System

Abstract

A system is provided.SOLUTION: A system comprising: means for collecting real-time communication data within an enterprise; means for analyzing the data using a natural language processing module and a voice recognition module; means for matching the analyzed data with a compliance violation detection rule set; means for generating an alert when a compliance violation is detected; means for storing the analysis results and violations in a database; means for periodically providing educational feedback to users; and means for administrator customization of the rule set.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] When offensive language or legally problematic messages are included in corporate communications, the problem can become serious before it is discovered. Traditional methods require manual monitoring of these messages and taking measures, which is inefficient and lacks real-time response. Furthermore, employees may not be adequately trained to understand what types of language violate compliance, which can result in similar violations being repeated. This exposes companies to legal risks and makes it difficult to maintain compliance. [Means for solving the problem]

[0005] The present invention provides a system that collects corporate communication data in real time and analyzes it using a natural language processing module and a speech recognition module. The system compares the analyzed data with a compliance violation detection rule set to detect offensive language or legally questionable messages. When a violation is detected, the system issues an alert to users and administrators, urging them to take immediate action. Furthermore, the analysis results and violation details are stored in a database, and educational feedback is periodically provided to users, improving employee compliance awareness. Administrators can customize the rule set to perform monitoring tailored to the specific circumstances of their company. This allows companies to prevent compliance violations in real time and mitigate legal risks.

[0006] "Internal communication data" refers to all text and voice messaging that takes place within a company.

[0007] "Real-time collection means" refers to a system or device that has the capability to immediately receive and begin processing transmitted data.

[0008] A "natural language processing module" refers to a program or algorithm that analyzes human language and understands its meaning and structure.

[0009] A "voice recognition module" refers to a program or algorithm that converts voice data into text data.

[0010] "Means of analysis" refers to the processes and techniques used to process received data and understand and classify its content.

[0011] A "compliance violation detection rule set" refers to a set of criteria or rules for determining whether a particular expression or content violates compliance.

[0012] "Alert methods" refers to the functions or systems that issue warnings and send notifications to appropriate users and administrators when violations are detected.

[0013] "Means for storing in a database" refers to a system or method for recording analysis results and violation details in digital form so that they can be referenced later.

[0014] "Means for providing educational feedback to users" refers to features or services that periodically provide information to users to help them learn about past violations and the reasons for them.

[0015] "Means for administrators to customize rule sets" refers to interfaces and functions that allow administrators to change and adjust violation detection criteria and rules according to the needs and circumstances of their company. [Brief explanation of the drawings]

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

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

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

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

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

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

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

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

[0024] [First embodiment]

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

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

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

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

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

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

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

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

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

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

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

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

[0037] The system according to the present invention monitors communication data within a company in real time to detect and prevent compliance violations. An embodiment of this system will be described below.

[0038] System Configuration

[0039] The system mainly consists of the following elements:

[0040] Server: Responsible for data collection, analysis, violation detection, alert generation, and data storage.

[0041] Terminal: A device used by users and administrators, such as a PC or smartphone.

[0042] User: An employee of a company who sends and receives messages within the system.

[0043] Administrator: A person responsible for customizing the system's rule set.

[0044] Program processing

[0045] 1. Data collection

[0046] The server receives data in real time from chat applications, mail servers, and voice communication systems used within the company, and depending on the format (text, voice), sends the data to the corresponding analysis queue.

[0047] 2. Data Analysis

[0048] The natural language processing module running on the server analyzes the collected text data. This module tokenizes sentences, tags parts of speech, and analyzes sentence structure to understand the meaning of the message. For example, a message such as "Your work is always slow" is identified as a negative expression.

[0049] The speech data is converted into text by a speech recognition module and subjected to similar analysis. For example, a user's utterance "Your work is always slow" is converted into text and then analyzed by a natural language processing module.

[0050] 3. Detecting non-compliance

[0051] The server then compares the parsed data against a predefined set of rules to detect compliance violations. These rules define offensive or legally problematic content. For example, phrases containing keywords such as "useless" or "incompetent" are identified as violations.

[0052] 4. Issue an alert

[0053] If a violation is detected, the server immediately issues an alert, which is sent to the message sender's device and the administrator's device. The alert includes the violating message, sender, and time of sending.

[0054] 5. Data Recording

[0055] The server records the detected violations and the analysis results in a database. This data is used for future analysis and audits. For example, a message such as "Your work is always slow" is recorded as a violation.

[0056] 6. Providing educational feedback

[0057] On the device, users receive periodic reports and real-time notifications, which help them learn which expressions are non-compliant. For example, users can view their non-compliance messages and the reasons for their actions.

[0058] 7. Customizing the ruleset

[0059] Administrators can customize the rule set of the AI ​​module from their own devices, which is necessary to adapt to the specific circumstances of the company, for example by adding new keywords based on new internal rules.

[0060] Specific examples

[0061] Employee A sends a message in chat saying, "Your work is always slow."

[0062] The server receives the message and analyzes it using a natural language processing module. Based on the analysis results, the message is determined to be negative and offensive.

[0063] A violation is detected based on the rule set, and the server immediately issues an alert, notifying employee A and the administrator.

[0064] The message content and reason for the violation are stored in a database.

[0065] At a later date, Employee A will check his / her violation message through a regular report and be warned.

[0066] In this way, the system of the present invention can enforce compliance within a company and manage risk in real time.

[0067] The processing flow will be explained below.

[0068] Step 1:

[0069] The server receives data in real time from the company's chat applications, mail servers, and voice communication systems, and depending on the format (text or voice), adds the data to a text analysis queue or a voice analysis queue.

[0070] Step 2:

[0071] The server passes the text data to a natural language processing (NLP) module, which tokenizes the received text data, tags it with parts of speech, and performs sentence structure analysis to understand the meaning and intent of the message.

[0072] Step 3:

[0073] The server passes the speech data to a speech recognition module, which converts it into text data. This text data is then analyzed by an NLP module, which performs tokenization, part-of-speech tagging, and sentence structure analysis, just like any other text data.

[0074] Step 4:

[0075] The server then compares the parsed text data against a set of compliance detection rules, which contain keywords and phrases that identify offensive or legally questionable content, such as "useless" or "incompetent."

[0076] Step 5:

[0077] The server generates an alert if a violation is detected based on the matching results. The generated alert includes information such as the content of the violating message, the sender, and the time of sending. The alert is sent to the terminals of the sender of the violating message and the administrator.

[0078] Step 6:

[0079] The server stores the detected violation data and its analysis results in a database, including the original message, analysis results, and detailed information about the violation, making it available for future analysis and audits.

[0080] Step 7:

[0081] On the device, users receive periodic reports and real-time notifications that allow them to learn how their behavior is non-compliant, for example, to see why their message was deemed "offensive."

[0082] Step 8:

[0083] From their own devices, administrators can customize the rule sets of the AI ​​module, changing or adding violation detection criteria and rules according to the specific needs and circumstances of the company. For example, this could include adding new keywords that are considered "offensive."

[0084] This specific processing step allows businesses to effectively detect compliance violations in real time and respond immediately.

[0085] Example 1

[0086] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0087] The volume of communication data within a company is enormous, making it difficult to detect compliance violations from that data. It is also difficult to detect in real time inappropriate or offensive language used unconsciously by employees in communications and take appropriate action. A lack of employee training feedback can lead to similar problems recurring. Furthermore, properly managing different rule sets for each company is also a challenge.

[0088] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0089] In this invention, the server includes means for collecting communication data within a company in real time, means for analyzing the data using a natural language processing module and a voice recognition module, means for comparing the analyzed data with a compliance violation detection rule set, means for saving the analysis results and violation details in a storage device, means for visualizing violation detection notifications and analysis results in real time, means for periodically providing educational feedback to users, and means for an administrator to change the rule set. This makes it possible to effectively monitor a wide range of communication data within a company, detect inappropriate expressions and offensive language in real time, and take prompt action. In addition, providing appropriate feedback and education to employees can help improve compliance awareness.

[0090] "Intra-company communications data" refers to data that includes all communications that take place within a company, including text messages, voice messages, emails, etc.

[0091] A "natural language processing module" is a computer program that analyzes collected text data and performs tokenization, part-of-speech tagging, sentence structure analysis, etc.

[0092] A "voice recognition module" is a computer program for converting voice data into text data.

[0093] A "compliance violation detection rule set" is a collection of pre-defined rules for identifying offensive language or legally questionable content.

[0094] "Notification" refers to alerts or messages sent to relevant parties when a compliance violation is detected.

[0095] "Storage device" refers to a database or storage medium for storing analysis results and violation details.

[0096] "Educational feedback" refers to regular reports and notifications to inform users of violations and the reasons for them, and to encourage improvement.

[0097] "Administrator" refers to the person responsible for customizing and managing the rule sets within the system.

[0098] "Means of collecting data in real time" refers to systems and programs that instantly acquire communication data generated within a company.

[0099] "Visualization means" refers to a system or program for graphically displaying analysis results and detected violations.

[0100] The system according to the present invention monitors communication data within a company in real time to detect and prevent compliance violations. An embodiment of the system according to the present invention will be described in detail below.

[0101] System Configuration

[0102] This system consists of a server, a terminal, and a user. The detailed roles of each element are shown below.

[0103] server

[0104] The server is primarily responsible for the following functions:

[0105] It collects data in real time from chat applications, mail servers, and voice communication systems used within a company.

[0106] Examples of specific hardware or software used: Slack API, Microsoft Exchange Server, Google Speech-to-Text API.

[0107] Example: Employee A sends a message on Slack saying, "Your work is always slow."

[0108] The collected data is analyzed using a natural language processing module, which includes tokenization, part-of-speech tagging, and sentence structure analysis.

[0109] Examples of software used: SpaCy, Google Cloud Natural Language API.

[0110] Example: The message "You're always slow at work" is identified as a negative expression.

[0111] The voice data is converted into text using a voice recognition module and analyzed in the same way.

[0112] Example: Voice data is converted into text by a speech recognition module, such as "Your work is always slow."

[0113] The parsed data is then compared against a set of rules to detect non-compliance, including those that contain offensive language or legally questionable content.

[0114] Examples of software used: Custom rule engine, regular expressions (RegEx).

[0115] Example: The expression "you" is identified as a violation.

[0116] If a violation is detected, a notification is issued, which is sent in real time to the message sender's device and to the administrator's device.

[0117] Examples of software used: Twilio, SMTP (sending email).

[0118] Example: Employee A and Manager B receive a notification.

[0119] The details of the detected violation and the analysis results are saved in a recording device.

[0120] Examples of software used: MySQL, PostgreSQL.

[0121] Example: The message "Your work is always slow" and the analysis results are saved.

[0122] Terminal

[0123] Terminals are devices used by users and administrators that provide the following functionality:

[0124] Receive real-time notifications.

[0125] Learn which expressions are non-compliant through regular reports.

[0126] Examples of software used: Tableau, Power BI.

[0127] Example: Employee A checks his / her violation messages and the reasons for them in a regular report.

[0128] Administrators customize the system's compliance violation detection rule set.

[0129] Examples of software used: Web browser, custom management tools.

[0130] Example: Administrator B adds a new keyword "incompetent" to the ruleset.

[0131] User

[0132] The user is an employee of the company and performs the following functions:

[0133] Send and receive messages within the system.

[0134] Example: Employee A sends a message on Slack.

[0135] Receive regular reports and real-time notifications to increase compliance awareness.

[0136] Example: Employee A checks the reason for his / her violation message in the regular report and makes improvements.

[0137] Examples and prompts

[0138] Examples:

[0139] Employee A sends a message on Slack saying, "Your work is always slow."

[0140] The server receives the message and analyzes it using a natural language processing module.

[0141] The message is deemed offensive and checked against a set of rules.

[0142] A violation is detected and the server issues a notification to Employee A and Manager B using Twilio.

[0143] The message and reason for the violation are stored in a MySQL database.

[0144] At a later date, Employee A checks his / her violation message in the Tableau report.

[0145] Administrator B uses a web browser to add a new keyword, "incompetent," to the rule set.

[0146] Example prompt sentence:

[0147] "Please explain the specific process by which an AI system monitors communication data within a company in real time and detects negative expressions. For example, please explain in detail how a message such as 'Your work is always slow' is processed."

[0148] In this way, the system of the present invention can enforce compliance within a company and manage risk in real time.

[0149] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0150] Step 1:

[0151] Data collection

[0152] The server collects data in real time from chat applications, mail servers, and voice communication systems used within the company.

[0153] Input: Message data from Slack, email data from the mail server, and voice data.

[0154] Output: Send the collected data to the text data analysis queue or the audio data analysis queue.

[0155] Specific operation: For example, if employee A sends a message on Slack saying, "Your work is always slow," the message is sent to the server in real time.

[0156] Step 2:

[0157] Data conversion (for audio data)

[0158] The server converts the collected voice data into text using a voice recognition module.

[0159] Input: Collected audio data.

[0160] Output: Audio data converted to text data.

[0161] Specific operation: For example, audio data such as "Your work is always slow" is converted into text using the Google Speech-to-Text API.

[0162] Step 3:

[0163] Data analysis

[0164] The server analyzes the text data using a natural language processing module, which performs tokenization, part-of-speech tagging, and sentence structure analysis.

[0165] Input: Text data (either collected directly or converted by a speech recognition module).

[0166] Output: The parsed data includes word tokens, parts of speech, and sentence structure information.

[0167] Specific operation: For example, the message "Your work is always slow" is parsed through SpaCy and tokenized into "you", "of", "work", "is", "always", and "slow".

[0168] Step 4:

[0169] Compliance violation detection

[0170] The server compares the parsed data against a set of compliance violation detection rules to determine whether a violation has occurred.

[0171] Input: Parsed text data.

[0172] Output: Violation detection results (details of the violation if there is one, no abnormality if there is no violation).

[0173] Specific behavior: For example, the analyzed message contains the offensive keyword "you," so it is detected as a violation.

[0174] Step 5:

[0175] Issuance of notifications

[0176] The server will issue a notification if a violation is detected, which will be sent to the message sender and an administrator.

[0177] Input: Violation detection result and violation content.

[0178] Output: Notification message (including details of non-compliance and time of sending).

[0179] Specific operation: For example, an alert will be sent via Twilio to employee A who sent a message saying, "Your work is always slow," and to his manager B.

[0180] Step 6:

[0181] Data recording

[0182] The server stores the details of the detected violation and the analysis results in a recording device.

[0183] Input: Violation detection results and their analysis.

[0184] Output: Violation records stored in a database.

[0185] Specific behavior: For example, the message "Your work is always slow" and the reason for the violation are recorded in a MySQL database.

[0186] Step 7:

[0187] Providing educational feedback

[0188] The device provides regular feedback to the user, helping them learn what expressions are considered non-compliant.

[0189] Input: Violation records stored in the database.

[0190] Output: Education report provided to the user.

[0191] Specific behavior: For example, Employee A checks his / her violation messages and the reasons for them through regular Tableau reports.

[0192] Step 8:

[0193] Customizing the ruleset

[0194] Administrators can customize rule sets from their own devices and adjust them to suit their company's specific circumstances.

[0195] Input: Ruleset change input by administrator.

[0196] Output: The updated ruleset.

[0197] Specific behavior: For example, Administrator B uses a web browser to add a new keyword, "incompetent," to the rule set.

[0198] (Application example 1)

[0199] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0200] As internal communication increases in modern companies, compliance is becoming more important than ever. However, many companies lack a system for monitoring internal communications in real time and immediately detecting violations, exposing them to the risk of serious compliance violations being overlooked. Even when violations are discovered, response can be delayed, potentially affecting the company's credibility and legal risk. Furthermore, a lack of educational feedback to help employees understand and improve their past violations can lead to repeated violations. To solve these problems, a system is needed that effectively monitors internal communications and immediately detects and notifies violations.

[0201] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0202] In this invention, the server includes means for collecting communication data within a company in real time, means for analyzing the data using a natural language processing module and a voice recognition module, means for comparing the analyzed data with a compliance violation detection rule set, means for issuing an alert when a compliance violation is detected, means for storing the analysis results and the details of the violation in a database, means for periodically providing educational feedback to users, means for an administrator to customize the rule set, means for delivering related notifications via an application installed on a smart device, means for notifying users and administrators of violation detection notifications and details, and means for generating periodic reports based on the violation detection data. This enables real-time monitoring of intra-company communications, immediate detection of compliance violations, prompt notification to target groups, and employee educational feedback.

[0203] "Internal Enterprise Communication Data" refers to the electronic data of all messages, communications, and discussions exchanged by employees within an enterprise.

[0204] "Means of collecting in real time" refers to a system that instantly collects communication data within a company and provides it for processing without delay.

[0205] "Natural language processing module" refers to software or algorithms for analyzing text data and understanding its meaning.

[0206] "Speech Recognition Module" refers to software or algorithms for converting voice data into text.

[0207] A "compliance violation detection rule set" is a set of rules for defining violation actions based on company rules and legal regulations.

[0208] "Means for issuing alerts" refers to a mechanism for immediately notifying relevant parties when a compliance violation is detected.

[0209] "Means for storing in a database" refers to a mechanism for permanently recording the analysis results and violation details and storing them for future reference and analysis.

[0210] "Means for providing educational feedback to users" refers to a mechanism for providing employees with information about compliance violations and advice for improvement.

