this system. Through this process, it is possible to objectively and fairly evaluate the performance of employees and work equipment and provide appropriate feedback.

[0210] 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.

[0211] The present invention is a system that combines an emotion engine with employee performance evaluation to take into account the employee's emotional state and provide more appropriate feedback. This is implemented through the following steps:

[0212] Data collection

[0213] The server automatically collects employee work data, including past performance data, current work status data, and work execution efficiency data, such as completed project tasks, achieved goals, login times, and task completion times. This data is obtained from various internal databases and external APIs.

[0214] Data Preprocessing

[0215] The collected data is pre-processed by the server, which includes the following operations:

[0216] 1. Unifying data formats: Converting data from different sources into a common format.

[0217] 2. Data Combination: Combining information from different data sources into a single data set.

[0218] 3. Missing Value Imputation: Properly impute missing data to create a consistent dataset.

[0219] 4. Normalization and scaling: Reduce the variability in the data and convert it into a format suitable for machine learning models.

[0220] Training the model

[0221] The server uses the preprocessed data to train machine learning models, including algorithms such as random forests, XGBoost, and neural networks, using historical performance data and corresponding evaluation results to enable the models to accurately predict employee performance.

[0222] Performance evaluation

[0223] Using the trained machine learning model, the server evaluates employee performance, using real-time updated work data as input to calculate an employee performance score, taking into account current project progress, daily work performance, and feedback from colleagues and superiors.

[0224] Emotion recognition by emotion engine

[0225] The server uses an emotion engine to recognize the employee's emotions. The emotion engine analyzes emotions from the user's facial expressions, voice tone, and text data. For example, the server recognizes facial expressions using a webcam and analyzes speaking tone from voice input to understand the employee's current emotional state.

[0226] feedback

[0227] The device provides feedback based on the evaluation results and the emotion recognition results of the emotion engine. The server sends the generated evaluation results and the emotion engine's analysis results to the device, which then notifies the employee's smartphone or computer. In addition to a score such as "This month's performance score: 85 / 100," emotion-based advice is also provided. If the emotion engine detects the user's stress level, special advice for stress reduction is also displayed.

[0228] Specific examples

[0229] As an example, consider the case of evaluating employee A, a salesperson.

[0230] 1. The server collects employee A's sales data, number of clients, and number of deals for the past year from the CRM system.

[0231] 2. The server appropriately preprocesses the collected data, imputes missing values, and combines the data.

[0232] 3. The server trains a machine learning model using the preprocessed data and generates a model suitable for evaluating A.

[0233] 4. The server uses the current month's sales data and customer feedback as inputs to calculate A's performance score in real time.

[0234] 5. The server recognizes employee A's emotions from his facial expressions and voice and analyzes them using an emotion engine.

[0235] 6. The device notifies Employee A of the evaluation results and emotion analysis results, and displays this month's score, specific suggestions for improvement, and emotion-based advice.

[0236] This system allows employees' performance to be evaluated objectively and fairly, and provides detailed feedback that takes emotions into consideration, which is expected to increase employee motivation and improve work performance.

[0237] The processing flow will be explained below.

[0238] Step 1:

[0239] The server collects employee work data. Specifically, it obtains task completion status from the project management system and sales data from the CRM system. It also simultaneously collects login and logout time data and the number of successful transactions. This data is obtained from an internal database or via an external API.

[0240] Step 2:

[0241] The server preprocesses the collected data. First, it standardizes the data format and combines data from different sources. For example, it sorts performance data chronologically and converts login times into a single timestamp. Next, it completes missing data and removes invalid data. Finally, it normalizes the data and scales it to make the data range uniform.

[0242] Step 3:

[0243] The server trains the machine learning model. It uses the preprocessed data to build a model using algorithms such as random forests and neural networks. This training requires past performance data and evaluation results, which the model uses to learn from and adjust parameters to predict employee performance.

[0244] Step 4:

[0245] The server uses the trained model to evaluate employee performance. It inputs data such as the progress of ongoing tasks, daily work performance, and feedback from colleagues and superiors into the model in real time to calculate the employee's performance score. This score is continuously updated to reflect the latest work situation.

[0246] Step 5:

[0247] The server uses an emotion engine to recognize employees' emotions. The emotion engine uses a webcam or microphone to analyze emotions from employees' facial expressions, voice tone, and even input text. For example, a facial recognition algorithm can determine whether an employee is smiling or stressed. A voice recognition algorithm can analyze an employee's emotional state from their tone of voice and choice of words.

[0248] Step 6:

[0249] The server integrates the evaluation results with the emotion recognition results from the emotion engine. It then takes into account the user's emotional state and adjusts the feedback accordingly. For example, for an employee who has received a high performance score but is feeling stressed, it can include advice on relaxation or a suggestion for a break.

[0250] Step 7:

[0251] The device notifies the employee of the evaluation results and tailored feedback. The evaluation results are displayed as "This month's performance score: 85 / 100," and advice based on analysis from the emotion engine is also provided. For example, "You achieved a high score, but it seems like you've been feeling stressed recently. I recommend you take a short break."

[0252] Step 8:

[0253] The user (employee) checks the notified evaluation results and feedback. Based on the feedback, they work to improve their own work, setting goals for the next month and improving their performance. At the same time, they also implement the presented advice for stress management to maintain their physical and mental health.

[0254] This processing flow allows employees' performance to be evaluated objectively and fairly, and detailed feedback is provided that takes emotions into consideration, which is expected to increase employee motivation, improve work performance, and maintain their physical and mental health.

[0255] Example 2

[0256] 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."

[0257] Conventional employee performance evaluation systems evaluate employees based solely on performance data and do not take into account the emotional state or mental stress of employees, which means that they are unable to provide appropriate feedback, resulting in insufficient improvement of employee motivation or work performance. Furthermore, because the content of the feedback is uniform, there is also the issue of not being able to provide advice tailored to the individual circumstances and emotions of each employee.

[0258] 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 means for collecting employee work data, means for preprocessing the collected data, means for training a machine learning model based on the preprocessed data, means for evaluating employee performance using the trained machine learning model, emotion recognition means for recognizing the employee's emotional state, and means for feeding back the evaluation result and the emotion recognition result to the employee. This enables detailed feedback that takes into account not only the employee's performance data but also their emotional state, making it possible to provide effective advice for increasing motivation and improving work.

[0259] "Employee business data" refers to all information generated in the course of an employee's work, including, in particular, past performance data, current work status data, and work performance efficiency data.

[0260] "Preprocessing" refers to a series of operations performed to prepare collected data in a format that is easier to analyze, including standardizing data formats, combining data, imputing missing values, and normalizing and scaling.

[0261] A "machine learning model" refers to an algorithm that is trained on collected data and is used for pattern recognition and inference.

[0262] "Evaluating performance" refers to quantitatively evaluating an employee's work performance using a trained machine learning model.

[0263] "Emotion recognition" refers to the process of analyzing emotions from employees' facial expressions, voice tones, and text data, which allows you to understand their current emotional state.

[0264] "Feedback" refers to information and advice provided to employees based on the evaluation results and emotion recognition results, including specific suggestions for improving performance and emotion-based advice.

[0265] The present invention is a system that combines an emotion engine with employee performance evaluation to provide more appropriate feedback that takes into account the employee's emotional state. The system collects employee work data, preprocesses the data, and trains a machine learning model based on the preprocessed data. The trained machine learning model is used to evaluate the employee's performance, and emotion recognition means is used to recognize the employee's emotional state. Based on this, the evaluation results and emotion recognition results are fed back to the employee.

[0266] Hardware and Software

[0267] The system uses the following major hardware and software:

[0268] 1. Server:

[0269] Perform data collection, data preprocessing, machine learning model training, and performance evaluation.

[0270] Usage example:

[0271] MySQL or PostgreSQL as database management systems (DBMS).

[0272] Machine learning libraries include scikit-learn, TensorFlow, and XGBoost.

[0273] 2. Terminal:

[0274] Use employees' smartphones or computers as devices for providing feedback.

[0275] 3. Emotion recognition equipment:

[0276] A microphone for voice input and a webcam for facial recognition.

[0277] Usage example:

[0278] Voice analysis software includes Amazon Polly and Google Cloud Speech-to-Text.

[0279] OpenCV is used as a facial expression recognition library.

[0280] Data processing and calculation

[0281] Here, we will explain the specific data processing and data calculations performed by the system:

[0282] 1. Data Collection:

[0283] The server collects data on employees' past performance, current work status, and work execution efficiency. This data is obtained from internal databases and external APIs, such as from project management tools and CRM systems.

[0284] 2. Data Preprocessing:

[0285] The server preprocesses the collected data, including unifying data formats, combining data, imputing missing values, and normalizing and scaling. For example, it converts CSV files of different formats into a single table-format data set and imputes missing values ​​with the average value.

[0286] 3. Training the machine learning model:

[0287] The server uses the preprocessed data to train machine learning models, using algorithms such as random forests, XGBoost, and neural networks. For example, the model is trained using past performance data and evaluation results.

[0288] 4. Performance Evaluation:

[0289] Using the trained machine learning model, the server evaluates employee performance and calculates a performance score using real-time performance data as input. For example, the score takes into account the progress of current projects and the percentage of tasks completed.

[0290] 5. Emotion recognition:

[0291] The server uses emotion recognition means to recognize the employee's emotions, for example, by capturing the employee's facial expressions with a webcam and analyzing the employee's voice tone to recognize the employee's current emotional state.

[0292] 6. Feedback:

[0293] The server then sends the generated evaluation results and emotion recognition results to the device. The device then notifies the employee's smartphone or computer, providing specific performance evaluations and emotion-based advice. For example, in addition to a score such as "This month's performance score: 85 / 100," feedback such as "You're experiencing increased stress, so we recommend you take a break" is displayed.

[0294] Specific examples

[0295] Here is a specific example of how to evaluate Employee A, a salesperson:

[0296] 1. The server collects employee A's sales data, number of clients, and number of deals for the past year from the CRM system.

[0297] 2. The server preprocesses the collected data, imputes missing values, combines data, and standardizes formats.

[0298] 3. The server trains the machine learning model using the preprocessed data and generates an evaluation model for Employee A.

[0299] 4. The server uses the current month's sales data and customer feedback as inputs to calculate Employee A's performance score in real time.

[0300] 5. The server recognizes employee A's emotions from his facial expressions and voice and analyzes them using an emotion engine.

[0301] 6. The device notifies Employee A of the evaluation results and emotion analysis results, and displays advice such as "This month's performance score: 85 / 100" and "Your stress is increasing, so we recommend you take a break."

[0302] Example prompts for generative AI models

[0303] Below are some example prompts to input to a generative AI model:

[0304] "What kind of feedback should we provide to Employee A in the month when his performance is below average? Please consider specific advice and provide suggestions that also reflect the results of the emotion engine analysis."

[0305] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0306] Step 1:

[0307] The server collects employee work data, specifically employee past performance data, current work status data, and work execution efficiency data, from internal databases and external APIs. The input in this step is data from CRM systems and project management tools, and the output is to compile this data into a single dataset.

[0308] Step 2:

[0309] The server pre-processes the collected data, which includes the following steps:

[0310] 1. Unification of data formats: Convert data of different formats into a common format (input: various data formats, output: unified data format).

[0311] 2. Data Combination: Combining data from different sources into a single dataset (input: individual data items, output: cohesive dataset).

[0312] 3. Missing value imputation: impute missing data in an appropriate way (input: partially missing data, output: complete dataset with imputed missing values).

[0313] 4. Normalization and scaling: Reduce the variability of the data and convert it into a format suitable for machine learning models (input: unnormalized data, output: normalized and scaled data).

[0314] Step 3:

[0315] The server uses the preprocessed data to train a machine learning model, specifically by following these steps:

[0316] 1. Data Split: Split the data into training and test sets (input: preprocessed data, output: training and test sets).

[0317] 2. Model selection: Choose an algorithm suitable for the task (input: candidate algorithms, output: selected algorithm).

[0318] 3. Training: The selected algorithm is trained on the data (input: training set, output: trained model).

[0319] 4. Hyperparameter tuning: Adjusting parameters to improve model performance (input: model and candidate parameters, output: optimized model).

[0320] Step 4:

[0321] The server evaluates employee performance using a trained machine learning model, including:

[0322] 1. Data input: Input the current month's performance data and customer feedback into the model (input: real-time performance data, output: predicted score).

[0323] 2. Score calculation: The model calculates the employee's performance score based on the input data (input: current performance data, output: performance score).

[0324] Step 5:

[0325] The server uses the emotion recognition means to recognize the employee's emotion, specifically by the following steps:

[0326] 1. Voice input analysis: Record the employee's voice and analyze the voice data (input: voice data, output: voice analysis results).

[0327] 2. Facial expression recognition: Capture employees' facial expressions using a webcam and analyze their emotions (input: facial expression data, output: facial expression analysis results).

[0328] 3. Text analysis: Analyzes text data entered by employees and infers their emotions (input: text data, output: text analysis results).

[0329] Step 6:

[0330] The server sends the generated evaluation results and emotion recognition results to the terminal, which then notifies the employee's smartphone or computer. Specifically, the process is as follows:

[0331] 1. Displaying evaluation results: Notifying employees of their performance scores and specific performance evaluations (Input: Evaluation results, Output: Notification message).

[0332] 2. Emotion-based advice: Provides emotion-based advice based on emotion recognition results (input: emotion analysis results, output: advice message).

[0333] (Application example 2)

[0334] 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."

[0335] The problem is that performance evaluation and feedback of employees cannot take into account the emotional state of employees, making it difficult to properly manage the impact of the evaluation results on employees' psychology. In addition, it is not possible to provide detailed advice based on the employee's emotional state, making it difficult to simultaneously improve performance and employee satisfaction.

[0336] The specific processing by the specific 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 employee work data, means for preprocessing the collected data, means for training a machine learning model based on the preprocessed data, means for evaluating employee performance using the trained machine learning model, an emotion engine including emotion analysis means for analyzing the emotional state of the employee at the time of performance evaluation, and means for feeding back the evaluation results and emotion analysis results to the employee. As a result, employee performance evaluations are performed taking into account the emotional state of the employee, the impact of the evaluation results on employee psychology is appropriately managed, and it is possible to achieve both improved performance and improved employee satisfaction.

[0337] "Employee business data" refers to various data generated when employees perform their work, including past performance data, current work status data, and work performance efficiency data.

[0338] "Preprocessing" refers to data processing techniques to prepare collected data for use as input data for machine learning models, and includes standardizing data formats, combining data, filling in missing values, and normalizing and scaling.

[0339] A "machine learning model" is an algorithm that learns patterns and features based on large amounts of data and makes predictions and classifications, and includes random forests, XGBoost, neural networks, and others.

[0340] "Performance appraisal" is a method of using a trained machine learning model to evaluate an employee's performance and efficiency in carrying out their work as a number or score.

[0341] "Emotion analysis means" refers to a method for analyzing emotions from an employee's facial expressions, voice tone, text data, etc., and includes an emotion engine using a camera or microphone.

[0342] The "emotion engine" is a system that performs facial expression recognition and voice analysis to grasp the user's emotional state.

[0343] "Feedback" refers to the act of notifying employees of the results of their performance evaluation and sentiment analysis, and providing them with advice based on this.

[0344] "Real-time feedback" is a feature that provides immediate advice based on evaluation results and emotional state at the moment an employee is working.

[0345] "Smart glasses" are wearable devices that are equipped with a camera, display, and voice recognition functions and are worn by the user.

[0346] The employee performance evaluation system of this invention objectively evaluates employee performance by collecting and preprocessing employee work data and training a machine learning model. It also uses emotion analysis means to understand the employee's emotional state and provides feedback based on the evaluation results and advice based on their emotions, thereby achieving both employee satisfaction and improved performance.

[0347] This system uses the following hardware and software.

[0348] Hardware and software:

[0349] 1. Server:

[0350] Data collection:

[0351] Employee past performance data, current work status data, and work execution efficiency data are collected from various internal databases and external APIs.

[0352] Data preprocessing:

[0353] Preprocessing includes standardizing data formats, combining data, imputing missing values, and normalizing and scaling to convert the collected data into a consistent form.

[0354] Training the machine learning model:

[0355] Train a machine learning model (e.g., Random Forest, XGBoost, or Neural Network) using the preprocessed data.

[0356] 2. Smart Glasses:

[0357] Emotion analysis:

[0358] The smart glasses use a camera and microphone to analyze the facial expressions and tone of voice of customers and employees in real time, and estimate their emotional state using an emotion analysis model (such as CNN).

[0359] Real-time feedback:

[0360] The smart glasses display real-time feedback from superiors and colleagues, advice based on performance data, and advice based on emotions.

[0361] How to use:

[0362] 1. Data Collection and Preprocessing:

[0363] The server automatically collects and preprocesses employee work data, including standardizing data formats and filling in missing values.

[0364] 2. Training the machine learning model:

[0365] The preprocessed data is used to train machine learning models, using algorithms such as random forests and neural networks.

[0366] 3. Performance Evaluation:

[0367] A trained machine learning model evaluates employee performance using current work situation data as input.

[0368] 4. Emotion analysis:

[0369] The smart glasses are equipped with a camera and microphone that analyze the facial expressions and voices of employees and customers in real time, estimating their emotional state and sending it to a server.

[0370] 5. Providing Feedback:

[0371] Using the evaluation results and sentiment analysis, feedback is displayed on the smart glasses screen to the employee, including advice based on performance evaluation and customized advice based on emotions.

[0372] Examples:

[0373] For example, when an employee wears smart glasses while serving customers in a brick-and-mortar store, the server collects and preprocesses the employee's work data to train an evaluation model. The smart glasses analyze the customer's facial expressions and tone of voice using emotion analysis tools, and display the customer's emotional state to the employee in real time. This allows the employee to understand the customer's emotional state and provide appropriate service, further improving customer satisfaction.

