system

Abstract

We provide the system. [Solution] A natural language processing method that analyzes workers' text data to quantify their emotions, A machine learning method for detecting anomalies in workers' work efficiency data and attendance data, A deep learning means for integrating the quantified emotion data and anomaly detection data to evaluate risk, A means of recommending countermeasures to high-risk workers, A means of notifying the administrator of the aforementioned recommendation, A system that includes a means of analyzing resident participation data in local activities and recommending measures for regional revitalization.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, 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

Summary of the Invention

Problems to be Solved by the Invention

[0004] In personnel management, it is an important issue to improve employee engagement and reduce the turnover rate. However, with conventional methods, it takes a long time to analyze a huge amount of data, and it is difficult to timely detect abnormalities in employees' emotions and productivity. Also, it has not been easy to quickly identify high-risk employees from these data and propose appropriate countermeasures. For this reason, it has been difficult to make a prompt and effective response at the site, and as a result, there have been many cases where the workplace environment has not been improved and the happiness of employees has not been enhanced.

Means for Solving the Problems

[0005] This invention provides a means for analyzing text data obtained from employees using a natural language processing model and quantifying their emotions. It also includes a means for analyzing employee productivity data and attendance data using a machine learning model to detect anomalies. Furthermore, it includes a means for integrating this quantified emotion data and anomaly detection data using a deep learning model to identify and evaluate high-risk employees. For identified high-risk employees, a recommendation system notifies managers of specific countermeasures, thereby enabling rapid response and improving the workplace environment.

[0006] "Natural language processing means" refers to technology that analyzes employees' text data and quantifies their emotions.

[0007] "Machine learning methods" refer to technologies that analyze employee productivity data and attendance data to detect anomalies in the data.

[0008] "Deep learning methods" refer to advanced analytical techniques that integrate emotional data and anomaly detection data to identify high-risk employees.

[0009] "Recommendation methods" refer to technologies that propose optimal countermeasures for high-risk employees and notify managers accordingly.

[0010] "Notification methods" refer to technologies used to inform administrators of analysis results and recommendation information. [Brief explanation of the drawing]

[0011] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4]This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11] This is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] This is a sequence diagram showing the processing flow of the data processing system in Example 2, which incorporates an emotion engine. [Figure 14] This is a sequence diagram showing the processing flow of the data processing system in Application Example 2, which combines an emotion engine. [Modes for carrying out the invention]

[0012] Hereinafter, an example of an embodiment of the system relating to the technology of this disclosure will be described with reference to the attached drawings.

[0013] First, let's explain the terminology used in the following explanation.

[0014] In the following embodiments, the labeled processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0015] In the following embodiments, the labeled RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0016] In the following embodiments, the labeled storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, and the like.

[0017] In the following embodiments, the labeled communication I / F (Interface) is an interface including a communication processor and 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), or Bluetooth (registered trademark), and the like.

[0018] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0019] [First Embodiment]

[0020] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0021] As shown in Figure 1, the 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.

[0022] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0023] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0024] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and 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.

[0025] 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 perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0026] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0027] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0028] As shown in Figure 2, in the data processing device 12, a specific processing 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" related to the technology of this 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 according to the specific processing program 56 executed on the RAM 30.

[0029] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0030] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0031] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0032] This invention relates to an innovative system for improving employee engagement. The server first collects various employee data from various systems within the company. This includes productivity data, attendance data, and the results of regularly conducted engagement surveys.

[0033] The server applies natural language processing (NLP) models to analyze collected text data, quantifying employee sentiment. This makes it possible to identify emotional tendencies such as positive, negative, and neutral. Next, the server uses machine learning (ML) models to analyze productivity and attendance data, detecting hidden anomalies within this data. These anomalies may include, for example, sudden declines in performance or frequent tardiness.

[0034] The server then uses deep learning technology to integrate emotional data and anomaly data, and uses this to assess employee risk. This assessment can identify employees who are likely to leave or who may be experiencing stress. The server then recommends the most appropriate course of action for these high-risk employees. The recommended course of action is notified to the administrator's terminal, allowing the administrator to take appropriate measures for the employee at the right time.

[0035] By enabling HR personnel to implement appropriate countermeasures, it becomes possible to improve employee engagement and reduce turnover. For example, suppose a server analyzes an employee's text data and detects that the employee is dissatisfied with their work. Simultaneously, if frequent tardiness is detected in the attendance data, the server identifies the employee as high-risk and recommends the implementation of "flexible working hours" and "mental health support." The user, upon receiving a notification on their device, can then schedule a meeting with the employee based on this information.

[0036] The following describes the processing flow.

[0037] Step 1:

[0038] The server collects employee productivity data, attendance data, and engagement survey results from various company data sources. The collected data is stored in a structured database.

[0039] Step 2:

[0040] The server applies a natural language processing (NLP) model to analyze text-based survey responses. From the analyzed data, it quantifies employees' emotions based on tendencies such as positive, negative, or neutral, and records the results in a database.

[0041] Step 3:

[0042] The server uses machine learning (ML) models to analyze productivity and attendance data and detect anomalies. In particular, it detects significant changes compared to past performance and attendance records and records the anomalies in the database.

[0043] Step 4:

[0044] The server uses deep learning models to integrate and analyze emotional and anomalous data. This helps identify employees at high risk of leaving the company or those who appear to be experiencing stress.

[0045] Step 5:

[0046] The server generates appropriate countermeasures for each high-risk employee, such as offering flexible working hours or scheduling counseling sessions. These recommended countermeasures are optimized by referencing an existing database of response strategies.

[0047] Step 6:

[0048] The server notifies the administrator's terminal of the generated recommended actions. The administrator, as a user, checks the notification on their terminal and begins preparing to take the recommended actions.

[0049] Step 7:

[0050] Based on the recommendations they receive, users can schedule meetings with employees and, if necessary, take concrete actions such as adjusting job responsibilities or implementing support plans. This can lead to improvements in the work environment and increased employee engagement.

[0051] (Example 1)

[0052] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0053] Managing employee engagement and reducing turnover are crucial for many organizations. However, quickly identifying changes in employee emotions or abnormalities in work performance can be difficult, sometimes leading to delays in taking appropriate action. There is a need for a system that accurately captures changes in employee emotional states and performance and enables early intervention.

[0054] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0055] In this invention, the server includes a natural language processing means for analyzing employee descriptive data and quantifying emotions, a machine learning means for detecting anomalies in employee work data and time management data, and a deep learning means for integrating the quantified emotion data and anomaly detection data to evaluate risk. This makes it possible to comprehensively evaluate the state of employees, identify risks early, and take countermeasures.

[0056] "Natural language processing" refers to information technology that analyzes text data to quantify the emotions and meanings contained within a document.

[0057] "Machine learning methods" are technologies that analyze employee work data and time management data to detect deviations from normal patterns.

[0058] "Deep learning" is an artificial intelligence technology that integrates quantified emotion data and anomaly detection data to accurately assess employee risk.

[0059] "Generation method" refers to information generation technology used to recommend appropriate countermeasures to employees identified as high-risk.

[0060] "Information provision means" refers to notification technology that provides administrators with necessary information in real time, enabling prompt responses.

[0061] This invention is a comprehensive system for improving the working environment for employees and reducing employee turnover. The server utilizes the company's database system to collect employee work data, time management data, and text data obtained from engagement surveys. A general-purpose database management system can be used in this process.

[0062] The server uses software such as "NLTK" and "spaCy" for natural language processing to analyze text data and quantify emotions such as positive, negative, and neutral. For anomaly detection using machine learning, "Scikit-learn" is used, applying models such as linear regression and random forest. Deep learning technologies such as "TENSORFLOW®" and "Keras" are used to integrate emotion data and anomaly detection data to perform highly accurate risk assessments.

[0063] As a concrete example, suppose the server analyzes the text data of an employee in the general affairs department and detects from the results that the employee is dissatisfied with their job. At the same time, if the attendance data shows frequent tardiness, that employee is identified as high risk. Based on these results, the server recommends "implementing flexible working hours" and "providing mental health support" and promptly notifies the administrator's terminal.

[0064] Administrators, as users, can receive notifications from the server and consider conducting interviews with employees or improving the work environment. For example, based on input such as, "Please tell me effective intervention methods to improve employee engagement," the generating AI model provides recommendations. In this way, the entire system works together effectively, enabling rapid responses.

[0065] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0066] Step 1:

[0067] The server collects employee work data, time management data, and text data obtained from engagement surveys from the company's database system. It takes various data stored in the company's database as input and processes that data into a format suitable for the next analysis step as output. Specifically, it uses SQL queries to extract necessary data from databases such as MySQL® and PostgreSQL.

[0068] Step 2:

[0069] The server applies natural language processing (NLP) to the collected text data to quantify emotions. It uses the text data processed in step 1 as input and generates an emotion score for each employee as output. Specific operations include tokenizing the text using "NLTK" or "spaCy" and applying an emotion analysis algorithm.

[0070] Step 3:

[0071] The server performs anomaly detection using machine learning (ML) based on business data and time management data. It uses numerical data obtained in Step 1 as input and generates an anomaly score as output. In this process, the server identifies anomalies using linear regression and random forests implemented with "Scikit-learn".

[0072] Step 4:

[0073] The server integrates emotion scores and anomaly scores and performs risk assessment using deep learning techniques. It uses the output data from steps 2 and 3 as input and generates risk assessment results indicating dwell time and stress levels as output. Specifically, it uses neural networks built with TensorFlow or Keras.

[0074] Step 5:

[0075] Based on the risk assessment results, the server identifies high-risk employees and recommends optimal countermeasures using a generative AI model. It uses the risk assessment data obtained in step 4 as input and generates recommendations for improvement measures as output. Specifically, it instructs the generative AI model using the prompt "Please tell me effective intervention methods to improve employee engagement."

[0076] Step 6:

[0077] The server notifies the administrator's terminal of the recommended course of action. Using the recommendations generated in step 5 as input, it creates a notification that the administrator can review as output. Specifically, it utilizes notification APIs such as "Slack" and "MICROSOFT® TEAMS®" to send information to the administrator in real time.

[0078] (Application Example 1)

[0079] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0080] In modern businesses and communities, there is a need to improve the engagement of individual workers and residents. However, traditional methods make it difficult to detect abnormalities in workers' emotions and productivity early on and to respond appropriately. Furthermore, understanding the level of participation in community activities and using that understanding to revitalize local communities is insufficient. Therefore, there is a need for a system that analyzes worker data to efficiently recommend countermeasures and supports resident activities in local communities.

[0081] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0082] In this invention, the server includes a natural language processing means for analyzing workers' text data and quantifying their emotions, a machine learning means for detecting anomalies in workers' work efficiency data and attendance data, and a means for analyzing resident participation data in community activities and recommending measures for community revitalization. This makes it possible to understand the behavioral trends of workers and residents and provide appropriate support aimed at improving engagement.

[0083] A "worker" refers to an individual who engages in specific tasks or duties and receives a salary as compensation for those tasks.

[0084] "Text data" refers to a collection of data in which character information is represented in an electronic format.

[0085] "Natural language processing methods" refer to technologies and methods for understanding and processing human language using computers.

[0086] "Quantifying emotions" is the process of quantitatively evaluating emotions and expressing them as numerical values.

[0087] "Work efficiency data" refers to information regarding the efficiency and productivity of workers when performing their tasks.

[0088] "Attendance data" refers to data that shows information related to employees' arrival and departure times and working hours.

[0089] "Machine learning methods" refer to computational techniques used to learn rules and patterns from data and perform predictions and classifications.

[0090] "Detecting anomalies" means identifying and detecting patterns or behaviors that are different from the norm.

[0091] "Community activities" refer to social or cultural activities that take place within a specific region or community.

[0092] "Resident participation data" refers to data that shows records and information about residents' participation in activities and events held in the region or community.

[0093] "Regional revitalization" refers to initiatives and processes aimed at invigorating the economic and cultural activities of a local community and improving its quality of life.

[0094] "Recommending countermeasures" means suggesting solutions or actions that are appropriate for a specific situation.

[0095] The system implementing this invention is built with the aim of improving the engagement of workers and residents within a smart city. The server collects workers' text data, work efficiency data, and attendance data, and uses natural language processing technology (e.g., NLTK, spaCy) to quantify their emotions. In this process, emotions are classified as positive, negative, or neutral. Furthermore, machine learning (e.g., scikit-learn) is applied to detect anomalies in the work efficiency data and attendance data.

[0096] The server simultaneously acquires participation data from residents within the area and analyzes it to revitalize community activities. Based on the resident participation data, it evaluates participation trends in community activities and generates recommendations for events that will attract residents' interest. The recommended actions are notified to the administrator's terminal or smartphone, supporting action plans for providing appropriate assistance.

[0097] For example, if a server detects a decline in residents' event participation in a particular area, and emotional data indicates a decrease in participation motivation, it can then suggest new events that residents might be interested in. An example of a prompt for the generative AI model would be, "Please provide feedback on recent community events as part of the citizen feedback." This makes it possible to appropriately understand the needs of residents and workers and improve engagement.

[0098] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0099] Step 1:

[0100] The server collects employee text data, work efficiency data, and attendance data from various systems within the company. Based on this input data, natural language processing technology is used to quantify the employees' emotions from the text data. The output is emotion data classified as positive, negative, or neutral.