[0211] "Means for administrators to customize rule sets" refers to a mechanism by which administrators can modify, add, or remove compliance violation detection rule sets according to the specific needs of their enterprise.

[0212] "Smart devices" refers to portable electronic devices with internet connectivity, including smartphones and tablets.

[0213] "Application" means a software program designed to perform a specific job or task.

[0214] "Means for delivering notifications" refers to a mechanism for sending real-time notifications to users and administrators via smart devices, etc.

[0215] "Means for notifying users and administrators of detected violations and providing details of those violations" refers to a mechanism for sending notifications containing detailed information to users and administrators when a violation is detected.

[0216] "Means for generating periodic reports" refers to a system for aggregating violation data over a fixed period and creating an analysis report.

[0217] In order to put the present invention into practice, a system is constructed that collects communication data within a company in real time and analyzes that data. The configuration and processing procedure of this system are described below.

[0218] System Configuration

[0219] The system consists of the following main components:

[0220] 1. Server

[0221] Data collection module: Works with internal company communication tools (chat apps and email servers) to collect data in real time.

[0222] Natural Language Processing (NLP) module: Analyzes collected text data and understands the meaning of messages. Available software includes SpaCy and Google Cloud Natural Language API.

[0223] Speech recognition module: Converts voice data into text. Uses Google Cloud Speech-to-Text API, etc.

[0224] Violation Detection Module: Compares the parsed data against a set of compliance violation detection rules.

[0225] Alerting module: Issues alerts and notifies users and administrators when violations are detected. Notification services include Firebase Cloud Messaging (FCM).

[0226] Data storage module: Analysis results and violation details are saved in a database. SQLite and PostgreSQL are available.

[0227] Report Generation Module: Generates regular reports based on violation data and provides feedback.

[0228] 2. Terminal

[0229] Smart devices (smartphones and tablets): The devices through which users access and receive notifications.

[0230] Application: Software installed on smart devices that provides real-time notifications, checks violations, and provides feedback.

[0231] Processing flow

[0232] 1. Data collection

[0233] The server's data collection module works in conjunction with each communication tool within the company to collect text and voice data in real time.

[0234] 2. Data Analysis

[0235] Natural language processing and speech recognition modules analyze the collected data and understand its meaning. For example, a message like "You're always slow at work" is identified as a negative expression.

[0236] 3. Detecting non-compliance

[0237] The server compares the parsed data against a set of compliance violation detection rules to determine whether there are any violations.

[0238] 4. Issue an alert

[0239] If a violation is detected, the alert module will be activated and send a notification to the user and administrator, such as "A compliance violation has been detected. From: Employee X, Content: 'Your work is always slow', Time: YYYY-MM-DD HH:MM."

[0240] 5. Data storage and report generation

[0241] The data storage module stores the details of violations and their analysis results in a database. Based on the stored data, the report generation module creates regular reports and provides educational feedback to employees.

[0242] Specific examples

[0243] Example of input prompt: "During a new project discussion, a message was sent stating, 'Your work is always slow.' Please determine whether this message is in compliance and issue an alert if necessary."

[0244] Example notification: "A compliance violation has been detected. From: Employee X, Content: 'Your work is always slow', Time: YYYY-MM-DD HH:MM"

[0245] In this way, the present invention provides real-time monitoring of corporate communication data, immediate detection of compliance violations, and prompt notification and educational feedback to target audiences.

[0246] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0247] Step 1:

[0248] The server connects with the company's internal communication tools and collects data in real time. The main inputs are employee chat messages and voice conversations. This data is sent to the server as raw data and added to a queue for further processing. The output is stored in the queue as text data or voice data.

[0249] Step 2:

[0250] The server's natural language processing (NLP) module analyzes the text data stored in the queue. The main input is the text data stored in the queue. The NLP module performs sentence tokenization (breaking down each word), part-of-speech tagging (identifying the grammatical role of each word), and sentence structure analysis (understanding the meaning of the sentence). For example, the message "Your work is always slow" is recognized as an offensive expression. The output is the analyzed text data and its semantic analysis results.

[0251] Step 3:

[0252] The server's speech recognition module converts the voice data stored in the queue into text. The main input is voice data. The speech recognition module (for example, Google Cloud Speech-to-Text API) analyzes the voice data and obtains text data. This text data is passed to the NLP module and analyzed again. The output is the text data converted from the voice data.

[0253] Step 4:

[0254] The server compares the analyzed data with a compliance violation detection rule set. The main input is the analyzed text data. The rule set defines offensive expressions and legally problematic keywords. For example, it includes keywords such as "incompetent" and "useless." This comparison determines whether or not a violation has occurred. The output is a judgment result indicating whether or not a violation has been detected.

[0255] Step 5:

[0256] If a violation is detected, the server's alert module is activated. The main input is the violation detection result. The alert module sends a notification containing the details of the violation to the user's and administrator's devices. Specifically, the notification is sent using Firebase Cloud Messaging (FCM). The output is a notification message sent to the user and administrator. For example, the notification may be in the format "A compliance violation has been detected. Sender: Employee X, Content: 'Your work is always slow', Time: YYYY-MM-DD HH:MM."

[0257] Step 6:

[0258] The server saves the details of violations and the analysis results in a database. The main inputs are the details of violations and the analysis results. This data is stored in a database system such as SQLite or PostgreSQL. The saved data is used for later analysis and audits. The output is the data recorded in the database.

[0259] Step 7:

[0260] The server's report generation module creates periodic reports based on the stored violation data. The main input is the violation data recorded in the database. The report generation module aggregates the violation data and creates an analysis report, which allows companies to provide educational feedback to employees. The output is a periodic analysis report. Specifically, a report in the format of "Violation report for the past month" is created.

[0261] Through the above processing steps, the present invention realizes real-time monitoring within a company, immediate detection and response to compliance violations, data storage, and provision of educational feedback.

[0262] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0263] The system according to the present invention monitors communication data within a company in real time and analyzes compliance violations and user sentiment. An embodiment of this system will be described below.

[0264] System Configuration

[0265] The system mainly consists of the following elements:

[0266] Server: Responsible for data collection, analysis, violation detection, emotion recognition, alert issuance, and data storage.

[0267] Terminal: A device used by users and administrators, such as a PC or smartphone.

[0268] User: An employee of a company who sends and receives messages within the system.

[0269] Administrator: The person responsible for customizing the system's rule set and emotion engine.

[0270] Program processing

[0271] 1. Data collection

[0272] The server receives data in real time from chat applications, mail servers, and voice communication systems used within the company. Depending on the format (text or voice), the data is added to a text analysis queue or a voice analysis queue.

[0273] 2. Data Analysis

[0274] The server passes the text data to a natural language processing (NLP) module, which tokenizes the received text data, tags it with parts of speech, and performs sentence structure analysis to understand the meaning and intent of the message.

[0275] The speech data is converted to text data by a speech recognition module and undergoes the same analysis as other text data: tokenization, part-of-speech tagging, and sentence structure analysis.

[0276] 3. Emotion Analysis

[0277] The server passes the parsed text data to an emotion engine to identify the user's emotion. This emotion engine contains algorithms that recognize the emotional tone of the text (e.g., anger, sadness, joy, etc.). For example, the message "I'm so angry" is identified as an emotion of anger.

[0278] 4. Detecting non-compliance

[0279] The server then checks the parsed data against a set of compliance detection rules, which contain keywords and phrases that identify offensive or legally questionable content, such as "useless" or "incompetent."

[0280] 5. Issuing an Alert

[0281] When a violation is detected, the server generates an alert, taking into account the emotional data. The generated alert includes information such as the content of the violating message, the emotional tone, the sender, and the time of sending. The alert is sent to the terminals of the sender of the violating message and the administrator.

[0282] 6. Data Recording

[0283] The server stores the detected violation data, analysis results, and emotional data in a database. The stored data includes the original message, analysis results, detailed violation information, and emotional tone, making it available for future analysis and audits.

[0284] 7. Providing educational feedback

[0285] On the device, users receive periodic reports and real-time notifications that allow them to learn how their behavior is non-compliant, for example, to see why their message was deemed "offensive."

[0286] 8. Customizing the ruleset

[0287] Administrators can customize the rulesets of the AI ​​modules and sentiment engines from their own devices to address the specific needs and circumstances of their companies, for example by adding new keywords that they consider "offensive."

[0288] Specific examples

[0289] Employee A sends a message in chat saying, "Your work is always slow. It really pisses me off."

[0290] The server receives the message and analyzes it with the NLP module. Based on the analysis results, the message is determined to be negative and aggressive.

[0291] The emotion engine identifies "I'm angry" as anger.

[0292] A violation is detected based on the rule set, and the server immediately issues an alert, notifying employee A and the administrator.

[0293] The message content, reason for violation, and emotional tone are stored in a database.

[0294] At a later date, Employee A will review his / her violation messages, the reasons for them, and the emotional tone of the messages through regular reports, and take appropriate action.

[0295] This specific process allows the system of the present invention to strengthen compliance and sentiment management within a company and manage risk in real time.

[0296] The processing flow will be explained below.

[0297] Step 1:

[0298] The server receives data in real time from the company's chat applications, mail servers, and voice communication systems, and depending on the format (text or voice), adds the data to a text analysis queue or a voice analysis queue.

[0299] Step 2:

[0300] The server passes the text data to a natural language processing (NLP) module, which tokenizes the received text data, tags it with parts of speech, and performs sentence structure analysis. This allows the meaning and intent of the message to be understood. For example, a message such as "You're always slow at work" is broken down into a series of words, and the role of each word is analyzed.

[0301] Step 3:

[0302] The server passes the voice data to a speech recognition module, which converts it into text data. The converted text data is then analyzed by an NLP module. For example, a voice message such as "You're always slow at work" is converted into written text, which is then analyzed.

[0303] Step 4:

[0304] The server passes the parsed text data to an emotion engine to identify the user's emotion. The emotion engine identifies the emotional tone of the text (e.g., anger, sadness, joy, etc.). For example, if the sentence "I'm so angry" is sent, it will be recognized as "anger."

[0305] Step 5:

[0306] The server compares the analyzed data and sentiment data with a set of rules to detect compliance violations. The rules include keywords and phrases that identify offensive or legally problematic content. For example, the server checks whether the message contains phrases like "useless" or "incompetent." A message like "You're always slow at work, it pisses me off" would be considered a violation.

[0307] Step 6:

[0308] When a violation is detected, the server generates an alert, taking into account emotional data. The alert includes the content of the violating message, its emotional tone (e.g., "anger"), the sender, and the time of sending. The alert is sent to the terminals of the sender of the violating message and the administrator.

[0309] Step 7:

[0310] The server stores the detected violation data, its analysis results, and emotional data in a database. The stored data includes the original message, analysis results, detailed information about the violation, and emotional tone. For example, "Your work is always slow. It makes me so angry" is stored, and the analysis results and the emotional tone of "anger" are recorded.

[0311] Step 8:

[0312] On the device, users receive periodic reports and real-time notifications, which help them learn how their actions are violating compliance. For example, users can see their violation messages, the reasons for their violations, and their emotional tone.

[0313] Step 9:

[0314] Administrators can customize the rulesets for the AI ​​module and emotion engine from their devices, for example adding new keywords that are considered "aggressive" or emotional tones (such as "sadness" or "joy").

[0315] This specific processing step allows enterprises to accurately monitor and quickly respond to compliance violations and users' emotional states in real time.

[0316] Example 2

[0317] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0318] Communications within companies are increasing every day, and some of them may contain offensive language or inappropriate content, which increases the company's compliance risks. It is also difficult to grasp changes in employee emotions in real time and take appropriate action. To solve these issues, a system is needed that monitors communication data in real time, analyzes emotions, and detects compliance violations.

[0319] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0320] In this invention, the server includes means for collecting communication data within a company in real time, means for analyzing the data using a natural language processing module and a voice recognition module, means for identifying user emotions from the analyzed data using an emotion engine, means for comparing the analyzed data with a compliance violation detection rule set, means for issuing an alert including the content, emotional tone, sender, and sending time of the violation message when a compliance violation is detected, means for storing the analysis results and the violation details in a database, means for periodically providing educational feedback to users, and means for an administrator to customize the rule set. This makes it possible to monitor communications within a company in real time, detect inappropriate expressions and changes in employee emotions early, and take appropriate measures.

[0321] "Internal communications data" refers to all forms of data that take place within a company, including emails, chat messages, voice calls, video conferences, etc.

[0322] "Real-time" refers to data being processed the instant it is generated or with very little delay.

[0323] A "natural language processing module" refers to a software module that allows a computer to analyze and understand the language that humans use on a daily basis.

[0324] "Speech recognition module" refers to a software module for converting voice data into text data.

[0325] An "emotion engine" refers to a system that includes algorithms for analyzing text or speech and identifying the emotional tones (e.g., joy, anger, sadness, etc.) contained within it.

[0326] "Rule Set" refers to the list of keywords or phrases and associated rules used to determine non-compliance.

[0327] An "alert" refers to a message or warning signal that notifies a user or administrator when the system detects an abnormality or problem.

[0328] "Database" refers to a system for storing information in an organized manner and for efficiently searching and managing it.

[0329] "Educational feedback" refers to a mechanism that provides users with an evaluation of their actions and suggestions for improvement.

[0330] An "administrator" is a person who has the authority to configure and operate the system and is responsible for customizing the rule set and monitoring the system.

[0331] The system based on this invention monitors communication data within a company in real time and analyzes compliance violations and user sentiment. This system is mainly composed of three elements: a server, a terminal, and a user.

[0332] System Configuration

[0333] 1. Server:

[0334] This is the core of the system, responsible for data collection, analysis, violation detection, emotion recognition, alert generation, and data storage. The server contains the following modules:

[0335] Natural Language Processing (NLP) module: Analyzes communication data using software tools such as SpaCy and NLTK.

[0336] Speech recognition module: Converts voice data into text using a module such as Google Speech-to-Text.

[0337] Emotion engine: Identifies the emotional tone of a user's text data using tools such as IBM Watson Tone Analyzer.

[0338] Database: A storage system for storing analysis results, violation details, and emotional tones.

[0339] 2. Terminal:

[0340] These are devices used by users and administrators, such as PCs and smartphones, to receive necessary reports and notifications.

[0341] 3. User:

[0342] These are company employees who send and receive messages within this system, and communicate via chat, email, etc. in the course of their daily work.

[0343] Specific examples

[0344] Example of employee A sending a message in chat saying, "Your work is always slow. It really pisses me off."

[0345] The server receives this message and adds it to the text analysis queue.

[0346] The NLP module tokenizes the message and performs part-of-speech tagging and sentence structure analysis.

[0347] The emotion engine identifies "annoyed" as the emotion "anger."

[0348] A non-compliance rule set identifies "Your work is always slow" as an offensive expression.

[0349] The server generates an alert containing details of the violation and emotional tone, and notifies Employee A and the manager.

[0350] This message, along with the analysis results, violation details, and emotional tone, is saved in a database.

[0351] At a later date, Employee A can review the periodic report on his / her own device and learn the reason for the violation and the emotional tone.

[0352] Prompt Sentence Examples

[0353] "Develop a system that performs employee sentiment analysis. The system collects chat data in real time, analyzes the text using natural language processing, and identifies emotional tones. It then detects compliance violations and sends alerts to managers if violations are detected. It also stores the emotional tones and violation data in a database so employees can receive feedback."

[0354] This system allows companies to monitor communication content in real time, detect and respond to inappropriate language and changes in employee sentiment early on, and provides employees with the opportunity to review and improve their own behavior through feedback.

[0355] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0356] Step 1: Collect data

[0357] The server receives data in real time from chat applications, mail servers, and voice communication systems used within the company, such as Slack, Microsoft Teams, email, and Zoom.

[0358] Input: Chat messages, emails, voice data

[0359] Output: Add data to the text analysis queue and speech analysis queue

[0360] How it works: The server sorts data received from each system into the appropriate analysis queue depending on its format (text or audio).

[0361] Step 2: Analyzing the text data

[0362] The server sends the data in the text analysis queue to a natural language processing (NLP) module, which performs tokenization, part-of-speech tagging, and sentence structure analysis of the text data.

[0363] Input: Text data

[0364] Output: Tokenized data, tagged data, sentence structure analysis results

[0365] How it works: It splits incoming text data into words and phrases, tags them with part-of-speech tags, and analyzes the structure of entire sentences.

[0366] Step 3: Analyzing the audio data

[0367] The server sends the speech data added to the speech analysis queue to a speech recognition module, which converts the speech data into text data that is then analyzed in a subsequent step.

[0368] Input: Audio data

[0369] Output: Text data

[0370] How it works: Converts speech to text and feeds it into an NLP module for further analysis.

[0371] Step 4: Sentiment Analysis

[0372] The server passes the text data analyzed by the NLP module to an emotion engine to identify the user's emotion. The emotion engine detects the emotional tone of the text (e.g., anger, sadness, joy, etc.).

[0373] Input: Parsed text data

[0374] Output: Emotional tone identification result

[0375] How it works: The emotion engine analyzes text and recognizes emotional tone, for example identifying the message "I'm so angry" as "anger."

[0376] Step 5: Detecting non-compliance

[0377] The server then compares the parsed text data against a set of compliance detection rules, which contain keywords and phrases to identify offensive or legally problematic content.