[0374] Example prompt sentence:

[0375] "Please propose a system that analyzes the emotions of customers from their facial expressions and tone of voice, and provides appropriate customer service advice based on that."

[0376] This invention allows employees' performance to be evaluated fairly and objectively, and provides feedback that takes emotions into account, making it possible to achieve both improved performance and increased employee satisfaction.

[0377] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0378] Step 1:

[0379] Data collection

[0380] The server collects data on employees' past performance, current work status, and work efficiency from various internal databases and external APIs. The input is employee work-related data, and the output is business data converted into a unified data format.

[0381] Step 2:

[0382] Data Preprocessing

[0383] The server preprocesses the collected business data by standardizing data formats, combining data, imputing missing values, normalizing and scaling it. The input is the collected business data, and the output is a preprocessed, consistent dataset.

[0384] Step 3:

[0385] Training a machine learning model

[0386] The server uses the preprocessed data to train a machine learning model (e.g., Random Forest, XGBoost, Neural Network, etc.) The input is the preprocessed dataset and the output is the trained machine learning model.

[0387] Step 4:

[0388] Performance evaluation

[0389] The server evaluates employee performance using the trained machine learning model, using current work situation data as input and the employee performance score as output. The input is the current work situation data and the output is the evaluated performance score.

[0390] Step 5:

[0391] Emotion analysis

[0392] The smart glasses use a camera and microphone to analyze the facial expressions and tone of voice of customers and employees in real time. The input is camera video and audio data, and the output is the emotion analysis results, which are sent to a server.

[0393] Step 6:

[0394] Real-time feedback

[0395] The server sends the evaluation results and emotion analysis results to the smart glasses, which then provide feedback to the employee. The smart glasses display shows feedback from superiors and colleagues, advice based on performance data, and advice based on emotions. The input is the evaluation results and emotion analysis results, and the output is real-time feedback display.

[0396] Step 7:

[0397] Providing Feedback

[0398] Based on the evaluation results and sentiment analysis, more specific advice is displayed on the smart glasses, allowing the user to adjust their interactions with the customer. The input is feedback information, and the output is specific advice provided to employees.

[0399] 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.

[0400] 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.

[0401] 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.

[0402] [Second embodiment]

[0403] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0404] 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.

[0405] 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).

[0406] 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.

[0407] 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.

[0408] 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).

[0409] 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.

[0410] 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.

[0411] 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.

[0412] 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.

[0413] 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.

[0414] 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."

[0415] The present invention is a system for objectively evaluating employee performance using AI technology, and is implemented through the following steps.

[0416] Data collection

[0417] The server automatically collects employee work data, including past performance data, current work status data, and work execution efficiency data, such as completed project tasks, achieved goals, login times, and task completion times. This data is obtained from various internal databases and external APIs.

[0418] Data Preprocessing

[0419] The collected data is pre-processed by the server, which includes the following operations:

[0420] 1. Unifying data formats: Converting data from different sources into a common format.

[0421] 2. Data Combination: Combining information from different data sources into a single data set.

[0422] 3. Missing Value Imputation: Properly impute missing data to create a consistent dataset.

[0423] 4. Normalization and scaling: Reduce the variability in the data and convert it into a format suitable for machine learning models.

[0424] Training the model

[0425] The server uses the preprocessed data to train machine learning models, including algorithms such as random forests, XGBoost, and neural networks, using historical performance data and corresponding evaluation results to enable the models to accurately predict employee performance.

[0426] Performance evaluation

[0427] Using the trained machine learning model, the server evaluates employee performance, using real-time updated work data as input to calculate an employee performance score, taking into account current project progress, daily work performance, and feedback from colleagues and superiors.

[0428] feedback

[0429] The evaluation results are fed back to the employee via a device. For example, an employee may receive a notification on their smartphone or computer saying, "This month's performance score is 80 points." In addition, specific advice on how to improve performance based on the evaluation results and information useful for setting goals for the next time are also provided.

[0430] Specific examples

[0431] As an example, consider the case of evaluating employee A, a salesperson.

[0432] 1. The server collects employee A's sales data, number of clients, and number of deals for the past year from the CRM system.

[0433] 2. The server appropriately preprocesses the collected data, imputes missing values, and combines the data.

[0434] 3. The server trains a machine learning model using the preprocessed data and generates a model suitable for evaluating A.

[0435] 4. The server uses the current month's sales data and customer feedback as inputs to calculate A's performance score in real time.

[0436] 5. The device notifies Employee A of the evaluation results and displays this month's score and specific suggestions for improvement.

[0437] The above is the basic form for carrying out the invention. This system allows employee performance evaluation to be carried out objectively and fairly, which is expected to improve employee motivation and the overall productivity of the company.

[0438] The processing flow will be explained below.

[0439] Step 1:

[0440] The server collects employee work data. Specifically, it obtains data on past performance, current work status, and work execution efficiency from databases and APIs. For example, it collects task completion status from a project management system, sales data from a CRM system, and login time data from a time attendance management system.

[0441] Step 2:

[0442] The server preprocesses the collected data. First, it standardizes the data format and converts data obtained from different systems into a common format. Next, it links the data and integrates data from different sources into a single dataset. If there is incomplete data or missing values, it fills them in to create a consistent dataset. It also performs normalization and scaling to reduce data variability.

[0443] Step 3:

[0444] The server trains the machine learning model. It uses the preprocessed data to build a model using algorithms such as random forest, XGBoost, and neural networks. The training dataset includes past performance data and evaluation results. Based on this data, the model adjusts parameters to predict employee performance.

[0445] Step 4:

[0446] The server uses a trained machine learning model to evaluate employee performance. It inputs data on the progress of ongoing tasks, daily work performance data, and feedback from colleagues and superiors into the model in real time to calculate an employee performance score. This score is continuously updated to reflect the latest work situation.

[0447] Step 5:

[0448] The device then provides feedback on the evaluation results to the employee. The server then sends the generated evaluation results and feedback to the device, which then notifies the employee on their smartphone or computer. For example, the device displays a score such as "This month's performance score: 85 / 100," as well as specific areas for improvement and advice on setting goals for the next time.

[0449] Step 6:

[0450] Users (employees) check the evaluation results and feedback and use them to improve their work. Based on the evaluation results, they clarify goals and areas for improvement for the next month and create plans to improve their own work performance.

[0451] This processing flow ensures that employee performance evaluations are conducted objectively and fairly, leading to increased employee motivation and improved corporate productivity.

[0452] Example 1

[0453] 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."

[0454] In today's workplaces, employee performance evaluations often contain many subjective elements, making it difficult to provide fair and objective evaluations. Furthermore, the process from data collection to evaluation is cumbersome, making it difficult to carry out the evaluation efficiently. Furthermore, the lack of specific feedback or advice based on the evaluation results means that employees are not provided with information that will help them improve their performance or set goals.

[0455] 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.

[0456] In this invention, the server includes means for collecting employee work data, means for preprocessing the collected data, means for training a machine learning model based on the preprocessed data, means for evaluating employee performance using the trained machine learning model, and means for providing feedback on the evaluation results to the employee, thereby enabling objective and fair evaluation of employee performance and providing efficient data processing and specific feedback.

[0457] "Employee business data" is a general term for various data generated in the course of an employee's work, and specifically includes project completion tasks, goals achieved, login times, and the time required to complete tasks.

[0458] "Means of collection" refers to a mechanism for automatically or manually obtaining necessary business data from a server or database, including database queries and API requests.

[0459] "Preprocessing measures" are a series of operations to convert collected data into a form suitable for analysis and model training, including data format unification, data merging, missing value imputation, normalization, and scaling.

[0460] "Means for training a machine learning model" means the process of using pre-processed data to build a machine learning model based on a specific algorithm and then optimizing the model, including, for example, the use of random forests or neural networks.

[0461] The "means for evaluating employee performance" refers to a system that utilizes a trained machine learning model to calculate employee performance scores based on collected work data, and includes real-time data entry and score calculation using the model.

[0462] "Means of providing feedback" refers to methods for notifying employees of the evaluation results and providing information such as specific suggestions for improvement and assistance with setting goals for the next time, and includes notification APIs and notification functions on devices.

[0463] Unifying "formats" refers to the process of converting data obtained from different sources into a common format, including, for example, unifying date formats and numerical units.

[0464] "Scaling" is the process of adjusting for variability in data and transforming it so that machine learning algorithms can function effectively, and includes techniques such as standardization and normalization.

[0465] A "machine learning model" is an algorithm that learns specific patterns and features from data and uses them to make predictions and classifications from future data, and includes random forests, XGBoost, neural networks, and more.

[0466] A "performance score" is an indicator used by machine learning models to quantify or evaluate employee performance, and serves as the basis for performance evaluation and feedback.

[0467] The present invention is a system for objectively and fairly evaluating employee performance, which includes a number of processing steps. The system includes a server, a terminal, and a user.

[0468] Server Roles

[0469] Data collection

[0470] The server is responsible for automatically collecting employee business data, including information such as:

[0471] Completed tasks in a project

[0472] Goals achieved

[0473] Login Time

[0474] Time required to complete the task

[0475] The server retrieves this data from internal databases or external APIs, accessing the data using SQL queries or RESTful APIs.

[0476] Data Preprocessing

[0477] The collected data is pre-processed by the server in the following ways:

[0478] 1. Data format unification: Converting data from different sources into a common format. For example, converting different date formats into the standard ISO 8601 format.

[0479] 2. Data Combining: Combining information from different data sources into a single dataset. A specific example is "Merging Data Frames Using the pandas Library."

[0480] 3. Missing value imputation: Properly impute missing data to create a consistent dataset. For example, imputing missing values ​​with the mean or median, or imputing using k-nearest neighbors, are some of the methods.

[0481] 4. Normalization and scaling: Transform the data into a format suitable for machine learning models. Specifically, "Standardize the data using scikit-learn's StandardScaler."

[0482] Training the model

[0483] During the training process, the server uses the preprocessed data to train machine learning models. The algorithms used in this system include Random Forest, XGBoost, and neural networks. Training is performed using past performance data and corresponding evaluation results.

[0484] The specific steps are:

[0485] We build a model using scikit-learn's RandomForestClassifier and perform 10-fold cross-validation.

[0486] Use GridSearchCV to tune hyperparameters and select the best model.

[0487] Performance evaluation

[0488] Using the trained machine learning model, the server evaluates employee performance in real time using the following data:

[0489] Current month's sales data

[0490] Customer Feedback

[0491] Daily business performance

[0492] A specific example is the action "predict a performance score using the predict method."

[0493] Device Role

[0494] feedback

[0495] The terminal is responsible for notifying the user of the evaluation results. Notification is done via the employee's smartphone or computer. Specifically, this involves the process of calling a notification API and sending the evaluation score.

[0496] Based on the evaluation results, the system will also display specific suggestions for improvement and information that will help employees set their next goal. For example, for an employee who received a score of 80, the system will display advice such as "Set your goal for next month at 90 and take an online course to improve specific skills."

[0497] Specific examples

[0498] For example, to evaluate Employee A in sales:

[0499] 1. The server collects Employee A's sales data, number of clients, and number of deals for the past year from the CRM system. Specifically, it executes the SQL query "SELECT FROM SalesData WHERE EmployeeID = 'A'".

[0500] 2. The server reads the collected data as a data frame using the pandas library, imputes missing values, and combines the data.

[0501] 3. The server uses the preprocessed data to train a model using scikit-learn's RandomForestClassifier and optimizes the model through 10-fold cross-validation.

[0502] 4. The server uses the model to calculate a performance score based on the current month's sales data and customer feedback.

[0503] 5. The device notifies Employee A of the evaluation results and displays the score and specific improvement suggestions.

[0504] Prompt Sentence Examples

[0505] Example prompts to input to the generative AI model:

[0506] "Evaluate Employee A's performance based on sales data, number of clients, and number of deals closed over the past year. Also consider the current month's sales data and customer feedback."

[0507] The above is the basic form for carrying out the present invention. This system allows employee performance evaluations to be carried out objectively and fairly, which is expected to improve employee motivation and corporate productivity.

[0508] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0509] Step 1: Data collection

[0510] The server automatically collects employee work data. Specifically, the server periodically retrieves data from a database or external API. For example, the server uses SQL queries to retrieve "project completed tasks," "goals achieved," "login time," and "time required to complete tasks" from the database. The server also calls RESTful APIs to retrieve customer feedback and industry-standard data from external services. This data serves as input data for performance evaluation. The collected work data is generated as output.

[0511] Step 2: Data Preprocessing

[0512] The server pre-processes the collected data in the following ways:

[0513] 1. Unifying data formats: If input data has different formats, the server converts them into a common format. For example, it converts all different date formats into ISO 8601 format.

[0514] 2. Data Combining: The server combines the information obtained from different data sources into a single dataset. Here, we use the pandas library to perform operations to merge multiple data frames.

[0515] 3. Missing Value Imputation: The server imputes missing data using the mean, median, or k-nearest neighbors.

[0516] 4. Normalization and scaling: The server scales the collected data to reduce variability and make it easier for machine learning algorithms to work. For example, it standardizes the data using scikit-learn's StandardScaler.

[0517] This outputs a preprocessed, consistent dataset.

[0518] Step 3: Train the machine learning model

[0519] The server trains a machine learning model based on the preprocessed data. It uses the preprocessed performance data and the corresponding evaluation results as input data. The server executes the following specific algorithms:

[0520] Random Forest: We build a model using scikit-learn's RandomForestClassifier. Specifically, we create multiple decision trees using the training data and combine them to generate the final model.

[0521] Cross-validation: The server performs 10-fold cross-validation during the training process to evaluate the accuracy of the model, and also uses GridSearchCV to optimize the hyperparameters.

[0522] This outputs a trained machine learning model.

[0523] Step 4: Performance evaluation

[0524] The server evaluates employee performance using a trained machine learning model. As input data, it uses current job data and the individual performance data of the employee being evaluated. The server performs the following specific tasks:

[0525] Evaluation Data Entry: Collect current month sales data, customer feedback, and daily operational performance data.

[0526] Input data into the model: The server inputs these data into the model and uses the predict method to calculate a performance score.

[0527] This outputs a real-time performance evaluation score for each employee.

[0528] Step 5: Feedback

[0529] The terminal feeds back the evaluation results from the server to the employee. It uses the performance evaluation score and related feedback information as input data. Specifically, it performs the following processes:

[0530] Notification sending: The device calls the notification API and sends the evaluation results to the employee's smartphone or computer. For example, an employee with an evaluation score of 80 will be notified that "Your performance score this month is 80 points."

[0531] Providing improvement suggestions: Based on the evaluation results, the device will display specific improvement suggestions and support for setting goals for the next time. For example, it may suggest, "Set your goal for next month at 90 points and take an online course to improve a specific skill."

[0532] This allows employees to receive performance evaluations and specific feedback.

[0533] Prompt Sentence Examples

[0534] Example prompts to input to the generative AI model:

[0535] "Evaluate Employee A's performance based on sales data, number of clients, and number of deals closed over the past year. Also consider the current month's sales data and customer feedback."

[0536] (Application example 1)

[0537] 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."

[0538] Employee performance evaluations tend to be subjective and often lack transparency and fairness. Furthermore, the lack of a method for properly evaluating the operating status of work equipment makes efficient management difficult. This has led to a growing need for a system that comprehensively evaluates the performance of employees and work equipment and provides feedback to managers.

[0539] 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.

[0540] In this invention, the server includes means for collecting employee work data, means for preprocessing the collected data, means for training a machine learning model based on the preprocessed data, means for evaluating employee performance using the trained machine learning model, means for providing feedback on the evaluation results to the employee, means for collecting labor equipment operation data, means for preprocessing the collected labor equipment operation data, means for training a machine learning model based on the preprocessed labor equipment operation data, means for evaluating the labor equipment performance using the trained machine learning model, and means for providing feedback on the evaluation results to a manager. This makes it possible to objectively and fairly evaluate the performance of employees and labor equipment and provide appropriate feedback to the manager.

[0541] An "employee" is an individual employed by an organization or company to perform work.

[0542] "Business data" refers to various information generated by employees in the course of performing their work, including completed project tasks, goals achieved, login times, and the time required to complete tasks.

[0543] "Labor equipment" refers to equipment used in factories and businesses that performs physical work, such as machinery and robots.

[0544] "Operation data" refers to information generated when labor equipment is operating, and specifically includes operating time, number of completed tasks, maintenance logs, error occurrence status, etc.

[0545] "Preprocessing" refers to the process of converting collected data into a format suitable for analysis and machine learning, and includes unifying data formats, combining data, imputing missing values, and normalizing and scaling data.

[0546] A "machine learning model" is an algorithm that is trained using collected and preprocessed data to identify specific patterns and relationships and use them to make predictions and evaluations.

[0547] "Training" refers to the process of training an algorithm with data to build a machine learning model.

[0548] "Feedback" refers to the act of notifying employees and managers of evaluation results and providing advice on how to improve and enhance performance based on those results.

[0549] "Evaluation" refers to calculating the performance and accomplishments of employees and work equipment based on collected data and trained machine learning models, and judging their performance.

[0550] "Supervisor" refers to the individual responsible for the operation and supervision of employees and work equipment, and who receives evaluations and feedback.

[0551] 1. Data Collection

[0552] The server automatically collects work data from employees and work equipment. Collected data includes the number of tasks completed by employees, goals achieved, login times, and the time required to complete tasks. It also includes operational data for work equipment, such as operating hours, number of completed tasks, maintenance logs, and error occurrence status. This data is obtained from various internal databases and external APIs.

[0553] 2. Data Preprocessing

[0554] The server pre-processes the collected data, which includes the following operations:

[0555] Data format unification: Converting data from different sources into a common format.

[0556] Data Combination: Combining information from different data sources into a single data set.