[0101] Step 2:

[0102] The server detects anomalies by running machine learning algorithms using quantified sentiment data, operational efficiency data, and attendance data. Specifically, it compares current data with past data to detect sudden performance declines and attendance irregularities. The output provides information on whether or not an anomaly was detected.

[0103] Step 3:

[0104] The server integrates sentiment data and anomaly detection data and uses deep learning technology to perform a risk assessment of workers. This risk assessment identifies workers who are likely to leave the company or who may be experiencing stress. The output is a list indicating the level of risk.

[0105] Step 4:

[0106] The server generates recommended words using an AI model that creates countermeasures for identified high-risk workers. Specifically, concrete measures such as "flexible working hours" and "mental health support" are suggested. The output is a list of recommended countermeasures.

[0107] Step 5:

[0108] The server notifies the administrator's terminal of the recommended countermeasures. The terminal helps the administrator, upon receiving the notification, take appropriate action regarding the worker in the right time. The output is the notification message received by the administrator.

[0109] Step 6:

[0110] The server collects participation data from local residents and analyzes the data to evaluate participation trends. This analysis detects declines in the frequency of participation in local events or a decrease in interest. The output is the result of the participation analysis.

[0111] Step 7:

[0112] The server generates recommendation text using an AI model to create new events designed to attract residents' interest. Specifically, it can suggest local cultural activities and sports events. The output is a list of recommended event suggestions.

[0113] Step 8:

[0114] Based on these results, the server presents specific event proposals to residents and sends notifications aimed at revitalizing the community. These notifications are sent to residents' and community's devices. The output is a notification of event proposals to residents.

[0115] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0116] This invention relates to a system that combines an emotion engine with the aim of improving employee engagement. It aims to improve the workplace environment by comprehensively evaluating employee conditions and recommending appropriate countermeasures to managers.

[0117] First, the server collects diverse data about employees, including productivity data, attendance data, and engagement survey results. Simultaneously, an emotion engine analyzes the user's voice and facial expressions to recognize emotions in real time. This process identifies the user's emotional state, and this data is integrated with the survey results.

[0118] Next, the server analyzes the descriptive data from the survey through a natural language processing (NLP) model to quantify employee emotions. This allows for the identification of specific emotional tendencies and a deeper understanding of employee states. Furthermore, machine learning (ML) models are used to analyze productivity and attendance data and detect anomalous patterns.

[0119] The analyzed sentiment data and anomaly detection data are integrated on the server using deep learning technology, and a risk assessment is performed. User sentiment information obtained by the sentiment engine is reflected in the recommendation of countermeasures, improving the accuracy of the recommendations.

[0120] As a concrete example, suppose an employee's productivity drops sharply, and a survey indicates dissatisfaction. In this case, the emotion engine recognizes the user's anxiety and stress, and the server recommends countermeasures, including suggestions for flexible working hours and stress management workshops. These recommendations are sent from the server to the administrator's terminal, allowing the user to take appropriate action based on the employee's situation.

[0121] Thus, the system of the present invention allows for a more accurate understanding of employees' conditions and enables timely and effective responses. This makes it possible to improve overall workplace engagement and productivity.

[0122] The following describes the processing flow.

[0123] Step 1:

[0124] The server collects employee productivity data, attendance data, and engagement survey results from various data sources. This data is automatically retrieved from daily operations systems and HR systems and stored in a database.

[0125] Step 2:

[0126] The emotion engine analyzes the user's voice and facial expression data to identify their emotional state in real time. The emotion engine acquires this data through microphones and cameras and performs analysis using emotion recognition algorithms.

[0127] Step 3:

[0128] The server uses a natural language processing (NLP) model to analyze the descriptive data from the survey. In this process, employee emotions are quantified as positive, negative, or neutral and recorded in a database. This allows for a quantitative evaluation of emotional tendencies.

[0129] Step 4:

[0130] The server uses machine learning (ML) models to analyze productivity and attendance data. It compares this data with historical data to detect anomalous patterns that deviate from the normal range. Once an anomaly is identified, the information is recorded in the database.

[0131] Step 5:

[0132] The server utilizes deep learning technology to integrate sentiment data and anomaly detection data, and uses this information to perform risk assessments of employees. User information from the sentiment engine is also reflected in the risk assessment, improving the accuracy of the decisions.

[0133] Step 6:

[0134] The server generates countermeasures for high-risk employees. These countermeasures include suggestions for adjusting working hours and stress management programs, and are tailored to take into account the user's emotional information recognized by the emotion engine.

[0135] Step 7:

[0136] The server notifies the administrator's terminal of the generated countermeasures. The administrator user reviews the recommendations on their terminal and takes specific actions as needed, such as scheduling interviews with employees.

[0137] Step 8:

[0138] Based on recommendations, users provide appropriate support, such as direct interaction with employees. Prompt action contributes to reducing employee stress and improving the work environment.

[0139] (Example 2)

[0140] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0141] Understanding the relationship between employees' emotional states and workplace performance, and identifying optimal countermeasures, is crucial for improving the work environment and increasing productivity. However, conventional technologies have not adequately performed precise emotional analysis or anomaly detection, resulting in incomplete countermeasures. To address these challenges, there is a need for multifaceted emotional assessment of employees, automated risk assessment, and the provision of appropriate countermeasures.

[0142] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0143] In this invention, the server includes means for analyzing employees' voices and facial expressions to identify their emotions, natural language processing means for analyzing employees' text data to quantify their emotions, and machine learning means for detecting anomalies in employees' productivity data and attendance data. This enables a multifaceted analysis of employees' emotions and work data, allowing for risk assessment and the proposal of optimal countermeasures.

[0144] "Means for identifying emotions by analyzing voice and facial expressions" refers to technology that analyzes an employee's voice and facial expressions to identify their emotional state in real time.

[0145] "Natural language processing method that analyzes text data to quantify emotions" is a technology that analyzes text data written by employees, extracts emotions from its content, and expresses them as quantitative numerical values.

[0146] "Machine learning methods for detecting anomalies in productivity data and attendance data" refers to machine learning algorithms used to analyze data related to employee productivity and working hours to detect unusual patterns or anomalies.

[0147] A "deep learning method for integrating data and assessing risk" is a technique that uses deep learning technology to integrate information obtained from multiple data sources and comprehensively assess employee-related risks.

[0148] "Methods for proposing countermeasures to high-risk employees" refers to technologies for generating and proposing specific and appropriate countermeasures to high-risk employees identified from integrated data.

[0149] The "means of notifying administrators of proposals" refers to a function that notifies administrators of the generated proposed countermeasures and aims to encourage appropriate action.

[0150] This invention relates to a system for improving employee engagement in the workplace environment. This system combines voice and facial expression analysis technology, natural language processing technology, machine learning technology, and deep learning technology to comprehensively evaluate employees' emotional states and recommend appropriate countermeasures to managers, thereby improving the workplace.

[0151] Hardware and software

[0152] The server is equipped with sensors and cameras to collect employee voices and facial expressions in real time. Acoustic analysis software is used for voice analysis, and video analysis software is used for facial expression analysis. Generative AI models are used for natural language processing to convert employee text data into sentiment scores. Machine learning algorithms are introduced to detect anomalies in productivity and attendance data. A deep learning framework is used to integrate this data and propose risk assessments and countermeasures.

[0153] Specific example

[0154] As a concrete example, consider a scenario where the server detects a sharp decline in employee A's productivity within a certain project. In this case, the administrator receives information via a terminal that employee A had recently shown signs of stress in voice analysis. Furthermore, analysis reveals that dissatisfaction has been extracted from the free-response answers of a survey. Based on this information, the server recommends implementing mental health counseling and introducing flexible working hours for employee A.

[0155] Example of a prompt

[0156] By using prompts such as "How can you reduce stress from the current workplace atmosphere?" as input to the AI ​​model, more specific and individualized solutions can be generated.

[0157] This system allows for a comprehensive assessment of employee emotions and performance, enabling responses tailored to individual needs, thereby improving overall workplace engagement and productivity.

[0158] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0159] Step 1:

[0160] The server collects employee voice and facial expression data using sensors and cameras. Inputs include audio files and video footage captured in real time from microphones and cameras. Audio data is analyzed using acoustic analysis software, and facial expression data is processed using video analysis software. The output is digital data indicating the employee's emotional state.

[0161] Step 2:

[0162] The server analyzes employee text data using a generated AI model. The input consists of written survey responses and communication logs. Using natural language processing technology, it extracts emotions from this text data and outputs a numerical emotion score.

[0163] Step 3:

[0164] The server collects employee productivity and attendance data and analyzes it using machine learning algorithms. Inputs include work hours, project progress, and attendance records. Based on this data, it detects anomalous patterns and provides anomaly detection results as output.

[0165] Step 4:

[0166] The server integrates analyzed sentiment data and anomaly detection data using a deep learning framework to assess risk. Inputs are quantified sentiment scores and anomaly detection results. Outputs are information identifying high-risk employees as a result of the risk assessment.

[0167] Step 5:

[0168] The server generates countermeasures based on the risk assessment results and notifies the administrator. For example, the prompt "What interventions would be effective for stress management in this situation?" is input into the AI ​​model to obtain appropriate recommendations. The output is a list of countermeasures sent to the administrator's terminal.

[0169] In this way, the system comprehensively assesses the state of employees and helps optimize the work environment.

[0170] (Application Example 2)

[0171] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0172] In modern workplaces, it is crucial to understand workers' emotions and stress levels in real time and take appropriate measures. However, traditional management methods have made it difficult to accurately grasp workers' emotional states, sometimes delaying the optimization of the work environment. This can lead to problems such as decreased worker productivity and increased turnover. Against this backdrop, there is a need for a system that can quickly and accurately capture workers' emotions and immediately adjust the environment.

[0173] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0174] In this invention, the server includes natural language processing means for analyzing the worker's language data and quantifying their emotions, machine learning means for detecting anomalies in the worker's work data and work schedule data, and deep learning means for integrating the quantified emotion data and anomaly detection data to evaluate risk. This makes it possible to grasp the worker's emotional state in real time and quickly adjust the work environment as needed.

[0175] "Linguistic data" refers to text information generated by workers and serves as fundamental data for analyzing emotional states and thought patterns.

[0176] "Natural language processing" is a technology that allows computers to understand and analyze human language, and is used to quantify emotions.

[0177] "Work data" refers to activity information related to a worker's tasks, and is used to evaluate work efficiency and productivity.

[0178] "Work data" refers to data that shows the working hours and attendance status of workers, and is used for attendance management and pattern anomaly detection.

[0179] "Machine learning" is a technology in which computers learn from data and automatically discover patterns and rules, and is used to detect anomalies.

[0180] "Deep learning" is a technique that uses multi-layered artificial neural networks to perform complex data analysis, and is used to assess risk.

[0181] "Emotion recognition" is a technology that identifies the emotional state of workers from their facial expressions and voice, providing information necessary for environmental adjustments.

[0182] "Environmental adjustment" is the process of optimizing the work environment according to the emotional state and work situation of the workers, with the aim of improving productivity and comfort.

[0183] The server analyzes the worker's language data and uses natural language processing technology to quantify their emotions. In this process, the text data entered by the worker is analyzed in real time, and an emotion score such as positive, negative, or neutral is generated. In addition, work data and attendance data are analyzed using machine learning, and anomalies are detected by comparing them with past data.

[0184] Furthermore, the server utilizes deep learning to integrate sentiment data and anomaly detection data to assess risk. The deep learning model performs a detailed analysis of employee states and identifies high-risk situations.

[0185] Emotion recognition sensors use cameras and microphones to collect the worker's facial expressions and voice, and analyze this data. This allows for a real-time understanding of the user's emotional state. Based on the emotional information collected in this way, the server suggests adjustments to the work environment, specifically such as adjusting the work pace or recommending breaks.

[0186] On the user's device, recommendation information sent from the server is visualized, allowing administrators to make concrete decisions based on this information.

[0187] For example, if a worker's face is monitored by a camera and they show a tired expression for a while, the server can recognize that emotional pattern and recommend to the manager, "Please suggest that the worker take a short break."

[0188] An example of a prompt might be, "What environmental adjustments are recommended based on the employee's monitored emotional state?" This prompt is used in the generative AI model to highly customize the response to the user's needs.

[0189] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0190] Step 1:

[0191] The server collects real-time facial and audio data from the camera and microphone. This data is used as input to analyze the user's emotional state through an emotion recognition sensor. The server preprocesses the collected data using image and audio processing techniques and converts it into a format suitable for the next process.

[0192] Step 2:

[0193] The server applies an emotion recognition algorithm to the pre-processed data to identify the user's emotional state. Specifically, it uses image processing techniques to analyze the user's facial expressions and voice analysis techniques to analyze their tone of voice. This generates a real-time emotion score. This emotion score is stored in a database and used as input for the next step.

[0194] Step 3:

[0195] The server collects worker language data and analyzes it using natural language processing techniques. This data is entered in text format, and the server uses a language model to assign sentiment labels such as positive, negative, and neutral. The resulting quantified sentiment data is then integrated with work data and performance data.

[0196] Step 4:

[0197] The server utilizes machine learning algorithms to analyze worker work data and attendance data. Past work history and attendance data are used as input. Anomaly detection algorithms identify unusual patterns and anomalies, which are then output as anomaly detection data.