[0378] Input: Parsed text data, emotional tone

[0379] Output: Whether or not there is a violation, detailed violation information

[0380] How it works: The server checks the text data against a set of rules to see if there are any violations, such as whether it contains keywords like "useless" or "incompetent."

[0381] Step 6: Issue an alert

[0382] When a violation is detected, the server generates an alert, taking into account the emotional data. The generated alert includes information such as the content, emotional tone, sender, and time of the violating message.

[0383] Input: Violation details, emotional tone

[0384] Output: Alert message

[0385] What it does: When a violation is detected, an alert is generated to notify the sender of the offending message and an administrator. The alert includes details such as the content and emotional tone of the offending message.

[0386] Step 7: Record the data

[0387] The server stores the data on which violations were detected, along with the analysis results and emotion data, in a database for future analysis and auditing.

[0388] Input: Violation details, emotional tone, message content

[0389] Output: Saved entries in the database

[0390] How it works: Analysis results and violation information are recorded in a database and stored for future reference and verification.

[0391] Step 8: Providing educational feedback

[0392] The device displays regular reports and real-time notifications to users, allowing them to learn and improve how their behavior is out of compliance.

[0393] Input: Analysis results, violation information, emotional tone

[0394] Output: Educational feedback report, real-time notifications

[0395] Action: Generate and provide feedback to the user based on their behavior and violations, for example informing them of the specific reasons why something was deemed "offensive."

[0396] Step 9: Customizing the ruleset

[0397] Administrators can use their devices to customize the rulesets of the AI ​​modules and emotion engines to meet the specific needs and circumstances of their companies.

[0398] Input: New rule or keyword

[0399] Output: Updated ruleset

[0400] How it works: Administrators adapt the ruleset to suit their company's needs and apply it to the system, which may include adding new keywords that are considered "offensive."

[0401] (Application example 2)

[0402] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0403] In recent years, inappropriate behavior and emotional changes between employees have often been the cause of trouble in corporate communications. In particular, in environments such as factories, where cooperation between employees and robots is necessary, communication is a crucial element. However, there is a lack of means to monitor this communication in real time and prevent problems before they occur. Therefore, there is a growing need for a system that can analyze all communication data, including emotions, in real time, identify compliance violations and negative emotional states, and immediately issue alerts.

[0404] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0405] In this invention, the server includes a means for collecting communication data within the company in real time, a means for analyzing the data using a natural language analysis module and a voice recognition module, and a means for comparing the analyzed data with a compliance violation detection rule set. This allows for real-time monitoring of communications between employees and robots and issuing warnings at appropriate times. The system also analyzes the user's emotional state and provides feedback based on that analysis, creating a healthier communication environment.

[0406] Definitions of important words

[0407] "Communication data" refers to data including text data and voice data generated within a company, as well as the results of sentiment analysis.

[0408] "Real-time" refers to immediate processing or analysis of data the moment it is generated or transmitted.

[0409] "Natural language analysis" refers to the process of mechanically interpreting text or audio data through tokenization, part-of-speech tagging, sentence structure analysis, etc.

[0410] "Speech recognition module" refers to a computer program or system for converting voice data into text data.

[0411] A "compliance violation detection rule set" refers to a collection of rules for detecting expressions that may violate legal or corporate standards based on specific keywords or phrases.

[0412] "Warning" refers to a notification or alert issued when non-compliance or a negative emotional state is detected.

[0413] "Database" refers to a system or structured collection of information for storing collected and analyzed data.

[0414] "Educational Feedback" refers to reports or guidance provided to users to help them learn about their communication and non-compliance behaviors.

[0415] "Customizing rule sets" refers to the act of an administrator tailoring compliance violation detection rules and sentiment analysis algorithms to the specific needs and circumstances of the company.

[0416] "User emotional state" refers to the user's emotional tone (e.g., anger, sadness, joy, etc.) derived from analyzed text data and voice data.

[0417] "Form for carrying out the invention" of the specification

[0418] System Configuration

[0419] The system according to the present invention comprises the following components:

[0420] Server: Responsible for data collection, analysis, violation detection, emotion recognition, warning issuance, and data storage. Specifically, it is equipped with a natural language analysis module, a voice recognition module, a compliance violation detection rule set, and an emotion analysis engine.

[0421] Terminal: A device used by employees and managers, such as a PC, a smartphone, or an application installed on a robot in a factory.

[0422] User: An employee of a company who sends and receives messages within the system.

[0423] Administrator: The person responsible for customizing the system's rule sets and sentiment analysis engine.

[0424] Explanation of program processing

[0425] The server uses a communications API to collect internal company communication data (text and voice data) in real time. This data is converted into text data through a natural language analysis module and a speech recognition module, and then analyzed based on the extracted tokens. In particular, the NLP (natural language processing) module performs tokenization, part-of-speech tagging, and sentence structure analysis to understand the meaning and intent of the message.

[0426] Meanwhile, the voice data is converted into text data by a speech recognition module. The analyzed text data is then passed to a sentiment analysis engine, which identifies emotional tones such as anger, sadness, joy, etc. If a user's message is "I'm so angry," the emotional tone will be recognized as "anger."

[0427] The analyzed data is compared against a set of compliance violation detection rules, which include offensive or legally questionable language. If such language is detected, the server immediately issues an alert. The alert includes information about the content, emotional tone, sender, and time of the offending message, and this information is sent to the user and administrator who detected the violation.

[0428] The data is stored in a database for permanent archival, including the original message, analysis results, violation details, and emotional tone, making it available for future analysis and audits.

[0429] At the device, users receive regular educational feedback on their communication behavior, including how they were non-compliant and their emotional state at the time. Administrators can customize the rule set and optimize the system by adding new keywords and phrases as needed.

[0430] Specific examples

[0431] For example, suppose employee A sends a message saying, "You're always slow at work. It pisses me off." In this case, the server would process it as follows:

[0432] 1. Messages are received in real time and analyzed by the NLP module.

[0433] 2. The sentiment analysis engine identifies the phrase "I'm angry" as anger.

[0434] 3. Based on the rule set, this message is detected as non-compliant.

[0435] 4. The server immediately issues an alert and notifies Employee A and the administrator.

[0436] Prompt Sentence Examples

[0437] Monitor chat logs between Employee A and Employee B in real time to detect compliance violations and sentiment changes. Start your analysis with the following chat logs:

[0438] "Employee A: You're always slow at work. It really pisses me off."

[0439] The system monitors communication between employees and robots in real time and provides immediate and appropriate feedback to support safe and efficient operations within the factory.

[0440] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0441] Program processing flow

[0442] Step 1:

[0443] The server collects communication data (text data and voice data) within the company in real time. Specifically, it uses a communication API to obtain messages and call content and adds them to a data queue. The input data is raw text or voice data, and the output is data added to the analysis queue.

[0444] Step 2:

[0445] The server passes the collected text data to a natural language analysis module, which tokenizes the data, tags it with parts of speech, and analyzes sentence structure. The input data is text data, and the output is tokenized analyzed data. The analyzed data is used to understand the meaning and intent of the message.

[0446] Step 3:

[0447] The server passes the collected voice data to a voice recognition module and converts it into text data. The voice recognition module analyzes the voice data and converts it into text. The input data is voice data and the output is text data. The converted text data is further analyzed by a natural language analysis module.

[0448] Step 4:

[0449] The server passes the parsed text data to the sentiment analysis engine, which analyzes the emotional tone of the text to identify the user's emotional state. The input data is the parsed text data, and the output is the identified emotional tone. For example, the message "I'm so angry" is recognized as "Anger."

[0450] Step 5:

[0451] The server compares the analyzed data against a set of compliance violation detection rules. The rule set contains keywords and phrases that determine whether a violation has occurred. The input data is the analyzed and sentiment analysis data, and the output is a determination of whether or not there is a compliance violation. If a violation is detected, information about it is generated.

[0452] Step 6:

[0453] The server immediately issues a warning if a violation is detected. The warning includes the content, emotional tone, sender, and time of sending of the violating message. The input data is the compliance violation judgment result, and the output is the generation and notification of a warning message. The notification is sent to the user who detected the violation and the administrator.

[0454] Step 7:

[0455] The server stores the analysis results and violation details in a database. The stored data includes the original message, analysis results, violation details, and emotional tone. The input data is the analyzed and violation detection data, and the output is storage in the database.

[0456] Step 8:

[0457] The device periodically provides educational feedback to the user, including how the user violated compliance and their emotional state at the time. The input data are the analysis results stored in the database, and the output is a report of the educational feedback.

[0458] Step 9:

[0459] Administrators use a terminal to customize the rule set. They can add new keywords and phrases. The input is the administrator's configuration information, and the output is an updated rule set. This allows the system to adapt to the specific needs and circumstances of the company.

[0460] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0461] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0462] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0463] [Second embodiment]

[0464] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0465] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0466] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0467] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0468] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0469] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0470] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0471] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0472] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0473] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0474] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0475] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0476] The system according to the present invention monitors communication data within a company in real time to detect and prevent compliance violations. An embodiment of this system will be described below.

[0477] System Configuration

[0478] The system mainly consists of the following elements:

[0479] Server: Responsible for data collection, analysis, violation detection, alert generation, and data storage.

[0480] Terminal: A device used by users and administrators, such as a PC or smartphone.

[0481] User: An employee of a company who sends and receives messages within the system.

[0482] Administrator: A person responsible for customizing the system's rule set.

[0483] Program processing

[0484] 1. Data collection

[0485] The server receives data in real time from chat applications, mail servers, and voice communication systems used within the company, and depending on the format (text, voice), sends the data to the corresponding analysis queue.

[0486] 2. Data Analysis

[0487] The natural language processing module running on the server analyzes the collected text data. This module tokenizes sentences, tags parts of speech, and analyzes sentence structure to understand the meaning of the message. For example, a message such as "Your work is always slow" is identified as a negative expression.

[0488] The speech data is converted into text by a speech recognition module and subjected to similar analysis. For example, a user's utterance "Your work is always slow" is converted into text and then analyzed by a natural language processing module.

[0489] 3. Detecting non-compliance

[0490] The server then compares the parsed data against a predefined set of rules to detect compliance violations. These rules define offensive or legally problematic content. For example, phrases containing keywords such as "useless" or "incompetent" are identified as violations.

[0491] 4. Issue an alert

[0492] If a violation is detected, the server immediately issues an alert, which is sent to the message sender's device and the administrator's device. The alert includes the violating message, sender, and time of sending.

[0493] 5. Data Recording

[0494] The server records the detected violations and the analysis results in a database. This data is used for future analysis and audits. For example, a message such as "Your work is always slow" is recorded as a violation.

[0495] 6. Providing educational feedback

[0496] On the device, users receive periodic reports and real-time notifications, which help them learn which expressions are non-compliant. For example, users can view their non-compliance messages and the reasons for their actions.

[0497] 7. Customizing the ruleset

[0498] Administrators can customize the rule set of the AI ​​module from their own devices, which is necessary to adapt to the specific circumstances of the company, for example by adding new keywords based on new internal rules.

[0499] Specific examples

[0500] Employee A sends a message in chat saying, "Your work is always slow."

[0501] The server receives the message and analyzes it using a natural language processing module. Based on the analysis results, the message is determined to be negative and offensive.

[0502] A violation is detected based on the rule set, and the server immediately issues an alert, notifying employee A and the administrator.

[0503] The message content and reason for the violation are stored in a database.

[0504] At a later date, Employee A will check his / her violation message through a regular report and be warned.

[0505] In this way, the system of the present invention can enforce compliance within a company and manage risk in real time.

[0506] The processing flow will be explained below.

[0507] Step 1:

[0508] The server receives data in real time from the company's chat applications, mail servers, and voice communication systems, and depending on the format (text or voice), adds the data to a text analysis queue or a voice analysis queue.

[0509] Step 2:

[0510] The server passes the text data to a natural language processing (NLP) module, which tokenizes the received text data, tags it with parts of speech, and performs sentence structure analysis to understand the meaning and intent of the message.

[0511] Step 3:

[0512] The server passes the speech data to a speech recognition module, which converts it into text data. This text data is then analyzed by an NLP module, which performs tokenization, part-of-speech tagging, and sentence structure analysis, just like any other text data.

[0513] Step 4:

[0514] The server then compares the parsed text data against a set of compliance detection rules, which contain keywords and phrases that identify offensive or legally questionable content, such as "useless" or "incompetent."

[0515] Step 5:

[0516] The server generates an alert if a violation is detected based on the matching results. The generated alert includes information such as the content of the violating message, the sender, and the time of sending. The alert is sent to the terminals of the sender of the violating message and the administrator.

[0517] Step 6:

[0518] The server stores the detected violation data and its analysis results in a database, including the original message, analysis results, and detailed information about the violation, making it available for future analysis and audits.

[0519] Step 7:

[0520] On the device, users receive periodic reports and real-time notifications that allow them to learn how their behavior is non-compliant, for example, to see why their message was deemed "offensive."

[0521] Step 8:

[0522] From their own devices, administrators can customize the rule sets of the AI ​​module, changing or adding violation detection criteria and rules according to the specific needs and circumstances of the company. For example, this could include adding new keywords that are considered "offensive."

[0523] This specific processing step allows businesses to effectively detect compliance violations in real time and respond immediately.

[0524] Example 1

[0525] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0526] The volume of communication data within a company is enormous, making it difficult to detect compliance violations from that data. It is also difficult to detect in real time inappropriate or offensive language used unconsciously by employees in communications and take appropriate action. A lack of employee training feedback can lead to similar problems recurring. Furthermore, properly managing different rule sets for each company is also a challenge.

[0527] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0528] In this invention, the server includes means for collecting communication data within a company in real time, means for analyzing the data using a natural language processing module and a voice recognition module, means for comparing the analyzed data with a compliance violation detection rule set, means for saving the analysis results and violation details in a storage device, means for visualizing violation detection notifications and analysis results in real time, means for periodically providing educational feedback to users, and means for an administrator to change the rule set. This makes it possible to effectively monitor a wide range of communication data within a company, detect inappropriate expressions and offensive language in real time, and take prompt action. In addition, providing appropriate feedback and education to employees can help improve compliance awareness.

[0529] "Intra-company communications data" refers to data that includes all communications that take place within a company, including text messages, voice messages, emails, etc.

[0530] A "natural language processing module" is a computer program that analyzes collected text data and performs tokenization, part-of-speech tagging, sentence structure analysis, etc.

[0531] A "voice recognition module" is a computer program for converting voice data into text data.

[0532] A "compliance violation detection rule set" is a collection of pre-defined rules for identifying offensive language or legally questionable content.

[0533] "Notification" refers to alerts or messages sent to relevant parties when a compliance violation is detected.

[0534] "Storage device" refers to a database or storage medium for storing analysis results and violation details.

[0535] "Educational feedback" refers to regular reports and notifications to inform users of violations and the reasons for them, and to encourage improvement.

[0536] "Administrator" refers to the person responsible for customizing and managing the rule sets within the system.

[0537] "Means of collecting data in real time" refers to systems and programs that instantly acquire communication data generated within a company.

[0538] "Visualization means" refers to a system or program for graphically displaying analysis results and detected violations.

[0539] The system according to the present invention monitors communication data within a company in real time to detect and prevent compliance violations. An embodiment of the system according to the present invention will be described in detail below.

[0540] System Configuration

[0541] This system consists of a server, a terminal, and a user. The detailed roles of each element are shown below.

[0542] server

[0543] The server is primarily responsible for the following functions:

[0544] It collects data in real time from chat applications, mail servers, and voice communication systems used within a company.

[0545] Examples of specific hardware or software used: Slack API, Microsoft Exchange Server, Google Speech-to-Text API.

[0546] Example: Employee A sends a message on Slack saying, "Your work is always slow."

[0547] The collected data is analyzed using a natural language processing module, which includes tokenization, part-of-speech tagging, and sentence structure analysis.

[0548] Examples of software used: SpaCy, Google Cloud Natural Language API.

[0549] Example: The message "You're always slow at work" is identified as a negative expression.

[0550] The voice data is converted into text using a voice recognition module and analyzed in the same way.

[0551] Example: Voice data is converted into text by a speech recognition module, such as "Your work is always slow."

[0552] The parsed data is then compared against a set of rules to detect non-compliance, including those that contain offensive language or legally questionable content.

[0553] Examples of software used: Custom rule engine, regular expressions (RegEx).

[0554] Example: The expression "you" is identified as a violation.

[0555] If a violation is detected, a notification is issued, which is sent in real time to the message sender's device and to the administrator's device.

[0556] Examples of software used: Twilio, SMTP (sending email).

[0557] Example: Employee A and Manager B receive a notification.

[0558] The details of the detected violation and the analysis results are saved in a recording device.

[0559] Examples of software used: MySQL, PostgreSQL.

[0560] Example: The message "Your work is always slow" and the analysis results are saved.

[0561] Terminal

[0562] Terminals are devices used by users and administrators that provide the following functionality:

[0563] Receive real-time notifications.

[0564] Learn which expressions are non-compliant through regular reports.

[0565] Examples of software used: Tableau, Power BI.

[0566] Example: Employee A checks his / her violation messages and the reasons for them in a regular report.

[0567] Administrators customize the system's compliance violation detection rule set.

[0568] Examples of software used: Web browser, custom management tools.

[0569] Example: Administrator B adds a new keyword "incompetent" to the ruleset.

[0570] User

[0571] The user is an employee of the company and performs the following functions:

[0572] Send and receive messages within the system.