[0557] Missing Value Imputation: Properly impute missing data to create a consistent dataset.

[0558] Normalization and scaling: Reduce the variability in the data and convert it into a format suitable for machine learning models.

[0559] 3. Training the model

[0560] The server uses the preprocessed data to train machine learning models, including algorithms such as random forests, XGBoost, and neural networks. Training uses historical performance data and corresponding evaluation results to enable the models to accurately predict employee and equipment performance.

[0561] 4. Performance Evaluation

[0562] Using the trained machine learning model, the server evaluates the performance of employees and work equipment. This evaluation uses real-time updated work data as input and calculates a performance score for employees and work equipment, taking into account, for example, the progress of current projects, daily work performance, feedback from colleagues and superiors, and the operation status of work equipment.

[0563] 5. Feedback

[0564] The evaluation results are fed back to employees and managers via terminals. Employees are notified on their smartphones or computers with a message such as, "This month's performance score is 80 points." Managers are provided with performance evaluations and improvement suggestions for work equipment. For example, the message may read, "Robot X's operating efficiency this month is 90%."

[0565] For example, let's assume that you are evaluating the performance of a welding robot, an example of a robot used in a factory. Past operation data (operating time, number of completed tasks, error log) for robot ID "robot1" is collected and a model is trained. Based on the training results, the current operation data is evaluated and feedback is sent to the manager. An example of a specific prompt sentence would be to evaluate performance based on the operation data "Robot ID: robot1, operating time: 120 hours, number of completed tasks: 200, error log: 5". The performance score is calculated as 85%, and feedback is sent to the manager.

[0566] The above is the basic form for carrying out the invention. This system allows for objective and fair evaluation of employee and work equipment performance, which is expected to improve employee motivation and overall company productivity.

[0567] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0568] Step 1:

[0569] The server collects work data from employees and work equipment. Specifically, it obtains the number of tasks completed by employees, the goals achieved, login time, and the time required to complete a task using an internal database or external API. Similarly, it collects the operating time of work equipment, the number of tasks completed, maintenance logs, and error occurrence status. This allows input data to be acquired.

[0570] Input: Employee work data and work equipment operation data

[0571] Output: Collected business and operational data

[0572] Step 2:

[0573] The server preprocesses the collected data. First, it converts the data formats obtained from different sources into a common format. Next, it integrates the information obtained from the different data sources into a single dataset and imputes missing values ​​appropriately. Finally, it normalizes and scales the data to reduce variability. This prepares the data in a format suitable for machine learning models.

[0574] Input: Collected business and operational data

[0575] Output: Preprocessed dataset

[0576] Step 3:

[0577] The server uses the preprocessed data to train machine learning models, specifically algorithms such as random forests, XGBoost, and neural networks, to accurately predict employee and equipment performance using past performance data and corresponding evaluation results.

[0578] Input: Preprocessed dataset

[0579] Output: A trained machine learning model

[0580] Step 4:

[0581] The server uses trained machine learning models to evaluate the performance of employees and work equipment. It inputs current work and operational data and calculates a performance score based on the model. For example, it takes into account the employee's current project progress, daily work performance, feedback from colleagues and superiors, and the operation status of work equipment.

[0582] Input: Current business and operational data

[0583] Output: Performance score

[0584] Step 5:

[0585] The terminal provides feedback on the evaluation results to employees and managers. Employees are notified via smartphone or computer in the form of, for example, "This month's performance score is 80 points." Managers are provided with the performance evaluation results of work equipment and suggestions for improvement. For example, the display might say, "Robot X's operating efficiency this month is 90%."

[0586] Input: Performance Score

[0587] Output: Feedback notification (provided to employee and manager)

[0588] These are the processing steps of this system. Through this process, it is possible to objectively and fairly evaluate the performance of employees and work equipment and provide appropriate feedback.

[0589] 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.

[0590] The present invention is a system that combines an emotion engine with employee performance evaluation to take into account the employee's emotional state and provide more appropriate feedback. This is implemented through the following steps:

[0591] Data collection

[0592] The server automatically collects employee work data, including past performance data, current work status data, and work execution efficiency data, such as completed project tasks, achieved goals, login times, and task completion times. This data is obtained from various internal databases and external APIs.

[0593] Data Preprocessing

[0594] The collected data is pre-processed by the server, which includes the following operations:

[0595] 1. Unifying data formats: Converting data from different sources into a common format.

[0596] 2. Data Combination: Combining information from different data sources into a single data set.

[0597] 3. Missing Value Imputation: Properly impute missing data to create a consistent dataset.

[0598] 4. Normalization and scaling: Reduce the variability in the data and convert it into a format suitable for machine learning models.

[0599] Training the model

[0600] The server uses the preprocessed data to train machine learning models, including algorithms such as random forests, XGBoost, and neural networks, using historical performance data and corresponding evaluation results to enable the models to accurately predict employee performance.

[0601] Performance evaluation

[0602] Using the trained machine learning model, the server evaluates employee performance, using real-time updated work data as input to calculate an employee performance score, taking into account current project progress, daily work performance, and feedback from colleagues and superiors.

[0603] Emotion recognition by emotion engine

[0604] The server uses an emotion engine to recognize the employee's emotions. The emotion engine analyzes emotions from the user's facial expressions, voice tone, and text data. For example, the server recognizes facial expressions using a webcam and analyzes speaking tone from voice input to understand the employee's current emotional state.

[0605] feedback

[0606] The device provides feedback based on the evaluation results and the emotion recognition results of the emotion engine. The server sends the generated evaluation results and the emotion engine's analysis results to the device, which then notifies the employee's smartphone or computer. In addition to a score such as "This month's performance score: 85 / 100," emotion-based advice is also provided. If the emotion engine detects the user's stress level, special advice for stress reduction is also displayed.

[0607] Specific examples

[0608] As an example, consider the case of evaluating employee A, a salesperson.

[0609] 1. The server collects employee A's sales data, number of clients, and number of deals for the past year from the CRM system.

[0610] 2. The server appropriately preprocesses the collected data, imputes missing values, and combines the data.

[0611] 3. The server trains a machine learning model using the preprocessed data and generates a model suitable for evaluating A.

[0612] 4. The server uses the current month's sales data and customer feedback as inputs to calculate A's performance score in real time.

[0613] 5. The server recognizes employee A's emotions from his facial expressions and voice and analyzes them using an emotion engine.

[0614] 6. The device notifies Employee A of the evaluation results and emotion analysis results, and displays this month's score, specific suggestions for improvement, and emotion-based advice.

[0615] This system allows employees' performance to be evaluated objectively and fairly, and provides detailed feedback that takes emotions into consideration, which is expected to increase employee motivation and improve work performance.

[0616] The processing flow will be explained below.

[0617] Step 1:

[0618] The server collects employee work data. Specifically, it obtains task completion status from the project management system and sales data from the CRM system. It also simultaneously collects login and logout time data and the number of successful transactions. This data is obtained from an internal database or via an external API.

[0619] Step 2:

[0620] The server preprocesses the collected data. First, it standardizes the data format and combines data from different sources. For example, it sorts performance data chronologically and converts login times into a single timestamp. Next, it completes missing data and removes invalid data. Finally, it normalizes the data and scales it to make the data range uniform.

[0621] Step 3:

[0622] The server trains the machine learning model. It uses the preprocessed data to build a model using algorithms such as random forests and neural networks. This training requires past performance data and evaluation results, which the model uses to learn from and adjust parameters to predict employee performance.

[0623] Step 4:

[0624] The server uses the trained model to evaluate employee performance. It inputs data such as the progress of ongoing tasks, daily work performance, and feedback from colleagues and superiors into the model in real time to calculate the employee's performance score. This score is continuously updated to reflect the latest work situation.

[0625] Step 5:

[0626] The server uses an emotion engine to recognize employees' emotions. The emotion engine uses a webcam or microphone to analyze emotions from employees' facial expressions, voice tone, and even input text. For example, a facial recognition algorithm can determine whether an employee is smiling or stressed. A voice recognition algorithm can analyze an employee's emotional state from their tone of voice and choice of words.

[0627] Step 6:

[0628] The server integrates the evaluation results with the emotion recognition results from the emotion engine. It then takes into account the user's emotional state and adjusts the feedback accordingly. For example, for an employee who has received a high performance score but is feeling stressed, it can include advice on relaxation or a suggestion for a break.

[0629] Step 7:

[0630] The device notifies the employee of the evaluation results and tailored feedback. The evaluation results are displayed as "This month's performance score: 85 / 100," and advice based on analysis from the emotion engine is also provided. For example, "You achieved a high score, but it seems like you've been feeling stressed recently. I recommend you take a short break."

[0631] Step 8:

[0632] The user (employee) checks the notified evaluation results and feedback. Based on the feedback, they work to improve their own work, setting goals for the next month and improving their performance. At the same time, they also implement the presented advice for stress management to maintain their physical and mental health.

[0633] This processing flow allows employees' performance to be evaluated objectively and fairly, and detailed feedback is provided that takes emotions into consideration, which is expected to increase employee motivation, improve work performance, and maintain their physical and mental health.

[0634] Example 2

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

[0636] Conventional employee performance evaluation systems evaluate employees based solely on performance data and do not take into account the emotional state or mental stress of employees, which means that they are unable to provide appropriate feedback, resulting in insufficient improvement of employee motivation or work performance. Furthermore, because the content of the feedback is uniform, there is also the issue of not being able to provide advice tailored to the individual circumstances and emotions of each employee.

[0637] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for collecting employee work data, means for preprocessing the collected data, means for training a machine learning model based on the preprocessed data, means for evaluating employee performance using the trained machine learning model, emotion recognition means for recognizing the employee's emotional state, and means for feeding back the evaluation result and the emotion recognition result to the employee. This enables detailed feedback that takes into account not only the employee's performance data but also their emotional state, making it possible to provide effective advice for increasing motivation and improving work.

[0638] "Employee business data" refers to all information generated in the course of an employee's work, including, in particular, past performance data, current work status data, and work performance efficiency data.

[0639] "Preprocessing" refers to a series of operations performed to prepare collected data in a format that is easier to analyze, including standardizing data formats, combining data, imputing missing values, and normalizing and scaling.

[0640] A "machine learning model" refers to an algorithm that is trained on collected data and is used for pattern recognition and inference.

[0641] "Evaluating performance" refers to quantitatively evaluating an employee's work performance using a trained machine learning model.

[0642] "Emotion recognition" refers to the process of analyzing emotions from employees' facial expressions, voice tones, and text data, which allows you to understand their current emotional state.

[0643] "Feedback" refers to information and advice provided to employees based on the evaluation results and emotion recognition results, including specific suggestions for improving performance and emotion-based advice.

[0644] The present invention is a system that combines an emotion engine with employee performance evaluation to provide more appropriate feedback that takes into account the employee's emotional state. The system collects employee work data, preprocesses the data, and trains a machine learning model based on the preprocessed data. The trained machine learning model is used to evaluate the employee's performance, and emotion recognition means is used to recognize the employee's emotional state. Based on this, the evaluation results and emotion recognition results are fed back to the employee.

[0645] Hardware and Software

[0646] The system uses the following major hardware and software:

[0647] 1. Server:

[0648] Perform data collection, data preprocessing, machine learning model training, and performance evaluation.

[0649] Usage example:

[0650] MySQL or PostgreSQL as database management systems (DBMS).

[0651] Machine learning libraries include scikit-learn, TensorFlow, and XGBoost.

[0652] 2. Terminal:

[0653] Use employees' smartphones or computers as devices for providing feedback.

[0654] 3. Emotion recognition equipment:

[0655] A microphone for voice input and a webcam for facial recognition.

[0656] Usage example:

[0657] Voice analysis software includes Amazon Polly and Google Cloud Speech-to-Text.

[0658] OpenCV is used as a facial expression recognition library.

[0659] Data processing and calculation

[0660] Here, we will explain the specific data processing and data calculations performed by the system:

[0661] 1. Data Collection:

[0662] The server collects data on employees' past performance, current work status, and work execution efficiency. This data is obtained from internal databases and external APIs, such as from project management tools and CRM systems.

[0663] 2. Data Preprocessing:

[0664] The server preprocesses the collected data, including unifying data formats, combining data, imputing missing values, and normalizing and scaling. For example, it converts CSV files of different formats into a single table-format data set and imputes missing values ​​with the average value.

[0665] 3. Training the machine learning model:

[0666] The server uses the preprocessed data to train machine learning models, using algorithms such as random forests, XGBoost, and neural networks. For example, the model is trained using past performance data and evaluation results.

[0667] 4. Performance Evaluation:

[0668] Using the trained machine learning model, the server evaluates employee performance and calculates a performance score using real-time performance data as input. For example, the score takes into account the progress of current projects and the percentage of tasks completed.

[0669] 5. Emotion recognition:

[0670] The server uses emotion recognition means to recognize the employee's emotions, for example, by capturing the employee's facial expressions with a webcam and analyzing the employee's voice tone to recognize the employee's current emotional state.

[0671] 6. Feedback:

[0672] The server then sends the generated evaluation results and emotion recognition results to the device. The device then notifies the employee's smartphone or computer, providing specific performance evaluations and emotion-based advice. For example, in addition to a score such as "This month's performance score: 85 / 100," feedback such as "You're experiencing increased stress, so we recommend you take a break" is displayed.

[0673] Specific examples

[0674] Here is a specific example of how to evaluate Employee A, a salesperson:

[0675] 1. The server collects employee A's sales data, number of clients, and number of deals for the past year from the CRM system.

[0676] 2. The server preprocesses the collected data, imputes missing values, combines data, and standardizes formats.

[0677] 3. The server trains the machine learning model using the preprocessed data and generates an evaluation model for Employee A.

[0678] 4. The server uses the current month's sales data and customer feedback as inputs to calculate Employee A's performance score in real time.

[0679] 5. The server recognizes employee A's emotions from his facial expressions and voice and analyzes them using an emotion engine.

[0680] 6. The device notifies Employee A of the evaluation results and emotion analysis results, and displays advice such as "This month's performance score: 85 / 100" and "Your stress is increasing, so we recommend you take a break."

[0681] Example prompts for generative AI models

[0682] Below are some example prompts to input to a generative AI model:

[0683] "What kind of feedback should we provide to Employee A in the month when his performance is below average? Please consider specific advice and provide suggestions that also reflect the results of the emotion engine analysis."

[0684] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0685] Step 1:

[0686] The server collects employee work data, specifically employee past performance data, current work status data, and work execution efficiency data, from internal databases and external APIs. The input in this step is data from CRM systems and project management tools, and the output is to compile this data into a single dataset.

[0687] Step 2:

[0688] The server pre-processes the collected data, which includes the following steps:

[0689] 1. Unification of data formats: Convert data of different formats into a common format (input: various data formats, output: unified data format).

[0690] 2. Data Combination: Combining data from different sources into a single dataset (input: individual data items, output: cohesive dataset).

[0691] 3. Missing value imputation: impute missing data in an appropriate way (input: partially missing data, output: complete dataset with imputed missing values).

[0692] 4. Normalization and scaling: Reduce the variability of the data and convert it into a format suitable for machine learning models (input: unnormalized data, output: normalized and scaled data).

[0693] Step 3:

[0694] The server uses the preprocessed data to train a machine learning model, specifically by following these steps:

[0695] 1. Data Split: Split the data into training and test sets (input: preprocessed data, output: training and test sets).

[0696] 2. Model selection: Choose an algorithm suitable for the task (input: candidate algorithms, output: selected algorithm).

[0697] 3. Training: The selected algorithm is trained on the data (input: training set, output: trained model).

[0698] 4. Hyperparameter tuning: Adjusting parameters to improve model performance (input: model and candidate parameters, output: optimized model).

[0699] Step 4:

[0700] The server evaluates employee performance using a trained machine learning model, including:

[0701] 1. Data input: Input the current month's performance data and customer feedback into the model (input: real-time performance data, output: predicted score).

[0702] 2. Score calculation: The model calculates the employee's performance score based on the input data (input: current performance data, output: performance score).

[0703] Step 5:

[0704] The server uses the emotion recognition means to recognize the employee's emotion, specifically by the following steps:

[0705] 1. Voice input analysis: Record the employee's voice and analyze the voice data (input: voice data, output: voice analysis results).

[0706] 2. Facial expression recognition: Capture employees' facial expressions using a webcam and analyze their emotions (input: facial expression data, output: facial expression analysis results).

[0707] 3. Text analysis: Analyzes text data entered by employees and infers their emotions (input: text data, output: text analysis results).

[0708] Step 6:

[0709] The server sends the generated evaluation results and emotion recognition results to the terminal, which then notifies the employee's smartphone or computer. Specifically, the process is as follows:

[0710] 1. Displaying evaluation results: Notifying employees of their performance scores and specific performance evaluations (Input: Evaluation results, Output: Notification message).

[0711] 2. Emotion-based advice: Provides emotion-based advice based on emotion recognition results (input: emotion analysis results, output: advice message).

[0712] (Application example 2)

[0713] 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."

[0714] The problem is that performance evaluation and feedback of employees cannot take into account the emotional state of employees, making it difficult to properly manage the impact of the evaluation results on employees' psychology. In addition, it is not possible to provide detailed advice based on the employee's emotional state, making it difficult to simultaneously improve performance and employee satisfaction.

[0715] The specific processing by the specific 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 employee work data, means for preprocessing the collected data, means for training a machine learning model based on the preprocessed data, means for evaluating employee performance using the trained machine learning model, an emotion engine including emotion analysis means for analyzing the emotional state of the employee at the time of performance evaluation, and means for feeding back the evaluation results and emotion analysis results to the employee. As a result, employee performance evaluations are performed taking into account the emotional state of the employee, the impact of the evaluation results on employee psychology is appropriately managed, and it is possible to achieve both improved performance and improved employee satisfaction.