[0198] Step 5:

[0199] The server integrates anomaly detection data and quantified sentiment data using deep learning techniques. In this step, a statistical model is used to perform a risk assessment. The output consists of a risk score and recommended actions based on it, which are stored in a database.

[0200] Step 6:

[0201] Based on the generated risk score and recommended actions, the server generates a prompt message and inputs it into the AI ​​model. The prompt message is structured in the format of "What environmental adjustments are recommended based on the employee's monitored emotional state?" and serves as the basis for the AI ​​model to generate specific countermeasures.

[0202] Step 7:

[0203] The server sends the generated recommended actions to the administrator's terminal. The administrator can use this information to adjust the work environment and provide feedback to workers. The outputted information helps users take quick and effective action.

[0204] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0205] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0206] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0207] [Second Embodiment]

[0208] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0209] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0210] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0211] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0212] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0213] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0214] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0215] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0216] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0217] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0218] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0219] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. 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".

[0220] This invention relates to an innovative system for improving employee engagement. The server first collects various employee data from various systems within the company. This includes productivity data, attendance data, and the results of regularly conducted engagement surveys.

[0221] The server applies natural language processing (NLP) models to analyze collected text data, quantifying employee sentiment. This makes it possible to identify emotional tendencies such as positive, negative, and neutral. Next, the server uses machine learning (ML) models to analyze productivity and attendance data, detecting hidden anomalies within this data. These anomalies may include, for example, sudden declines in performance or frequent tardiness.

[0222] The server then uses deep learning technology to integrate emotional data and anomaly data, and uses this to assess employee risk. This assessment can identify employees who are likely to leave or who may be experiencing stress. The server then recommends the most appropriate course of action for these high-risk employees. The recommended course of action is notified to the administrator's terminal, allowing the administrator to take appropriate measures for the employee at the right time.

[0223] By enabling HR personnel to implement appropriate countermeasures, it becomes possible to improve employee engagement and reduce turnover. For example, suppose a server analyzes an employee's text data and detects that the employee is dissatisfied with their work. Simultaneously, if frequent tardiness is detected in the attendance data, the server identifies the employee as high-risk and recommends the implementation of "flexible working hours" and "mental health support." The user, upon receiving a notification on their device, can then schedule a meeting with the employee based on this information.

[0224] The following describes the processing flow.

[0225] Step 1:

[0226] The server collects employee productivity data, attendance data, and engagement survey results from various company data sources. The collected data is stored in a structured database.

[0227] Step 2:

[0228] The server applies a natural language processing (NLP) model to analyze text-based survey responses. From the analyzed data, it quantifies employees' emotions based on tendencies such as positive, negative, or neutral, and records the results in a database.

[0229] Step 3:

[0230] The server uses machine learning (ML) models to analyze productivity and attendance data and detect anomalies. In particular, it detects significant changes compared to past performance and attendance records and records the anomalies in the database.

[0231] Step 4:

[0232] The server uses deep learning models to integrate and analyze emotional and anomaly data. This helps identify employees at high risk of leaving the company or those who appear to be experiencing stress.

[0233] Step 5:

[0234] The server generates appropriate countermeasures for each high-risk employee, such as suggesting flexible working hours or scheduling counseling sessions. The recommended countermeasures are optimized by referencing an existing database of response strategies.

[0235] Step 6:

[0236] The server notifies the administrator's terminal of the generated action recommendations. The administrator, as a user, checks the notification on their terminal and begins preparing to take the recommended action.

[0237] Step 7:

[0238] Based on the recommendations they receive, users can schedule meetings with employees and, if necessary, take concrete actions such as adjusting job responsibilities or implementing support plans. This can lead to improvements in the work environment and increased employee engagement.

[0239] (Example 1)

[0240] Next, we will describe Example 1. 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."

[0241] Managing employee engagement and reducing turnover are crucial for many organizations. However, quickly identifying changes in employee emotions or abnormalities in work performance can be difficult, sometimes leading to delays in taking appropriate action. There is a need for a system that accurately captures changes in employee emotional states and performance and enables early intervention.

[0242] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0243] In this invention, the server includes a natural language processing means for analyzing employee descriptive data and quantifying emotions, a machine learning means for detecting anomalies in employee work data and time management data, and a deep learning means for integrating the quantified emotion data and anomaly detection data to evaluate risk. This makes it possible to comprehensively evaluate the state of employees, identify risks early, and take countermeasures.

[0244] "Natural language processing" refers to information technology that analyzes text data to quantify the emotions and meanings contained within a document.

[0245] "Machine learning methods" are technologies that analyze employee work data and time management data to detect deviations from normal patterns.

[0246] "Deep learning" is an artificial intelligence technology that integrates quantified emotion data and anomaly detection data to accurately assess employee risk.

[0247] "Generation method" refers to information generation technology used to recommend appropriate countermeasures to employees identified as high-risk.

[0248] "Information provision means" refers to notification technology that provides administrators with necessary information in real time, enabling prompt responses.

[0249] This invention is a comprehensive system for improving the working environment for employees and reducing employee turnover. The server utilizes the company's database system to collect employee work data, time management data, and text data obtained from engagement surveys. A general-purpose database management system can be used in this process.

[0250] The server uses software such as "NLTK" and "spaCy" for natural language processing to analyze text data and quantify emotions such as positive, negative, and neutral. For anomaly detection using machine learning, "Scikit-learn" is used, applying models such as linear regression and random forest. By using deep learning techniques such as "TensorFlow" and "Keras," emotion data and anomaly detection data are integrated to perform highly accurate risk assessments.

[0251] As a concrete example, suppose the server analyzes the text data of an employee in the general affairs department and detects from the results that the employee is dissatisfied with their job. At the same time, if the attendance data shows frequent tardiness, that employee is identified as high risk. Based on these results, the server recommends "implementing flexible working hours" and "providing mental health support" and promptly notifies the administrator's terminal.

[0252] Administrators, as users, can receive notifications from the server and consider conducting interviews with employees or improving the work environment. For example, based on input such as, "Please tell me effective intervention methods to improve employee engagement," the generating AI model provides recommendations. In this way, the entire system works together effectively, enabling rapid responses.

[0253] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0254] Step 1:

[0255] The server collects employee work data, time management data, and text data obtained from engagement surveys from the company's database system. It takes various data stored in the company's database as input and processes that data into a format suitable for the next analysis step as output. Specifically, it uses SQL queries to extract necessary data from databases such as MySQL and PostgreSQL.

[0256] Step 2:

[0257] The server applies natural language processing (NLP) to the collected text data to quantify emotions. It uses the text data processed in step 1 as input and generates an emotion score for each employee as output. Specific operations include tokenizing the text using "NLTK" or "spaCy" and applying an emotion analysis algorithm.

[0258] Step 3:

[0259] The server performs anomaly detection using machine learning (ML) based on business data and time management data. It uses numerical data obtained in Step 1 as input and generates an anomaly score as output. In this process, the server identifies anomalies using linear regression and random forests implemented with "Scikit-learn".

[0260] Step 4:

[0261] The server integrates emotion scores and anomaly scores and performs risk assessment using deep learning techniques. It uses the output data from steps 2 and 3 as input and generates risk assessment results indicating dwell time and stress levels as output. Specifically, it uses neural networks built with TensorFlow or Keras.

[0262] Step 5:

[0263] Based on the risk assessment results, the server identifies high-risk employees and recommends optimal countermeasures using a generative AI model. It uses the risk assessment data obtained in step 4 as input and generates recommendations for improvement measures as output. Specifically, it instructs the generative AI model using the prompt "Please tell me effective intervention methods to improve employee engagement."

[0264] Step 6:

[0265] The server notifies the administrator's terminal of the recommended course of action. Using the recommendations generated in step 5 as input, it creates a notification that the administrator can review as output. Specifically, it utilizes notification APIs such as "Slack" and "Microsoft Teams" to send information to the administrator in real time.

[0266] (Application Example 1)

[0267] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0268] In modern businesses and communities, there is a need to improve the engagement of individual workers and residents. However, traditional methods make it difficult to detect abnormalities in workers' emotions and productivity early on and to respond appropriately. Furthermore, understanding the level of participation in community activities and using that understanding to revitalize local communities is insufficient. Therefore, there is a need for a system that analyzes worker data to efficiently recommend countermeasures and supports resident activities in local communities.

[0269] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0270] In this invention, the server includes a natural language processing means for analyzing workers' text data and quantifying their emotions, a machine learning means for detecting anomalies in workers' work efficiency data and attendance data, and a means for analyzing resident participation data in community activities and recommending measures for community revitalization. This makes it possible to understand the behavioral trends of workers and residents and provide appropriate support aimed at improving engagement.

[0271] A "worker" refers to an individual who engages in specific tasks or duties and receives a salary as compensation for those tasks.

[0272] "Text data" refers to a collection of data in which character information is represented in an electronic format.

[0273] "Natural language processing methods" refer to technologies and methods for understanding and processing human language using computers.

[0274] "Quantifying emotions" is the process of quantitatively evaluating emotions and expressing them as numerical values.

[0275] "Work efficiency data" refers to information regarding the efficiency and productivity of workers when performing their tasks.

[0276] "Attendance data" refers to data that shows information related to employees' arrival and departure times and working hours.

[0277] "Machine learning methods" refer to computational techniques used to learn rules and patterns from data and perform predictions and classifications.

[0278] "Detecting anomalies" means identifying and detecting patterns or behaviors that are different from the norm.

[0279] "Community activities" refer to social or cultural activities that take place within a specific region or community.

[0280] "Resident participation data" refers to data that shows records and information about residents' participation in activities and events held in the region or community.

[0281] "Regional revitalization" refers to initiatives and processes aimed at invigorating the economic and cultural activities of a local community and improving its quality of life.

[0282] "Recommending countermeasures" means suggesting solutions or actions that are appropriate for a specific situation.

[0283] The system implementing this invention is built with the aim of improving the engagement of workers and residents within a smart city. The server collects workers' text data, work efficiency data, and attendance data, and uses natural language processing technology (e.g., NLTK, spaCy) to quantify their emotions. In this process, emotions are classified as positive, negative, or neutral. Furthermore, machine learning (e.g., scikit-learn) is applied to detect anomalies in the work efficiency data and attendance data.

[0284] The server simultaneously acquires the participation data of the residents within the region and conducts analysis for the activation of regional activities. Based on the residents' participation data, it evaluates the participation trends in regional activities and generates recommendations for events that can attract the interest of the residents. The recommended countermeasures are notified to the administrator's terminal or smartphone, supporting the action plan for providing appropriate assistance.

[0285] As a specific example, when the server detects that the frequency of residents' event participation in a specific region is decreasing and the decline in participation motivation is indicated from the sentiment data as the cause, it can be mentioned that new events that the residents are interested in are proposed. Examples of prompt sentences for the generation AI model include "Please describe your opinions on recent citizen events as the content of feedback from citizens." By doing so, it becomes possible to appropriately grasp the needs of residents and workers and improve engagement.

[0286] The flow of the specific process in Application Example 1 will be described using FIG. 12.

[0287] Step 1:

[0288] The server collects the text data, work efficiency data, and attendance data of workers from each system within the company. Based on this input data, using natural language processing technology, it processes the text data to quantify the sentiment of the workers. The output is sentiment data classified as positive, negative, or neutral.

[0289] Step 2:

[0290] The server uses the quantified sentiment data, work efficiency data, and attendance data to detect anomalies by executing a machine learning algorithm. Specifically, it compares with past data to detect a sharp decline in performance or irregularities in attendance. As the output, information on whether an anomaly has been detected is obtained.

[0291] Step 3:

[0292] The server integrates sentiment data and anomaly detection data and uses deep learning technology to perform a risk assessment of workers. This risk assessment identifies workers who are likely to leave the company or who may be experiencing stress. The output is a list indicating the level of risk.

[0293] Step 4:

[0294] The server generates recommended words using an AI model that creates countermeasures for identified high-risk workers. Specifically, concrete measures such as "flexible working hours" and "mental health support" are suggested. The output is a list of recommended countermeasures.

[0295] Step 5:

[0296] The server notifies the administrator's terminal of the recommended countermeasures. The terminal helps the administrator, upon receiving the notification, take appropriate action regarding the worker in the right time. The output is the notification message received by the administrator.

[0297] Step 6:

[0298] The server collects participation data from local residents and analyzes the data to evaluate participation trends. This analysis detects declines in the frequency of participation in local events and interest. The output is the result of the participation analysis.

[0299] Step 7:

[0300] The server generates recommendation text using an AI model to create new events designed to attract residents' interest. Specifically, it can suggest local cultural activities and sports events. The output is a list of recommended event suggestions.

[0301] Step 8:

[0302] Based on these results, the server presents specific event plans for residents and sends notifications aimed at regional revitalization. These notifications are sent to the terminals of residents and communities. The output is the notification of the event plan to the residents.

[0303] Furthermore, an emotion engine for estimating the user's emotion may be combined. That is, the specific processing unit 290 may estimate the user's emotion using the emotion specific model 59 and perform specific processing using the user's emotion.

[0304] The present invention relates to a system that combines an emotion engine for the purpose of improving employee engagement. By comprehensively evaluating the state of employees and recommending appropriate countermeasures to managers, it aims to improve the workplace environment.