[0573] Example: Employee A sends a message on Slack.

[0574] Receive regular reports and real-time notifications to increase compliance awareness.

[0575] Example: Employee A checks the reason for his / her violation message in the regular report and makes improvements.

[0576] Examples and prompts

[0577] Examples:

[0578] Employee A sends a message on Slack saying, "Your work is always slow."

[0579] The server receives the message and analyzes it using a natural language processing module.

[0580] The message is deemed offensive and checked against a set of rules.

[0581] A violation is detected and the server issues a notification to Employee A and Manager B using Twilio.

[0582] The message and reason for the violation are stored in a MySQL database.

[0583] At a later date, Employee A checks his / her violation message in the Tableau report.

[0584] Administrator B uses a web browser to add a new keyword, "incompetent," to the rule set.

[0585] Example prompt sentence:

[0586] "Please explain the specific process by which an AI system monitors communication data within a company in real time and detects negative expressions. For example, please explain in detail how a message such as 'Your work is always slow' is processed."

[0587] In this way, the system of the present invention can enforce compliance within a company and manage risk in real time.

[0588] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0589] Step 1:

[0590] Data collection

[0591] The server collects data in real time from chat applications, mail servers, and voice communication systems used within the company.

[0592] Input: Message data from Slack, email data from the mail server, and voice data.

[0593] Output: Send the collected data to the text data analysis queue or the audio data analysis queue.

[0594] Specific operation: For example, if employee A sends a message on Slack saying, "Your work is always slow," the message is sent to the server in real time.

[0595] Step 2:

[0596] Data conversion (for audio data)

[0597] The server converts the collected voice data into text using a voice recognition module.

[0598] Input: Collected audio data.

[0599] Output: Audio data converted to text data.

[0600] Specific operation: For example, audio data such as "Your work is always slow" is converted into text using the Google Speech-to-Text API.

[0601] Step 3:

[0602] Data analysis

[0603] The server analyzes the text data using a natural language processing module, which performs tokenization, part-of-speech tagging, and sentence structure analysis.

[0604] Input: Text data (either collected directly or converted by a speech recognition module).

[0605] Output: The parsed data includes word tokens, parts of speech, and sentence structure information.

[0606] Specific operation: For example, the message "Your work is always slow" is parsed through SpaCy and tokenized into "you", "of", "work", "is", "always", and "slow".

[0607] Step 4:

[0608] Compliance violation detection

[0609] The server compares the parsed data against a set of compliance violation detection rules to determine whether a violation has occurred.

[0610] Input: Parsed text data.

[0611] Output: Violation detection results (details of the violation if there is one, no abnormality if there is no violation).

[0612] Specific behavior: For example, the analyzed message contains the offensive keyword "you," so it is detected as a violation.

[0613] Step 5:

[0614] Issuance of notifications

[0615] The server will issue a notification if a violation is detected, which will be sent to the message sender and an administrator.

[0616] Input: Violation detection result and violation content.

[0617] Output: Notification message (including details of non-compliance and time of sending).

[0618] Specific operation: For example, an alert will be sent via Twilio to employee A who sent a message saying, "Your work is always slow," and to his manager B.

[0619] Step 6:

[0620] Data recording

[0621] The server stores the details of the detected violation and the analysis results in a recording device.

[0622] Input: Violation detection results and their analysis.

[0623] Output: Violation records stored in a database.

[0624] Specific behavior: For example, the message "Your work is always slow" and the reason for the violation are recorded in a MySQL database.

[0625] Step 7:

[0626] Providing educational feedback

[0627] The device provides regular feedback to the user, helping them learn what expressions are considered non-compliant.

[0628] Input: Violation records stored in the database.

[0629] Output: Education report provided to the user.

[0630] Specific behavior: For example, Employee A checks his / her violation messages and the reasons for them through regular Tableau reports.

[0631] Step 8:

[0632] Customizing the ruleset

[0633] Administrators can customize rule sets from their own devices and adjust them to suit their company's specific circumstances.

[0634] Input: Ruleset change input by administrator.

[0635] Output: The updated ruleset.

[0636] Specific behavior: For example, Administrator B uses a web browser to add a new keyword, "incompetent," to the rule set.

[0637] (Application example 1)

[0638] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0639] As internal communication increases in modern companies, compliance is becoming more important than ever. However, many companies lack a system for monitoring internal communications in real time and immediately detecting violations, exposing them to the risk of serious compliance violations being overlooked. Even when violations are discovered, response can be delayed, potentially affecting the company's credibility and legal risk. Furthermore, a lack of educational feedback to help employees understand and improve their past violations can lead to repeated violations. To solve these problems, a system is needed that effectively monitors internal communications and immediately detects and notifies violations.

[0640] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0641] In this invention, the server includes means for collecting communication data within a company in real time, means for analyzing the data using a natural language processing module and a voice recognition module, means for comparing the analyzed data with a compliance violation detection rule set, means for issuing an alert when a compliance violation is detected, means for storing the analysis results and the details of the violation in a database, means for periodically providing educational feedback to users, means for an administrator to customize the rule set, means for delivering related notifications via an application installed on a smart device, means for notifying users and administrators of violation detection notifications and details, and means for generating periodic reports based on the violation detection data. This enables real-time monitoring of intra-company communications, immediate detection of compliance violations, prompt notification to target groups, and employee educational feedback.

[0642] "Internal Enterprise Communication Data" refers to the electronic data of all messages, communications, and discussions exchanged by employees within an enterprise.

[0643] "Means of collecting in real time" refers to a system that instantly collects communication data within a company and provides it for processing without delay.

[0644] "Natural language processing module" refers to software or algorithms for analyzing text data and understanding its meaning.

[0645] "Speech Recognition Module" refers to software or algorithms for converting voice data into text.

[0646] A "compliance violation detection rule set" is a set of rules for defining violation actions based on company rules and legal regulations.

[0647] "Means for issuing alerts" refers to a mechanism for immediately notifying relevant parties when a compliance violation is detected.

[0648] "Means for storing in a database" refers to a mechanism for permanently recording the analysis results and violation details and storing them for future reference and analysis.

[0649] "Means for providing educational feedback to users" refers to a mechanism for providing employees with information about compliance violations and advice for improvement.

[0650] "Means for administrators to customize rule sets" refers to a mechanism by which administrators can modify, add, or remove compliance violation detection rule sets according to the specific needs of their enterprise.

[0651] "Smart devices" refers to portable electronic devices with internet connectivity, including smartphones and tablets.

[0652] "Application" means a software program designed to perform a specific job or task.

[0653] "Means for delivering notifications" refers to a mechanism for sending real-time notifications to users and administrators via smart devices, etc.

[0654] "Means for notifying users and administrators of detected violations and providing details of those violations" refers to a mechanism for sending notifications containing detailed information to users and administrators when a violation is detected.

[0655] "Means for generating periodic reports" refers to a system for aggregating violation data over a fixed period and creating an analysis report.

[0656] In order to put the present invention into practice, a system is constructed that collects communication data within a company in real time and analyzes that data. The configuration and processing procedure of this system are described below.

[0657] System Configuration

[0658] The system consists of the following main components:

[0659] 1. Server

[0660] Data collection module: Works with internal company communication tools (chat apps and email servers) to collect data in real time.

[0661] Natural Language Processing (NLP) module: Analyzes collected text data and understands the meaning of messages. Available software includes SpaCy and Google Cloud Natural Language API.

[0662] Speech recognition module: Converts voice data into text. Uses Google Cloud Speech-to-Text API, etc.

[0663] Violation Detection Module: Compares the parsed data against a set of compliance violation detection rules.

[0664] Alerting module: Issues alerts and notifies users and administrators when violations are detected. Notification services include Firebase Cloud Messaging (FCM).

[0665] Data storage module: Analysis results and violation details are saved in a database. SQLite and PostgreSQL are available.

[0666] Report Generation Module: Generates regular reports based on violation data and provides feedback.

[0667] 2. Terminal

[0668] Smart devices (smartphones and tablets): The devices through which users access and receive notifications.

[0669] Application: Software installed on smart devices that provides real-time notifications, checks violations, and provides feedback.

[0670] Processing flow

[0671] 1. Data collection

[0672] The server's data collection module works in conjunction with each communication tool within the company to collect text and voice data in real time.

[0673] 2. Data Analysis

[0674] Natural language processing and speech recognition modules analyze the collected data and understand its meaning. For example, a message like "You're always slow at work" is identified as a negative expression.

[0675] 3. Detecting non-compliance

[0676] The server compares the parsed data against a set of compliance violation detection rules to determine whether there are any violations.

[0677] 4. Issue an alert

[0678] If a violation is detected, the alert module will be activated and send a notification to the user and administrator, such as "A compliance violation has been detected. From: Employee X, Content: 'Your work is always slow', Time: YYYY-MM-DD HH:MM."

[0679] 5. Data storage and report generation

[0680] The data storage module stores the details of violations and their analysis results in a database. Based on the stored data, the report generation module creates regular reports and provides educational feedback to employees.

[0681] Specific examples

[0682] Example of input prompt: "During a new project discussion, a message was sent stating, 'Your work is always slow.' Please determine whether this message is in compliance and issue an alert if necessary."

[0683] Example notification: "A compliance violation has been detected. From: Employee X, Content: 'Your work is always slow', Time: YYYY-MM-DD HH:MM"

[0684] In this way, the present invention provides real-time monitoring of corporate communication data, immediate detection of compliance violations, and prompt notification and educational feedback to target audiences.

[0685] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0686] Step 1:

[0687] The server connects with the company's internal communication tools and collects data in real time. The main inputs are employee chat messages and voice conversations. This data is sent to the server as raw data and added to a queue for further processing. The output is stored in the queue as text data or voice data.

[0688] Step 2:

[0689] The server's natural language processing (NLP) module analyzes the text data stored in the queue. The main input is the text data stored in the queue. The NLP module performs sentence tokenization (breaking down each word), part-of-speech tagging (identifying the grammatical role of each word), and sentence structure analysis (understanding the meaning of the sentence). For example, the message "Your work is always slow" is recognized as an offensive expression. The output is the analyzed text data and its semantic analysis results.

[0690] Step 3:

[0691] The server's speech recognition module converts the voice data stored in the queue into text. The main input is voice data. The speech recognition module (for example, Google Cloud Speech-to-Text API) analyzes the voice data and obtains text data. This text data is passed to the NLP module and analyzed again. The output is the text data converted from the voice data.

[0692] Step 4:

[0693] The server compares the analyzed data with a compliance violation detection rule set. The main input is the analyzed text data. The rule set defines offensive expressions and legally problematic keywords. For example, it includes keywords such as "incompetent" and "useless." This comparison determines whether or not a violation has occurred. The output is a judgment result indicating whether or not a violation has been detected.

[0694] Step 5:

[0695] If a violation is detected, the server's alert module is activated. The main input is the violation detection result. The alert module sends a notification containing the details of the violation to the user's and administrator's devices. Specifically, the notification is sent using Firebase Cloud Messaging (FCM). The output is a notification message sent to the user and administrator. For example, the notification may be in the format "A compliance violation has been detected. Sender: Employee X, Content: 'Your work is always slow', Time: YYYY-MM-DD HH:MM."

[0696] Step 6:

[0697] The server saves the details of violations and the analysis results in a database. The main inputs are the details of violations and the analysis results. This data is stored in a database system such as SQLite or PostgreSQL. The saved data is used for later analysis and audits. The output is the data recorded in the database.

[0698] Step 7:

[0699] The server's report generation module creates periodic reports based on the stored violation data. The main input is the violation data recorded in the database. The report generation module aggregates the violation data and creates an analysis report, which allows companies to provide educational feedback to employees. The output is a periodic analysis report. Specifically, a report in the format of "Violation report for the past month" is created.

[0700] Through the above processing steps, the present invention realizes real-time monitoring within a company, immediate detection and response to compliance violations, data storage, and provision of educational feedback.

[0701] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0702] The system according to the present invention monitors communication data within a company in real time and analyzes compliance violations and user sentiment. An embodiment of this system will be described below.

[0703] System Configuration

[0704] The system mainly consists of the following elements:

[0705] Server: Responsible for data collection, analysis, violation detection, emotion recognition, alert issuance, and data storage.

[0706] Terminal: A device used by users and administrators, such as a PC or smartphone.

[0707] User: An employee of a company who sends and receives messages within the system.

[0708] Administrator: The person responsible for customizing the system's rule set and emotion engine.

[0709] Program processing

[0710] 1. Data collection

[0711] The server receives data in real time from chat applications, mail servers, and voice communication systems used within the company. Depending on the format (text or voice), the data is added to a text analysis queue or a voice analysis queue.

[0712] 2. Data Analysis

[0713] The server passes the text data to a natural language processing (NLP) module, which tokenizes the received text data, tags it with parts of speech, and performs sentence structure analysis to understand the meaning and intent of the message.

[0714] The speech data is converted to text data by a speech recognition module and undergoes the same analysis as other text data: tokenization, part-of-speech tagging, and sentence structure analysis.

[0715] 3. Emotion Analysis

[0716] The server passes the parsed text data to an emotion engine to identify the user's emotion. This emotion engine contains algorithms that recognize the emotional tone of the text (e.g., anger, sadness, joy, etc.). For example, the message "I'm so angry" is identified as an emotion of anger.

[0717] 4. Detecting non-compliance

[0718] The server then checks the parsed data against a set of compliance detection rules, which contain keywords and phrases that identify offensive or legally questionable content, such as "useless" or "incompetent."

[0719] 5. Issuing an Alert

[0720] When a violation is detected, the server generates an alert, taking into account the emotional data. The generated alert includes information such as the content of the violating message, the emotional tone, the sender, and the time of sending. The alert is sent to the terminals of the sender of the violating message and the administrator.

[0721] 6. Data Recording

[0722] The server stores the detected violation data, analysis results, and emotional data in a database. The stored data includes the original message, analysis results, detailed violation information, and emotional tone, making it available for future analysis and audits.

[0723] 7. Providing educational feedback

[0724] On the device, users receive periodic reports and real-time notifications that allow them to learn how their behavior is non-compliant, for example, to see why their message was deemed "offensive."

[0725] 8. Customizing the ruleset

[0726] Administrators can customize the rulesets of the AI ​​modules and sentiment engines from their own devices to address the specific needs and circumstances of their companies, for example by adding new keywords that they consider "offensive."

[0727] Specific examples

[0728] Employee A sends a message in chat saying, "Your work is always slow. It really pisses me off."

[0729] The server receives the message and analyzes it with the NLP module. Based on the analysis results, the message is determined to be negative and aggressive.

[0730] The emotion engine identifies "I'm angry" as anger.

[0731] A violation is detected based on the rule set, and the server immediately issues an alert, notifying employee A and the administrator.

[0732] The message content, reason for violation, and emotional tone are stored in a database.

[0733] At a later date, Employee A will review his / her violation messages, the reasons for them, and the emotional tone of the messages through regular reports, and take appropriate action.

[0734] This specific process allows the system of the present invention to strengthen compliance and sentiment management within a company and manage risk in real time.

[0735] The processing flow will be explained below.

[0736] Step 1:

[0737] The server receives data in real time from the company's chat applications, mail servers, and voice communication systems, and depending on the format (text or voice), adds the data to a text analysis queue or a voice analysis queue.

[0738] Step 2:

[0739] The server passes the text data to a natural language processing (NLP) module, which tokenizes the received text data, tags it with parts of speech, and performs sentence structure analysis. This allows the meaning and intent of the message to be understood. For example, a message such as "You're always slow at work" is broken down into a series of words, and the role of each word is analyzed.

[0740] Step 3:

[0741] The server passes the voice data to a speech recognition module, which converts it into text data. The converted text data is then analyzed by an NLP module. For example, a voice message such as "You're always slow at work" is converted into written text, which is then analyzed.

[0742] Step 4:

[0743] The server passes the parsed text data to an emotion engine to identify the user's emotion. The emotion engine identifies the emotional tone of the text (e.g., anger, sadness, joy, etc.). For example, if the sentence "I'm so angry" is sent, it will be recognized as "anger."

[0744] Step 5:

[0745] The server compares the analyzed data and sentiment data with a set of rules to detect compliance violations. The rules include keywords and phrases that identify offensive or legally problematic content. For example, the server checks whether the message contains phrases like "useless" or "incompetent." A message like "You're always slow at work, it pisses me off" would be considered a violation.

[0746] Step 6:

[0747] When a violation is detected, the server generates an alert, taking into account emotional data. The alert includes the content of the violating message, its emotional tone (e.g., "anger"), the sender, and the time of sending. The alert is sent to the terminals of the sender of the violating message and the administrator.

[0748] Step 7:

[0749] The server stores the detected violation data, its analysis results, and emotional data in a database. The stored data includes the original message, analysis results, detailed information about the violation, and emotional tone. For example, "Your work is always slow. It makes me so angry" is stored, and the analysis results and the emotional tone of "anger" are recorded.

[0750] Step 8:

[0751] On the device, users receive periodic reports and real-time notifications, which help them learn how their actions are violating compliance. For example, users can see their violation messages, the reasons for their violations, and their emotional tone.

[0752] Step 9:

[0753] Administrators can customize the rulesets for the AI ​​module and emotion engine from their devices, for example adding new keywords that are considered "aggressive" or emotional tones (such as "sadness" or "joy").