[0716] "Employee business data" refers to various data generated when employees perform their work, including past performance data, current work status data, and work performance efficiency data.

[0717] "Preprocessing" refers to data processing techniques to prepare collected data for use as input data for machine learning models, and includes standardizing data formats, combining data, filling in missing values, and normalizing and scaling.

[0718] A "machine learning model" is an algorithm that learns patterns and features based on large amounts of data and makes predictions and classifications, and includes random forests, XGBoost, neural networks, and others.

[0719] "Performance appraisal" is a method of using a trained machine learning model to evaluate an employee's performance and efficiency in carrying out their work as a number or score.

[0720] "Emotion analysis means" refers to a method for analyzing emotions from an employee's facial expressions, voice tone, text data, etc., and includes an emotion engine using a camera or microphone.

[0721] The "emotion engine" is a system that performs facial expression recognition and voice analysis to grasp the user's emotional state.

[0722] "Feedback" refers to the act of notifying employees of the results of their performance evaluation and sentiment analysis, and providing them with advice based on this.

[0723] "Real-time feedback" is a feature that provides immediate advice based on evaluation results and emotional state at the moment an employee is working.

[0724] "Smart glasses" are wearable devices that are equipped with a camera, display, and voice recognition functions and are worn by the user.

[0725] The employee performance evaluation system of this invention objectively evaluates employee performance by collecting and preprocessing employee work data and training a machine learning model. It also uses emotion analysis means to understand the employee's emotional state and provides feedback based on the evaluation results and advice based on their emotions, thereby achieving both employee satisfaction and improved performance.

[0726] This system uses the following hardware and software.

[0727] Hardware and software:

[0728] 1. Server:

[0729] Data collection:

[0730] Employee past performance data, current work status data, and work execution efficiency data are collected from various internal databases and external APIs.

[0731] Data preprocessing:

[0732] Preprocessing includes standardizing data formats, combining data, imputing missing values, and normalizing and scaling to convert the collected data into a consistent form.

[0733] Training the machine learning model:

[0734] Train a machine learning model (e.g., Random Forest, XGBoost, or Neural Network) using the preprocessed data.

[0735] 2. Smart Glasses:

[0736] Emotion analysis:

[0737] The smart glasses use a camera and microphone to analyze the facial expressions and tone of voice of customers and employees in real time, and estimate their emotional state using an emotion analysis model (such as CNN).

[0738] Real-time feedback:

[0739] The smart glasses display real-time feedback from superiors and colleagues, advice based on performance data, and advice based on emotions.

[0740] How to use:

[0741] 1. Data Collection and Preprocessing:

[0742] The server automatically collects and preprocesses employee work data, including standardizing data formats and filling in missing values.

[0743] 2. Training the machine learning model:

[0744] The preprocessed data is used to train machine learning models, using algorithms such as random forests and neural networks.

[0745] 3. Performance Evaluation:

[0746] A trained machine learning model evaluates employee performance using current work situation data as input.

[0747] 4. Emotion analysis:

[0748] The smart glasses are equipped with a camera and microphone that analyze the facial expressions and voices of employees and customers in real time, estimating their emotional state and sending it to a server.

[0749] 5. Providing Feedback:

[0750] Using the evaluation results and sentiment analysis, feedback is displayed on the smart glasses screen to the employee, including advice based on performance evaluation and customized advice based on emotions.

[0751] Examples:

[0752] For example, when an employee wears smart glasses while serving customers in a brick-and-mortar store, the server collects and preprocesses the employee's work data to train an evaluation model. The smart glasses analyze the customer's facial expressions and tone of voice using emotion analysis tools, and display the customer's emotional state to the employee in real time. This allows the employee to understand the customer's emotional state and provide appropriate service, further improving customer satisfaction.

[0753] Example prompt sentence:

[0754] "Please propose a system that analyzes the emotions of customers from their facial expressions and tone of voice, and provides appropriate customer service advice based on that."

[0755] This invention allows employees' performance to be evaluated fairly and objectively, and provides feedback that takes emotions into account, making it possible to achieve both improved performance and increased employee satisfaction.

[0756] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0757] Step 1:

[0758] Data collection

[0759] The server collects data on employees' past performance, current work status, and work efficiency from various internal databases and external APIs. The input is employee work-related data, and the output is business data converted into a unified data format.

[0760] Step 2:

[0761] Data Preprocessing

[0762] The server preprocesses the collected business data by standardizing data formats, combining data, imputing missing values, normalizing and scaling it. The input is the collected business data, and the output is a preprocessed, consistent dataset.

[0763] Step 3:

[0764] Training a machine learning model

[0765] The server uses the preprocessed data to train a machine learning model (e.g., Random Forest, XGBoost, Neural Network, etc.) The input is the preprocessed dataset and the output is the trained machine learning model.

[0766] Step 4:

[0767] Performance evaluation

[0768] The server evaluates employee performance using the trained machine learning model, using current work situation data as input and the employee performance score as output. The input is the current work situation data and the output is the evaluated performance score.

[0769] Step 5:

[0770] Emotion analysis

[0771] The smart glasses use a camera and microphone to analyze the facial expressions and tone of voice of customers and employees in real time. The input is camera video and audio data, and the output is the emotion analysis results, which are sent to a server.

[0772] Step 6:

[0773] Real-time feedback

[0774] The server sends the evaluation results and emotion analysis results to the smart glasses, which then provide feedback to the employee. The smart glasses display shows feedback from superiors and colleagues, advice based on performance data, and advice based on emotions. The input is the evaluation results and emotion analysis results, and the output is real-time feedback display.

[0775] Step 7:

[0776] Providing Feedback

[0777] Based on the evaluation results and sentiment analysis, more specific advice is displayed on the smart glasses, allowing the user to adjust their interactions with the customer. The input is feedback information, and the output is specific advice provided to employees.

[0778] 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.

[0779] 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.

[0780] 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.

[0781] [Third embodiment]

[0782] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0783] 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.

[0784] 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).

[0785] 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.

[0786] 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.

[0787] 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).

[0788] 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.

[0789] 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.

[0790] 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.

[0791] 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.

[0792] 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.

[0793] 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."

[0794] The present invention is a system for objectively evaluating employee performance using AI technology, and is implemented through the following steps.

[0795] Data collection

[0796] The server automatically collects employee work data, including past performance data, current work status data, and work execution efficiency data, such as completed project tasks, achieved goals, login times, and task completion times. This data is obtained from various internal databases and external APIs.

[0797] Data Preprocessing

[0798] The collected data is pre-processed by the server, which includes the following operations:

[0799] 1. Unifying data formats: Converting data from different sources into a common format.

[0800] 2. Data Combination: Combining information from different data sources into a single data set.

[0801] 3. Missing Value Imputation: Properly impute missing data to create a consistent dataset.

[0802] 4. Normalization and scaling: Reduce the variability in the data and convert it into a format suitable for machine learning models.

[0803] Training the model

[0804] The server uses the preprocessed data to train machine learning models, including algorithms such as random forests, XGBoost, and neural networks, using historical performance data and corresponding evaluation results to enable the models to accurately predict employee performance.

[0805] Performance evaluation

[0806] Using the trained machine learning model, the server evaluates employee performance, using real-time updated work data as input to calculate an employee performance score, taking into account current project progress, daily work performance, and feedback from colleagues and superiors.

[0807] feedback

[0808] The evaluation results are fed back to the employee via a device. For example, an employee may receive a notification on their smartphone or computer saying, "This month's performance score is 80 points." In addition, specific advice on how to improve performance based on the evaluation results and information useful for setting goals for the next time are also provided.

[0809] Specific examples

[0810] As an example, consider the case of evaluating employee A, a salesperson.

[0811] 1. The server collects employee A's sales data, number of clients, and number of deals for the past year from the CRM system.

[0812] 2. The server appropriately preprocesses the collected data, imputes missing values, and combines the data.

[0813] 3. The server trains a machine learning model using the preprocessed data and generates a model suitable for evaluating A.

[0814] 4. The server uses the current month's sales data and customer feedback as inputs to calculate A's performance score in real time.

[0815] 5. The device notifies Employee A of the evaluation results and displays this month's score and specific suggestions for improvement.

[0816] The above is the basic form for carrying out the invention. This system allows employee performance evaluation to be carried out objectively and fairly, which is expected to improve employee motivation and the overall productivity of the company.

[0817] The processing flow will be explained below.

[0818] Step 1:

[0819] The server collects employee work data. Specifically, it obtains data on past performance, current work status, and work execution efficiency from databases and APIs. For example, it collects task completion status from a project management system, sales data from a CRM system, and login time data from a time attendance management system.

[0820] Step 2:

[0821] The server preprocesses the collected data. First, it standardizes the data format and converts data obtained from different systems into a common format. Next, it links the data and integrates data from different sources into a single dataset. If there is incomplete data or missing values, it fills them in to create a consistent dataset. It also performs normalization and scaling to reduce data variability.

[0822] Step 3:

[0823] The server trains the machine learning model. It uses the preprocessed data to build a model using algorithms such as random forest, XGBoost, and neural networks. The training dataset includes past performance data and evaluation results. Based on this data, the model adjusts parameters to predict employee performance.

[0824] Step 4:

[0825] The server uses a trained machine learning model to evaluate employee performance. It inputs data on the progress of ongoing tasks, daily work performance data, and feedback from colleagues and superiors into the model in real time to calculate an employee performance score. This score is continuously updated to reflect the latest work situation.

[0826] Step 5:

[0827] The device then provides feedback on the evaluation results to the employee. The server then sends the generated evaluation results and feedback to the device, which then notifies the employee on their smartphone or computer. For example, the device displays a score such as "This month's performance score: 85 / 100," as well as specific areas for improvement and advice on setting goals for the next time.

[0828] Step 6:

[0829] Users (employees) check the evaluation results and feedback and use them to improve their work. Based on the evaluation results, they clarify goals and areas for improvement for the next month and create plans to improve their own work performance.

[0830] This processing flow ensures that employee performance evaluations are conducted objectively and fairly, leading to increased employee motivation and improved corporate productivity.

[0831] Example 1

[0832] 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."

[0833] In today's workplaces, employee performance evaluations often contain many subjective elements, making it difficult to provide fair and objective evaluations. Furthermore, the process from data collection to evaluation is cumbersome, making it difficult to carry out the evaluation efficiently. Furthermore, the lack of specific feedback or advice based on the evaluation results means that employees are not provided with information that will help them improve their performance or set goals.

[0834] 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.

[0835] In this invention, the server includes means for collecting employee work data, means for preprocessing the collected data, means for training a machine learning model based on the preprocessed data, means for evaluating employee performance using the trained machine learning model, and means for providing feedback on the evaluation results to the employee, thereby enabling objective and fair evaluation of employee performance and providing efficient data processing and specific feedback.

[0836] "Employee business data" is a general term for various data generated in the course of an employee's work, and specifically includes project completion tasks, goals achieved, login times, and the time required to complete tasks.

[0837] "Means of collection" refers to a mechanism for automatically or manually obtaining necessary business data from a server or database, including database queries and API requests.

[0838] "Preprocessing measures" are a series of operations to convert collected data into a form suitable for analysis and model training, including data format unification, data merging, missing value imputation, normalization, and scaling.

[0839] "Means for training a machine learning model" means the process of using pre-processed data to build a machine learning model based on a specific algorithm and then optimizing the model, including, for example, the use of random forests or neural networks.

[0840] The "means for evaluating employee performance" refers to a system that utilizes a trained machine learning model to calculate employee performance scores based on collected work data, and includes real-time data entry and score calculation using the model.

[0841] "Means of providing feedback" refers to methods for notifying employees of the evaluation results and providing information such as specific suggestions for improvement and assistance with setting goals for the next time, and includes notification APIs and notification functions on devices.

[0842] Unifying "formats" refers to the process of converting data obtained from different sources into a common format, including, for example, unifying date formats and numerical units.

[0843] "Scaling" is the process of adjusting for variability in data and transforming it so that machine learning algorithms can function effectively, and includes techniques such as standardization and normalization.

[0844] A "machine learning model" is an algorithm that learns specific patterns and features from data and uses them to make predictions and classifications from future data, and includes random forests, XGBoost, neural networks, and more.

[0845] A "performance score" is an indicator used by machine learning models to quantify or evaluate employee performance, and serves as the basis for performance evaluation and feedback.

[0846] The present invention is a system for objectively and fairly evaluating employee performance, which includes a number of processing steps. The system includes a server, a terminal, and a user.

[0847] Server Roles

[0848] Data collection

[0849] The server is responsible for automatically collecting employee business data, including information such as:

[0850] Completed tasks in a project

[0851] Goals achieved

[0852] Login Time

[0853] Time required to complete the task

[0854] The server retrieves this data from internal databases or external APIs, accessing the data using SQL queries or RESTful APIs.

[0855] Data Preprocessing

[0856] The collected data is pre-processed by the server in the following ways:

[0857] 1. Data format unification: Converting data from different sources into a common format. For example, converting different date formats into the standard ISO 8601 format.

[0858] 2. Data Combining: Combining information from different data sources into a single dataset. A specific example is "Merging Data Frames Using the pandas Library."

[0859] 3. Missing value imputation: Properly impute missing data to create a consistent dataset. For example, imputing missing values ​​with the mean or median, or imputing using k-nearest neighbors, are some of the methods.

[0860] 4. Normalization and scaling: Transform the data into a format suitable for machine learning models. Specifically, "Standardize the data using scikit-learn's StandardScaler."

[0861] Training the model

[0862] During the training process, the server uses the preprocessed data to train machine learning models. The algorithms used in this system include Random Forest, XGBoost, and neural networks. Training is performed using past performance data and corresponding evaluation results.

[0863] The specific steps are:

[0864] We build a model using scikit-learn's RandomForestClassifier and perform 10-fold cross-validation.

[0865] Use GridSearchCV to tune hyperparameters and select the best model.

[0866] Performance evaluation

[0867] Using the trained machine learning model, the server evaluates employee performance in real time using the following data:

[0868] Current month's sales data

[0869] Customer Feedback

[0870] Daily business performance

[0871] A specific example is the action "predict a performance score using the predict method."

[0872] Device Role

[0873] feedback

[0874] The terminal is responsible for notifying the user of the evaluation results. Notification is done via the employee's smartphone or computer. Specifically, this involves the process of calling a notification API and sending the evaluation score.

[0875] Based on the evaluation results, the system will also display specific suggestions for improvement and information that will help employees set their next goal. For example, for an employee who received a score of 80, the system will display advice such as "Set your goal for next month at 90 and take an online course to improve specific skills."

[0876] Specific examples

[0877] For example, to evaluate Employee A in sales:

[0878] 1. The server collects Employee A's sales data, number of clients, and number of deals for the past year from the CRM system. Specifically, it executes the SQL query "SELECT FROM SalesData WHERE EmployeeID = 'A'".

[0879] 2. The server reads the collected data as a data frame using the pandas library, imputes missing values, and combines the data.

[0880] 3. The server uses the preprocessed data to train a model using scikit-learn's RandomForestClassifier and optimizes the model through 10-fold cross-validation.

[0881] 4. The server uses the model to calculate a performance score based on the current month's sales data and customer feedback.

[0882] 5. The device notifies Employee A of the evaluation results and displays the score and specific improvement suggestions.

[0883] Prompt Sentence Examples

[0884] Example prompts to input to the generative AI model:

[0885] "Evaluate Employee A's performance based on sales data, number of clients, and number of deals closed over the past year. Also consider the current month's sales data and customer feedback."

[0886] The above is the basic form for carrying out the present invention. This system allows employee performance evaluations to be carried out objectively and fairly, which is expected to improve employee motivation and corporate productivity.

[0887] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0888] Step 1: Data collection

[0889] The server automatically collects employee work data. Specifically, the server periodically retrieves data from a database or external API. For example, the server uses SQL queries to retrieve "project completed tasks," "goals achieved," "login time," and "time required to complete tasks" from the database. The server also calls RESTful APIs to retrieve customer feedback and industry-standard data from external services. This data serves as input data for performance evaluation. The collected work data is generated as output.

[0890] Step 2: Data Preprocessing

[0891] The server pre-processes the collected data in the following ways:

[0892] 1. Unifying data formats: If input data has different formats, the server converts them into a common format. For example, it converts all different date formats into ISO 8601 format.

[0893] 2. Data Combining: The server combines the information obtained from different data sources into a single dataset. Here, we use the pandas library to perform operations to merge multiple data frames.

[0894] 3. Missing Value Imputation: The server imputes missing data using the mean, median, or k-nearest neighbors.

[0895] 4. Normalization and scaling: The server scales the collected data to reduce variability and make it easier for machine learning algorithms to work. For example, it standardizes the data using scikit-learn's StandardScaler.

[0896] This outputs a preprocessed, consistent dataset.

[0897] Step 3: Train the machine learning model

[0898] The server trains a machine learning model based on the preprocessed data. It uses the preprocessed performance data and the corresponding evaluation results as input data. The server executes the following specific algorithms:

[0899] Random Forest: We build a model using scikit-learn's RandomForestClassifier. Specifically, we create multiple decision trees using the training data and combine them to generate the final model.

[0900] Cross-validation: The server performs 10-fold cross-validation during the training process to evaluate the accuracy of the model, and also uses GridSearchCV to optimize the hyperparameters.

[0901] This outputs a trained machine learning model.

[0902] Step 4: Performance evaluation

[0903] The server evaluates employee performance using a trained machine learning model. As input data, it uses current job data and the individual performance data of the employee being evaluated. The server performs the following specific tasks:

[0904] Evaluation Data Entry: Collect current month sales data, customer feedback, and daily operational performance data.

[0905] Input data into the model: The server inputs these data into the model and uses the predict method to calculate a performance score.

[0906] This outputs a real-time performance evaluation score for each employee.