[0305] First, the server collects various data related to employees. This includes productivity data, attendance data, and the results of engagement surveys. In parallel, the emotion engine analyzes the user's voice and expressions to recognize emotions in real time. By this operation, the user's emotional state is identified, and the data is integrated with the survey results.

[0306] Next, the server analyzes the descriptive data of the survey through a natural language processing (NLP) model to quantify the emotions of employees. This enables the identification of specific emotional trends and a deeper understanding of the state of employees. Furthermore, machine learning (ML) models are used to analyze productivity and attendance data to detect abnormal patterns.

[0307] The analyzed emotion data and anomaly detection data are integrated on the server using deep learning technology for risk assessment. The user's emotion information obtained by the emotion engine is reflected in the recommendation of countermeasures to improve the accuracy of the recommendation.

[0308] As a concrete example, suppose an employee's productivity drops sharply, and a survey indicates dissatisfaction. In this case, the emotion engine recognizes the user's anxiety and stress, and the server recommends countermeasures, including suggestions for flexible working hours and stress management workshops. These recommendations are sent from the server to the administrator's terminal, allowing the user to take appropriate action based on the employee's situation.

[0309] Thus, the system of the present invention allows for a more accurate understanding of employees' conditions and enables timely and effective responses. This makes it possible to improve overall workplace engagement and productivity.

[0310] The following describes the processing flow.

[0311] Step 1:

[0312] The server collects employee productivity data, attendance data, and engagement survey results from various data sources. This data is automatically retrieved from daily operations systems and HR systems and stored in a database.

[0313] Step 2:

[0314] The emotion engine analyzes the user's voice and facial expression data to identify their emotional state in real time. The emotion engine acquires this data through microphones and cameras and performs analysis using emotion recognition algorithms.

[0315] Step 3:

[0316] The server uses a natural language processing (NLP) model to analyze the descriptive data from the survey. In this process, employee emotions are quantified as positive, negative, or neutral and recorded in a database. This allows for a quantitative evaluation of emotional tendencies.

[0317] Step 4:

[0318] The server uses machine learning (ML) models to analyze productivity and attendance data. It compares this data with historical data to detect anomalous patterns that deviate from the normal range. Once an anomaly is identified, the information is recorded in the database.

[0319] Step 5:

[0320] The server utilizes deep learning technology to integrate sentiment data and anomaly detection data, and uses this information to perform risk assessments of employees. User information from the sentiment engine is also reflected in the risk assessment, improving the accuracy of the decisions.

[0321] Step 6:

[0322] The server generates countermeasures for high-risk employees. These countermeasures include adjustments to working hours and suggestions for stress management programs, and are tailored to take into account the user's emotional information recognized by the emotion engine.

[0323] Step 7:

[0324] The server notifies the administrator's terminal of the generated countermeasures. The administrator user reviews the recommendations on their terminal and takes specific actions as needed, such as scheduling interviews with employees.

[0325] Step 8:

[0326] Based on recommendations, users provide appropriate support, such as direct interaction with employees. Prompt action contributes to reducing employee stress and improving the work environment.

[0327] (Example 2)

[0328] Next, we will describe Example 2. 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".

[0329] Understanding the relationship between employees' emotional states and workplace performance, and identifying optimal countermeasures, is crucial for improving the work environment and increasing productivity. However, conventional technologies have not adequately performed precise emotional analysis or anomaly detection, resulting in incomplete countermeasures. To address these challenges, there is a need for multifaceted emotional assessment of employees, automated risk assessment, and the provision of appropriate countermeasures.

[0330] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0331] In this invention, the server includes means for analyzing employees' voices and facial expressions to identify their emotions, natural language processing means for analyzing employees' text data to quantify their emotions, and machine learning means for detecting anomalies in employees' productivity data and attendance data. This enables a multifaceted analysis of employees' emotions and work data, allowing for risk assessment and the proposal of optimal countermeasures.

[0332] "Means for identifying emotions by analyzing voice and facial expressions" refers to technology that analyzes an employee's voice and facial expressions to identify their emotional state in real time.

[0333] "Natural language processing method that analyzes text data to quantify emotions" is a technology that analyzes text data written by employees, extracts emotions from its content, and expresses them as quantitative numerical values.

[0334] "Machine learning methods for detecting anomalies in productivity data and attendance data" refers to machine learning algorithms used to analyze data related to employee productivity and working hours to detect unusual patterns or anomalies.

[0335] A "deep learning method for integrating data and assessing risk" is a technique that uses deep learning technology to integrate information obtained from multiple data sources and comprehensively assess employee-related risks.

[0336] "Methods for proposing countermeasures to high-risk employees" refers to technologies for generating and proposing specific and appropriate countermeasures to high-risk employees identified from integrated data.

[0337] The "means of notifying administrators of proposals" refers to a function that notifies administrators of the generated proposed countermeasures and aims to encourage appropriate action.

[0338] This invention relates to a system for improving employee engagement in the workplace environment. This system combines voice and facial expression analysis technology, natural language processing technology, machine learning technology, and deep learning technology to comprehensively evaluate employees' emotional states and recommend appropriate countermeasures to managers, thereby improving the workplace.

[0339] Hardware and software

[0340] The server is equipped with sensors and cameras to collect employee voices and facial expressions in real time. Acoustic analysis software is used for voice analysis, and video analysis software is used for facial expression analysis. Generative AI models are used for natural language processing to convert employee text data into sentiment scores. Machine learning algorithms are introduced to detect anomalies in productivity and attendance data. A deep learning framework is used to integrate this data and propose risk assessments and countermeasures.

[0341] Specific example

[0342] As a concrete example, consider a scenario where the server detects a sharp decline in employee A's productivity within a certain project. In this case, the administrator receives information via a terminal that employee A had recently shown signs of stress in voice analysis. Furthermore, analysis reveals that dissatisfaction has been extracted from the free-response answers of a survey. Based on this information, the server recommends implementing mental health counseling and introducing flexible working hours for employee A.

[0343] Example of a prompt

[0344] By using prompts such as "How can you reduce stress from the current workplace atmosphere?" as input to the AI ​​model, more specific and individualized solutions can be generated.

[0345] This system allows for a comprehensive assessment of employee emotions and performance, enabling responses tailored to individual needs, thereby improving overall workplace engagement and productivity.

[0346] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0347] Step 1:

[0348] The server collects employee voice and facial expression data using sensors and cameras. Inputs include audio files and video footage captured in real time from microphones and cameras. Audio data is analyzed using acoustic analysis software, and facial expression data is processed using video analysis software. The output is digital data indicating the employee's emotional state.

[0349] Step 2:

[0350] The server analyzes employee text data using a generated AI model. The input consists of written survey responses and communication logs. Using natural language processing technology, it extracts emotions from this text data and outputs a numerical emotion score.

[0351] Step 3:

[0352] The server collects employee productivity and attendance data and analyzes it using machine learning algorithms. Inputs include work hours, project progress, and attendance records. Based on this data, it detects anomalous patterns and provides anomaly detection results as output.

[0353] Step 4:

[0354] The server integrates analyzed sentiment data and anomaly detection data using a deep learning framework to assess risk. Inputs are quantified sentiment scores and anomaly detection results. Outputs are information identifying high-risk employees as a result of the risk assessment.

[0355] Step 5:

[0356] The server generates countermeasures based on the risk assessment results and notifies the administrator. For example, the prompt "What interventions would be effective for stress management in this situation?" is input into the AI ​​model to obtain appropriate recommendations. The output is a list of countermeasures sent to the administrator's terminal.

[0357] In this way, the system comprehensively assesses the state of employees and helps optimize the work environment.

[0358] (Application Example 2)

[0359] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0360] In modern workplaces, it is crucial to understand workers' emotions and stress levels in real time and take appropriate measures. However, traditional management methods have made it difficult to accurately grasp workers' emotional states, sometimes delaying the optimization of the work environment. This can lead to problems such as decreased worker productivity and increased turnover. Against this backdrop, there is a need for a system that can quickly and accurately capture workers' emotions and immediately adjust the environment.

[0361] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0362] In this invention, the server includes natural language processing means for analyzing the worker's language data and quantifying their emotions, machine learning means for detecting anomalies in the worker's work data and work schedule data, and deep learning means for integrating the quantified emotion data and anomaly detection data to evaluate risk. This makes it possible to grasp the worker's emotional state in real time and quickly adjust the work environment as needed.

[0363] "Linguistic data" refers to text information generated by workers and serves as fundamental data for analyzing emotional states and thought patterns.

[0364] "Natural language processing" is a technology that allows computers to understand and analyze human language, and is used to quantify emotions.

[0365] "Work data" refers to activity information related to a worker's tasks, and is used to evaluate work efficiency and productivity.

[0366] "Work data" refers to data that shows the working hours and attendance status of workers, and is used for attendance management and pattern anomaly detection.

[0367] "Machine learning" is a technology in which computers learn from data and automatically discover patterns and rules, and is used to detect anomalies.

[0368] "Deep learning" is a technique that uses multi-layered artificial neural networks to perform complex data analysis, and is used to assess risk.

[0369] "Emotion recognition" is a technology that identifies the emotional state of workers from their facial expressions and voice, providing information necessary for environmental adjustments.

[0370] "Environmental adjustment" is the process of optimizing the work environment according to the emotional state and work situation of the workers, with the aim of improving productivity and comfort.

[0371] The server analyzes the worker's language data and uses natural language processing technology to quantify their emotions. In this process, the text data entered by the worker is analyzed in real time, and an emotion score such as positive, negative, or neutral is generated. In addition, work data and attendance data are analyzed using machine learning, and anomalies are detected by comparing them with past data.

[0372] Furthermore, the server utilizes deep learning to integrate sentiment data and anomaly detection data to assess risk. The deep learning model performs a detailed analysis of employee states and identifies high-risk situations.

[0373] Emotion recognition sensors use cameras and microphones to collect the worker's facial expressions and voice, and analyze this data. This allows for a real-time understanding of the user's emotional state. Based on the emotional information collected in this way, the server suggests adjustments to the work environment, specifically such as adjusting the work pace or recommending breaks.

[0374] On the user's device, recommendation information sent from the server is visualized, allowing administrators to make concrete decisions based on this information.

[0375] For example, if a worker's face is monitored by a camera and they show a tired expression for a while, the server can recognize that emotional pattern and recommend to the manager, "Please suggest that the worker take a short break."

[0376] An example of a prompt might be, "What environmental adjustments are recommended based on the employee's monitored emotional state?" This prompt is used in the generative AI model to highly customize the response to the user's needs.

[0377] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0378] Step 1:

[0379] The server collects real-time facial and audio data from the camera and microphone. This data is used as input to analyze the user's emotional state through an emotion recognition sensor. The server preprocesses the collected data using image and audio processing techniques and converts it into a format suitable for the next process.

[0380] Step 2:

[0381] The server applies an emotion recognition algorithm to the pre-processed data to identify the user's emotional state. Specifically, it uses image processing techniques to analyze the user's facial expressions and voice analysis techniques to analyze their tone of voice. This generates a real-time emotion score. This emotion score is stored in a database and used as input for the next step.

[0382] Step 3:

[0383] The server collects worker language data and analyzes it using natural language processing techniques. This data is entered in text format, and the server uses a language model to assign sentiment labels such as positive, negative, and neutral. The resulting quantified sentiment data is then integrated with work data and performance data.

[0384] Step 4:

[0385] The server uses machine learning algorithms to analyze worker work data and attendance data. Past work history and attendance data are used as input. Anomaly detection algorithms identify unusual patterns and anomalies, which are then output as anomaly detection data.

[0386] Step 5:

[0387] The server integrates anomaly detection data and quantified sentiment data using deep learning techniques. In this step, a statistical model is used to perform a risk assessment. The output consists of a risk score and recommended actions based on it, which are stored in a database.

[0388] Step 6:

[0389] Based on the generated risk score and recommended actions, the server generates a prompt message and inputs it into the AI ​​model. The prompt message is structured in the format of "What environmental adjustments are recommended based on the employee's monitored emotional state?" and serves as the basis for the AI ​​model to generate specific countermeasures.

[0390] Step 7:

[0391] The server sends the generated recommended actions to the administrator's terminal. The administrator can use this information to adjust the work environment and provide feedback to workers. The outputted information helps users take quick and effective action.

[0392] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0393] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0394] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0395] [Third Embodiment]

[0396] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0397] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0398] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0399] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0400] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0401] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0402] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0403] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0404] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0405] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0406] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0407] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0408] This invention relates to an innovative system for improving employee engagement. The server first collects various employee data from various systems within the company. This includes productivity data, attendance data, and the results of regularly conducted engagement surveys.

[0409] The server applies natural language processing (NLP) models to analyze collected text data, quantifying employee sentiment. This makes it possible to identify emotional tendencies such as positive, negative, and neutral. Next, the server uses machine learning (ML) models to analyze productivity and attendance data, detecting hidden anomalies within this data. These anomalies may include, for example, sudden declines in performance or frequent tardiness.

[0410] The server then uses deep learning technology to integrate emotional data and anomaly data, and uses this to assess employee risk. This assessment can identify employees who are likely to leave or who may be experiencing stress. The server then recommends the most appropriate course of action for these high-risk employees. The recommended course of action is notified to the administrator's terminal, allowing the administrator to take appropriate measures for the employee at the right time.