[0754] This specific processing step allows enterprises to accurately monitor and quickly respond to compliance violations and users' emotional states in real time.

[0755] Example 2

[0756] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0757] Communications within companies are increasing every day, and some of them may contain offensive language or inappropriate content, which increases the company's compliance risks. It is also difficult to grasp changes in employee emotions in real time and take appropriate action. To solve these issues, a system is needed that monitors communication data in real time, analyzes emotions, and detects compliance violations.

[0758] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0759] In this invention, the server includes means for collecting communication data within a company in real time, means for analyzing the data using a natural language processing module and a voice recognition module, means for identifying user emotions from the analyzed data using an emotion engine, means for comparing the analyzed data with a compliance violation detection rule set, means for issuing an alert including the content, emotional tone, sender, and sending time of the violation message when a compliance violation is detected, means for storing the analysis results and the violation details in a database, means for periodically providing educational feedback to users, and means for an administrator to customize the rule set. This makes it possible to monitor communications within a company in real time, detect inappropriate expressions and changes in employee emotions early, and take appropriate measures.

[0760] "Internal communications data" refers to all forms of data that take place within a company, including emails, chat messages, voice calls, video conferences, etc.

[0761] "Real-time" refers to data being processed the instant it is generated or with very little delay.

[0762] A "natural language processing module" refers to a software module that allows a computer to analyze and understand the language that humans use on a daily basis.

[0763] "Speech recognition module" refers to a software module for converting voice data into text data.

[0764] An "emotion engine" refers to a system that includes algorithms for analyzing text or speech and identifying the emotional tones (e.g., joy, anger, sadness, etc.) contained within it.

[0765] "Rule Set" refers to the list of keywords or phrases and associated rules used to determine non-compliance.

[0766] An "alert" refers to a message or warning signal that notifies a user or administrator when the system detects an abnormality or problem.

[0767] "Database" refers to a system for storing information in an organized manner and for efficiently searching and managing it.

[0768] "Educational feedback" refers to a mechanism that provides users with an evaluation of their actions and suggestions for improvement.

[0769] An "administrator" is a person who has the authority to configure and operate the system and is responsible for customizing the rule set and monitoring the system.

[0770] The system based on this invention monitors communication data within a company in real time and analyzes compliance violations and user sentiment. This system is mainly composed of three elements: a server, a terminal, and a user.

[0771] System Configuration

[0772] 1. Server:

[0773] This is the core of the system, responsible for data collection, analysis, violation detection, emotion recognition, alert generation, and data storage. The server contains the following modules:

[0774] Natural Language Processing (NLP) module: Analyzes communication data using software tools such as SpaCy and NLTK.

[0775] Speech recognition module: Converts voice data into text using a module such as Google Speech-to-Text.

[0776] Emotion engine: Identifies the emotional tone of a user's text data using tools such as IBM Watson Tone Analyzer.

[0777] Database: A storage system for storing analysis results, violation details, and emotional tones.

[0778] 2. Terminal:

[0779] These are devices used by users and administrators, such as PCs and smartphones, to receive necessary reports and notifications.

[0780] 3. User:

[0781] These are company employees who send and receive messages within this system, and communicate via chat, email, etc. in the course of their daily work.

[0782] Specific examples

[0783] Example of employee A sending a message in chat saying, "Your work is always slow. It really pisses me off."

[0784] The server receives this message and adds it to the text analysis queue.

[0785] The NLP module tokenizes the message and performs part-of-speech tagging and sentence structure analysis.

[0786] The emotion engine identifies "annoyed" as the emotion "anger."

[0787] A non-compliance rule set identifies "Your work is always slow" as an offensive expression.

[0788] The server generates an alert containing details of the violation and emotional tone, and notifies Employee A and the manager.

[0789] This message, along with the analysis results, violation details, and emotional tone, is saved in a database.

[0790] At a later date, Employee A can review the periodic report on his / her own device and learn the reason for the violation and the emotional tone.

[0791] Prompt Sentence Examples

[0792] "Develop a system that performs employee sentiment analysis. The system collects chat data in real time, analyzes the text using natural language processing, and identifies emotional tones. It then detects compliance violations and sends alerts to managers if violations are detected. It also stores the emotional tones and violation data in a database so employees can receive feedback."

[0793] This system allows companies to monitor communication content in real time, detect and respond to inappropriate language and changes in employee sentiment early on, and provides employees with the opportunity to review and improve their own behavior through feedback.

[0794] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0795] Step 1: Collect data

[0796] The server receives data in real time from chat applications, mail servers, and voice communication systems used within the company, such as Slack, Microsoft Teams, email, and Zoom.

[0797] Input: Chat messages, emails, voice data

[0798] Output: Add data to the text analysis queue and speech analysis queue

[0799] How it works: The server sorts data received from each system into the appropriate analysis queue depending on its format (text or audio).

[0800] Step 2: Analyzing the text data

[0801] The server sends the data in the text analysis queue to a natural language processing (NLP) module, which performs tokenization, part-of-speech tagging, and sentence structure analysis of the text data.

[0802] Input: Text data

[0803] Output: Tokenized data, tagged data, sentence structure analysis results

[0804] How it works: It splits incoming text data into words and phrases, tags them with part-of-speech tags, and analyzes the structure of entire sentences.

[0805] Step 3: Analyzing the audio data

[0806] The server sends the speech data added to the speech analysis queue to a speech recognition module, which converts the speech data into text data that is then analyzed in a subsequent step.

[0807] Input: Audio data

[0808] Output: Text data

[0809] How it works: Converts speech to text and feeds it into an NLP module for further analysis.

[0810] Step 4: Sentiment Analysis

[0811] The server passes the text data analyzed by the NLP module to an emotion engine to identify the user's emotion. The emotion engine detects the emotional tone of the text (e.g., anger, sadness, joy, etc.).

[0812] Input: Parsed text data

[0813] Output: Emotional tone identification result

[0814] How it works: The emotion engine analyzes text and recognizes emotional tone, for example identifying the message "I'm so angry" as "anger."

[0815] Step 5: Detecting non-compliance

[0816] The server then compares the parsed text data against a set of compliance detection rules, which contain keywords and phrases to identify offensive or legally problematic content.

[0817] Input: Parsed text data, emotional tone

[0818] Output: Whether or not there is a violation, detailed violation information

[0819] How it works: The server checks the text data against a set of rules to see if there are any violations, such as whether it contains keywords like "useless" or "incompetent."

[0820] Step 6: Issue an alert

[0821] When a violation is detected, the server generates an alert, taking into account the emotional data. The generated alert includes information such as the content, emotional tone, sender, and time of the violating message.

[0822] Input: Violation details, emotional tone

[0823] Output: Alert message

[0824] What it does: When a violation is detected, an alert is generated to notify the sender of the offending message and an administrator. The alert includes details such as the content and emotional tone of the offending message.

[0825] Step 7: Record the data

[0826] The server stores the data on which violations were detected, along with the analysis results and emotion data, in a database for future analysis and auditing.

[0827] Input: Violation details, emotional tone, message content

[0828] Output: Saved entries in the database

[0829] How it works: Analysis results and violation information are recorded in a database and stored for future reference and verification.

[0830] Step 8: Providing educational feedback

[0831] The device displays regular reports and real-time notifications to users, allowing them to learn and improve how their behavior is out of compliance.

[0832] Input: Analysis results, violation information, emotional tone

[0833] Output: Educational feedback report, real-time notifications

[0834] Action: Generate and provide feedback to the user based on their behavior and violations, for example informing them of the specific reasons why something was deemed "offensive."

[0835] Step 9: Customizing the ruleset

[0836] Administrators can use their devices to customize the rulesets of the AI ​​modules and emotion engines to meet the specific needs and circumstances of their companies.

[0837] Input: New rule or keyword

[0838] Output: Updated ruleset

[0839] How it works: Administrators adapt the ruleset to suit their company's needs and apply it to the system, which may include adding new keywords that are considered "offensive."

[0840] (Application example 2)

[0841] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0842] In recent years, inappropriate behavior and emotional changes between employees have often been the cause of trouble in corporate communications. In particular, in environments such as factories, where cooperation between employees and robots is necessary, communication is a crucial element. However, there is a lack of means to monitor this communication in real time and prevent problems before they occur. Therefore, there is a growing need for a system that can analyze all communication data, including emotions, in real time, identify compliance violations and negative emotional states, and immediately issue alerts.

[0843] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0844] In this invention, the server includes a means for collecting communication data within the company in real time, a means for analyzing the data using a natural language analysis module and a voice recognition module, and a means for comparing the analyzed data with a compliance violation detection rule set. This allows for real-time monitoring of communications between employees and robots and issuing warnings at appropriate times. The system also analyzes the user's emotional state and provides feedback based on that analysis, creating a healthier communication environment.

[0845] Definitions of important words

[0846] "Communication data" refers to data including text data and voice data generated within a company, as well as the results of sentiment analysis.

[0847] "Real-time" refers to immediate processing or analysis of data the moment it is generated or transmitted.

[0848] "Natural language analysis" refers to the process of mechanically interpreting text or audio data through tokenization, part-of-speech tagging, sentence structure analysis, etc.

[0849] "Speech recognition module" refers to a computer program or system for converting voice data into text data.

[0850] A "compliance violation detection rule set" refers to a collection of rules for detecting expressions that may violate legal or corporate standards based on specific keywords or phrases.

[0851] "Warning" refers to a notification or alert issued when non-compliance or a negative emotional state is detected.

[0852] "Database" refers to a system or structured collection of information for storing collected and analyzed data.

[0853] "Educational Feedback" refers to reports or guidance provided to users to help them learn about their communication and non-compliance behaviors.

[0854] "Customizing rule sets" refers to the act of an administrator tailoring compliance violation detection rules and sentiment analysis algorithms to the specific needs and circumstances of the company.

[0855] "User emotional state" refers to the user's emotional tone (e.g., anger, sadness, joy, etc.) derived from analyzed text data and voice data.

[0856] "Form for carrying out the invention" of the specification

[0857] System Configuration

[0858] The system according to the present invention comprises the following components:

[0859] Server: Responsible for data collection, analysis, violation detection, emotion recognition, warning issuance, and data storage. Specifically, it is equipped with a natural language analysis module, a voice recognition module, a compliance violation detection rule set, and an emotion analysis engine.

[0860] Terminal: A device used by employees and managers, such as a PC, a smartphone, or an application installed on a robot in a factory.

[0861] User: An employee of a company who sends and receives messages within the system.

[0862] Administrator: The person responsible for customizing the system's rule sets and sentiment analysis engine.

[0863] Explanation of program processing

[0864] The server uses a communications API to collect internal company communication data (text and voice data) in real time. This data is converted into text data through a natural language analysis module and a speech recognition module, and then analyzed based on the extracted tokens. In particular, the NLP (natural language processing) module performs tokenization, part-of-speech tagging, and sentence structure analysis to understand the meaning and intent of the message.

[0865] Meanwhile, the voice data is converted into text data by a speech recognition module. The analyzed text data is then passed to a sentiment analysis engine, which identifies emotional tones such as anger, sadness, joy, etc. If a user's message is "I'm so angry," the emotional tone will be recognized as "anger."

[0866] The analyzed data is compared against a set of compliance violation detection rules, which include offensive or legally questionable language. If such language is detected, the server immediately issues an alert. The alert includes information about the content, emotional tone, sender, and time of the offending message, and this information is sent to the user and administrator who detected the violation.

[0867] The data is stored in a database for permanent archival, including the original message, analysis results, violation details, and emotional tone, making it available for future analysis and audits.

[0868] At the device, users receive regular educational feedback on their communication behavior, including how they were non-compliant and their emotional state at the time. Administrators can customize the rule set and optimize the system by adding new keywords and phrases as needed.

[0869] Specific examples

[0870] For example, suppose employee A sends a message saying, "You're always slow at work. It pisses me off." In this case, the server would process it as follows:

[0871] 1. Messages are received in real time and analyzed by the NLP module.

[0872] 2. The sentiment analysis engine identifies the phrase "I'm angry" as anger.

[0873] 3. Based on the rule set, this message is detected as non-compliant.

[0874] 4. The server immediately issues an alert and notifies Employee A and the administrator.

[0875] Prompt Sentence Examples

[0876] Monitor chat logs between Employee A and Employee B in real time to detect compliance violations and sentiment changes. Start your analysis with the following chat logs:

[0877] "Employee A: You're always slow at work. It really pisses me off."

[0878] The system monitors communication between employees and robots in real time and provides immediate and appropriate feedback to support safe and efficient operations within the factory.

[0879] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0880] Program processing flow

[0881] Step 1:

[0882] The server collects communication data (text data and voice data) within the company in real time. Specifically, it uses a communication API to obtain messages and call content and adds them to a data queue. The input data is raw text or voice data, and the output is data added to the analysis queue.

[0883] Step 2:

[0884] The server passes the collected text data to a natural language analysis module, which tokenizes the data, tags it with parts of speech, and analyzes sentence structure. The input data is text data, and the output is tokenized analyzed data. The analyzed data is used to understand the meaning and intent of the message.

[0885] Step 3:

[0886] The server passes the collected voice data to a voice recognition module and converts it into text data. The voice recognition module analyzes the voice data and converts it into text. The input data is voice data and the output is text data. The converted text data is further analyzed by a natural language analysis module.

[0887] Step 4:

[0888] The server passes the parsed text data to the sentiment analysis engine, which analyzes the emotional tone of the text to identify the user's emotional state. The input data is the parsed text data, and the output is the identified emotional tone. For example, the message "I'm so angry" is recognized as "Anger."

[0889] Step 5:

[0890] The server compares the analyzed data against a set of compliance violation detection rules. The rule set contains keywords and phrases that determine whether a violation has occurred. The input data is the analyzed and sentiment analysis data, and the output is a determination of whether or not there is a compliance violation. If a violation is detected, information about it is generated.

[0891] Step 6:

[0892] The server immediately issues a warning if a violation is detected. The warning includes the content, emotional tone, sender, and time of sending of the violating message. The input data is the compliance violation judgment result, and the output is the generation and notification of a warning message. The notification is sent to the user who detected the violation and the administrator.

[0893] Step 7:

[0894] The server stores the analysis results and violation details in a database. The stored data includes the original message, analysis results, violation details, and emotional tone. The input data is the analyzed and violation detection data, and the output is storage in the database.

[0895] Step 8:

[0896] The device periodically provides educational feedback to the user, including how the user violated compliance and their emotional state at the time. The input data are the analysis results stored in the database, and the output is a report of the educational feedback.

[0897] Step 9:

[0898] Administrators use a terminal to customize the rule set. They can add new keywords and phrases. The input is the administrator's configuration information, and the output is an updated rule set. This allows the system to adapt to the specific needs and circumstances of the company.

[0899] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0900] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0901] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0902] [Third embodiment]

[0903] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0904] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0905] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0906] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0907] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0908] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0909] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0910] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0911] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0912] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0913] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0914] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0915] The system according to the present invention monitors communication data within a company in real time to detect and prevent compliance violations. An embodiment of this system will be described below.

[0916] System Configuration

[0917] The system mainly consists of the following elements:

[0918] Server: Responsible for data collection, analysis, violation detection, alert generation, and data storage.

[0919] Terminal: A device used by users and administrators, such as a PC or smartphone.

[0920] User: An employee of a company who sends and receives messages within the system.

[0921] Administrator: A person responsible for customizing the system's rule set.

[0922] Program processing

[0923] 1. Data collection

[0924] The server receives data in real time from chat applications, mail servers, and voice communication systems used within the company, and depending on the format (text, voice), sends the data to the corresponding analysis queue.

[0925] 2. Data Analysis

[0926] The natural language processing module running on the server analyzes the collected text data. This module tokenizes sentences, tags parts of speech, and analyzes sentence structure to understand the meaning of the message. For example, a message such as "Your work is always slow" is identified as a negative expression.

[0927] The speech data is converted into text by a speech recognition module and subjected to similar analysis. For example, a user's utterance "Your work is always slow" is converted into text and then analyzed by a natural language processing module.

[0928] 3. Detecting non-compliance

[0929] The server then compares the parsed data against a predefined set of rules to detect compliance violations. These rules define offensive or legally problematic content. For example, phrases containing keywords such as "useless" or "incompetent" are identified as violations.

[0930] 4. Issue an alert

[0931] If a violation is detected, the server immediately issues an alert, which is sent to the message sender's device and the administrator's device. The alert includes the violating message, sender, and time of sending.

[0932] 5. Data Recording

[0933] The server records the detected violations and the analysis results in a database. This data is used for future analysis and audits. For example, a message such as "Your work is always slow" is recorded as a violation.

[0934] 6. Providing educational feedback

[0935] On the device, users receive periodic reports and real-time notifications, which help them learn which expressions are non-compliant. For example, users can view their non-compliance messages and the reasons for their actions.

[0936] 7. Customizing the ruleset

[0937] Administrators can customize the rule set of the AI ​​module from their own devices, which is necessary to adapt to the specific circumstances of the company, for example by adding new keywords based on new internal rules.

[0938] Specific examples

[0939] Employee A sends a message in chat saying, "Your work is always slow."

[0940] The server receives the message and analyzes it using a natural language processing module. Based on the analysis results, the message is determined to be negative and offensive.

[0941] A violation is detected based on the rule set, and the server immediately issues an alert, notifying employee A and the administrator.