[0907] Step 5: Feedback

[0908] The terminal feeds back the evaluation results from the server to the employee. It uses the performance evaluation score and related feedback information as input data. Specifically, it performs the following processes:

[0909] Notification sending: The device calls the notification API and sends the evaluation results to the employee's smartphone or computer. For example, an employee with an evaluation score of 80 will be notified that "Your performance score this month is 80 points."

[0910] Providing improvement suggestions: Based on the evaluation results, the device will display specific improvement suggestions and support for setting goals for the next time. For example, it may suggest, "Set your goal for next month at 90 points and take an online course to improve a specific skill."

[0911] This allows employees to receive performance evaluations and specific feedback.

[0912] Prompt Sentence Examples

[0913] Example prompts to input to the generative AI model:

[0914] "Evaluate Employee A's performance based on sales data, number of clients, and number of deals closed over the past year. Also consider the current month's sales data and customer feedback."

[0915] (Application example 1)

[0916] 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."

[0917] Employee performance evaluations tend to be subjective and often lack transparency and fairness. Furthermore, the lack of a method for properly evaluating the operating status of work equipment makes efficient management difficult. This has led to a growing need for a system that comprehensively evaluates the performance of employees and work equipment and provides feedback to managers.

[0918] 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.

[0919] In this invention, the server includes means for collecting employee work data, means for preprocessing the collected data, means for training a machine learning model based on the preprocessed data, means for evaluating employee performance using the trained machine learning model, means for providing feedback on the evaluation results to the employee, means for collecting labor equipment operation data, means for preprocessing the collected labor equipment operation data, means for training a machine learning model based on the preprocessed labor equipment operation data, means for evaluating the labor equipment performance using the trained machine learning model, and means for providing feedback on the evaluation results to a manager. This makes it possible to objectively and fairly evaluate the performance of employees and labor equipment and provide appropriate feedback to the manager.

[0920] An "employee" is an individual employed by an organization or company to perform work.

[0921] "Business data" refers to various information generated by employees in the course of performing their work, including completed project tasks, goals achieved, login times, and the time required to complete tasks.

[0922] "Labor equipment" refers to equipment used in factories and businesses that performs physical work, such as machinery and robots.

[0923] "Operation data" refers to information generated when labor equipment is operating, and specifically includes operating time, number of completed tasks, maintenance logs, error occurrence status, etc.

[0924] "Preprocessing" refers to the process of converting collected data into a format suitable for analysis and machine learning, and includes unifying data formats, combining data, imputing missing values, and normalizing and scaling data.

[0925] A "machine learning model" is an algorithm that is trained using collected and preprocessed data to identify specific patterns and relationships and use them to make predictions and evaluations.

[0926] "Training" refers to the process of training an algorithm with data to build a machine learning model.

[0927] "Feedback" refers to the act of notifying employees and managers of evaluation results and providing advice on how to improve and enhance performance based on those results.

[0928] "Evaluation" refers to calculating the performance and accomplishments of employees and work equipment based on collected data and trained machine learning models, and judging their performance.

[0929] "Supervisor" refers to the individual responsible for the operation and supervision of employees and work equipment, and who receives evaluations and feedback.

[0930] 1. Data Collection

[0931] The server automatically collects work data from employees and work equipment. Collected data includes the number of tasks completed by employees, goals achieved, login times, and the time required to complete tasks. It also includes operational data for work equipment, such as operating hours, number of completed tasks, maintenance logs, and error occurrence status. This data is obtained from various internal databases and external APIs.

[0932] 2. Data Preprocessing

[0933] The server pre-processes the collected data, which includes the following operations:

[0934] Data format unification: Converting data from different sources into a common format.

[0935] Data Combination: Combining information from different data sources into a single data set.

[0936] Missing Value Imputation: Properly impute missing data to create a consistent dataset.

[0937] Normalization and scaling: Reduce the variability in the data and convert it into a format suitable for machine learning models.

[0938] 3. Training the model

[0939] The server uses the preprocessed data to train machine learning models, including algorithms such as random forests, XGBoost, and neural networks. Training uses historical performance data and corresponding evaluation results to enable the models to accurately predict employee and equipment performance.

[0940] 4. Performance Evaluation

[0941] Using the trained machine learning model, the server evaluates the performance of employees and work equipment. This evaluation uses real-time updated work data as input and calculates a performance score for employees and work equipment, taking into account, for example, the progress of current projects, daily work performance, feedback from colleagues and superiors, and the operation status of work equipment.

[0942] 5. Feedback

[0943] The evaluation results are fed back to employees and managers via terminals. Employees are notified on their smartphones or computers with a message such as, "This month's performance score is 80 points." Managers are provided with performance evaluations and improvement suggestions for work equipment. For example, the message may read, "Robot X's operating efficiency this month is 90%."

[0944] For example, let's assume that you are evaluating the performance of a welding robot, an example of a robot used in a factory. Past operation data (operating time, number of completed tasks, error log) for robot ID "robot1" is collected and a model is trained. Based on the training results, the current operation data is evaluated and feedback is sent to the manager. An example of a specific prompt sentence would be to evaluate performance based on the operation data "Robot ID: robot1, operating time: 120 hours, number of completed tasks: 200, error log: 5". The performance score is calculated as 85%, and feedback is sent to the manager.

[0945] The above is the basic form for carrying out the invention. This system allows for objective and fair evaluation of employee and work equipment performance, which is expected to improve employee motivation and overall company productivity.

[0946] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0947] Step 1:

[0948] The server collects work data from employees and work equipment. Specifically, it obtains the number of tasks completed by employees, the goals achieved, login time, and the time required to complete a task using an internal database or external API. Similarly, it collects the operating time of work equipment, the number of tasks completed, maintenance logs, and error occurrence status. This allows input data to be acquired.

[0949] Input: Employee work data and work equipment operation data

[0950] Output: Collected business and operational data

[0951] Step 2:

[0952] The server preprocesses the collected data. First, it converts the data formats obtained from different sources into a common format. Next, it integrates the information obtained from the different data sources into a single dataset and imputes missing values ​​appropriately. Finally, it normalizes and scales the data to reduce variability. This prepares the data in a format suitable for machine learning models.

[0953] Input: Collected business and operational data

[0954] Output: Preprocessed dataset

[0955] Step 3:

[0956] The server uses the preprocessed data to train machine learning models, specifically algorithms such as random forests, XGBoost, and neural networks, to accurately predict employee and equipment performance using past performance data and corresponding evaluation results.

[0957] Input: Preprocessed dataset

[0958] Output: A trained machine learning model

[0959] Step 4:

[0960] The server uses trained machine learning models to evaluate the performance of employees and work equipment. It inputs current work and operational data and calculates a performance score based on the model. For example, it takes into account the employee's current project progress, daily work performance, feedback from colleagues and superiors, and the operation status of work equipment.

[0961] Input: Current business and operational data

[0962] Output: Performance score

[0963] Step 5:

[0964] The terminal provides feedback on the evaluation results to employees and managers. Employees are notified via smartphone or computer in the form of, for example, "This month's performance score is 80 points." Managers are provided with the performance evaluation results of work equipment and suggestions for improvement. For example, the display might say, "Robot X's operating efficiency this month is 90%."

[0965] Input: Performance Score

[0966] Output: Feedback notification (provided to employee and manager)

[0967] These are the processing steps of this system. Through this process, it is possible to objectively and fairly evaluate the performance of employees and work equipment and provide appropriate feedback.

[0968] 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.

[0969] The present invention is a system that combines an emotion engine with employee performance evaluation to take into account the employee's emotional state and provide more appropriate feedback. This is implemented through the following steps:

[0970] Data collection

[0971] The server automatically collects employee work data, including past performance data, current work status data, and work execution efficiency data, such as completed project tasks, achieved goals, login times, and task completion times. This data is obtained from various internal databases and external APIs.

[0972] Data Preprocessing

[0973] The collected data is pre-processed by the server, which includes the following operations:

[0974] 1. Unifying data formats: Converting data from different sources into a common format.

[0975] 2. Data Combination: Combining information from different data sources into a single data set.

[0976] 3. Missing Value Imputation: Properly impute missing data to create a consistent dataset.

[0977] 4. Normalization and scaling: Reduce the variability in the data and convert it into a format suitable for machine learning models.

[0978] Training the model

[0979] The server uses the preprocessed data to train machine learning models, including algorithms such as random forests, XGBoost, and neural networks, using historical performance data and corresponding evaluation results to enable the models to accurately predict employee performance.

[0980] Performance evaluation

[0981] Using the trained machine learning model, the server evaluates employee performance, using real-time updated work data as input to calculate an employee performance score, taking into account current project progress, daily work performance, and feedback from colleagues and superiors.

[0982] Emotion recognition by emotion engine

[0983] The server uses an emotion engine to recognize the employee's emotions. The emotion engine analyzes emotions from the user's facial expressions, voice tone, and text data. For example, the server recognizes facial expressions using a webcam and analyzes speaking tone from voice input to understand the employee's current emotional state.

[0984] feedback

[0985] The device provides feedback based on the evaluation results and the emotion recognition results of the emotion engine. The server sends the generated evaluation results and the emotion engine's analysis results to the device, which then notifies the employee's smartphone or computer. In addition to a score such as "This month's performance score: 85 / 100," emotion-based advice is also provided. If the emotion engine detects the user's stress level, special advice for stress reduction is also displayed.

[0986] Specific examples

[0987] As an example, consider the case of evaluating employee A, a salesperson.

[0988] 1. The server collects employee A's sales data, number of clients, and number of deals for the past year from the CRM system.

[0989] 2. The server appropriately preprocesses the collected data, imputes missing values, and combines the data.

[0990] 3. The server trains a machine learning model using the preprocessed data and generates a model suitable for evaluating A.

[0991] 4. The server uses the current month's sales data and customer feedback as inputs to calculate A's performance score in real time.

[0992] 5. The server recognizes employee A's emotions from his facial expressions and voice and analyzes them using an emotion engine.

[0993] 6. The device notifies Employee A of the evaluation results and emotion analysis results, and displays this month's score, specific suggestions for improvement, and emotion-based advice.

[0994] This system allows employees' performance to be evaluated objectively and fairly, and provides detailed feedback that takes emotions into consideration, which is expected to increase employee motivation and improve work performance.

[0995] The processing flow will be explained below.

[0996] Step 1:

[0997] The server collects employee work data. Specifically, it obtains task completion status from the project management system and sales data from the CRM system. It also simultaneously collects login and logout time data and the number of successful transactions. This data is obtained from an internal database or via an external API.

[0998] Step 2:

[0999] The server preprocesses the collected data. First, it standardizes the data format and combines data from different sources. For example, it sorts performance data chronologically and converts login times into a single timestamp. Next, it completes missing data and removes invalid data. Finally, it normalizes the data and scales it to make the data range uniform.

[1000] Step 3:

[1001] The server trains the machine learning model. It uses the preprocessed data to build a model using algorithms such as random forests and neural networks. This training requires past performance data and evaluation results, which the model uses to learn from and adjust parameters to predict employee performance.

[1002] Step 4:

[1003] The server uses the trained model to evaluate employee performance. It inputs data such as the progress of ongoing tasks, daily work performance, and feedback from colleagues and superiors into the model in real time to calculate the employee's performance score. This score is continuously updated to reflect the latest work situation.

[1004] Step 5:

[1005] The server uses an emotion engine to recognize employees' emotions. The emotion engine uses a webcam or microphone to analyze emotions from employees' facial expressions, voice tone, and even input text. For example, a facial recognition algorithm can determine whether an employee is smiling or stressed. A voice recognition algorithm can analyze an employee's emotional state from their tone of voice and choice of words.

[1006] Step 6:

[1007] The server integrates the evaluation results with the emotion recognition results from the emotion engine. It then takes into account the user's emotional state and adjusts the feedback accordingly. For example, for an employee who has received a high performance score but is feeling stressed, it can include advice on relaxation or a suggestion for a break.

[1008] Step 7:

[1009] The device notifies the employee of the evaluation results and tailored feedback. The evaluation results are displayed as "This month's performance score: 85 / 100," and advice based on analysis from the emotion engine is also provided. For example, "You achieved a high score, but it seems like you've been feeling stressed recently. I recommend you take a short break."

[1010] Step 8:

[1011] The user (employee) checks the notified evaluation results and feedback. Based on the feedback, they work to improve their own work, setting goals for the next month and improving their performance. At the same time, they also implement the presented advice for stress management to maintain their physical and mental health.

[1012] This processing flow allows employees' performance to be evaluated objectively and fairly, and detailed feedback is provided that takes emotions into consideration, which is expected to increase employee motivation, improve work performance, and maintain their physical and mental health.

[1013] Example 2

[1014] 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."

[1015] Conventional employee performance evaluation systems evaluate employees based solely on performance data and do not take into account the emotional state or mental stress of employees, which means that they are unable to provide appropriate feedback, resulting in insufficient improvement of employee motivation or work performance. Furthermore, because the content of the feedback is uniform, there is also the issue of not being able to provide advice tailored to the individual circumstances and emotions of each employee.

[1016] 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 means for collecting employee work data, means for preprocessing the collected data, means for training a machine learning model based on the preprocessed data, means for evaluating employee performance using the trained machine learning model, emotion recognition means for recognizing the employee's emotional state, and means for feeding back the evaluation result and the emotion recognition result to the employee. This enables detailed feedback that takes into account not only the employee's performance data but also their emotional state, making it possible to provide effective advice for increasing motivation and improving work.

[1017] "Employee business data" refers to all information generated in the course of an employee's work, including, in particular, past performance data, current work status data, and work performance efficiency data.

[1018] "Preprocessing" refers to a series of operations performed to prepare collected data in a format that is easier to analyze, including standardizing data formats, combining data, imputing missing values, and normalizing and scaling.

[1019] A "machine learning model" refers to an algorithm that is trained on collected data and is used for pattern recognition and inference.

[1020] "Evaluating performance" refers to quantitatively evaluating an employee's work performance using a trained machine learning model.

[1021] "Emotion recognition" refers to the process of analyzing emotions from employees' facial expressions, voice tones, and text data, which allows you to understand their current emotional state.

[1022] "Feedback" refers to information and advice provided to employees based on the evaluation results and emotion recognition results, including specific suggestions for improving performance and emotion-based advice.

[1023] The present invention is a system that combines an emotion engine with employee performance evaluation to provide more appropriate feedback that takes into account the employee's emotional state. The system collects employee work data, preprocesses the data, and trains a machine learning model based on the preprocessed data. The trained machine learning model is used to evaluate the employee's performance, and emotion recognition means is used to recognize the employee's emotional state. Based on this, the evaluation results and emotion recognition results are fed back to the employee.

[1024] Hardware and Software

[1025] The system uses the following major hardware and software:

[1026] 1. Server:

[1027] Perform data collection, data preprocessing, machine learning model training, and performance evaluation.

[1028] Usage example:

[1029] MySQL or PostgreSQL as database management systems (DBMS).

[1030] Machine learning libraries include scikit-learn, TensorFlow, and XGBoost.

[1031] 2. Terminal:

[1032] Use employees' smartphones or computers as devices for providing feedback.

[1033] 3. Emotion recognition equipment:

[1034] A microphone for voice input and a webcam for facial recognition.

[1035] Usage example:

[1036] Voice analysis software includes Amazon Polly and Google Cloud Speech-to-Text.

[1037] OpenCV is used as a facial expression recognition library.

[1038] Data processing and calculation

[1039] Here, we will explain the specific data processing and data calculations performed by the system:

[1040] 1. Data Collection:

[1041] The server collects data on employees' past performance, current work status, and work execution efficiency. This data is obtained from internal databases and external APIs, such as from project management tools and CRM systems.

[1042] 2. Data Preprocessing:

[1043] The server preprocesses the collected data, including unifying data formats, combining data, imputing missing values, and normalizing and scaling. For example, it converts CSV files of different formats into a single table-format data set and imputes missing values ​​with the average value.

[1044] 3. Training the machine learning model:

[1045] The server uses the preprocessed data to train machine learning models, using algorithms such as random forests, XGBoost, and neural networks. For example, the model is trained using past performance data and evaluation results.

[1046] 4. Performance Evaluation:

[1047] Using the trained machine learning model, the server evaluates employee performance and calculates a performance score using real-time performance data as input. For example, the score takes into account the progress of current projects and the percentage of tasks completed.

[1048] 5. Emotion recognition:

[1049] The server uses emotion recognition means to recognize the employee's emotions, for example, by capturing the employee's facial expressions with a webcam and analyzing the employee's voice tone to recognize the employee's current emotional state.

[1050] 6. Feedback:

[1051] The server then sends the generated evaluation results and emotion recognition results to the device. The device then notifies the employee's smartphone or computer, providing specific performance evaluations and emotion-based advice. For example, in addition to a score such as "This month's performance score: 85 / 100," feedback such as "You're experiencing increased stress, so we recommend you take a break" is displayed.

[1052] Specific examples

[1053] Here is a specific example of how to evaluate Employee A, a salesperson:

[1054] 1. The server collects employee A's sales data, number of clients, and number of deals for the past year from the CRM system.

[1055] 2. The server preprocesses the collected data, imputes missing values, combines data, and standardizes formats.

[1056] 3. The server trains the machine learning model using the preprocessed data and generates an evaluation model for Employee A.

[1057] 4. The server uses the current month's sales data and customer feedback as inputs to calculate Employee A's performance score in real time.

[1058] 5. The server recognizes employee A's emotions from his facial expressions and voice and analyzes them using an emotion engine.

[1059] 6. The device notifies Employee A of the evaluation results and emotion analysis results, and displays advice such as "This month's performance score: 85 / 100" and "Your stress is increasing, so we recommend you take a break."

[1060] Example prompts for generative AI models

[1061] Below are some example prompts to input to a generative AI model:

[1062] "What kind of feedback should we provide to Employee A in the month when his performance is below average? Please consider specific advice and provide suggestions that also reflect the results of the emotion engine analysis."