[0411] By enabling HR personnel to implement appropriate countermeasures, it becomes possible to improve employee engagement and reduce turnover. For example, suppose a server analyzes an employee's text data and detects that the employee is dissatisfied with their work. Simultaneously, if frequent tardiness is detected in the attendance data, the server identifies the employee as high-risk and recommends the implementation of "flexible working hours" and "mental health support." The user, upon receiving a notification on their device, can then schedule a meeting with the employee based on this information.

[0412] The following describes the processing flow.

[0413] Step 1:

[0414] The server collects employee productivity data, attendance data, and engagement survey results from various company data sources. The collected data is stored in a structured database.

[0415] Step 2:

[0416] The server applies a natural language processing (NLP) model to analyze text-based survey responses. From the analyzed data, it quantifies employees' emotions based on tendencies such as positive, negative, or neutral, and records the results in a database.

[0417] Step 3:

[0418] The server uses machine learning (ML) models to analyze productivity and attendance data and detect anomalies. In particular, it detects significant changes compared to past performance and attendance records and records the anomalies in the database.

[0419] Step 4:

[0420] The server uses deep learning models to integrate and analyze emotional and anomaly data. This helps identify employees at high risk of leaving the company or those who appear to be experiencing stress.

[0421] Step 5:

[0422] The server generates appropriate countermeasures for each high-risk employee, such as suggesting flexible working hours or scheduling counseling sessions. The recommended countermeasures are optimized by referencing an existing database of response strategies.

[0423] Step 6:

[0424] The server notifies the administrator's terminal of the generated action recommendations. The administrator, as a user, checks the notification on their terminal and begins preparing to take the recommended action.

[0425] Step 7:

[0426] Based on the recommendations they receive, users can schedule meetings with employees and, if necessary, take concrete actions such as adjusting job responsibilities or implementing support plans. This can lead to improvements in the work environment and increased employee engagement.

[0427] (Example 1)

[0428] Next, we will describe Example 1. 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."

[0429] Managing employee engagement and reducing turnover are crucial for many organizations. However, quickly identifying changes in employee emotions or abnormalities in work performance can be difficult, sometimes leading to delays in taking appropriate action. There is a need for a system that accurately captures changes in employee emotional states and performance and enables early intervention.

[0430] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0431] In this invention, the server includes a natural language processing means for analyzing employee descriptive data and quantifying emotions, a machine learning means for detecting anomalies in employee work data and time management data, and a deep learning means for integrating the quantified emotion data and anomaly detection data to evaluate risk. This makes it possible to comprehensively evaluate the state of employees, identify risks early, and take countermeasures.

[0432] "Natural language processing" refers to information technology that analyzes text data to quantify the emotions and meanings contained within a document.

[0433] "Machine learning methods" are technologies that analyze employee work data and time management data to detect deviations from normal patterns.

[0434] "Deep learning" is an artificial intelligence technology that integrates quantified emotion data and anomaly detection data to accurately assess employee risk.

[0435] "Generation method" refers to information generation technology used to recommend appropriate countermeasures to employees identified as high-risk.

[0436] "Information provision means" refers to notification technology that provides administrators with necessary information in real time, enabling prompt responses.

[0437] This invention is a comprehensive system for improving the working environment for employees and reducing employee turnover. The server utilizes the company's database system to collect employee work data, time management data, and text data obtained from engagement surveys. A general-purpose database management system can be used in this process.

[0438] The server uses software such as "NLTK" and "spaCy" for natural language processing to analyze text data and quantify emotions such as positive, negative, and neutral. For anomaly detection using machine learning, "Scikit-learn" is used, applying models such as linear regression and random forest. By using deep learning techniques such as "TensorFlow" and "Keras," emotion data and anomaly detection data are integrated to perform highly accurate risk assessments.

[0439] As a concrete example, suppose the server analyzes the text data of an employee in the general affairs department and detects from the results that the employee is dissatisfied with their job. At the same time, if the attendance data shows frequent tardiness, that employee is identified as high risk. Based on these results, the server recommends "implementing flexible working hours" and "providing mental health support" and promptly notifies the administrator's terminal.

[0440] Administrators, as users, can receive notifications from the server and consider conducting interviews with employees or improving the work environment. For example, based on input such as, "Please tell me effective intervention methods to improve employee engagement," the generating AI model provides recommendations. In this way, the entire system works together effectively, enabling rapid responses.

[0441] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0442] Step 1:

[0443] The server collects employee work data, time management data, and text data obtained from engagement surveys from the company's database system. It takes various data stored in the company's database as input and processes that data into a format suitable for the next analysis step as output. Specifically, it uses SQL queries to extract necessary data from databases such as MySQL and PostgreSQL.

[0444] Step 2:

[0445] The server applies natural language processing (NLP) to the collected text data to quantify emotions. It uses the text data processed in step 1 as input and generates an emotion score for each employee as output. Specific operations include tokenizing the text using "NLTK" or "spaCy" and applying an emotion analysis algorithm.

[0446] Step 3:

[0447] The server performs anomaly detection using machine learning (ML) based on business data and time management data. It uses numerical data obtained in Step 1 as input and generates an anomaly score as output. In this process, the server identifies anomalies using linear regression and random forests implemented with "Scikit-learn".

[0448] Step 4:

[0449] The server integrates emotion scores and anomaly scores and performs risk assessment using deep learning techniques. It uses the output data from steps 2 and 3 as input and generates risk assessment results indicating dwell time and stress levels as output. Specifically, it uses neural networks built with TensorFlow or Keras.

[0450] Step 5:

[0451] Based on the risk assessment results, the server identifies high-risk employees and recommends optimal countermeasures using a generative AI model. It uses the risk assessment data obtained in step 4 as input and generates recommendations for improvement measures as output. Specifically, it instructs the generative AI model using the prompt "Please tell me effective intervention methods to improve employee engagement."

[0452] Step 6:

[0453] The server notifies the administrator's terminal of the recommended course of action. Using the recommendations generated in step 5 as input, it creates a notification that the administrator can review as output. Specifically, it utilizes notification APIs such as "Slack" and "Microsoft Teams" to send information to the administrator in real time.

[0454] (Application Example 1)

[0455] Next, we will explain Application Example 1. In the following explanation, 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."

[0456] In modern businesses and communities, there is a need to improve the engagement of individual workers and residents. However, traditional methods make it difficult to detect abnormalities in workers' emotions and productivity early on and to respond appropriately. Furthermore, understanding the level of participation in community activities and using that understanding to revitalize local communities is insufficient. Therefore, there is a need for a system that analyzes worker data to efficiently recommend countermeasures and supports resident activities in local communities.

[0457] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0458] In this invention, the server includes a natural language processing means for analyzing workers' text data and quantifying their emotions, a machine learning means for detecting anomalies in workers' work efficiency data and attendance data, and a means for analyzing resident participation data in community activities and recommending measures for community revitalization. This makes it possible to understand the behavioral trends of workers and residents and provide appropriate support aimed at improving engagement.

[0459] A "worker" refers to an individual who engages in specific tasks or duties and receives a salary as compensation for those tasks.

[0460] "Text data" refers to a collection of data in which character information is represented in an electronic format.

[0461] "Natural language processing methods" refer to technologies and methods for understanding and processing human language using computers.

[0462] "Quantifying emotions" is the process of quantitatively evaluating emotions and expressing them as numerical values.

[0463] "Work efficiency data" refers to information regarding the efficiency and productivity of workers when performing their tasks.

[0464] "Attendance data" refers to data that shows information related to employees' arrival and departure times and working hours.

[0465] "Machine learning methods" refer to computational techniques used to learn rules and patterns from data and perform predictions and classifications.

[0466] "Detecting anomalies" means identifying and detecting patterns or behaviors that are different from the norm.

[0467] "Community activities" refer to social or cultural activities that take place within a specific region or community.

[0468] "Resident participation data" refers to data that shows records and information about residents' participation in activities and events held in the region or community.

[0469] "Regional revitalization" refers to initiatives and processes aimed at invigorating the economic and cultural activities of a local community and improving its quality of life.

[0470] "Recommending countermeasures" means suggesting solutions or actions that are appropriate for a specific situation.

[0471] The system implementing this invention is built with the aim of improving the engagement of workers and residents within a smart city. The server collects workers' text data, work efficiency data, and attendance data, and uses natural language processing technology (e.g., NLTK, spaCy) to quantify their emotions. In this process, emotions are classified as positive, negative, or neutral. Furthermore, machine learning (e.g., scikit-learn) is applied to detect anomalies in the work efficiency data and attendance data.

[0472] The server simultaneously acquires participation data from residents within the area and analyzes it to revitalize community activities. Based on the resident participation data, it evaluates participation trends in community activities and generates recommendations for events that will attract residents' interest. The recommended actions are notified to the administrator's terminal or smartphone, supporting action plans for providing appropriate assistance.

[0473] For example, if a server detects a decline in residents' event participation in a particular area, and emotional data indicates a decrease in participation motivation, it can then suggest new events that residents might be interested in. An example of a prompt for the generative AI model would be, "Please provide feedback on recent community events as part of the citizen feedback." This makes it possible to appropriately understand the needs of residents and workers and improve engagement.

[0474] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0475] Step 1:

[0476] The server collects employee text data, work efficiency data, and attendance data from various systems within the company. Based on this input data, natural language processing technology is used to quantify the employees' emotions from the text data. The output is emotion data classified as positive, negative, or neutral.

[0477] Step 2:

[0478] The server detects anomalies by running machine learning algorithms using quantified sentiment data, operational efficiency data, and attendance data. Specifically, it compares current data with past data to detect sudden performance declines and attendance irregularities. The output provides information on whether or not an anomaly was detected.

[0479] Step 3:

[0480] The server integrates sentiment data and anomaly detection data and uses deep learning technology to perform a risk assessment of workers. This risk assessment identifies workers who are likely to leave the company or who may be experiencing stress. The output is a list indicating the level of risk.

[0481] Step 4:

[0482] The server generates recommended words using an AI model that creates countermeasures for identified high-risk workers. Specifically, concrete measures such as "flexible working hours" and "mental health support" are suggested. The output is a list of recommended countermeasures.

[0483] Step 5:

[0484] The server notifies the administrator's terminal of the recommended countermeasures. The terminal helps the administrator, upon receiving the notification, take appropriate action regarding the worker in the right time. The output is the notification message received by the administrator.

[0485] Step 6:

[0486] The server collects participation data from local residents and analyzes the data to evaluate participation trends. This analysis detects declines in the frequency of participation in local events and interest. The output is the result of the participation analysis.

[0487] Step 7:

[0488] The server generates recommendation text using an AI model to create new events designed to attract residents' interest. Specifically, it can suggest local cultural activities and sports events. The output is a list of recommended event suggestions.

[0489] Step 8:

[0490] Based on these results, the server presents specific event proposals to residents and sends notifications aimed at revitalizing the community. These notifications are sent to residents' and community's devices. The output is a notification of event proposals to residents.

[0491] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0492] This invention relates to a system that combines an emotion engine with the aim of improving employee engagement. It aims to improve the workplace environment by comprehensively evaluating employee conditions and recommending appropriate countermeasures to managers.

[0493] First, the server collects diverse data about employees, including productivity data, attendance data, and engagement survey results. Simultaneously, an emotion engine analyzes the user's voice and facial expressions to recognize emotions in real time. This process identifies the user's emotional state, and this data is integrated with the survey results.

[0494] Next, the server analyzes the descriptive data from the survey through a natural language processing (NLP) model to quantify employee emotions. This allows for the identification of specific emotional tendencies and a deeper understanding of employee states. Furthermore, machine learning (ML) models are used to analyze productivity and attendance data and detect anomalous patterns.

[0495] The analyzed sentiment data and anomaly detection data are integrated on the server using deep learning technology, and a risk assessment is performed. User sentiment information obtained by the sentiment engine is reflected in the recommendation of countermeasures, improving the accuracy of the recommendations.

[0496] As a concrete example, suppose an employee's productivity drops sharply, and a survey indicates dissatisfaction. In this case, the emotion engine recognizes the user's anxiety and stress, and the server recommends countermeasures, including suggestions for flexible working hours and stress management workshops. These recommendations are sent from the server to the administrator's terminal, allowing the user to take appropriate action based on the employee's situation.

[0497] Thus, the system of the present invention allows for a more accurate understanding of employees' conditions and enables timely and effective responses. This makes it possible to improve overall workplace engagement and productivity.

[0498] The following describes the processing flow.

[0499] Step 1:

[0500] The server collects employee productivity data, attendance data, and engagement survey results from various data sources. This data is automatically retrieved from daily operations systems and HR systems and stored in a database.

[0501] Step 2:

[0502] The emotion engine analyzes the user's voice and facial expression data to identify their emotional state in real time. The emotion engine acquires this data through microphones and cameras and performs analysis using emotion recognition algorithms.

[0503] Step 3:

[0504] The server uses a natural language processing (NLP) model to analyze the descriptive data from the survey. In this process, employee emotions are quantified as positive, negative, or neutral and recorded in a database. This allows for a quantitative evaluation of emotional tendencies.

[0505] Step 4:

[0506] The server uses machine learning (ML) models to analyze productivity and attendance data. It compares this data with historical data to detect anomalous patterns that deviate from the normal range. Once an anomaly is identified, the information is recorded in the database.