[0942] The message content and reason for the violation are stored in a database.

[0943] At a later date, Employee A will check his / her violation message through a regular report and be warned.

[0944] In this way, the system of the present invention can enforce compliance within a company and manage risk in real time.

[0945] The processing flow will be explained below.

[0946] Step 1:

[0947] The server receives data in real time from the company's chat applications, mail servers, and voice communication systems, and depending on the format (text or voice), adds the data to a text analysis queue or a voice analysis queue.

[0948] Step 2:

[0949] The server passes the text data to a natural language processing (NLP) module, which tokenizes the received text data, tags it with parts of speech, and performs sentence structure analysis to understand the meaning and intent of the message.

[0950] Step 3:

[0951] The server passes the speech data to a speech recognition module, which converts it into text data. This text data is then analyzed by an NLP module, which performs tokenization, part-of-speech tagging, and sentence structure analysis, just like any other text data.

[0952] Step 4:

[0953] The server then compares the parsed text data against a set of compliance detection rules, which contain keywords and phrases that identify offensive or legally questionable content, such as "useless" or "incompetent."

[0954] Step 5:

[0955] The server generates an alert if a violation is detected based on the matching results. The generated alert includes information such as the content of the violating message, the sender, and the time of sending. The alert is sent to the terminals of the sender of the violating message and the administrator.

[0956] Step 6:

[0957] The server stores the detected violation data and its analysis results in a database, including the original message, analysis results, and detailed information about the violation, making it available for future analysis and audits.

[0958] Step 7:

[0959] On the device, users receive periodic reports and real-time notifications that allow them to learn how their behavior is non-compliant, for example, to see why their message was deemed "offensive."

[0960] Step 8:

[0961] From their own devices, administrators can customize the rule sets of the AI ​​module, changing or adding violation detection criteria and rules according to the specific needs and circumstances of the company. For example, this could include adding new keywords that are considered "offensive."

[0962] This specific processing step allows businesses to effectively detect compliance violations in real time and respond immediately.

[0963] Example 1

[0964] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0965] The volume of communication data within a company is enormous, making it difficult to detect compliance violations from that data. It is also difficult to detect in real time inappropriate or offensive language used unconsciously by employees in communications and take appropriate action. A lack of employee training feedback can lead to similar problems recurring. Furthermore, properly managing different rule sets for each company is also a challenge.

[0966] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0967] In this invention, the server includes means for collecting communication data within a company in real time, means for analyzing the data using a natural language processing module and a voice recognition module, means for comparing the analyzed data with a compliance violation detection rule set, means for saving the analysis results and violation details in a storage device, means for visualizing violation detection notifications and analysis results in real time, means for periodically providing educational feedback to users, and means for an administrator to change the rule set. This makes it possible to effectively monitor a wide range of communication data within a company, detect inappropriate expressions and offensive language in real time, and take prompt action. In addition, providing appropriate feedback and education to employees can help improve compliance awareness.

[0968] "Intra-company communications data" refers to data that includes all communications that take place within a company, including text messages, voice messages, emails, etc.

[0969] A "natural language processing module" is a computer program that analyzes collected text data and performs tokenization, part-of-speech tagging, sentence structure analysis, etc.

[0970] A "voice recognition module" is a computer program for converting voice data into text data.

[0971] A "compliance violation detection rule set" is a collection of pre-defined rules for identifying offensive language or legally questionable content.

[0972] "Notification" refers to alerts or messages sent to relevant parties when a compliance violation is detected.

[0973] "Storage device" refers to a database or storage medium for storing analysis results and violation details.

[0974] "Educational feedback" refers to regular reports and notifications to inform users of violations and the reasons for them, and to encourage improvement.

[0975] "Administrator" refers to the person responsible for customizing and managing the rule sets within the system.

[0976] "Means of collecting data in real time" refers to systems and programs that instantly acquire communication data generated within a company.

[0977] "Visualization means" refers to a system or program for graphically displaying analysis results and detected violations.

[0978] The system according to the present invention monitors communication data within a company in real time to detect and prevent compliance violations. An embodiment of the system according to the present invention will be described in detail below.

[0979] System Configuration

[0980] This system consists of a server, a terminal, and a user. The detailed roles of each element are shown below.

[0981] server

[0982] The server is primarily responsible for the following functions:

[0983] It collects data in real time from chat applications, mail servers, and voice communication systems used within a company.

[0984] Examples of specific hardware or software used: Slack API, Microsoft Exchange Server, Google Speech-to-Text API.

[0985] Example: Employee A sends a message on Slack saying, "Your work is always slow."

[0986] The collected data is analyzed using a natural language processing module, which includes tokenization, part-of-speech tagging, and sentence structure analysis.

[0987] Examples of software used: SpaCy, Google Cloud Natural Language API.

[0988] Example: The message "You're always slow at work" is identified as a negative expression.

[0989] The voice data is converted into text using a voice recognition module and analyzed in the same way.

[0990] Example: Voice data is converted into text by a speech recognition module, such as "Your work is always slow."

[0991] The parsed data is then compared against a set of rules to detect non-compliance, including those that contain offensive language or legally questionable content.

[0992] Examples of software used: Custom rule engine, regular expressions (RegEx).

[0993] Example: The expression "you" is identified as a violation.

[0994] If a violation is detected, a notification is issued, which is sent in real time to the message sender's device and to the administrator's device.

[0995] Examples of software used: Twilio, SMTP (sending email).

[0996] Example: Employee A and Manager B receive a notification.

[0997] The details of the detected violation and the analysis results are saved in a recording device.

[0998] Examples of software used: MySQL, PostgreSQL.

[0999] Example: The message "Your work is always slow" and the analysis results are saved.

[1000] Terminal

[1001] Terminals are devices used by users and administrators that provide the following functionality:

[1002] Receive real-time notifications.

[1003] Learn which expressions are non-compliant through regular reports.

[1004] Examples of software used: Tableau, Power BI.

[1005] Example: Employee A checks his / her violation messages and the reasons for them in a regular report.

[1006] Administrators customize the system's compliance violation detection rule set.

[1007] Examples of software used: Web browser, custom management tools.

[1008] Example: Administrator B adds a new keyword "incompetent" to the ruleset.

[1009] User

[1010] The user is an employee of the company and performs the following functions:

[1011] Send and receive messages within the system.

[1012] Example: Employee A sends a message on Slack.

[1013] Receive regular reports and real-time notifications to increase compliance awareness.

[1014] Example: Employee A checks the reason for his / her violation message in the regular report and makes improvements.

[1015] Examples and prompts

[1016] Examples:

[1017] Employee A sends a message on Slack saying, "Your work is always slow."

[1018] The server receives the message and analyzes it using a natural language processing module.

[1019] The message is deemed offensive and checked against a set of rules.

[1020] A violation is detected and the server issues a notification to Employee A and Manager B using Twilio.

[1021] The message and reason for the violation are stored in a MySQL database.

[1022] At a later date, Employee A checks his / her violation message in the Tableau report.

[1023] Administrator B uses a web browser to add a new keyword, "incompetent," to the rule set.

[1024] Example prompt sentence:

[1025] "Please explain the specific process by which an AI system monitors communication data within a company in real time and detects negative expressions. For example, please explain in detail how a message such as 'Your work is always slow' is processed."

[1026] In this way, the system of the present invention can enforce compliance within a company and manage risk in real time.

[1027] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1028] Step 1:

[1029] Data collection

[1030] The server collects data in real time from chat applications, mail servers, and voice communication systems used within the company.

[1031] Input: Message data from Slack, email data from the mail server, and voice data.

[1032] Output: Send the collected data to the text data analysis queue or the audio data analysis queue.

[1033] Specific operation: For example, if employee A sends a message on Slack saying, "Your work is always slow," the message is sent to the server in real time.

[1034] Step 2:

[1035] Data conversion (for audio data)

[1036] The server converts the collected voice data into text using a voice recognition module.

[1037] Input: Collected audio data.

[1038] Output: Audio data converted to text data.

[1039] Specific operation: For example, audio data such as "Your work is always slow" is converted into text using the Google Speech-to-Text API.

[1040] Step 3:

[1041] Data analysis

[1042] The server analyzes the text data using a natural language processing module, which performs tokenization, part-of-speech tagging, and sentence structure analysis.

[1043] Input: Text data (either collected directly or converted by a speech recognition module).

[1044] Output: The parsed data includes word tokens, parts of speech, and sentence structure information.

[1045] Specific operation: For example, the message "Your work is always slow" is parsed through SpaCy and tokenized into "you", "of", "work", "is", "always", and "slow".

[1046] Step 4:

[1047] Compliance violation detection

[1048] The server compares the parsed data against a set of compliance violation detection rules to determine whether a violation has occurred.

[1049] Input: Parsed text data.

[1050] Output: Violation detection results (details of the violation if there is one, no abnormality if there is no violation).

[1051] Specific behavior: For example, the analyzed message contains the offensive keyword "you," so it is detected as a violation.

[1052] Step 5:

[1053] Issuance of notifications

[1054] The server will issue a notification if a violation is detected, which will be sent to the message sender and an administrator.

[1055] Input: Violation detection result and violation content.

[1056] Output: Notification message (including details of non-compliance and time of sending).

[1057] Specific operation: For example, an alert will be sent via Twilio to employee A who sent a message saying, "Your work is always slow," and to his manager B.

[1058] Step 6:

[1059] Data recording

[1060] The server stores the details of the detected violation and the analysis results in a recording device.

[1061] Input: Violation detection results and their analysis.

[1062] Output: Violation records stored in a database.

[1063] Specific behavior: For example, the message "Your work is always slow" and the reason for the violation are recorded in a MySQL database.

[1064] Step 7:

[1065] Providing educational feedback

[1066] The device provides regular feedback to the user, helping them learn what expressions are considered non-compliant.

[1067] Input: Violation records stored in the database.

[1068] Output: Education report provided to the user.

[1069] Specific behavior: For example, Employee A checks his / her violation messages and the reasons for them through regular Tableau reports.

[1070] Step 8:

[1071] Customizing the ruleset

[1072] Administrators can customize rule sets from their own devices and adjust them to suit their company's specific circumstances.

[1073] Input: Ruleset change input by administrator.

[1074] Output: The updated ruleset.

[1075] Specific behavior: For example, Administrator B uses a web browser to add a new keyword, "incompetent," to the rule set.

[1076] (Application example 1)

[1077] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1078] As internal communication increases in modern companies, compliance is becoming more important than ever. However, many companies lack a system for monitoring internal communications in real time and immediately detecting violations, exposing them to the risk of serious compliance violations being overlooked. Even when violations are discovered, response can be delayed, potentially affecting the company's credibility and legal risk. Furthermore, a lack of educational feedback to help employees understand and improve their past violations can lead to repeated violations. To solve these problems, a system is needed that effectively monitors internal communications and immediately detects and notifies violations.

[1079] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1080] In this invention, the server includes means for collecting communication data within a company in real time, means for analyzing the data using a natural language processing module and a voice recognition module, means for comparing the analyzed data with a compliance violation detection rule set, means for issuing an alert when a compliance violation is detected, means for storing the analysis results and the details of the violation in a database, means for periodically providing educational feedback to users, means for an administrator to customize the rule set, means for delivering related notifications via an application installed on a smart device, means for notifying users and administrators of violation detection notifications and details, and means for generating periodic reports based on the violation detection data. This enables real-time monitoring of intra-company communications, immediate detection of compliance violations, prompt notification to target groups, and employee educational feedback.

[1081] "Internal Enterprise Communication Data" refers to the electronic data of all messages, communications, and discussions exchanged by employees within an enterprise.

[1082] "Means of collecting in real time" refers to a system that instantly collects communication data within a company and provides it for processing without delay.

[1083] "Natural language processing module" refers to software or algorithms for analyzing text data and understanding its meaning.

[1084] "Speech Recognition Module" refers to software or algorithms for converting voice data into text.

[1085] A "compliance violation detection rule set" is a set of rules for defining violation actions based on company rules and legal regulations.

[1086] "Means for issuing alerts" refers to a mechanism for immediately notifying relevant parties when a compliance violation is detected.

[1087] "Means for storing in a database" refers to a mechanism for permanently recording the analysis results and violation details and storing them for future reference and analysis.

[1088] "Means for providing educational feedback to users" refers to a mechanism for providing employees with information about compliance violations and advice for improvement.

[1089] "Means for administrators to customize rule sets" refers to a mechanism by which administrators can modify, add, or remove compliance violation detection rule sets according to the specific needs of their enterprise.

[1090] "Smart devices" refers to portable electronic devices with internet connectivity, including smartphones and tablets.

[1091] "Application" means a software program designed to perform a specific job or task.

[1092] "Means for delivering notifications" refers to a mechanism for sending real-time notifications to users and administrators via smart devices, etc.

[1093] "Means for notifying users and administrators of detected violations and providing details of those violations" refers to a mechanism for sending notifications containing detailed information to users and administrators when a violation is detected.

[1094] "Means for generating periodic reports" refers to a system for aggregating violation data over a fixed period and creating an analysis report.

[1095] In order to put the present invention into practice, a system is constructed that collects communication data within a company in real time and analyzes that data. The configuration and processing procedure of this system are described below.

[1096] System Configuration

[1097] The system consists of the following main components:

[1098] 1. Server

[1099] Data collection module: Works with internal company communication tools (chat apps and email servers) to collect data in real time.

[1100] Natural Language Processing (NLP) module: Analyzes collected text data and understands the meaning of messages. Available software includes SpaCy and Google Cloud Natural Language API.

[1101] Speech recognition module: Converts voice data into text. Uses Google Cloud Speech-to-Text API, etc.

[1102] Violation Detection Module: Compares the parsed data against a set of compliance violation detection rules.

[1103] Alerting module: Issues alerts and notifies users and administrators when violations are detected. Notification services include Firebase Cloud Messaging (FCM).

[1104] Data storage module: Analysis results and violation details are saved in a database. SQLite and PostgreSQL are available.

[1105] Report Generation Module: Generates regular reports based on violation data and provides feedback.

[1106] 2. Terminal

[1107] Smart devices (smartphones and tablets): The devices through which users access and receive notifications.

[1108] Application: Software installed on smart devices that provides real-time notifications, checks violations, and provides feedback.

[1109] Processing flow

[1110] 1. Data collection

[1111] The server's data collection module works in conjunction with each communication tool within the company to collect text and voice data in real time.

[1112] 2. Data Analysis

[1113] Natural language processing and speech recognition modules analyze the collected data and understand its meaning. For example, a message like "You're always slow at work" is identified as a negative expression.

[1114] 3. Detecting non-compliance

[1115] The server compares the parsed data against a set of compliance violation detection rules to determine whether there are any violations.

[1116] 4. Issue an alert

[1117] If a violation is detected, the alert module will be activated and send a notification to the user and administrator, such as "A compliance violation has been detected. From: Employee X, Content: 'Your work is always slow', Time: YYYY-MM-DD HH:MM."

[1118] 5. Data storage and report generation

[1119] The data storage module stores the details of violations and their analysis results in a database. Based on the stored data, the report generation module creates regular reports and provides educational feedback to employees.

[1120] Specific examples

[1121] Example of input prompt: "During a new project discussion, a message was sent stating, 'Your work is always slow.' Please determine whether this message is in compliance and issue an alert if necessary."

[1122] Example notification: "A compliance violation has been detected. From: Employee X, Content: 'Your work is always slow', Time: YYYY-MM-DD HH:MM"

[1123] In this way, the present invention provides real-time monitoring of corporate communication data, immediate detection of compliance violations, and prompt notification and educational feedback to target audiences.

[1124] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1125] Step 1:

[1126] The server connects with the company's internal communication tools and collects data in real time. The main inputs are employee chat messages and voice conversations. This data is sent to the server as raw data and added to a queue for further processing. The output is stored in the queue as text data or voice data.

[1127] Step 2:

[1128] The server's natural language processing (NLP) module analyzes the text data stored in the queue. The main input is the text data stored in the queue. The NLP module performs sentence tokenization (breaking down each word), part-of-speech tagging (identifying the grammatical role of each word), and sentence structure analysis (understanding the meaning of the sentence). For example, the message "Your work is always slow" is recognized as an offensive expression. The output is the analyzed text data and its semantic analysis results.

[1129] Step 3:

[1130] The server's speech recognition module converts the voice data stored in the queue into text. The main input is voice data. The speech recognition module (for example, Google Cloud Speech-to-Text API) analyzes the voice data and obtains text data. This text data is passed to the NLP module and analyzed again. The output is the text data converted from the voice data.

[1131] Step 4:

[1132] The server compares the analyzed data with a compliance violation detection rule set. The main input is the analyzed text data. The rule set defines offensive expressions and legally problematic keywords. For example, it includes keywords such as "incompetent" and "useless." This comparison determines whether or not a violation has occurred. The output is a judgment result indicating whether or not a violation has been detected.

[1133] Step 5:

[1134] If a violation is detected, the server's alert module is activated. The main input is the violation detection result. The alert module sends a notification containing the details of the violation to the user's and administrator's devices. Specifically, the notification is sent using Firebase Cloud Messaging (FCM). The output is a notification message sent to the user and administrator. For example, the notification may be in the format "A compliance violation has been detected. Sender: Employee X, Content: 'Your work is always slow', Time: YYYY-MM-DD HH:MM."