[1063] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1064] Step 1:

[1065] The server collects employee work data, specifically employee past performance data, current work status data, and work execution efficiency data, from internal databases and external APIs. The input in this step is data from CRM systems and project management tools, and the output is to compile this data into a single dataset.

[1066] Step 2:

[1067] The server pre-processes the collected data, which includes the following steps:

[1068] 1. Unification of data formats: Convert data of different formats into a common format (input: various data formats, output: unified data format).

[1069] 2. Data Combination: Combining data from different sources into a single dataset (input: individual data items, output: cohesive dataset).

[1070] 3. Missing value imputation: impute missing data in an appropriate way (input: partially missing data, output: complete dataset with imputed missing values).

[1071] 4. Normalization and scaling: Reduce the variability of the data and convert it into a format suitable for machine learning models (input: unnormalized data, output: normalized and scaled data).

[1072] Step 3:

[1073] The server uses the preprocessed data to train a machine learning model, specifically by following these steps:

[1074] 1. Data Split: Split the data into training and test sets (input: preprocessed data, output: training and test sets).

[1075] 2. Model selection: Choose an algorithm suitable for the task (input: candidate algorithms, output: selected algorithm).

[1076] 3. Training: The selected algorithm is trained on the data (input: training set, output: trained model).

[1077] 4. Hyperparameter tuning: Adjusting parameters to improve model performance (input: model and candidate parameters, output: optimized model).

[1078] Step 4:

[1079] The server evaluates employee performance using a trained machine learning model, including:

[1080] 1. Data input: Input the current month's performance data and customer feedback into the model (input: real-time performance data, output: predicted score).

[1081] 2. Score calculation: The model calculates the employee's performance score based on the input data (input: current performance data, output: performance score).

[1082] Step 5:

[1083] The server uses the emotion recognition means to recognize the employee's emotion, specifically by the following steps:

[1084] 1. Voice input analysis: Record the employee's voice and analyze the voice data (input: voice data, output: voice analysis results).

[1085] 2. Facial expression recognition: Capture employees' facial expressions using a webcam and analyze their emotions (input: facial expression data, output: facial expression analysis results).

[1086] 3. Text analysis: Analyzes text data entered by employees and infers their emotions (input: text data, output: text analysis results).

[1087] Step 6:

[1088] The server sends the generated evaluation results and emotion recognition results to the terminal, which then notifies the employee's smartphone or computer. Specifically, the process is as follows:

[1089] 1. Displaying evaluation results: Notifying employees of their performance scores and specific performance evaluations (Input: Evaluation results, Output: Notification message).

[1090] 2. Emotion-based advice: Provides emotion-based advice based on emotion recognition results (input: emotion analysis results, output: advice message).

[1091] (Application example 2)

[1092] 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."

[1093] The problem is that performance evaluation and feedback of employees cannot take into account the emotional state of employees, making it difficult to properly manage the impact of the evaluation results on employees' psychology. In addition, it is not possible to provide detailed advice based on the employee's emotional state, making it difficult to simultaneously improve performance and employee satisfaction.

[1094] The specific processing by the specific 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 employee work data, means for preprocessing the collected data, means for training a machine learning model based on the preprocessed data, means for evaluating employee performance using the trained machine learning model, an emotion engine including emotion analysis means for analyzing the emotional state of the employee at the time of performance evaluation, and means for feeding back the evaluation results and emotion analysis results to the employee. As a result, employee performance evaluations are performed taking into account the emotional state of the employee, the impact of the evaluation results on employee psychology is appropriately managed, and it is possible to achieve both improved performance and improved employee satisfaction.

[1095] "Employee business data" refers to various data generated when employees perform their work, including past performance data, current work status data, and work performance efficiency data.

[1096] "Preprocessing" refers to data processing techniques to prepare collected data for use as input data for machine learning models, and includes standardizing data formats, combining data, filling in missing values, and normalizing and scaling.

[1097] A "machine learning model" is an algorithm that learns patterns and features based on large amounts of data and makes predictions and classifications, and includes random forests, XGBoost, neural networks, and others.

[1098] "Performance appraisal" is a method of using a trained machine learning model to evaluate an employee's performance and efficiency in carrying out their work as a number or score.

[1099] "Emotion analysis means" refers to a method for analyzing emotions from an employee's facial expressions, voice tone, text data, etc., and includes an emotion engine using a camera or microphone.

[1100] The "emotion engine" is a system that performs facial expression recognition and voice analysis to grasp the user's emotional state.

[1101] "Feedback" refers to the act of notifying employees of the results of their performance evaluation and sentiment analysis, and providing them with advice based on this.

[1102] "Real-time feedback" is a feature that provides immediate advice based on evaluation results and emotional state at the moment an employee is working.

[1103] "Smart glasses" are wearable devices that are equipped with a camera, display, and voice recognition functions and are worn by the user.

[1104] The employee performance evaluation system of this invention objectively evaluates employee performance by collecting and preprocessing employee work data and training a machine learning model. It also uses emotion analysis means to understand the employee's emotional state and provides feedback based on the evaluation results and advice based on their emotions, thereby achieving both employee satisfaction and improved performance.

[1105] This system uses the following hardware and software.

[1106] Hardware and software:

[1107] 1. Server:

[1108] Data collection:

[1109] Employee past performance data, current work status data, and work execution efficiency data are collected from various internal databases and external APIs.

[1110] Data preprocessing:

[1111] Preprocessing includes standardizing data formats, combining data, imputing missing values, and normalizing and scaling to convert the collected data into a consistent form.

[1112] Training the machine learning model:

[1113] Train a machine learning model (e.g., Random Forest, XGBoost, or Neural Network) using the preprocessed data.

[1114] 2. Smart Glasses:

[1115] Emotion analysis:

[1116] The smart glasses use a camera and microphone to analyze the facial expressions and tone of voice of customers and employees in real time, and estimate their emotional state using an emotion analysis model (such as CNN).

[1117] Real-time feedback:

[1118] The smart glasses display real-time feedback from superiors and colleagues, advice based on performance data, and advice based on emotions.

[1119] How to use:

[1120] 1. Data Collection and Preprocessing:

[1121] The server automatically collects and preprocesses employee work data, including standardizing data formats and filling in missing values.

[1122] 2. Training the machine learning model:

[1123] The preprocessed data is used to train machine learning models, using algorithms such as random forests and neural networks.

[1124] 3. Performance Evaluation:

[1125] A trained machine learning model evaluates employee performance using current work situation data as input.

[1126] 4. Emotion analysis:

[1127] The smart glasses are equipped with a camera and microphone that analyze the facial expressions and voices of employees and customers in real time, estimating their emotional state and sending it to a server.

[1128] 5. Providing Feedback:

[1129] Using the evaluation results and sentiment analysis, feedback is displayed on the smart glasses screen to the employee, including advice based on performance evaluation and customized advice based on emotions.

[1130] Examples:

[1131] For example, when an employee wears smart glasses while serving customers in a brick-and-mortar store, the server collects and preprocesses the employee's work data to train an evaluation model. The smart glasses analyze the customer's facial expressions and tone of voice using emotion analysis tools, and display the customer's emotional state to the employee in real time. This allows the employee to understand the customer's emotional state and provide appropriate service, further improving customer satisfaction.

[1132] Example prompt sentence:

[1133] "Please propose a system that analyzes the emotions of customers from their facial expressions and tone of voice, and provides appropriate customer service advice based on that."

[1134] This invention allows employees' performance to be evaluated fairly and objectively, and provides feedback that takes emotions into account, making it possible to achieve both improved performance and increased employee satisfaction.

[1135] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1136] Step 1:

[1137] Data collection

[1138] The server collects data on employees' past performance, current work status, and work efficiency from various internal databases and external APIs. The input is employee work-related data, and the output is business data converted into a unified data format.

[1139] Step 2:

[1140] Data Preprocessing

[1141] The server preprocesses the collected business data by standardizing data formats, combining data, imputing missing values, normalizing and scaling it. The input is the collected business data, and the output is a preprocessed, consistent dataset.

[1142] Step 3:

[1143] Training a machine learning model

[1144] The server uses the preprocessed data to train a machine learning model (e.g., Random Forest, XGBoost, Neural Network, etc.) The input is the preprocessed dataset and the output is the trained machine learning model.

[1145] Step 4:

[1146] Performance evaluation

[1147] The server evaluates employee performance using the trained machine learning model, using current work situation data as input and the employee performance score as output. The input is the current work situation data and the output is the evaluated performance score.

[1148] Step 5:

[1149] Emotion analysis

[1150] The smart glasses use a camera and microphone to analyze the facial expressions and tone of voice of customers and employees in real time. The input is camera video and audio data, and the output is the emotion analysis results, which are sent to a server.

[1151] Step 6:

[1152] Real-time feedback

[1153] The server sends the evaluation results and emotion analysis results to the smart glasses, which then provide feedback to the employee. The smart glasses display shows feedback from superiors and colleagues, advice based on performance data, and advice based on emotions. The input is the evaluation results and emotion analysis results, and the output is real-time feedback display.

[1154] Step 7:

[1155] Providing Feedback

[1156] Based on the evaluation results and sentiment analysis, more specific advice is displayed on the smart glasses, allowing the user to adjust their interactions with the customer. The input is feedback information, and the output is specific advice provided to employees.

[1157] 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.

[1158] 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.

[1159] 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.

[1160] [Fourth embodiment]

[1161] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1162] 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.

[1163] 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).

[1164] 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.

[1165] 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.

[1166] 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).

[1167] 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.

[1168] 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.

[1169] 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.

[1170] 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.

[1171] 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.

[1172] 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.

[1173] 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."

[1174] The present invention is a system for objectively evaluating employee performance using AI technology, and is implemented through the following steps.

[1175] Data collection

[1176] The server automatically collects employee work data, including past performance data, current work status data, and work execution efficiency data, such as completed project tasks, achieved goals, login times, and task completion times. This data is obtained from various internal databases and external APIs.

[1177] Data Preprocessing

[1178] The collected data is pre-processed by the server, which includes the following operations:

[1179] 1. Unifying data formats: Converting data from different sources into a common format.

[1180] 2. Data Combination: Combining information from different data sources into a single data set.

[1181] 3. Missing Value Imputation: Properly impute missing data to create a consistent dataset.

[1182] 4. Normalization and scaling: Reduce the variability in the data and convert it into a format suitable for machine learning models.

[1183] Training the model

[1184] The server uses the preprocessed data to train machine learning models, including algorithms such as random forests, XGBoost, and neural networks, using historical performance data and corresponding evaluation results to enable the models to accurately predict employee performance.

[1185] Performance evaluation

[1186] Using the trained machine learning model, the server evaluates employee performance, using real-time updated work data as input to calculate an employee performance score, taking into account current project progress, daily work performance, and feedback from colleagues and superiors.

[1187] feedback

[1188] The evaluation results are fed back to the employee via a device. For example, an employee may receive a notification on their smartphone or computer saying, "This month's performance score is 80 points." In addition, specific advice on how to improve performance based on the evaluation results and information useful for setting goals for the next time are also provided.

[1189] Specific examples

[1190] As an example, consider the case of evaluating employee A, a salesperson.

[1191] 1. The server collects employee A's sales data, number of clients, and number of deals for the past year from the CRM system.

[1192] 2. The server appropriately preprocesses the collected data, imputes missing values, and combines the data.

[1193] 3. The server trains a machine learning model using the preprocessed data and generates a model suitable for evaluating A.

[1194] 4. The server uses the current month's sales data and customer feedback as inputs to calculate A's performance score in real time.

[1195] 5. The device notifies Employee A of the evaluation results and displays this month's score and specific suggestions for improvement.

[1196] The above is the basic form for carrying out the invention. This system allows employee performance evaluation to be carried out objectively and fairly, which is expected to improve employee motivation and the overall productivity of the company.

[1197] The processing flow will be explained below.

[1198] Step 1:

[1199] The server collects employee work data. Specifically, it obtains data on past performance, current work status, and work execution efficiency from databases and APIs. For example, it collects task completion status from a project management system, sales data from a CRM system, and login time data from a time attendance management system.

[1200] Step 2:

[1201] The server preprocesses the collected data. First, it standardizes the data format and converts data obtained from different systems into a common format. Next, it links the data and integrates data from different sources into a single dataset. If there is incomplete data or missing values, it fills them in to create a consistent dataset. It also performs normalization and scaling to reduce data variability.

[1202] Step 3:

[1203] The server trains the machine learning model. It uses the preprocessed data to build a model using algorithms such as random forest, XGBoost, and neural networks. The training dataset includes past performance data and evaluation results. Based on this data, the model adjusts parameters to predict employee performance.

[1204] Step 4:

[1205] The server uses a trained machine learning model to evaluate employee performance. It inputs data on the progress of ongoing tasks, daily work performance data, and feedback from colleagues and superiors into the model in real time to calculate an employee performance score. This score is continuously updated to reflect the latest work situation.

[1206] Step 5:

[1207] The device then provides feedback on the evaluation results to the employee. The server then sends the generated evaluation results and feedback to the device, which then notifies the employee on their smartphone or computer. For example, the device displays a score such as "This month's performance score: 85 / 100," as well as specific areas for improvement and advice on setting goals for the next time.

[1208] Step 6:

[1209] Users (employees) check the evaluation results and feedback and use them to improve their work. Based on the evaluation results, they clarify goals and areas for improvement for the next month and create plans to improve their own work performance.

[1210] This processing flow ensures that employee performance evaluations are conducted objectively and fairly, leading to increased employee motivation and improved corporate productivity.

[1211] Example 1

[1212] 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."

[1213] In today's workplaces, employee performance evaluations often contain many subjective elements, making it difficult to provide fair and objective evaluations. Furthermore, the process from data collection to evaluation is cumbersome, making it difficult to carry out the evaluation efficiently. Furthermore, the lack of specific feedback or advice based on the evaluation results means that employees are not provided with information that will help them improve their performance or set goals.

[1214] 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.

[1215] In this invention, the server includes means for collecting employee work data, means for preprocessing the collected data, means for training a machine learning model based on the preprocessed data, means for evaluating employee performance using the trained machine learning model, and means for providing feedback on the evaluation results to the employee, thereby enabling objective and fair evaluation of employee performance and providing efficient data processing and specific feedback.

[1216] "Employee business data" is a general term for various data generated in the course of an employee's work, and specifically includes project completion tasks, goals achieved, login times, and the time required to complete tasks.

[1217] "Means of collection" refers to a mechanism for automatically or manually obtaining necessary business data from a server or database, including database queries and API requests.

[1218] "Preprocessing measures" are a series of operations to convert collected data into a form suitable for analysis and model training, including data format unification, data merging, missing value imputation, normalization, and scaling.

[1219] "Means for training a machine learning model" means the process of using pre-processed data to build a machine learning model based on a specific algorithm and then optimizing the model, including, for example, the use of random forests or neural networks.

[1220] The "means for evaluating employee performance" refers to a system that utilizes a trained machine learning model to calculate employee performance scores based on collected work data, and includes real-time data entry and score calculation using the model.

[1221] "Means of providing feedback" refers to methods for notifying employees of the evaluation results and providing information such as specific suggestions for improvement and assistance with setting goals for the next time, and includes notification APIs and notification functions on devices.

[1222] Unifying "formats" refers to the process of converting data obtained from different sources into a common format, including, for example, unifying date formats and numerical units.

[1223] "Scaling" is the process of adjusting for variability in data and transforming it so that machine learning algorithms can function effectively, and includes techniques such as standardization and normalization.

[1224] A "machine learning model" is an algorithm that learns specific patterns and features from data and uses them to make predictions and classifications from future data, and includes random forests, XGBoost, neural networks, and more.

[1225] A "performance score" is an indicator used by machine learning models to quantify or evaluate employee performance, and serves as the basis for performance evaluation and feedback.

[1226] The present invention is a system for objectively and fairly evaluating employee performance, which includes a number of processing steps. The system includes a server, a terminal, and a user.

[1227] Server Roles

[1228] Data collection

[1229] The server is responsible for automatically collecting employee business data, including information such as:

[1230] Completed tasks in a project

[1231] Goals achieved

[1232] Login Time

[1233] Time required to complete the task

[1234] The server retrieves this data from internal databases or external APIs, accessing the data using SQL queries or RESTful APIs.

[1235] Data Preprocessing

[1236] The collected data is pre-processed by the server in the following ways:

[1237] 1. Data format unification: Converting data from different sources into a common format. For example, converting different date formats into the standard ISO 8601 format.

[1238] 2. Data Combining: Combining information from different data sources into a single dataset. A specific example is "Merging Data Frames Using the pandas Library."

[1239] 3. Missing value imputation: Properly impute missing data to create a consistent dataset. For example, imputing missing values ​​with the mean or median, or imputing using k-nearest neighbors, are some of the methods.

[1240] 4. Normalization and scaling: Transform the data into a format suitable for machine learning models. Specifically, "Standardize the data using scikit-learn's StandardScaler."

[1241] Training the model

[1242] During the training process, the server uses the preprocessed data to train machine learning models. The algorithms used in this system include Random Forest, XGBoost, and neural networks. Training is performed using past performance data and corresponding evaluation results.

[1243] The specific steps are:

[1244] We build a model using scikit-learn's RandomForestClassifier and perform 10-fold cross-validation.

[1245] Use GridSearchCV to tune hyperparameters and select the best model.

[1246] Performance evaluation

[1247] Using the trained machine learning model, the server evaluates employee performance in real time using the following data:

[1248] Current month's sales data

[1249] Customer Feedback

[1250] Daily business performance

[1251] A specific example is the action "predict a performance score using the predict method."

[1252] Device Role

[1253] feedback

[1254] The terminal is responsible for notifying the user of the evaluation results. Notification is done via the employee's smartphone or computer. Specifically, this involves the process of calling a notification API and sending the evaluation score.