[0507] Step 5:

[0508] The server utilizes deep learning technology to integrate sentiment data and anomaly detection data, and uses this information to perform risk assessments of employees. User information from the sentiment engine is also reflected in the risk assessment, improving the accuracy of the decisions.

[0509] Step 6:

[0510] The server generates countermeasures for high-risk employees. These countermeasures include adjustments to working hours and suggestions for stress management programs, and are tailored to take into account the user's emotional information recognized by the emotion engine.

[0511] Step 7:

[0512] The server notifies the administrator's terminal of the generated countermeasures. The administrator user reviews the recommendations on their terminal and takes specific actions as needed, such as scheduling interviews with employees.

[0513] Step 8:

[0514] Based on recommendations, users provide appropriate support, such as direct interaction with employees. Prompt action contributes to reducing employee stress and improving the work environment.

[0515] (Example 2)

[0516] Next, we will describe Example 2. 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."

[0517] Understanding the relationship between employees' emotional states and workplace performance, and identifying optimal countermeasures, is crucial for improving the work environment and increasing productivity. However, conventional technologies have not adequately performed precise emotional analysis or anomaly detection, resulting in incomplete countermeasures. To address these challenges, there is a need for multifaceted emotional assessment of employees, automated risk assessment, and the provision of appropriate countermeasures.

[0518] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0519] In this invention, the server includes means for analyzing employees' voices and facial expressions to identify their emotions, natural language processing means for analyzing employees' text data to quantify their emotions, and machine learning means for detecting anomalies in employees' productivity data and attendance data. This enables a multifaceted analysis of employees' emotions and work data, allowing for risk assessment and the proposal of optimal countermeasures.

[0520] "Means for identifying emotions by analyzing voice and facial expressions" refers to technology that analyzes an employee's voice and facial expressions to identify their emotional state in real time.

[0521] "Natural language processing method that analyzes text data to quantify emotions" is a technology that analyzes text data written by employees, extracts emotions from its content, and expresses them as quantitative numerical values.

[0522] "Machine learning methods for detecting anomalies in productivity data and attendance data" refers to machine learning algorithms used to analyze data related to employee productivity and working hours to detect unusual patterns or anomalies.

[0523] A "deep learning method for integrating data and assessing risk" is a technique that uses deep learning technology to integrate information obtained from multiple data sources and comprehensively assess employee-related risks.

[0524] "Methods for proposing countermeasures to high-risk employees" refers to technologies for generating and proposing specific and appropriate countermeasures to high-risk employees identified from integrated data.

[0525] The "means of notifying administrators of proposals" refers to a function that notifies administrators of the generated proposed countermeasures and aims to encourage appropriate action.

[0526] This invention relates to a system for improving employee engagement in the workplace environment. This system combines voice and facial expression analysis technology, natural language processing technology, machine learning technology, and deep learning technology to comprehensively evaluate employees' emotional states and recommend appropriate countermeasures to managers, thereby improving the workplace.

[0527] Hardware and software

[0528] The server is equipped with sensors and cameras to collect employee voices and facial expressions in real time. Acoustic analysis software is used for voice analysis, and video analysis software is used for facial expression analysis. Generative AI models are used for natural language processing to convert employee text data into sentiment scores. Machine learning algorithms are introduced to detect anomalies in productivity and attendance data. A deep learning framework is used to integrate this data and propose risk assessments and countermeasures.

[0529] Specific example

[0530] As a concrete example, consider a scenario where the server detects a sharp decline in employee A's productivity within a certain project. In this case, the administrator receives information via a terminal that employee A had recently shown signs of stress in voice analysis. Furthermore, analysis reveals that dissatisfaction has been extracted from the free-response answers of a survey. Based on this information, the server recommends implementing mental health counseling and introducing flexible working hours for employee A.

[0531] Example of a prompt

[0532] By using prompts such as "How can you reduce stress from the current workplace atmosphere?" as input to the AI ​​model, more specific and individualized solutions can be generated.

[0533] This system allows for a comprehensive assessment of employee emotions and performance, enabling responses tailored to individual needs, thereby improving overall workplace engagement and productivity.

[0534] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0535] Step 1:

[0536] The server collects employee voice and facial expression data using sensors and cameras. Inputs include audio files and video footage captured in real time from microphones and cameras. Audio data is analyzed using acoustic analysis software, and facial expression data is processed using video analysis software. The output is digital data indicating the employee's emotional state.

[0537] Step 2:

[0538] The server analyzes employee text data using a generated AI model. The input consists of written survey responses and communication logs. Using natural language processing technology, it extracts emotions from this text data and outputs a numerical emotion score.

[0539] Step 3:

[0540] The server collects employee productivity and attendance data and analyzes it using machine learning algorithms. Inputs include work hours, project progress, and attendance records. Based on this data, it detects anomalous patterns and provides anomaly detection results as output.

[0541] Step 4:

[0542] The server integrates analyzed sentiment data and anomaly detection data using a deep learning framework to assess risk. Inputs are quantified sentiment scores and anomaly detection results. Outputs are information identifying high-risk employees as a result of the risk assessment.

[0543] Step 5:

[0544] The server generates countermeasures based on the risk assessment results and notifies the administrator. For example, the prompt "What interventions would be effective for stress management in this situation?" is input into the AI ​​model to obtain appropriate recommendations. The output is a list of countermeasures sent to the administrator's terminal.

[0545] In this way, the system comprehensively assesses the state of employees and helps optimize the work environment.

[0546] (Application Example 2)

[0547] Next, we will explain application example 2. In the following explanation, 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."

[0548] In modern workplaces, it is crucial to understand workers' emotions and stress levels in real time and take appropriate measures. However, traditional management methods have made it difficult to accurately grasp workers' emotional states, sometimes delaying the optimization of the work environment. This can lead to problems such as decreased worker productivity and increased turnover. Against this backdrop, there is a need for a system that can quickly and accurately capture workers' emotions and immediately adjust the environment.

[0549] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0550] In this invention, the server includes natural language processing means for analyzing the worker's language data and quantifying their emotions, machine learning means for detecting anomalies in the worker's work data and work schedule data, and deep learning means for integrating the quantified emotion data and anomaly detection data to evaluate risk. This makes it possible to grasp the worker's emotional state in real time and quickly adjust the work environment as needed.

[0551] "Linguistic data" refers to text information generated by workers and serves as fundamental data for analyzing emotional states and thought patterns.

[0552] "Natural language processing" is a technology that allows computers to understand and analyze human language, and is used to quantify emotions.

[0553] "Work data" refers to activity information related to a worker's tasks, and is used to evaluate work efficiency and productivity.

[0554] "Work data" refers to data that shows the working hours and attendance status of workers, and is used for attendance management and pattern anomaly detection.

[0555] "Machine learning" is a technology in which computers learn from data and automatically discover patterns and rules, and is used to detect anomalies.

[0556] "Deep learning" is a technique that uses multi-layered artificial neural networks to perform complex data analysis, and is used to assess risk.

[0557] "Emotion recognition" is a technology that identifies the emotional state of workers from their facial expressions and voice, providing information necessary for environmental adjustments.

[0558] "Environmental adjustment" is the process of optimizing the work environment according to the emotional state and work situation of the workers, with the aim of improving productivity and comfort.

[0559] The server analyzes the worker's language data and uses natural language processing technology to quantify their emotions. In this process, the text data entered by the worker is analyzed in real time, and an emotion score such as positive, negative, or neutral is generated. In addition, work data and attendance data are analyzed using machine learning, and anomalies are detected by comparing them with past data.

[0560] Furthermore, the server utilizes deep learning to integrate sentiment data and anomaly detection data to assess risk. The deep learning model performs a detailed analysis of employee states and identifies high-risk situations.

[0561] Emotion recognition sensors use cameras and microphones to collect the worker's facial expressions and voice, and analyze this data. This allows for a real-time understanding of the user's emotional state. Based on the emotional information collected in this way, the server suggests adjustments to the work environment, specifically such as adjusting the work pace or recommending breaks.

[0562] On the user's device, recommendation information sent from the server is visualized, allowing administrators to make concrete decisions based on this information.

[0563] For example, if a worker's face is monitored by a camera and they show a tired expression for a while, the server can recognize that emotional pattern and recommend to the manager, "Please suggest that the worker take a short break."

[0564] An example of a prompt might be, "What environmental adjustments are recommended based on the employee's monitored emotional state?" This prompt is used in the generative AI model to highly customize the response to the user's needs.

[0565] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0566] Step 1:

[0567] The server collects real-time facial and audio data from the camera and microphone. This data is used as input to analyze the user's emotional state through an emotion recognition sensor. The server preprocesses the collected data using image and audio processing techniques and converts it into a format suitable for the next process.

[0568] Step 2:

[0569] The server applies an emotion recognition algorithm to the pre-processed data to identify the user's emotional state. Specifically, it uses image processing techniques to analyze the user's facial expressions and voice analysis techniques to analyze their tone of voice. This generates a real-time emotion score. This emotion score is stored in a database and used as input for the next step.

[0570] Step 3:

[0571] The server collects worker language data and analyzes it using natural language processing techniques. This data is entered in text format, and the server uses a language model to assign sentiment labels such as positive, negative, and neutral. The resulting quantified sentiment data is then integrated with work data and performance data.

[0572] Step 4:

[0573] The server uses machine learning algorithms to analyze worker work data and attendance data. Past work history and attendance data are used as input. Anomaly detection algorithms identify unusual patterns and anomalies, which are then output as anomaly detection data.

[0574] Step 5:

[0575] The server integrates anomaly detection data and quantified sentiment data using deep learning techniques. In this step, a statistical model is used to perform a risk assessment. The output consists of a risk score and recommended actions based on it, which are stored in a database.

[0576] Step 6:

[0577] Based on the generated risk score and recommended actions, the server generates a prompt message and inputs it into the AI ​​model. The prompt message is structured in the format of "What environmental adjustments are recommended based on the employee's monitored emotional state?" and serves as the basis for the AI ​​model to generate specific countermeasures.

[0578] Step 7:

[0579] The server sends the generated recommended actions to the administrator's terminal. The administrator can use this information to adjust the work environment and provide feedback to workers. The outputted information helps users take quick and effective action.

[0580] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0581] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0582] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0583] [Fourth Embodiment]

[0584] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0585] As shown in Figure 7, the 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.

[0586] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0587] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0588] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0589] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0590] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0591] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive 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 robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0592] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0593] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0594] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0595] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0596] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0597] This invention relates to an innovative system for improving employee engagement. The server first collects various employee data from various systems within the company. This includes productivity data, attendance data, and the results of regularly conducted engagement surveys.

[0598] The server applies natural language processing (NLP) models to analyze collected text data, quantifying employee sentiment. This makes it possible to identify emotional tendencies such as positive, negative, and neutral. Next, the server uses machine learning (ML) models to analyze productivity and attendance data, detecting hidden anomalies within this data. These anomalies may include, for example, sudden declines in performance or frequent tardiness.

[0599] The server then uses deep learning technology to integrate emotional data and anomaly data, and uses this to assess employee risk. This assessment can identify employees who are likely to leave or who may be experiencing stress. The server then recommends the most appropriate course of action for these high-risk employees. The recommended course of action is notified to the administrator's terminal, allowing the administrator to take appropriate measures for the employee at the right time.

[0600] By enabling HR personnel to implement appropriate countermeasures, it becomes possible to improve employee engagement and reduce turnover. For example, suppose a server analyzes an employee's text data and detects that the employee is dissatisfied with their work. Simultaneously, if frequent tardiness is detected in the attendance data, the server identifies the employee as high-risk and recommends the implementation of "flexible working hours" and "mental health support." The user, upon receiving a notification on their device, can then schedule a meeting with the employee based on this information.

[0601] The following describes the processing flow.

[0602] Step 1:

[0603] The server collects employee productivity data, attendance data, and engagement survey results from various company data sources. The collected data is stored in a structured database.

[0604] Step 2:

[0605] The server applies a natural language processing (NLP) model to analyze text-based survey responses. From the analyzed data, it quantifies employees' emotions based on tendencies such as positive, negative, or neutral, and records the results in a database.

[0606] Step 3:

[0607] The server uses machine learning (ML) models to analyze productivity and attendance data and detect anomalies. In particular, it detects significant changes compared to past performance and attendance records and records the anomalies in the database.

[0608] Step 4:

[0609] The server uses deep learning models to integrate and analyze emotional and anomaly data. This helps identify employees at high risk of leaving the company or those who appear to be experiencing stress.

[0610] Step 5:

[0611] The server generates appropriate countermeasures for each high-risk employee, such as suggesting flexible working hours or scheduling counseling sessions. The recommended countermeasures are optimized by referencing an existing database of response strategies.

[0612] Step 6:

[0613] The server notifies the administrator's terminal of the generated action recommendations. The administrator, as a user, checks the notification on their terminal and begins preparing to take the recommended action.

[0614] Step 7:

[0615] Based on the recommendations they receive, users can schedule meetings with employees and, if necessary, take concrete actions such as adjusting job responsibilities or implementing support plans. This can lead to improvements in the work environment and increased employee engagement.

[0616] (Example 1)

[0617] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0618] Managing employee engagement and reducing turnover are crucial for many organizations. However, quickly identifying changes in employee emotions or abnormalities in work performance can be difficult, sometimes leading to delays in taking appropriate action. There is a need for a system that accurately captures changes in employee emotional states and performance and enables early intervention.