[1135] Step 6:

[1136] The server saves the details of violations and the analysis results in a database. The main inputs are the details of violations and the analysis results. This data is stored in a database system such as SQLite or PostgreSQL. The saved data is used for later analysis and audits. The output is the data recorded in the database.

[1137] Step 7:

[1138] The server's report generation module creates periodic reports based on the stored violation data. The main input is the violation data recorded in the database. The report generation module aggregates the violation data and creates an analysis report, which allows companies to provide educational feedback to employees. The output is a periodic analysis report. Specifically, a report in the format of "Violation report for the past month" is created.

[1139] Through the above processing steps, the present invention realizes real-time monitoring within a company, immediate detection and response to compliance violations, data storage, and provision of educational feedback.

[1140] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1141] The system according to the present invention monitors communication data within a company in real time and analyzes compliance violations and user sentiment. An embodiment of this system will be described below.

[1142] System Configuration

[1143] The system mainly consists of the following elements:

[1144] Server: Responsible for data collection, analysis, violation detection, emotion recognition, alert issuance, and data storage.

[1145] Terminal: A device used by users and administrators, such as a PC or smartphone.

[1146] User: An employee of a company who sends and receives messages within the system.

[1147] Administrator: The person responsible for customizing the system's rule set and emotion engine.

[1148] Program processing

[1149] 1. Data collection

[1150] The server receives data in real time from chat applications, mail servers, and voice communication systems used within the company. Depending on the format (text or voice), the data is added to a text analysis queue or a voice analysis queue.

[1151] 2. Data Analysis

[1152] The server passes the text data to a natural language processing (NLP) module, which tokenizes the received text data, tags it with parts of speech, and performs sentence structure analysis to understand the meaning and intent of the message.

[1153] The speech data is converted to text data by a speech recognition module and undergoes the same analysis as other text data: tokenization, part-of-speech tagging, and sentence structure analysis.

[1154] 3. Emotion Analysis

[1155] The server passes the parsed text data to an emotion engine to identify the user's emotion. This emotion engine contains algorithms that recognize the emotional tone of the text (e.g., anger, sadness, joy, etc.). For example, the message "I'm so angry" is identified as an emotion of anger.

[1156] 4. Detecting non-compliance

[1157] The server then checks the parsed data against a set of compliance detection rules, which contain keywords and phrases that identify offensive or legally questionable content, such as "useless" or "incompetent."

[1158] 5. Issuing an Alert

[1159] When a violation is detected, the server generates an alert, taking into account the emotional data. The generated alert includes information such as the content of the violating message, the emotional tone, the sender, and the time of sending. The alert is sent to the terminals of the sender of the violating message and the administrator.

[1160] 6. Data Recording

[1161] The server stores the detected violation data, analysis results, and emotional data in a database. The stored data includes the original message, analysis results, detailed violation information, and emotional tone, making it available for future analysis and audits.

[1162] 7. Providing educational feedback

[1163] On the device, users receive periodic reports and real-time notifications that allow them to learn how their behavior is non-compliant, for example, to see why their message was deemed "offensive."

[1164] 8. Customizing the ruleset

[1165] Administrators can customize the rulesets of the AI ​​modules and sentiment engines from their own devices to address the specific needs and circumstances of their companies, for example by adding new keywords that they consider "offensive."

[1166] Specific examples

[1167] Employee A sends a message in chat saying, "Your work is always slow. It really pisses me off."

[1168] The server receives the message and analyzes it with the NLP module. Based on the analysis results, the message is determined to be negative and aggressive.

[1169] The emotion engine identifies "I'm angry" as anger.

[1170] A violation is detected based on the rule set, and the server immediately issues an alert, notifying employee A and the administrator.

[1171] The message content, reason for violation, and emotional tone are stored in a database.

[1172] At a later date, Employee A will review his / her violation messages, the reasons for them, and the emotional tone of the messages through regular reports, and take appropriate action.

[1173] This specific process allows the system of the present invention to strengthen compliance and sentiment management within a company and manage risk in real time.

[1174] The processing flow will be explained below.

[1175] Step 1:

[1176] The server receives data in real time from the company's chat applications, mail servers, and voice communication systems, and depending on the format (text or voice), adds the data to a text analysis queue or a voice analysis queue.

[1177] Step 2:

[1178] The server passes the text data to a natural language processing (NLP) module, which tokenizes the received text data, tags it with parts of speech, and performs sentence structure analysis. This allows the meaning and intent of the message to be understood. For example, a message such as "You're always slow at work" is broken down into a series of words, and the role of each word is analyzed.

[1179] Step 3:

[1180] The server passes the voice data to a speech recognition module, which converts it into text data. The converted text data is then analyzed by an NLP module. For example, a voice message such as "You're always slow at work" is converted into written text, which is then analyzed.

[1181] Step 4:

[1182] The server passes the parsed text data to an emotion engine to identify the user's emotion. The emotion engine identifies the emotional tone of the text (e.g., anger, sadness, joy, etc.). For example, if the sentence "I'm so angry" is sent, it will be recognized as "anger."

[1183] Step 5:

[1184] The server compares the analyzed data and sentiment data with a set of rules to detect compliance violations. The rules include keywords and phrases that identify offensive or legally problematic content. For example, the server checks whether the message contains phrases like "useless" or "incompetent." A message like "You're always slow at work, it pisses me off" would be considered a violation.

[1185] Step 6:

[1186] When a violation is detected, the server generates an alert, taking into account emotional data. The alert includes the content of the violating message, its emotional tone (e.g., "anger"), the sender, and the time of sending. The alert is sent to the terminals of the sender of the violating message and the administrator.

[1187] Step 7:

[1188] The server stores the detected violation data, its analysis results, and emotional data in a database. The stored data includes the original message, analysis results, detailed information about the violation, and emotional tone. For example, "Your work is always slow. It makes me so angry" is stored, and the analysis results and the emotional tone of "anger" are recorded.

[1189] Step 8:

[1190] On the device, users receive periodic reports and real-time notifications, which help them learn how their actions are violating compliance. For example, users can see their violation messages, the reasons for their violations, and their emotional tone.

[1191] Step 9:

[1192] Administrators can customize the rulesets for the AI ​​module and emotion engine from their devices, for example adding new keywords that are considered "aggressive" or emotional tones (such as "sadness" or "joy").

[1193] This specific processing step allows enterprises to accurately monitor and quickly respond to compliance violations and users' emotional states in real time.

[1194] Example 2

[1195] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1196] Communications within companies are increasing every day, and some of them may contain offensive language or inappropriate content, which increases the company's compliance risks. It is also difficult to grasp changes in employee emotions in real time and take appropriate action. To solve these issues, a system is needed that monitors communication data in real time, analyzes emotions, and detects compliance violations.

[1197] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1198] In this invention, the server includes means for collecting communication data within a company in real time, means for analyzing the data using a natural language processing module and a voice recognition module, means for identifying user emotions from the analyzed data using an emotion engine, means for comparing the analyzed data with a compliance violation detection rule set, means for issuing an alert including the content, emotional tone, sender, and sending time of the violation message when a compliance violation is detected, means for storing the analysis results and the violation details in a database, means for periodically providing educational feedback to users, and means for an administrator to customize the rule set. This makes it possible to monitor communications within a company in real time, detect inappropriate expressions and changes in employee emotions early, and take appropriate measures.

[1199] "Internal communications data" refers to all forms of data that take place within a company, including emails, chat messages, voice calls, video conferences, etc.

[1200] "Real-time" refers to data being processed the instant it is generated or with very little delay.

[1201] A "natural language processing module" refers to a software module that allows a computer to analyze and understand the language that humans use on a daily basis.

[1202] "Speech recognition module" refers to a software module for converting voice data into text data.

[1203] An "emotion engine" refers to a system that includes algorithms for analyzing text or speech and identifying the emotional tones (e.g., joy, anger, sadness, etc.) contained within it.

[1204] "Rule Set" refers to the list of keywords or phrases and associated rules used to determine non-compliance.

[1205] An "alert" refers to a message or warning signal that notifies a user or administrator when the system detects an abnormality or problem.

[1206] "Database" refers to a system for storing information in an organized manner and for efficiently searching and managing it.

[1207] "Educational feedback" refers to a mechanism that provides users with an evaluation of their actions and suggestions for improvement.

[1208] An "administrator" is a person who has the authority to configure and operate the system and is responsible for customizing the rule set and monitoring the system.

[1209] The system based on this invention monitors communication data within a company in real time and analyzes compliance violations and user sentiment. This system is mainly composed of three elements: a server, a terminal, and a user.

[1210] System Configuration

[1211] 1. Server:

[1212] This is the core of the system, responsible for data collection, analysis, violation detection, emotion recognition, alert generation, and data storage. The server contains the following modules:

[1213] Natural Language Processing (NLP) module: Analyzes communication data using software tools such as SpaCy and NLTK.

[1214] Speech recognition module: Converts voice data into text using a module such as Google Speech-to-Text.

[1215] Emotion engine: Identifies the emotional tone of a user's text data using tools such as IBM Watson Tone Analyzer.

[1216] Database: A storage system for storing analysis results, violation details, and emotional tones.

[1217] 2. Terminal:

[1218] These are devices used by users and administrators, such as PCs and smartphones, to receive necessary reports and notifications.

[1219] 3. User:

[1220] These are company employees who send and receive messages within this system, and communicate via chat, email, etc. in the course of their daily work.

[1221] Specific examples

[1222] Example of employee A sending a message in chat saying, "Your work is always slow. It really pisses me off."

[1223] The server receives this message and adds it to the text analysis queue.

[1224] The NLP module tokenizes the message and performs part-of-speech tagging and sentence structure analysis.

[1225] The emotion engine identifies "annoyed" as the emotion "anger."

[1226] A non-compliance rule set identifies "Your work is always slow" as an offensive expression.

[1227] The server generates an alert containing details of the violation and emotional tone, and notifies Employee A and the manager.

[1228] This message, along with the analysis results, violation details, and emotional tone, is saved in a database.

[1229] At a later date, Employee A can review the periodic report on his / her own device and learn the reason for the violation and the emotional tone.

[1230] Prompt Sentence Examples

[1231] "Develop a system that performs employee sentiment analysis. The system collects chat data in real time, analyzes the text using natural language processing, and identifies emotional tones. It then detects compliance violations and sends alerts to managers if violations are detected. It also stores the emotional tones and violation data in a database so employees can receive feedback."

[1232] This system allows companies to monitor communication content in real time, detect and respond to inappropriate language and changes in employee sentiment early on, and provides employees with the opportunity to review and improve their own behavior through feedback.

[1233] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1234] Step 1: Collect data

[1235] The server receives data in real time from chat applications, mail servers, and voice communication systems used within the company, such as Slack, Microsoft Teams, email, and Zoom.

[1236] Input: Chat messages, emails, voice data

[1237] Output: Add data to the text analysis queue and speech analysis queue

[1238] How it works: The server sorts data received from each system into the appropriate analysis queue depending on its format (text or audio).

[1239] Step 2: Analyzing the text data

[1240] The server sends the data in the text analysis queue to a natural language processing (NLP) module, which performs tokenization, part-of-speech tagging, and sentence structure analysis of the text data.

[1241] Input: Text data

[1242] Output: Tokenized data, tagged data, sentence structure analysis results

[1243] How it works: It splits incoming text data into words and phrases, tags them with part-of-speech tags, and analyzes the structure of entire sentences.

[1244] Step 3: Analyzing the audio data

[1245] The server sends the speech data added to the speech analysis queue to a speech recognition module, which converts the speech data into text data that is then analyzed in a subsequent step.

[1246] Input: Audio data

[1247] Output: Text data

[1248] How it works: Converts speech to text and feeds it into an NLP module for further analysis.

[1249] Step 4: Sentiment Analysis

[1250] The server passes the text data analyzed by the NLP module to an emotion engine to identify the user's emotion. The emotion engine detects the emotional tone of the text (e.g., anger, sadness, joy, etc.).

[1251] Input: Parsed text data

[1252] Output: Emotional tone identification result

[1253] How it works: The emotion engine analyzes text and recognizes emotional tone, for example identifying the message "I'm so angry" as "anger."

[1254] Step 5: Detecting non-compliance

[1255] The server then compares the parsed text data against a set of compliance detection rules, which contain keywords and phrases to identify offensive or legally problematic content.

[1256] Input: Parsed text data, emotional tone

[1257] Output: Whether or not there is a violation, detailed violation information

[1258] How it works: The server checks the text data against a set of rules to see if there are any violations, such as whether it contains keywords like "useless" or "incompetent."

[1259] Step 6: Issue an alert

[1260] When a violation is detected, the server generates an alert, taking into account the emotional data. The generated alert includes information such as the content, emotional tone, sender, and time of the violating message.

[1261] Input: Violation details, emotional tone

[1262] Output: Alert message

[1263] What it does: When a violation is detected, an alert is generated to notify the sender of the offending message and an administrator. The alert includes details such as the content and emotional tone of the offending message.

[1264] Step 7: Record the data

[1265] The server stores the data on which violations were detected, along with the analysis results and emotion data, in a database for future analysis and auditing.

[1266] Input: Violation details, emotional tone, message content

[1267] Output: Saved entries in the database

[1268] How it works: Analysis results and violation information are recorded in a database and stored for future reference and verification.

[1269] Step 8: Providing educational feedback

[1270] The device displays regular reports and real-time notifications to users, allowing them to learn and improve how their behavior is out of compliance.

[1271] Input: Analysis results, violation information, emotional tone

[1272] Output: Educational feedback report, real-time notifications

[1273] Action: Generate and provide feedback to the user based on their behavior and violations, for example informing them of the specific reasons why something was deemed "offensive."

[1274] Step 9: Customizing the ruleset

[1275] Administrators can use their devices to customize the rulesets of the AI ​​modules and emotion engines to meet the specific needs and circumstances of their companies.

[1276] Input: New rule or keyword

[1277] Output: Updated ruleset

[1278] How it works: Administrators adapt the ruleset to suit their company's needs and apply it to the system, which may include adding new keywords that are considered "offensive."

[1279] (Application example 2)

[1280] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1281] In recent years, inappropriate behavior and emotional changes between employees have often been the cause of trouble in corporate communications. In particular, in environments such as factories, where cooperation between employees and robots is necessary, communication is a crucial element. However, there is a lack of means to monitor this communication in real time and prevent problems before they occur. Therefore, there is a growing need for a system that can analyze all communication data, including emotions, in real time, identify compliance violations and negative emotional states, and immediately issue alerts.

[1282] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1283] In this invention, the server includes a means for collecting communication data within the company in real time, a means for analyzing the data using a natural language analysis module and a voice recognition module, and a means for comparing the analyzed data with a compliance violation detection rule set. This allows for real-time monitoring of communications between employees and robots and issuing warnings at appropriate times. The system also analyzes the user's emotional state and provides feedback based on that analysis, creating a healthier communication environment.

[1284] Definitions of important words

[1285] "Communication data" refers to data including text data and voice data generated within a company, as well as the results of sentiment analysis.

[1286] "Real-time" refers to immediate processing or analysis of data the moment it is generated or transmitted.

[1287] "Natural language analysis" refers to the process of mechanically interpreting text or audio data through tokenization, part-of-speech tagging, sentence structure analysis, etc.

[1288] "Speech recognition module" refers to a computer program or system for converting voice data into text data.

[1289] A "compliance violation detection rule set" refers to a collection of rules for detecting expressions that may violate legal or corporate standards based on specific keywords or phrases.

[1290] "Warning" refers to a notification or alert issued when non-compliance or a negative emotional state is detected.

[1291] "Database" refers to a system or structured collection of information for storing collected and analyzed data.

[1292] "Educational Feedback" refers to reports or guidance provided to users to help them learn about their communication and non-compliance behaviors.

[1293] "Customizing rule sets" refers to the act of an administrator tailoring compliance violation detection rules and sentiment analysis algorithms to the specific needs and circumstances of the company.

[1294] "User emotional state" refers to the user's emotional tone (e.g., anger, sadness, joy, etc.) derived from analyzed text data and voice data.

[1295] "Form for carrying out the invention" of the specification

[1296] System Configuration

[1297] The system according to the present invention comprises the following components:

[1298] Server: Responsible for data collection, analysis, violation detection, emotion recognition, warning issuance, and data storage. Specifically, it is equipped with a natural language analysis module, a voice recognition module, a compliance violation detection rule set, and an emotion analysis engine.

[1299] Terminal: A device used by employees and managers, such as a PC, a smartphone, or an application installed on a robot in a factory.

[1300] User: An employee of a company who sends and receives messages within the system.

[1301] Administrator: The person responsible for customizing the system's rule sets and sentiment analysis engine.

[1302] Explanation of program processing

[1303] The server uses a communications API to collect internal company communication data (text and voice data) in real time. This data is converted into text data through a natural language analysis module and a speech recognition module, and then analyzed based on the extracted tokens. In particular, the NLP (natural language processing) module performs tokenization, part-of-speech tagging, and sentence structure analysis to understand the meaning and intent of the message.

[1304] Meanwhile, the voice data is converted into text data by a speech recognition module. The analyzed text data is then passed to a sentiment analysis engine, which identifies emotional tones such as anger, sadness, joy, etc. If a user's message is "I'm so angry," the emotional tone will be recognized as "anger."