[1255] Based on the evaluation results, the system will also display specific suggestions for improvement and information that will help employees set their next goal. For example, for an employee who received a score of 80, the system will display advice such as "Set your goal for next month at 90 and take an online course to improve specific skills."

[1256] Specific examples

[1257] For example, to evaluate Employee A in sales:

[1258] 1. The server collects Employee A's sales data, number of clients, and number of deals for the past year from the CRM system. Specifically, it executes the SQL query "SELECT FROM SalesData WHERE EmployeeID = 'A'".

[1259] 2. The server reads the collected data as a data frame using the pandas library, imputes missing values, and combines the data.

[1260] 3. The server uses the preprocessed data to train a model using scikit-learn's RandomForestClassifier and optimizes the model through 10-fold cross-validation.

[1261] 4. The server uses the model to calculate a performance score based on the current month's sales data and customer feedback.

[1262] 5. The device notifies Employee A of the evaluation results and displays the score and specific improvement suggestions.

[1263] Prompt Sentence Examples

[1264] Example prompts to input to the generative AI model:

[1265] "Evaluate Employee A's performance based on sales data, number of clients, and number of deals closed over the past year. Also consider the current month's sales data and customer feedback."

[1266] The above is the basic form for carrying out the present invention. This system allows employee performance evaluations to be carried out objectively and fairly, which is expected to improve employee motivation and corporate productivity.

[1267] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1268] Step 1: Data collection

[1269] The server automatically collects employee work data. Specifically, the server periodically retrieves data from a database or external API. For example, the server uses SQL queries to retrieve "project completed tasks," "goals achieved," "login time," and "time required to complete tasks" from the database. The server also calls RESTful APIs to retrieve customer feedback and industry-standard data from external services. This data serves as input data for performance evaluation. The collected work data is generated as output.

[1270] Step 2: Data Preprocessing

[1271] The server pre-processes the collected data in the following ways:

[1272] 1. Unifying data formats: If input data has different formats, the server converts them into a common format. For example, it converts all different date formats into ISO 8601 format.

[1273] 2. Data Combining: The server combines the information obtained from different data sources into a single dataset. Here, we use the pandas library to perform operations to merge multiple data frames.

[1274] 3. Missing Value Imputation: The server imputes missing data using the mean, median, or k-nearest neighbors.

[1275] 4. Normalization and scaling: The server scales the collected data to reduce variability and make it easier for machine learning algorithms to work. For example, it standardizes the data using scikit-learn's StandardScaler.

[1276] This outputs a preprocessed, consistent dataset.

[1277] Step 3: Train the machine learning model

[1278] The server trains a machine learning model based on the preprocessed data. It uses the preprocessed performance data and the corresponding evaluation results as input data. The server executes the following specific algorithms:

[1279] Random Forest: We build a model using scikit-learn's RandomForestClassifier. Specifically, we create multiple decision trees using the training data and combine them to generate the final model.

[1280] Cross-validation: The server performs 10-fold cross-validation during the training process to evaluate the accuracy of the model, and also uses GridSearchCV to optimize the hyperparameters.

[1281] This outputs a trained machine learning model.

[1282] Step 4: Performance evaluation

[1283] The server evaluates employee performance using a trained machine learning model. As input data, it uses current job data and the individual performance data of the employee being evaluated. The server performs the following specific tasks:

[1284] Evaluation Data Entry: Collect current month sales data, customer feedback, and daily operational performance data.

[1285] Input data into the model: The server inputs these data into the model and uses the predict method to calculate a performance score.

[1286] This outputs a real-time performance evaluation score for each employee.

[1287] Step 5: Feedback

[1288] The terminal feeds back the evaluation results from the server to the employee. It uses the performance evaluation score and related feedback information as input data. Specifically, it performs the following processes:

[1289] Notification sending: The device calls the notification API and sends the evaluation results to the employee's smartphone or computer. For example, an employee with an evaluation score of 80 will be notified that "Your performance score this month is 80 points."

[1290] Providing improvement suggestions: Based on the evaluation results, the device will display specific improvement suggestions and support for setting goals for the next time. For example, it may suggest, "Set your goal for next month at 90 points and take an online course to improve a specific skill."

[1291] This allows employees to receive performance evaluations and specific feedback.

[1292] Prompt Sentence Examples

[1293] Example prompts to input to the generative AI model:

[1294] "Evaluate Employee A's performance based on sales data, number of clients, and number of deals closed over the past year. Also consider the current month's sales data and customer feedback."

[1295] (Application example 1)

[1296] 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."

[1297] Employee performance evaluations tend to be subjective and often lack transparency and fairness. Furthermore, the lack of a method for properly evaluating the operating status of work equipment makes efficient management difficult. This has led to a growing need for a system that comprehensively evaluates the performance of employees and work equipment and provides feedback to managers.

[1298] 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.

[1299] In this invention, the server includes means for collecting employee work data, means for preprocessing the collected data, means for training a machine learning model based on the preprocessed data, means for evaluating employee performance using the trained machine learning model, means for providing feedback on the evaluation results to the employee, means for collecting labor equipment operation data, means for preprocessing the collected labor equipment operation data, means for training a machine learning model based on the preprocessed labor equipment operation data, means for evaluating the labor equipment performance using the trained machine learning model, and means for providing feedback on the evaluation results to a manager. This makes it possible to objectively and fairly evaluate the performance of employees and labor equipment and provide appropriate feedback to the manager.

[1300] An "employee" is an individual employed by an organization or company to perform work.

[1301] "Business data" refers to various information generated by employees in the course of performing their work, including completed project tasks, goals achieved, login times, and the time required to complete tasks.

[1302] "Labor equipment" refers to equipment used in factories and businesses that performs physical work, such as machinery and robots.

[1303] "Operation data" refers to information generated when labor equipment is operating, and specifically includes operating time, number of completed tasks, maintenance logs, error occurrence status, etc.

[1304] "Preprocessing" refers to the process of converting collected data into a format suitable for analysis and machine learning, and includes unifying data formats, combining data, imputing missing values, and normalizing and scaling data.

[1305] A "machine learning model" is an algorithm that is trained using collected and preprocessed data to identify specific patterns and relationships and use them to make predictions and evaluations.

[1306] "Training" refers to the process of training an algorithm with data to build a machine learning model.

[1307] "Feedback" refers to the act of notifying employees and managers of evaluation results and providing advice on how to improve and enhance performance based on those results.

[1308] "Evaluation" refers to calculating the performance and accomplishments of employees and work equipment based on collected data and trained machine learning models, and judging their performance.

[1309] "Supervisor" refers to the individual responsible for the operation and supervision of employees and work equipment, and who receives evaluations and feedback.

[1310] 1. Data Collection

[1311] The server automatically collects work data from employees and work equipment. Collected data includes the number of tasks completed by employees, goals achieved, login times, and the time required to complete tasks. It also includes operational data for work equipment, such as operating hours, number of completed tasks, maintenance logs, and error occurrence status. This data is obtained from various internal databases and external APIs.

[1312] 2. Data Preprocessing

[1313] The server pre-processes the collected data, which includes the following operations:

[1314] Data format unification: Converting data from different sources into a common format.

[1315] Data Combination: Combining information from different data sources into a single data set.

[1316] Missing Value Imputation: Properly impute missing data to create a consistent dataset.

[1317] Normalization and scaling: Reduce the variability in the data and convert it into a format suitable for machine learning models.

[1318] 3. Training the model

[1319] The server uses the preprocessed data to train machine learning models, including algorithms such as random forests, XGBoost, and neural networks. Training uses historical performance data and corresponding evaluation results to enable the models to accurately predict employee and equipment performance.

[1320] 4. Performance Evaluation

[1321] Using the trained machine learning model, the server evaluates the performance of employees and work equipment. This evaluation uses real-time updated work data as input and calculates a performance score for employees and work equipment, taking into account, for example, the progress of current projects, daily work performance, feedback from colleagues and superiors, and the operation status of work equipment.

[1322] 5. Feedback

[1323] The evaluation results are fed back to employees and managers via terminals. Employees are notified on their smartphones or computers with a message such as, "This month's performance score is 80 points." Managers are provided with performance evaluations and improvement suggestions for work equipment. For example, the message may read, "Robot X's operating efficiency this month is 90%."

[1324] For example, let's assume that you are evaluating the performance of a welding robot, an example of a robot used in a factory. Past operation data (operating time, number of completed tasks, error log) for robot ID "robot1" is collected and a model is trained. Based on the training results, the current operation data is evaluated and feedback is sent to the manager. An example of a specific prompt sentence would be to evaluate performance based on the operation data "Robot ID: robot1, operating time: 120 hours, number of completed tasks: 200, error log: 5". The performance score is calculated as 85%, and feedback is sent to the manager.

[1325] The above is the basic form for carrying out the invention. This system allows for objective and fair evaluation of employee and work equipment performance, which is expected to improve employee motivation and overall company productivity.

[1326] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1327] Step 1:

[1328] The server collects work data from employees and work equipment. Specifically, it obtains the number of tasks completed by employees, the goals achieved, login time, and the time required to complete a task using an internal database or external API. Similarly, it collects the operating time of work equipment, the number of tasks completed, maintenance logs, and error occurrence status. This allows input data to be acquired.

[1329] Input: Employee work data and work equipment operation data

[1330] Output: Collected business and operational data

[1331] Step 2:

[1332] The server preprocesses the collected data. First, it converts the data formats obtained from different sources into a common format. Next, it integrates the information obtained from the different data sources into a single dataset and imputes missing values ​​appropriately. Finally, it normalizes and scales the data to reduce variability. This prepares the data in a format suitable for machine learning models.

[1333] Input: Collected business and operational data

[1334] Output: Preprocessed dataset

[1335] Step 3:

[1336] The server uses the preprocessed data to train machine learning models, specifically algorithms such as random forests, XGBoost, and neural networks, to accurately predict employee and equipment performance using past performance data and corresponding evaluation results.

[1337] Input: Preprocessed dataset

[1338] Output: A trained machine learning model

[1339] Step 4:

[1340] The server uses trained machine learning models to evaluate the performance of employees and work equipment. It inputs current work and operational data and calculates a performance score based on the model. For example, it takes into account the employee's current project progress, daily work performance, feedback from colleagues and superiors, and the operation status of work equipment.

[1341] Input: Current business and operational data

[1342] Output: Performance score

[1343] Step 5:

[1344] The terminal provides feedback on the evaluation results to employees and managers. Employees are notified via smartphone or computer in the form of, for example, "This month's performance score is 80 points." Managers are provided with the performance evaluation results of work equipment and suggestions for improvement. For example, the display might say, "Robot X's operating efficiency this month is 90%."

[1345] Input: Performance Score

[1346] Output: Feedback notification (provided to employee and manager)

[1347] These are the processing steps of this system. Through this process, it is possible to objectively and fairly evaluate the performance of employees and work equipment and provide appropriate feedback.

[1348] 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.

[1349] The present invention is a system that combines an emotion engine with employee performance evaluation to take into account the employee's emotional state and provide more appropriate feedback. This is implemented through the following steps:

[1350] Data collection

[1351] The server automatically collects employee work data, including past performance data, current work status data, and work execution efficiency data, such as completed project tasks, achieved goals, login times, and task completion times. This data is obtained from various internal databases and external APIs.

[1352] Data Preprocessing

[1353] The collected data is pre-processed by the server, which includes the following operations:

[1354] 1. Unifying data formats: Converting data from different sources into a common format.

[1355] 2. Data Combination: Combining information from different data sources into a single data set.

[1356] 3. Missing Value Imputation: Properly impute missing data to create a consistent dataset.

[1357] 4. Normalization and scaling: Reduce the variability in the data and convert it into a format suitable for machine learning models.

[1358] Training the model

[1359] The server uses the preprocessed data to train machine learning models, including algorithms such as random forests, XGBoost, and neural networks, using historical performance data and corresponding evaluation results to enable the models to accurately predict employee performance.

[1360] Performance evaluation

[1361] Using the trained machine learning model, the server evaluates employee performance, using real-time updated work data as input to calculate an employee performance score, taking into account current project progress, daily work performance, and feedback from colleagues and superiors.

[1362] Emotion recognition by emotion engine

[1363] The server uses an emotion engine to recognize the employee's emotions. The emotion engine analyzes emotions from the user's facial expressions, voice tone, and text data. For example, the server recognizes facial expressions using a webcam and analyzes speaking tone from voice input to understand the employee's current emotional state.

[1364] feedback

[1365] The device provides feedback based on the evaluation results and the emotion recognition results of the emotion engine. The server sends the generated evaluation results and the emotion engine's analysis results to the device, which then notifies the employee's smartphone or computer. In addition to a score such as "This month's performance score: 85 / 100," emotion-based advice is also provided. If the emotion engine detects the user's stress level, special advice for stress reduction is also displayed.

[1366] Specific examples

[1367] As an example, consider the case of evaluating employee A, a salesperson.

[1368] 1. The server collects employee A's sales data, number of clients, and number of deals for the past year from the CRM system.

[1369] 2. The server appropriately preprocesses the collected data, imputes missing values, and combines the data.

[1370] 3. The server trains a machine learning model using the preprocessed data and generates a model suitable for evaluating A.

[1371] 4. The server uses the current month's sales data and customer feedback as inputs to calculate A's performance score in real time.

[1372] 5. The server recognizes employee A's emotions from his facial expressions and voice and analyzes them using an emotion engine.

[1373] 6. The device notifies Employee A of the evaluation results and emotion analysis results, and displays this month's score, specific suggestions for improvement, and emotion-based advice.

[1374] This system allows employees' performance to be evaluated objectively and fairly, and provides detailed feedback that takes emotions into consideration, which is expected to increase employee motivation and improve work performance.

[1375] The processing flow will be explained below.

[1376] Step 1:

[1377] The server collects employee work data. Specifically, it obtains task completion status from the project management system and sales data from the CRM system. It also simultaneously collects login and logout time data and the number of successful transactions. This data is obtained from an internal database or via an external API.

[1378] Step 2:

[1379] The server preprocesses the collected data. First, it standardizes the data format and combines data from different sources. For example, it sorts performance data chronologically and converts login times into a single timestamp. Next, it completes missing data and removes invalid data. Finally, it normalizes the data and scales it to make the data range uniform.

[1380] Step 3:

[1381] The server trains the machine learning model. It uses the preprocessed data to build a model using algorithms such as random forests and neural networks. This training requires past performance data and evaluation results, which the model uses to learn from and adjust parameters to predict employee performance.

[1382] Step 4:

[1383] The server uses the trained model to evaluate employee performance. It inputs data such as the progress of ongoing tasks, daily work performance, and feedback from colleagues and superiors into the model in real time to calculate the employee's performance score. This score is continuously updated to reflect the latest work situation.

[1384] Step 5:

[1385] The server uses an emotion engine to recognize employees' emotions. The emotion engine uses a webcam or microphone to analyze emotions from employees' facial expressions, voice tone, and even input text. For example, a facial recognition algorithm can determine whether an employee is smiling or stressed. A voice recognition algorithm can analyze an employee's emotional state from their tone of voice and choice of words.

[1386] Step 6:

[1387] The server integrates the evaluation results with the emotion recognition results from the emotion engine. It then takes into account the user's emotional state and adjusts the feedback accordingly. For example, for an employee who has received a high performance score but is feeling stressed, it can include advice on relaxation or a suggestion for a break.

[1388] Step 7:

[1389] The device notifies the employee of the evaluation results and tailored feedback. The evaluation results are displayed as "This month's performance score: 85 / 100," and advice based on analysis from the emotion engine is also provided. For example, "You achieved a high score, but it seems like you've been feeling stressed recently. I recommend you take a short break."

[1390] Step 8:

[1391] The user (employee) checks the notified evaluation results and feedback. Based on the feedback, they work to improve their own work, setting goals for the next month and improving their performance. At the same time, they also implement the presented advice for stress management to maintain their physical and mental health.

[1392] This processing flow allows employees' performance to be evaluated objectively and fairly, and detailed feedback is provided that takes emotions into consideration, which is expected to increase employee motivation, improve work performance, and maintain their physical and mental health.

[1393] Example 2

[1394] 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."

[1395] Conventional employee performance evaluation systems evaluate employees based solely on performance data and do not take into account the emotional state or mental stress of employees, which means that they are unable to provide appropriate feedback, resulting in insufficient improvement of employee motivation or work performance. Furthermore, because the content of the feedback is uniform, there is also the issue of not being able to provide advice tailored to the individual circumstances and emotions of each employee.

[1396] 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 means for collecting employee work data, means for preprocessing the collected data, means for training a machine learning model based on the preprocessed data, means for evaluating employee performance using the trained machine learning model, emotion recognition means for recognizing the employee's emotional state, and means for feeding back the evaluation result and the emotion recognition result to the employee. This enables detailed feedback that takes into account not only the employee's performance data but also their emotional state, making it possible to provide effective advice for increasing motivation and improving work.

[1397] "Employee business data" refers to all information generated in the course of an employee's work, including, in particular, past performance data, current work status data, and work performance efficiency data.

[1398] "Preprocessing" refers to a series of operations performed to prepare collected data in a format that is easier to analyze, including standardizing data formats, combining data, imputing missing values, and normalizing and scaling.

[1399] A "machine learning model" refers to an algorithm that is trained on collected data and is used for pattern recognition and inference.

[1400] "Evaluating performance" refers to quantitatively evaluating an employee's work performance using a trained machine learning model.

[1401] "Emotion recognition" refers to the process of analyzing emotions from employees' facial expressions, voice tones, and text data, which allows you to understand their current emotional state.

[1402] "Feedback" refers to information and advice provided to employees based on the evaluation results and emotion recognition results, including specific suggestions for improving performance and emotion-based advice.