[0619] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0620] In this invention, the server includes a natural language processing means for analyzing employee descriptive data and quantifying emotions, a machine learning means for detecting anomalies in employee work data and time management data, and a deep learning means for integrating the quantified emotion data and anomaly detection data to evaluate risk. This makes it possible to comprehensively evaluate the state of employees, identify risks early, and take countermeasures.

[0621] "Natural language processing" refers to information technology that analyzes text data to quantify the emotions and meanings contained within a document.

[0622] "Machine learning methods" are technologies that analyze employee work data and time management data to detect deviations from normal patterns.

[0623] "Deep learning" is an artificial intelligence technology that integrates quantified emotion data and anomaly detection data to accurately assess employee risk.

[0624] "Generation method" refers to information generation technology used to recommend appropriate countermeasures to employees identified as high-risk.

[0625] "Information provision means" refers to notification technology that provides administrators with necessary information in real time, enabling prompt responses.

[0626] This invention is a comprehensive system for improving the working environment for employees and reducing employee turnover. The server utilizes the company's database system to collect employee work data, time management data, and text data obtained from engagement surveys. A general-purpose database management system can be used in this process.

[0627] The server uses software such as "NLTK" and "spaCy" for natural language processing to analyze text data and quantify emotions such as positive, negative, and neutral. For anomaly detection using machine learning, "Scikit-learn" is used, applying models such as linear regression and random forest. By using deep learning techniques such as "TensorFlow" and "Keras," emotion data and anomaly detection data are integrated to perform highly accurate risk assessments.

[0628] As a concrete example, suppose the server analyzes the text data of an employee in the general affairs department and detects from the results that the employee is dissatisfied with their job. At the same time, if the attendance data shows frequent tardiness, that employee is identified as high risk. Based on these results, the server recommends "implementing flexible working hours" and "providing mental health support" and promptly notifies the administrator's terminal.

[0629] Administrators, as users, can receive notifications from the server and consider conducting interviews with employees or improving the work environment. For example, based on input such as, "Please tell me effective intervention methods to improve employee engagement," the generating AI model provides recommendations. In this way, the entire system works together effectively, enabling rapid responses.

[0630] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0631] Step 1:

[0632] The server collects employee work data, time management data, and text data obtained from engagement surveys from the company's database system. It takes various data stored in the company's database as input and processes that data into a format suitable for the next analysis step as output. Specifically, it uses SQL queries to extract necessary data from databases such as MySQL and PostgreSQL.

[0633] Step 2:

[0634] The server applies natural language processing (NLP) to the collected text data to quantify emotions. It uses the text data processed in step 1 as input and generates an emotion score for each employee as output. Specific operations include tokenizing the text using "NLTK" or "spaCy" and applying an emotion analysis algorithm.

[0635] Step 3:

[0636] The server performs anomaly detection using machine learning (ML) based on business data and time management data. It uses numerical data obtained in Step 1 as input and generates an anomaly score as output. In this process, the server identifies anomalies using linear regression and random forests implemented with "Scikit-learn".

[0637] Step 4:

[0638] The server integrates emotion scores and anomaly scores and performs risk assessment using deep learning techniques. It uses the output data from steps 2 and 3 as input and generates risk assessment results indicating dwell time and stress levels as output. Specifically, it uses neural networks built with TensorFlow or Keras.

[0639] Step 5:

[0640] Based on the risk assessment results, the server identifies high-risk employees and recommends optimal countermeasures using a generative AI model. It uses the risk assessment data obtained in step 4 as input and generates recommendations for improvement measures as output. Specifically, it instructs the generative AI model using the prompt "Please tell me effective intervention methods to improve employee engagement."

[0641] Step 6:

[0642] The server notifies the administrator's terminal of the recommended course of action. Using the recommendations generated in step 5 as input, it creates a notification that the administrator can review as output. Specifically, it utilizes notification APIs such as "Slack" and "Microsoft Teams" to send information to the administrator in real time.

[0643] (Application Example 1)

[0644] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0645] In modern businesses and communities, there is a need to improve the engagement of individual workers and residents. However, traditional methods make it difficult to detect abnormalities in workers' emotions and productivity early on and to respond appropriately. Furthermore, understanding the level of participation in community activities and using that understanding to revitalize local communities is insufficient. Therefore, there is a need for a system that analyzes worker data to efficiently recommend countermeasures and supports resident activities in local communities.

[0646] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0647] In this invention, the server includes a natural language processing means for analyzing workers' text data and quantifying their emotions, a machine learning means for detecting anomalies in workers' work efficiency data and attendance data, and a means for analyzing resident participation data in community activities and recommending measures for community revitalization. This makes it possible to understand the behavioral trends of workers and residents and provide appropriate support aimed at improving engagement.

[0648] A "worker" refers to an individual who engages in specific tasks or duties and receives a salary as compensation for those tasks.

[0649] "Text data" refers to a collection of data in which character information is represented in an electronic format.

[0650] "Natural language processing methods" refer to technologies and methods for understanding and processing human language using computers.

[0651] "Quantifying emotions" is the process of quantitatively evaluating emotions and expressing them as numerical values.

[0652] "Work efficiency data" refers to information regarding the efficiency and productivity of workers when performing their tasks.

[0653] "Attendance data" refers to data that shows information related to employees' arrival and departure times and working hours.

[0654] "Machine learning methods" refer to computational techniques used to learn rules and patterns from data and perform predictions and classifications.

[0655] "Detecting anomalies" means identifying and detecting patterns or behaviors that are different from the norm.

[0656] "Community activities" refer to social or cultural activities that take place within a specific region or community.

[0657] "Resident participation data" refers to data that shows records and information about residents' participation in activities and events held in the region or community.

[0658] "Regional revitalization" refers to initiatives and processes aimed at invigorating the economic and cultural activities of a local community and improving its quality of life.

[0659] "Recommending countermeasures" means suggesting solutions or actions that are appropriate for a specific situation.

[0660] The system implementing this invention is built with the aim of improving the engagement of workers and residents within a smart city. The server collects workers' text data, work efficiency data, and attendance data, and uses natural language processing technology (e.g., NLTK, spaCy) to quantify their emotions. In this process, emotions are classified as positive, negative, or neutral. Furthermore, machine learning (e.g., scikit-learn) is applied to detect anomalies in the work efficiency data and attendance data.

[0661] The server simultaneously acquires participation data from residents within the area and analyzes it to revitalize community activities. Based on the resident participation data, it evaluates participation trends in community activities and generates recommendations for events that will attract residents' interest. The recommended actions are notified to the administrator's terminal or smartphone, supporting action plans for providing appropriate assistance.

[0662] For example, if a server detects a decline in residents' event participation in a particular area, and emotional data indicates a decrease in participation motivation, it can then suggest new events that residents might be interested in. An example of a prompt for the generative AI model would be, "Please provide feedback on recent community events as part of the citizen feedback." This makes it possible to appropriately understand the needs of residents and workers and improve engagement.

[0663] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0664] Step 1:

[0665] The server collects employee text data, work efficiency data, and attendance data from various systems within the company. Based on this input data, natural language processing technology is used to quantify the employees' emotions from the text data. The output is emotion data classified as positive, negative, or neutral.

[0666] Step 2:

[0667] The server detects anomalies by running machine learning algorithms using quantified sentiment data, operational efficiency data, and attendance data. Specifically, it compares current data with past data to detect sudden performance declines and attendance irregularities. The output provides information on whether or not an anomaly was detected.

[0668] Step 3:

[0669] The server integrates sentiment data and anomaly detection data and uses deep learning technology to perform a risk assessment of workers. This risk assessment identifies workers who are likely to leave the company or who may be experiencing stress. The output is a list indicating the level of risk.

[0670] Step 4:

[0671] The server generates recommended words using an AI model that creates countermeasures for identified high-risk workers. Specifically, concrete measures such as "flexible working hours" and "mental health support" are suggested. The output is a list of recommended countermeasures.

[0672] Step 5:

[0673] The server notifies the administrator's terminal of the recommended countermeasures. The terminal helps the administrator, upon receiving the notification, take appropriate action regarding the worker in the right time. The output is the notification message received by the administrator.

[0674] Step 6:

[0675] The server collects participation data from local residents and analyzes the data to evaluate participation trends. This analysis detects declines in the frequency of participation in local events and interest. The output is the result of the participation analysis.

[0676] Step 7:

[0677] The server generates recommendation text using an AI model to create new events designed to attract residents' interest. Specifically, it can suggest local cultural activities and sports events. The output is a list of recommended event suggestions.

[0678] Step 8:

[0679] Based on these results, the server presents specific event proposals to residents and sends notifications aimed at revitalizing the community. These notifications are sent to residents' and community's devices. The output is a notification of event proposals to residents.

[0680] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0681] This invention relates to a system that combines an emotion engine with the aim of improving employee engagement. It aims to improve the workplace environment by comprehensively evaluating employee conditions and recommending appropriate countermeasures to managers.

[0682] First, the server collects diverse data about employees, including productivity data, attendance data, and engagement survey results. Simultaneously, an emotion engine analyzes the user's voice and facial expressions to recognize emotions in real time. This process identifies the user's emotional state, and this data is integrated with the survey results.

[0683] Next, the server analyzes the descriptive data from the survey through a natural language processing (NLP) model to quantify employee emotions. This allows for the identification of specific emotional tendencies and a deeper understanding of employee states. Furthermore, machine learning (ML) models are used to analyze productivity and attendance data and detect anomalous patterns.

[0684] The analyzed sentiment data and anomaly detection data are integrated on the server using deep learning technology, and a risk assessment is performed. User sentiment information obtained by the sentiment engine is reflected in the recommendation of countermeasures, improving the accuracy of the recommendations.

[0685] As a concrete example, suppose an employee's productivity drops sharply, and a survey indicates dissatisfaction. In this case, the emotion engine recognizes the user's anxiety and stress, and the server recommends countermeasures, including suggestions for flexible working hours and stress management workshops. These recommendations are sent from the server to the administrator's terminal, allowing the user to take appropriate action based on the employee's situation.

[0686] Thus, the system of the present invention allows for a more accurate understanding of employees' conditions and enables timely and effective responses. This makes it possible to improve overall workplace engagement and productivity.

[0687] The following describes the processing flow.

[0688] Step 1:

[0689] The server collects employee productivity data, attendance data, and engagement survey results from various data sources. This data is automatically retrieved from daily operations systems and HR systems and stored in a database.

[0690] Step 2:

[0691] The emotion engine analyzes the user's voice and facial expression data to identify their emotional state in real time. The emotion engine acquires this data through microphones and cameras and performs analysis using emotion recognition algorithms.

[0692] Step 3:

[0693] The server uses a natural language processing (NLP) model to analyze the descriptive data from the survey. In this process, employee emotions are quantified as positive, negative, or neutral and recorded in a database. This allows for a quantitative evaluation of emotional tendencies.

[0694] Step 4:

[0695] The server uses machine learning (ML) models to analyze productivity and attendance data. It compares this data with historical data to detect anomalous patterns that deviate from the normal range. Once an anomaly is identified, the information is recorded in the database.

[0696] Step 5:

[0697] The server utilizes deep learning technology to integrate sentiment data and anomaly detection data, and uses this information to perform risk assessments of employees. User information from the sentiment engine is also reflected in the risk assessment, improving the accuracy of the decisions.

[0698] Step 6:

[0699] The server generates countermeasures for high-risk employees. These countermeasures include adjustments to working hours and suggestions for stress management programs, and are tailored to take into account the user's emotional information recognized by the emotion engine.

[0700] Step 7:

[0701] The server notifies the administrator's terminal of the generated countermeasures. The administrator user reviews the recommendations on their terminal and takes specific actions as needed, such as scheduling interviews with employees.

[0702] Step 8:

[0703] Based on recommendations, users provide appropriate support, such as direct interaction with employees. Prompt action contributes to reducing employee stress and improving the work environment.

[0704] (Example 2)

[0705] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0706] Understanding the relationship between employees' emotional states and workplace performance, and identifying optimal countermeasures, is crucial for improving the work environment and increasing productivity. However, conventional technologies have not adequately performed precise emotional analysis or anomaly detection, resulting in incomplete countermeasures. To address these challenges, there is a need for multifaceted emotional assessment of employees, automated risk assessment, and the provision of appropriate countermeasures.

[0707] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0708] In this invention, the server includes means for analyzing employees' voices and facial expressions to identify their emotions, natural language processing means for analyzing employees' text data to quantify their emotions, and machine learning means for detecting anomalies in employees' productivity data and attendance data. This enables a multifaceted analysis of employees' emotions and work data, allowing for risk assessment and the proposal of optimal countermeasures.

[0709] "Means for identifying emotions by analyzing voice and facial expressions" refers to technology that analyzes an employee's voice and facial expressions to identify their emotional state in real time.

[0710] "Natural language processing method that analyzes text data to quantify emotions" is a technology that analyzes text data written by employees, extracts emotions from its content, and expresses them as quantitative numerical values.

[0711] "Machine learning methods for detecting anomalies in productivity data and attendance data" refers to machine learning algorithms used to analyze data related to employee productivity and working hours to detect unusual patterns or anomalies.

[0712] A "deep learning method for integrating data and assessing risk" is a technique that uses deep learning technology to integrate information obtained from multiple data sources and comprehensively assess employee-related risks.