[1403] The present invention is a system that combines an emotion engine with employee performance evaluation to provide more appropriate feedback that takes into account the employee's emotional state. The system collects employee work data, preprocesses the data, and trains a machine learning model based on the preprocessed data. The trained machine learning model is used to evaluate the employee's performance, and emotion recognition means is used to recognize the employee's emotional state. Based on this, the evaluation results and emotion recognition results are fed back to the employee.

[1404] Hardware and Software

[1405] The system uses the following major hardware and software:

[1406] 1. Server:

[1407] Perform data collection, data preprocessing, machine learning model training, and performance evaluation.

[1408] Usage example:

[1409] MySQL or PostgreSQL as database management systems (DBMS).

[1410] Machine learning libraries include scikit-learn, TensorFlow, and XGBoost.

[1411] 2. Terminal:

[1412] Use employees' smartphones or computers as devices for providing feedback.

[1413] 3. Emotion recognition equipment:

[1414] A microphone for voice input and a webcam for facial recognition.

[1415] Usage example:

[1416] Voice analysis software includes Amazon Polly and Google Cloud Speech-to-Text.

[1417] OpenCV is used as a facial expression recognition library.

[1418] Data processing and calculation

[1419] Here, we will explain the specific data processing and data calculations performed by the system:

[1420] 1. Data Collection:

[1421] The server collects data on employees' past performance, current work status, and work execution efficiency. This data is obtained from internal databases and external APIs, such as from project management tools and CRM systems.

[1422] 2. Data Preprocessing:

[1423] The server preprocesses the collected data, including unifying data formats, combining data, imputing missing values, and normalizing and scaling. For example, it converts CSV files of different formats into a single table-format data set and imputes missing values ​​with the average value.

[1424] 3. Training the machine learning model:

[1425] The server uses the preprocessed data to train machine learning models, using algorithms such as random forests, XGBoost, and neural networks. For example, the model is trained using past performance data and evaluation results.

[1426] 4. Performance Evaluation:

[1427] Using the trained machine learning model, the server evaluates employee performance and calculates a performance score using real-time performance data as input. For example, the score takes into account the progress of current projects and the percentage of tasks completed.

[1428] 5. Emotion recognition:

[1429] The server uses emotion recognition means to recognize the employee's emotions, for example, by capturing the employee's facial expressions with a webcam and analyzing the employee's voice tone to recognize the employee's current emotional state.

[1430] 6. Feedback:

[1431] The server then sends the generated evaluation results and emotion recognition results to the device. The device then notifies the employee's smartphone or computer, providing specific performance evaluations and emotion-based advice. For example, in addition to a score such as "This month's performance score: 85 / 100," feedback such as "You're experiencing increased stress, so we recommend you take a break" is displayed.

[1432] Specific examples

[1433] Here is a specific example of how to evaluate Employee A, a salesperson:

[1434] 1. The server collects employee A's sales data, number of clients, and number of deals for the past year from the CRM system.

[1435] 2. The server preprocesses the collected data, imputes missing values, combines data, and standardizes formats.

[1436] 3. The server trains the machine learning model using the preprocessed data and generates an evaluation model for Employee A.

[1437] 4. The server uses the current month's sales data and customer feedback as inputs to calculate Employee A's performance score in real time.

[1438] 5. The server recognizes employee A's emotions from his facial expressions and voice and analyzes them using an emotion engine.

[1439] 6. The device notifies Employee A of the evaluation results and emotion analysis results, and displays advice such as "This month's performance score: 85 / 100" and "Your stress is increasing, so we recommend you take a break."

[1440] Example prompts for generative AI models

[1441] Below are some example prompts to input to a generative AI model:

[1442] "What kind of feedback should we provide to Employee A in the month when his performance is below average? Please consider specific advice and provide suggestions that also reflect the results of the emotion engine analysis."

[1443] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1444] Step 1:

[1445] The server collects employee work data, specifically employee past performance data, current work status data, and work execution efficiency data, from internal databases and external APIs. The input in this step is data from CRM systems and project management tools, and the output is to compile this data into a single dataset.

[1446] Step 2:

[1447] The server pre-processes the collected data, which includes the following steps:

[1448] 1. Unification of data formats: Convert data of different formats into a common format (input: various data formats, output: unified data format).

[1449] 2. Data Combination: Combining data from different sources into a single dataset (input: individual data items, output: cohesive dataset).

[1450] 3. Missing value imputation: impute missing data in an appropriate way (input: partially missing data, output: complete dataset with imputed missing values).

[1451] 4. Normalization and scaling: Reduce the variability of the data and convert it into a format suitable for machine learning models (input: unnormalized data, output: normalized and scaled data).

[1452] Step 3:

[1453] The server uses the preprocessed data to train a machine learning model, specifically by following these steps:

[1454] 1. Data Split: Split the data into training and test sets (input: preprocessed data, output: training and test sets).

[1455] 2. Model selection: Choose an algorithm suitable for the task (input: candidate algorithms, output: selected algorithm).

[1456] 3. Training: The selected algorithm is trained on the data (input: training set, output: trained model).

[1457] 4. Hyperparameter tuning: Adjusting parameters to improve model performance (input: model and candidate parameters, output: optimized model).

[1458] Step 4:

[1459] The server evaluates employee performance using a trained machine learning model, including:

[1460] 1. Data input: Input the current month's performance data and customer feedback into the model (input: real-time performance data, output: predicted score).

[1461] 2. Score calculation: The model calculates the employee's performance score based on the input data (input: current performance data, output: performance score).

[1462] Step 5:

[1463] The server uses the emotion recognition means to recognize the employee's emotion, specifically by the following steps:

[1464] 1. Voice input analysis: Record the employee's voice and analyze the voice data (input: voice data, output: voice analysis results).

[1465] 2. Facial expression recognition: Capture employees' facial expressions using a webcam and analyze their emotions (input: facial expression data, output: facial expression analysis results).

[1466] 3. Text analysis: Analyzes text data entered by employees and infers their emotions (input: text data, output: text analysis results).

[1467] Step 6:

[1468] The server sends the generated evaluation results and emotion recognition results to the terminal, which then notifies the employee's smartphone or computer. Specifically, the process is as follows:

[1469] 1. Displaying evaluation results: Notifying employees of their performance scores and specific performance evaluations (Input: Evaluation results, Output: Notification message).

[1470] 2. Emotion-based advice: Provides emotion-based advice based on emotion recognition results (input: emotion analysis results, output: advice message).

[1471] (Application example 2)

[1472] 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."

[1473] The problem is that performance evaluation and feedback of employees cannot take into account the emotional state of employees, making it difficult to properly manage the impact of the evaluation results on employees' psychology. In addition, it is not possible to provide detailed advice based on the employee's emotional state, making it difficult to simultaneously improve performance and employee satisfaction.

[1474] The specific processing by the specific 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 employee work data, means for preprocessing the collected data, means for training a machine learning model based on the preprocessed data, means for evaluating employee performance using the trained machine learning model, an emotion engine including emotion analysis means for analyzing the emotional state of the employee at the time of performance evaluation, and means for feeding back the evaluation results and emotion analysis results to the employee. As a result, employee performance evaluations are performed taking into account the emotional state of the employee, the impact of the evaluation results on employee psychology is appropriately managed, and it is possible to achieve both improved performance and improved employee satisfaction.

[1475] "Employee business data" refers to various data generated when employees perform their work, including past performance data, current work status data, and work performance efficiency data.

[1476] "Preprocessing" refers to data processing techniques to prepare collected data for use as input data for machine learning models, and includes standardizing data formats, combining data, filling in missing values, and normalizing and scaling.

[1477] A "machine learning model" is an algorithm that learns patterns and features based on large amounts of data and makes predictions and classifications, and includes random forests, XGBoost, neural networks, and others.

[1478] "Performance appraisal" is a method of using a trained machine learning model to evaluate an employee's performance and efficiency in carrying out their work as a number or score.

[1479] "Emotion analysis means" refers to a method for analyzing emotions from an employee's facial expressions, voice tone, text data, etc., and includes an emotion engine using a camera or microphone.

[1480] The "emotion engine" is a system that performs facial expression recognition and voice analysis to grasp the user's emotional state.

[1481] "Feedback" refers to the act of notifying employees of the results of their performance evaluation and sentiment analysis, and providing them with advice based on this.

[1482] "Real-time feedback" is a feature that provides immediate advice based on evaluation results and emotional state at the moment an employee is working.

[1483] "Smart glasses" are wearable devices that are equipped with a camera, display, and voice recognition functions and are worn by the user.

[1484] The employee performance evaluation system of this invention objectively evaluates employee performance by collecting and preprocessing employee work data and training a machine learning model. It also uses emotion analysis means to understand the employee's emotional state and provides feedback based on the evaluation results and advice based on their emotions, thereby achieving both employee satisfaction and improved performance.

[1485] This system uses the following hardware and software.

[1486] Hardware and software:

[1487] 1. Server:

[1488] Data collection:

[1489] Employee past performance data, current work status data, and work execution efficiency data are collected from various internal databases and external APIs.

[1490] Data preprocessing:

[1491] Preprocessing includes standardizing data formats, combining data, imputing missing values, and normalizing and scaling to convert the collected data into a consistent form.

[1492] Training the machine learning model:

[1493] Train a machine learning model (e.g., Random Forest, XGBoost, or Neural Network) using the preprocessed data.

[1494] 2. Smart Glasses:

[1495] Emotion analysis:

[1496] The smart glasses use a camera and microphone to analyze the facial expressions and tone of voice of customers and employees in real time, and estimate their emotional state using an emotion analysis model (such as CNN).

[1497] Real-time feedback:

[1498] The smart glasses display real-time feedback from superiors and colleagues, advice based on performance data, and advice based on emotions.

[1499] How to use:

[1500] 1. Data Collection and Preprocessing:

[1501] The server automatically collects and preprocesses employee work data, including standardizing data formats and filling in missing values.

[1502] 2. Training the machine learning model:

[1503] The preprocessed data is used to train machine learning models, using algorithms such as random forests and neural networks.

[1504] 3. Performance Evaluation:

[1505] A trained machine learning model evaluates employee performance using current work situation data as input.

[1506] 4. Emotion analysis:

[1507] The smart glasses are equipped with a camera and microphone that analyze the facial expressions and voices of employees and customers in real time, estimating their emotional state and sending it to a server.

[1508] 5. Providing Feedback:

[1509] Using the evaluation results and sentiment analysis, feedback is displayed on the smart glasses screen to the employee, including advice based on performance evaluation and customized advice based on emotions.

[1510] Examples:

[1511] For example, when an employee wears smart glasses while serving customers in a brick-and-mortar store, the server collects and preprocesses the employee's work data to train an evaluation model. The smart glasses analyze the customer's facial expressions and tone of voice using emotion analysis tools, and display the customer's emotional state to the employee in real time. This allows the employee to understand the customer's emotional state and provide appropriate service, further improving customer satisfaction.

[1512] Example prompt sentence:

[1513] "Please propose a system that analyzes the emotions of customers from their facial expressions and tone of voice, and provides appropriate customer service advice based on that."

[1514] This invention allows employees' performance to be evaluated fairly and objectively, and provides feedback that takes emotions into account, making it possible to achieve both improved performance and increased employee satisfaction.

[1515] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1516] Step 1:

[1517] Data collection

[1518] The server collects data on employees' past performance, current work status, and work efficiency from various internal databases and external APIs. The input is employee work-related data, and the output is business data converted into a unified data format.

[1519] Step 2:

[1520] Data Preprocessing

[1521] The server preprocesses the collected business data by standardizing data formats, combining data, imputing missing values, normalizing and scaling it. The input is the collected business data, and the output is a preprocessed, consistent dataset.

[1522] Step 3:

[1523] Training a machine learning model

[1524] The server uses the preprocessed data to train a machine learning model (e.g., Random Forest, XGBoost, Neural Network, etc.) The input is the preprocessed dataset and the output is the trained machine learning model.

[1525] Step 4:

[1526] Performance evaluation

[1527] The server evaluates employee performance using the trained machine learning model, using current work situation data as input and the employee performance score as output. The input is the current work situation data and the output is the evaluated performance score.

[1528] Step 5:

[1529] Emotion analysis

[1530] The smart glasses use a camera and microphone to analyze the facial expressions and tone of voice of customers and employees in real time. The input is camera video and audio data, and the output is the emotion analysis results, which are sent to a server.

[1531] Step 6:

[1532] Real-time feedback

[1533] The server sends the evaluation results and emotion analysis results to the smart glasses, which then provide feedback to the employee. The smart glasses display shows feedback from superiors and colleagues, advice based on performance data, and advice based on emotions. The input is the evaluation results and emotion analysis results, and the output is real-time feedback display.

[1534] Step 7:

[1535] Providing Feedback

[1536] Based on the evaluation results and sentiment analysis, more specific advice is displayed on the smart glasses, allowing the user to adjust their interactions with the customer. The input is feedback information, and the output is specific advice provided to employees.

[1537] 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.

[1538] 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.

[1539] 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.

[1540] 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.

[1541] 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.

[1542] 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.

[1543] 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).

[1544] 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.

[1545] 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."

[1546] 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.

[1547] 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).

[1548] 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.

[1549] 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.

[1550] 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.

[1551] 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.

[1552] 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.

[1553] 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.

[1554] 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.

[1555] 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.

[1556] 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.

[1557] 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.

[1558] The following is further disclosed regarding the above embodiment.

[1559] (Claim 1)

[1560] a means of collecting employee work data;

[1561] a means for pre-processing the collected data;

[1562] a means for training a machine learning model on the preprocessed data; and

[1563] a means for evaluating employee performance using a trained machine learning model;

[1564] A means of providing feedback on the results of the evaluation to employees;

[1565] A system including:

[1566] (Claim 2)

[1567] 2. The system of claim 1, wherein the evaluation means uses employee past performance data, current work situation data, and work performance efficiency data as inputs.

[1568] (Claim 3)

[1569] 10. The system of claim 1, wherein the feedback means provides advice for improving performance based on the evaluation results.

[1570] "Example 1"

[1571] (Claim 1)

[1572] a means of collecting employee work data;

[1573] a means for pre-processing the collected data;

[1574] a means for training a machine learning model on the preprocessed data; and

[1575] a means for evaluating employee performance using a trained machine learning model;

[1576] A means of providing feedback on the results of the evaluation to employees;

[1577] A system including:

[1578] (Claim 2)

[1579] 2. The system of claim 1, wherein the evaluation means uses employee past performance data, current work status data, and work execution efficiency data as input, and preprocesses the data, including standardizing formats, combining data, filling in missing values, normalizing, and scaling.

[1580] (Claim 3)

[1581] 2. The system according to claim 1, wherein the feedback means provides specific advice for improving performance and information useful for setting goals for the next time based on the evaluation results.

[1582] "Application Example 1"

[1583] (Claim 1)

[1584] a means of collecting employee work data;

[1585] a means for pre-processing the collected data;

[1586] a means for training a machine learning model on the preprocessed data; and

[1587] a means for evaluating employee performance using a trained machine learning model;

[1588] A means of providing feedback on the results of the evaluation to employees;

[1589] a means for collecting operating data of the work equipment;

[1590] A means for preprocessing the collected labor equipment operation data;

[1591] a means for training a machine learning model based on the preprocessed labor equipment operating data;

[1592] a means for evaluating the performance of work equipment using a trained machine learning model;

[1593] a means of providing feedback on the evaluation results to the manager;

[1594] A system including:

[1595] (Claim 2)

[1596] 2. The system of claim 1, wherein the evaluation means uses employee past performance data, current work situation data, and work performance efficiency data as inputs.

[1597] (Claim 3)

[1598] 10. The system of claim 1, wherein the feedback means provides advice for improving performance based on the evaluation results.

[1599] "Example 2: Combining Emotion Engines"

[1600] (Claim 1)

[1601] a means of collecting employee work data;

[1602] a means for pre-processing the collected data;

[1603] a means for training a machine learning model on the preprocessed data; and

[1604] a means for evaluating employee performance using a trained machine learning model;

[1605] an emotion recognition means for recognizing an employee's emotional state;

[1606] a means for providing feedback to employees on the evaluation results and emotion recognition results;

[1607] A system including:

[1608] (Claim 2)

[1609] 2. The system of claim 1, wherein the evaluation means uses employee past performance data, current work situation data, and work performance efficiency data as inputs.

[1610] (Claim 3)

[1611] 10. The system of claim 1, wherein the feedback means provides advice for improving performance based on the evaluation results and the emotion recognition results.

[1612] "Application example 2 when combining emotion engines"

[1613] (Claim 1)

[1614] a means of collecting employee work data;

[1615] a means for pre-processing the collected data;

[1616] a means for training a machine learning model on the preprocessed data; and

[1617] a means for evaluating employee performance using a trained machine learning model;

[1618] an emotion engine including an emotion analysis means for analyzing the employee's emotional state during a performance evaluation;

[1619] A means of providing feedback to employees on the results of evaluations and sentiment analysis;

[1620] A system including:

[1621] (Claim 2)

[1622] 2. The system of claim 1, wherein the evaluation means uses as inputs employee past performance data, current work situation data, work performance efficiency data, and employee emotion data.

[1623] (Claim 3)

[1624] 10. The system of claim 1, wherein the feedback means provides performance improvement advice and emotion-based advice based on the evaluation results and the emotional state. [Explanation of symbols]

[1625] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a means of collecting employee work data; a means for pre-processing the collected data; a means for training a machine learning model on the preprocessed data; and a means for evaluating employee performance using a trained machine learning model; A means of providing feedback on the results of the evaluation to employees; A system including:

2. 2. The system of claim 1, wherein the evaluation means uses employee past performance data, current work situation data, and work performance efficiency data as inputs.

3. The system according to claim 1 , wherein the feedback means provides advice for improving performance based on the evaluation results.

Images