[0713] "Methods for proposing countermeasures to high-risk employees" refers to technologies for generating and proposing specific and appropriate countermeasures to high-risk employees identified from integrated data.

[0714] The "means of notifying administrators of proposals" refers to a function that notifies administrators of the generated proposed countermeasures and aims to encourage appropriate action.

[0715] This invention relates to a system for improving employee engagement in the workplace environment. This system combines voice and facial expression analysis technology, natural language processing technology, machine learning technology, and deep learning technology to comprehensively evaluate employees' emotional states and recommend appropriate countermeasures to managers, thereby improving the workplace.

[0716] Hardware and software

[0717] The server is equipped with sensors and cameras to collect employee voices and facial expressions in real time. Acoustic analysis software is used for voice analysis, and video analysis software is used for facial expression analysis. Generative AI models are used for natural language processing to convert employee text data into sentiment scores. Machine learning algorithms are introduced to detect anomalies in productivity and attendance data. A deep learning framework is used to integrate this data and propose risk assessments and countermeasures.

[0718] Specific example

[0719] As a concrete example, consider a scenario where the server detects a sharp decline in employee A's productivity within a certain project. In this case, the administrator receives information via a terminal that employee A had recently shown signs of stress in voice analysis. Furthermore, analysis reveals that dissatisfaction has been extracted from the free-response answers of a survey. Based on this information, the server recommends implementing mental health counseling and introducing flexible working hours for employee A.

[0720] Example of a prompt

[0721] By using prompts such as "How can you reduce stress from the current workplace atmosphere?" as input to the AI ​​model, more specific and individualized solutions can be generated.

[0722] This system allows for a comprehensive assessment of employee emotions and performance, enabling responses tailored to individual needs, thereby improving overall workplace engagement and productivity.

[0723] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0724] Step 1:

[0725] The server collects employee voice and facial expression data using sensors and cameras. Inputs include audio files and video footage captured in real time from microphones and cameras. Audio data is analyzed using acoustic analysis software, and facial expression data is processed using video analysis software. The output is digital data indicating the employee's emotional state.

[0726] Step 2:

[0727] The server analyzes employee text data using a generated AI model. The input consists of written survey responses and communication logs. Using natural language processing technology, it extracts emotions from this text data and outputs a numerical emotion score.

[0728] Step 3:

[0729] The server collects employee productivity and attendance data and analyzes it using machine learning algorithms. Inputs include work hours, project progress, and attendance records. Based on this data, it detects anomalous patterns and provides anomaly detection results as output.

[0730] Step 4:

[0731] The server integrates analyzed sentiment data and anomaly detection data using a deep learning framework to assess risk. Inputs are quantified sentiment scores and anomaly detection results. Outputs are information identifying high-risk employees as a result of the risk assessment.

[0732] Step 5:

[0733] The server generates countermeasures based on the risk assessment results and notifies the administrator. For example, the prompt "What interventions would be effective for stress management in this situation?" is input into the AI ​​model to obtain appropriate recommendations. The output is a list of countermeasures sent to the administrator's terminal.

[0734] In this way, the system comprehensively assesses the state of employees and helps optimize the work environment.

[0735] (Application Example 2)

[0736] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0737] In modern workplaces, it is crucial to understand workers' emotions and stress levels in real time and take appropriate measures. However, traditional management methods have made it difficult to accurately grasp workers' emotional states, sometimes delaying the optimization of the work environment. This can lead to problems such as decreased worker productivity and increased turnover. Against this backdrop, there is a need for a system that can quickly and accurately capture workers' emotions and immediately adjust the environment.

[0738] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0739] In this invention, the server includes natural language processing means for analyzing the worker's language data and quantifying their emotions, machine learning means for detecting anomalies in the worker's work data and work schedule data, and deep learning means for integrating the quantified emotion data and anomaly detection data to evaluate risk. This makes it possible to grasp the worker's emotional state in real time and quickly adjust the work environment as needed.

[0740] "Linguistic data" refers to text information generated by workers and serves as fundamental data for analyzing emotional states and thought patterns.

[0741] "Natural language processing" is a technology that allows computers to understand and analyze human language, and is used to quantify emotions.

[0742] "Work data" refers to activity information related to a worker's tasks, and is used to evaluate work efficiency and productivity.

[0743] "Work data" refers to data that shows the working hours and attendance status of workers, and is used for attendance management and pattern anomaly detection.

[0744] "Machine learning" is a technology in which computers learn from data and automatically discover patterns and rules, and is used to detect anomalies.

[0745] "Deep learning" is a technique that uses multi-layered artificial neural networks to perform complex data analysis, and is used to assess risk.

[0746] "Emotion recognition" is a technology that identifies the emotional state of workers from their facial expressions and voice, providing information necessary for environmental adjustments.

[0747] "Environmental adjustment" is the process of optimizing the work environment according to the emotional state and work situation of the workers, with the aim of improving productivity and comfort.

[0748] The server analyzes the worker's language data and uses natural language processing technology to quantify their emotions. In this process, the text data entered by the worker is analyzed in real time, and an emotion score such as positive, negative, or neutral is generated. In addition, work data and attendance data are analyzed using machine learning, and anomalies are detected by comparing them with past data.

[0749] Furthermore, the server utilizes deep learning to integrate sentiment data and anomaly detection data to assess risk. The deep learning model performs a detailed analysis of employee states and identifies high-risk situations.

[0750] Emotion recognition sensors use cameras and microphones to collect the worker's facial expressions and voice, and analyze this data. This allows for a real-time understanding of the user's emotional state. Based on the emotional information collected in this way, the server suggests adjustments to the work environment, specifically such as adjusting the work pace or recommending breaks.

[0751] On the user's device, recommendation information sent from the server is visualized, allowing administrators to make concrete decisions based on this information.

[0752] For example, if a worker's face is monitored by a camera and they show a tired expression for a while, the server can recognize that emotional pattern and recommend to the manager, "Please suggest that the worker take a short break."

[0753] An example of a prompt might be, "What environmental adjustments are recommended based on the employee's monitored emotional state?" This prompt is used in the generative AI model to highly customize the response to the user's needs.

[0754] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0755] Step 1:

[0756] The server collects real-time facial and audio data from the camera and microphone. This data is used as input to analyze the user's emotional state through an emotion recognition sensor. The server preprocesses the collected data using image and audio processing techniques and converts it into a format suitable for the next process.

[0757] Step 2:

[0758] The server applies an emotion recognition algorithm to the pre-processed data to identify the user's emotional state. Specifically, it uses image processing techniques to analyze the user's facial expressions and voice analysis techniques to analyze their tone of voice. This generates a real-time emotion score. This emotion score is stored in a database and used as input for the next step.

[0759] Step 3:

[0760] The server collects worker language data and analyzes it using natural language processing techniques. This data is entered in text format, and the server uses a language model to assign sentiment labels such as positive, negative, and neutral. The resulting quantified sentiment data is then integrated with work data and performance data.

[0761] Step 4:

[0762] The server uses machine learning algorithms to analyze worker work data and attendance data. Past work history and attendance data are used as input. Anomaly detection algorithms identify unusual patterns and anomalies, which are then output as anomaly detection data.

[0763] Step 5:

[0764] The server integrates anomaly detection data and quantified sentiment data using deep learning techniques. In this step, a statistical model is used to perform a risk assessment. The output consists of a risk score and recommended actions based on it, which are stored in a database.

[0765] Step 6:

[0766] Based on the generated risk score and recommended actions, the server generates a prompt message and inputs it into the AI ​​model. The prompt message is structured in the format of "What environmental adjustments are recommended based on the employee's monitored emotional state?" and serves as the basis for the AI ​​model to generate specific countermeasures.

[0767] Step 7:

[0768] The server sends the generated recommended actions to the administrator's terminal. The administrator can use this information to adjust the work environment and provide feedback to workers. The outputted information helps users take quick and effective action.

[0769] 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 controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0770] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0771] 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 this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0772] Furthermore, the emotion identification model 59, acting 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 a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0773] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0774] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0775] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0776] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0777] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is 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 the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0778] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0779] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0780] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0781] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0782] 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.

[0783] Furthermore, it is not necessary to store the entirety 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 the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0784] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0785] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of 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). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0786] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0787] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0788] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0789] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted as being incorporated by reference.

[0790] The following is further disclosed regarding the embodiments described above.

[0791] (Claim 1)

[0792] A natural language processing method that analyzes employee text data to quantify emotions,

[0793] A machine learning method for detecting anomalies in employee productivity data and attendance data,

[0794] A deep learning means for integrating the quantified emotion data and anomaly detection data to evaluate risk,

[0795] A means of recommending countermeasures to high-risk employees,

[0796] A system including means for notifying the administrator of the aforementioned recommendations.

[0797] (Claim 2)

[0798] The system according to claim 1, characterized in that the natural language processing means has the function of identifying positive, negative, and neutral emotions from employee descriptive data.

[0799] (Claim 3)

[0800] The system according to claim 1, characterized in that the machine learning means includes a function to cooperate with a recruitment system to detect anomalies by comparing them with past data.

[0801] "Example 1"

[0802] (Claim 1)

[0803] A natural language processing method that analyzes employee descriptive data to quantify emotions,

[0804] A machine learning method for detecting anomalies in employee work data and time management data,

[0805] A deep learning means for integrating the quantified emotion data and anomaly detection data to evaluate risk,

[0806] A generation method for recommending countermeasures for high-risk employees,

[0807] A system that includes means of providing the aforementioned recommendations to the administrator.

[0808] (Claim 2)

[0809] The system according to claim 1, characterized in that the natural language processing means has the function of identifying positive, negative, and neutral emotions from employee descriptive data.

[0810] (Claim 3)

[0811] The system according to claim 1, characterized in that the machine learning means includes a function to cooperate with a human resources management system in order to detect anomalies by comparing them with past data.

[0812] "Application Example 1"

[0813] (Claim 1)

[0814] A natural language processing method that analyzes workers' text data to quantify their emotions,

[0815] A machine learning method for detecting anomalies in workers' work efficiency data and attendance data,

[0816] A deep learning means for integrating the quantified emotion data and anomaly detection data to evaluate risk,

[0817] A means of recommending countermeasures to high-risk workers,

[0818] A means of notifying the administrator of the aforementioned recommendation,

[0819] A system that includes a means of analyzing resident participation data in local activities and recommending measures for regional revitalization.

[0820] (Claim 2)

[0821] The system according to claim 1, characterized in that the natural language processing means has the function of identifying positive, negative, and neutral emotions from descriptive data of workers and residents.

[0822] (Claim 3)

[0823] The system according to claim 1, characterized in that the machine learning means has a function to cooperate with a recruitment system to detect abnormalities in workers by comparing them with past data, and evaluates the tendency of residents to participate in local activities.

[0824] "Example 2 of combining an emotion engine"

[0825] (Claim 1)

[0826] A means of identifying emotions by analyzing the voice and facial expressions of employees,

[0827] A natural language processing method that analyzes employee text data to quantify emotions,

[0828] A machine learning method for detecting anomalies in employee productivity data and attendance data,

[0829] A deep learning means for integrating the quantified emotion data and anomaly detection data to evaluate risk,

[0830] A means of proposing countermeasures for high-risk employees,

[0831] A system including means for notifying the administrator of the aforementioned proposal.

[0832] (Claim 2)

[0833] The system according to claim 1, characterized in that the natural language processing means has the function of identifying positive, negative, and neutral emotions from employee descriptive data.

[0834] (Claim 3)

[0835] The system according to claim 1, characterized in that the machine learning means has a function to cooperate with an information processing system to detect anomalies by comparing them with past data.

[0836] "Application example 2 when combining with an emotional engine"

[0837] (Claim 1)

[0838] A natural language processing method that analyzes employee language data to quantify emotions,

[0839] A machine learning method for detecting anomalies in worker work data and work schedule data,

[0840] A deep learning means for integrating the quantified emotion data and anomaly detection data to evaluate risk,

[0841] A means of recommending countermeasures to high-risk employees,

[0842] A means of notifying the administrator of the aforementioned recommendation,

[0843] An emotion recognition method that analyzes the facial expressions and voice data of workers to recognize their emotions,

[0844] A system that includes means for dynamically adjusting the work environment based on emotion recognition results.

[0845] (Claim 2)

[0846] The system according to claim 1, characterized in that the natural language processing means has the function of identifying positive, negative, and neutral emotions from the worker's descriptive data.

[0847] (Claim 3)

[0848] The system according to claim 1, characterized in that the machine learning means includes a function to cooperate with an applicant management system in order to detect anomalies by comparing them with past data. [Explanation of Symbols]

[0849] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A natural language processing method that analyzes workers' text data to quantify their emotions, A machine learning method for detecting anomalies in workers' work efficiency data and attendance data, A deep learning means for integrating the quantified emotion data and anomaly detection data to evaluate risk, A means of recommending countermeasures to high-risk workers, A means of notifying the administrator of the aforementioned recommendation, A system that includes a means of analyzing resident participation data in local activities and recommending measures for regional revitalization.

2. The system according to claim 1, characterized in that the natural language processing means has the function of identifying positive, negative, and neutral emotions from descriptive data of workers and residents.

3. The system according to claim 1, characterized in that the machine learning means has a function to cooperate with a recruitment system to detect abnormalities in workers by comparing them with past data, and evaluates the tendency of residents to participate in community activities.

Images