[1305] The analyzed data is compared against a set of compliance violation detection rules, which include offensive or legally questionable language. If such language is detected, the server immediately issues an alert. The alert includes information about the content, emotional tone, sender, and time of the offending message, and this information is sent to the user and administrator who detected the violation.

[1306] The data is stored in a database for permanent archival, including the original message, analysis results, violation details, and emotional tone, making it available for future analysis and audits.

[1307] At the device, users receive regular educational feedback on their communication behavior, including how they were non-compliant and their emotional state at the time. Administrators can customize the rule set and optimize the system by adding new keywords and phrases as needed.

[1308] Specific examples

[1309] For example, suppose employee A sends a message saying, "You're always slow at work. It pisses me off." In this case, the server would process it as follows:

[1310] 1. Messages are received in real time and analyzed by the NLP module.

[1311] 2. The sentiment analysis engine identifies the phrase "I'm angry" as anger.

[1312] 3. Based on the rule set, this message is detected as non-compliant.

[1313] 4. The server immediately issues an alert and notifies Employee A and the administrator.

[1314] Prompt Sentence Examples

[1315] Monitor chat logs between Employee A and Employee B in real time to detect compliance violations and sentiment changes. Start your analysis with the following chat logs:

[1316] "Employee A: You're always slow at work. It really pisses me off."

[1317] The system monitors communication between employees and robots in real time and provides immediate and appropriate feedback to support safe and efficient operations within the factory.

[1318] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1319] Program processing flow

[1320] Step 1:

[1321] The server collects communication data (text data and voice data) within the company in real time. Specifically, it uses a communication API to obtain messages and call content and adds them to a data queue. The input data is raw text or voice data, and the output is data added to the analysis queue.

[1322] Step 2:

[1323] The server passes the collected text data to a natural language analysis module, which tokenizes the data, tags it with parts of speech, and analyzes sentence structure. The input data is text data, and the output is tokenized analyzed data. The analyzed data is used to understand the meaning and intent of the message.

[1324] Step 3:

[1325] The server passes the collected voice data to a voice recognition module and converts it into text data. The voice recognition module analyzes the voice data and converts it into text. The input data is voice data and the output is text data. The converted text data is further analyzed by a natural language analysis module.

[1326] Step 4:

[1327] The server passes the parsed text data to the sentiment analysis engine, which analyzes the emotional tone of the text to identify the user's emotional state. The input data is the parsed text data, and the output is the identified emotional tone. For example, the message "I'm so angry" is recognized as "Anger."

[1328] Step 5:

[1329] The server compares the analyzed data against a set of compliance violation detection rules. The rule set contains keywords and phrases that determine whether a violation has occurred. The input data is the analyzed and sentiment analysis data, and the output is a determination of whether or not there is a compliance violation. If a violation is detected, information about it is generated.

[1330] Step 6:

[1331] The server immediately issues a warning if a violation is detected. The warning includes the content, emotional tone, sender, and time of sending of the violating message. The input data is the compliance violation judgment result, and the output is the generation and notification of a warning message. The notification is sent to the user who detected the violation and the administrator.

[1332] Step 7:

[1333] The server stores the analysis results and violation details in a database. The stored data includes the original message, analysis results, violation details, and emotional tone. The input data is the analyzed and violation detection data, and the output is storage in the database.

[1334] Step 8:

[1335] The device periodically provides educational feedback to the user, including how the user violated compliance and their emotional state at the time. The input data are the analysis results stored in the database, and the output is a report of the educational feedback.

[1336] Step 9:

[1337] Administrators use a terminal to customize the rule set. They can add new keywords and phrases. The input is the administrator's configuration information, and the output is an updated rule set. This allows the system to adapt to the specific needs and circumstances of the company.

[1338] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1339] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1340] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1341] [Fourth embodiment]

[1342] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1343] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1344] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1345] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1346] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1347] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1348] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1349] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1350] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1351] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1352] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1353] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1354] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1355] The system according to the present invention monitors communication data within a company in real time to detect and prevent compliance violations. An embodiment of this system will be described below.

[1356] System Configuration

[1357] The system mainly consists of the following elements:

[1358] Server: Responsible for data collection, analysis, violation detection, alert generation, and data storage.

[1359] Terminal: A device used by users and administrators, such as a PC or smartphone.

[1360] User: An employee of a company who sends and receives messages within the system.

[1361] Administrator: A person responsible for customizing the system's rule set.

[1362] Program processing

[1363] 1. Data collection

[1364] The server receives data in real time from chat applications, mail servers, and voice communication systems used within the company, and depending on the format (text, voice), sends the data to the corresponding analysis queue.

[1365] 2. Data Analysis

[1366] The natural language processing module running on the server analyzes the collected text data. This module tokenizes sentences, tags parts of speech, and analyzes sentence structure to understand the meaning of the message. For example, a message such as "Your work is always slow" is identified as a negative expression.

[1367] The speech data is converted into text by a speech recognition module and subjected to similar analysis. For example, a user's utterance "Your work is always slow" is converted into text and then analyzed by a natural language processing module.

[1368] 3. Detecting non-compliance

[1369] The server then compares the parsed data against a predefined set of rules to detect compliance violations. These rules define offensive or legally problematic content. For example, phrases containing keywords such as "useless" or "incompetent" are identified as violations.

[1370] 4. Issue an alert

[1371] If a violation is detected, the server immediately issues an alert, which is sent to the message sender's device and the administrator's device. The alert includes the violating message, sender, and time of sending.

[1372] 5. Data Recording

[1373] The server records the detected violations and the analysis results in a database. This data is used for future analysis and audits. For example, a message such as "Your work is always slow" is recorded as a violation.

[1374] 6. Providing educational feedback

[1375] On the device, users receive periodic reports and real-time notifications, which help them learn which expressions are non-compliant. For example, users can view their non-compliance messages and the reasons for their actions.

[1376] 7. Customizing the ruleset

[1377] Administrators can customize the rule set of the AI ​​module from their own devices, which is necessary to adapt to the specific circumstances of the company, for example by adding new keywords based on new internal rules.

[1378] Specific examples

[1379] Employee A sends a message in chat saying, "Your work is always slow."

[1380] The server receives the message and analyzes it using a natural language processing module. Based on the analysis results, the message is determined to be negative and offensive.

[1381] A violation is detected based on the rule set, and the server immediately issues an alert, notifying employee A and the administrator.

[1382] The message content and reason for the violation are stored in a database.

[1383] At a later date, Employee A will check his / her violation message through a regular report and be warned.

[1384] In this way, the system of the present invention can enforce compliance within a company and manage risk in real time.

[1385] The processing flow will be explained below.

[1386] Step 1:

[1387] The server receives data in real time from the company's chat applications, mail servers, and voice communication systems, and depending on the format (text or voice), adds the data to a text analysis queue or a voice analysis queue.

[1388] Step 2:

[1389] The server passes the text data to a natural language processing (NLP) module, which tokenizes the received text data, tags it with parts of speech, and performs sentence structure analysis to understand the meaning and intent of the message.

[1390] Step 3:

[1391] The server passes the speech data to a speech recognition module, which converts it into text data. This text data is then analyzed by an NLP module, which performs tokenization, part-of-speech tagging, and sentence structure analysis, just like any other text data.

[1392] Step 4:

[1393] The server then compares the parsed text data against a set of compliance detection rules, which contain keywords and phrases that identify offensive or legally questionable content, such as "useless" or "incompetent."

[1394] Step 5:

[1395] The server generates an alert if a violation is detected based on the matching results. The generated alert includes information such as the content of the violating message, the sender, and the time of sending. The alert is sent to the terminals of the sender of the violating message and the administrator.

[1396] Step 6:

[1397] The server stores the detected violation data and its analysis results in a database, including the original message, analysis results, and detailed information about the violation, making it available for future analysis and audits.

[1398] Step 7:

[1399] On the device, users receive periodic reports and real-time notifications that allow them to learn how their behavior is non-compliant, for example, to see why their message was deemed "offensive."

[1400] Step 8:

[1401] From their own devices, administrators can customize the rule sets of the AI ​​module, changing or adding violation detection criteria and rules according to the specific needs and circumstances of the company. For example, this could include adding new keywords that are considered "offensive."

[1402] This specific processing step allows businesses to effectively detect compliance violations in real time and respond immediately.

[1403] Example 1

[1404] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1405] The volume of communication data within a company is enormous, making it difficult to detect compliance violations from that data. It is also difficult to detect in real time inappropriate or offensive language used unconsciously by employees in communications and take appropriate action. A lack of employee training feedback can lead to similar problems recurring. Furthermore, properly managing different rule sets for each company is also a challenge.

[1406] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1407] In this invention, the server includes means for collecting communication data within a company in real time, means for analyzing the data using a natural language processing module and a voice recognition module, means for comparing the analyzed data with a compliance violation detection rule set, means for saving the analysis results and violation details in a storage device, means for visualizing violation detection notifications and analysis results in real time, means for periodically providing educational feedback to users, and means for an administrator to change the rule set. This makes it possible to effectively monitor a wide range of communication data within a company, detect inappropriate expressions and offensive language in real time, and take prompt action. In addition, providing appropriate feedback and education to employees can help improve compliance awareness.

[1408] "Intra-company communications data" refers to data that includes all communications that take place within a company, including text messages, voice messages, emails, etc.

[1409] A "natural language processing module" is a computer program that analyzes collected text data and performs tokenization, part-of-speech tagging, sentence structure analysis, etc.

[1410] A "voice recognition module" is a computer program for converting voice data into text data.

[1411] A "compliance violation detection rule set" is a collection of pre-defined rules for identifying offensive language or legally questionable content.

[1412] "Notification" refers to alerts or messages sent to relevant parties when a compliance violation is detected.

[1413] "Storage device" refers to a database or storage medium for storing analysis results and violation details.

[1414] "Educational feedback" refers to regular reports and notifications to inform users of violations and the reasons for them, and to encourage improvement.

[1415] "Administrator" refers to the person responsible for customizing and managing the rule sets within the system.

[1416] "Means of collecting data in real time" refers to systems and programs that instantly acquire communication data generated within a company.

[1417] "Visualization means" refers to a system or program for graphically displaying analysis results and detected violations.

[1418] The system according to the present invention monitors communication data within a company in real time to detect and prevent compliance violations. An embodiment of the system according to the present invention will be described in detail below.

[1419] System Configuration

[1420] This system consists of a server, a terminal, and a user. The detailed roles of each element are shown below.

[1421] server

[1422] The server is primarily responsible for the following functions:

[1423] It collects data in real time from chat applications, mail servers, and voice communication systems used within a company.

[1424] Examples of specific hardware or software used: Slack API, Microsoft Exchange Server, Google Speech-to-Text API.

[1425] Example: Employee A sends a message on Slack saying, "Your work is always slow."

[1426] The collected data is analyzed using a natural language processing module, which includes tokenization, part-of-speech tagging, and sentence structure analysis.

[1427] Examples of software used: SpaCy, Google Cloud Natural Language API.

[1428] Example: The message "You're always slow at work" is identified as a negative expression.

[1429] The voice data is converted into text using a voice recognition module and analyzed in the same way.

[1430] Example: Voice data is converted into text by a speech recognition module, such as "Your work is always slow."

[1431] The parsed data is then compared against a set of rules to detect non-compliance, including those that contain offensive language or legally questionable content.

[1432] Examples of software used: Custom rule engine, regular expressions (RegEx).

[1433] Example: The expression "you" is identified as a violation.

[1434] If a violation is detected, a notification is issued, which is sent in real time to the message sender's device and to the administrator's device.

[1435] Examples of software used: Twilio, SMTP (sending email).

[1436] Example: Employee A and Manager B receive a notification.

[1437] The details of the detected violation and the analysis results are saved in a recording device.

[1438] Examples of software used: MySQL, PostgreSQL.

[1439] Example: The message "Your work is always slow" and the analysis results are saved.

[1440] Terminal

[1441] Terminals are devices used by users and administrators that provide the following functionality:

[1442] Receive real-time notifications.

[1443] Learn which expressions are non-compliant through regular reports.

[1444] Examples of software used: Tableau, Power BI.

[1445] Example: Employee A checks his / her violation messages and the reasons for them in a regular report.

[1446] Administrators customize the system's compliance violation detection rule set.

[1447] Examples of software used: Web browser, custom management tools.

[1448] Example: Administrator B adds a new keyword "incompetent" to the ruleset.

[1449] User

[1450] The user is an employee of the company and performs the following functions:

[1451] Send and receive messages within the system.

[1452] Example: Employee A sends a message on Slack.

[1453] Receive regular reports and real-time notifications to increase compliance awareness.

[1454] Example: Employee A checks the reason for his / her violation message in the regular report and makes improvements.

[1455] Examples and prompts

[1456] Examples:

[1457] Employee A sends a message on Slack saying, "Your work is always slow."

[1458] The server receives the message and analyzes it using a natural language processing module.

[1459] The message is deemed offensive and checked against a set of rules.

[1460] A violation is detected and the server issues a notification to Employee A and Manager B using Twilio.

[1461] The message and reason for the violation are stored in a MySQL database.

[1462] At a later date, Employee A checks his / her violation message in the Tableau report.

[1463] Administrator B uses a web browser to add a new keyword, "incompetent," to the rule set.

[1464] Example prompt sentence:

[1465] "Please explain the specific process by which an AI system monitors communication data within a company in real time and detects negative expressions. For example, please explain in detail how a message such as 'Your work is always slow' is processed."

[1466] In this way, the system of the present invention can enforce compliance within a company and manage risk in real time.

[1467] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1468] Step 1:

[1469] Data collection

[1470] The server collects data in real time from chat applications, mail servers, and voice communication systems used within the company.

[1471] Input: Message data from Slack, email data from the mail server, and voice data.

[1472] Output: Send the collected data to the text data analysis queue or the audio data analysis queue.

[1473] Specific operation: For example, if employee A sends a message on Slack saying, "Your work is always slow," the message is sent to the server in real time.

[1474] Step 2:

[1475] Data conversion (for audio data)

[1476] The server converts the collected voice data into text using a voice recognition module.

[1477] Input: Collected audio data.

[1478] Output: Audio data converted to text data.

[1479] Specific operation: For example, audio data such as "Your work is always slow" is converted into text using the Google Speech-to-Text API.

[1480] Step 3:

[1481] Data analysis

[1482] The server analyzes the text data using a natural language processing module, which performs tokenization, part-of-speech tagging, and sentence structure analysis.

[1483] Input: Text data (either collected directly or converted by a speech recognition module).

[1484] Output: The parsed data includes word tokens, parts of speech, and sentence structure information.

[1485] Specific operation: For example, the message "Your work is always slow" is parsed through SpaCy and tokenized into "you", "of", "work", "is", "always", and "slow".

[1486] Step 4:

[1487] Compliance violation detection

[1488] The server compares the parsed data against a set of compliance violation detection rules to determine whether a violation has occurred.

[1489] Input: Parsed text data.

[1490] Output: Violation detection results (details of the violation if there is one, no abnormality if there is no violation).

[1491] Specific behavior: For example, the analyzed message contains the offensive keyword "you," so it is detected as a violation.

[1492] Step 5:

[1493] Issuance of notifications

[1494] The server will issue a notification if a violation is detected, which will be sent to the message sender and an administrator.

[1495] Input: Violation detection result and violation content.

[1496] Output: Notification message (including details of non-compliance and time of sending).

[1497] Specific operation: For example, an alert will be sent via Twilio to employee A who sent a message saying, "Your work is always slow," and to his manager B.

[1498] Step 6:

[1499] Data recording

[1500] The server stores the details of the detected violation and the analysis results in a recording device.

[1501] Input: Violation detection results and their analysis.

[1502] Output: Violation records stored in a database.

[1503] Specific behavior: For example, the message "Your work is always slow" and the reason for the violation are recorded in a MySQL database.

[1504] Step 7:

[1505] Providing educational feedback

[1506] The device provides regular feedback to the user, helping them learn what expressions are considered non-compliant.

[1507] Input: Violation records stored in the database.

[1508] Output: Education report provided to the user.

[1509] Specific behavior: For example, Employee A checks his / her violation messages and the reasons for them through regular Tableau reports.

[1510] Step 8:

[1511] Customizing the ruleset

[1512] Administrators can customize rule sets from their own devices and adjust them to suit their company's specific circumstances.

[1513] Input: Ruleset change input by administrator.

[1514] Output: The updated ruleset.

[1515] Specific behavior: For example, Administrator B uses a web browser to add a new keyword, "incompetent," to the rule set.

[1516] (Application example 1)

[1517] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1518] As internal communication increases in modern companies, compliance is becoming more important than ever. However, many companies lack a system for monitoring internal communications in real time and immediately detecting violations, exposing them to the risk of serious compliance violations being overlooked. Even when violations are discovered, response can be delayed, potentially affecting the company's credibility and legal risk. Furthermore, a lack of educational feedback to help employees understand and improve their past violations can lead to repeated violations. To solve these problems, a system is needed that effectively monitors internal communications and immediately detects and notifies violations.

[1519] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1520] In this invention, the server includes means for collecting communication data within a company in real time, means for analyzing the data using a natural language processing module and a voice recognition module, means for comparing the analyzed data with a compliance violation detection rule set, means for issuing an alert when a compliance violation is detected, means for storing the analysis results and the details of the violation in a database, means for periodically providing educational feedback to users, means for an administrator to customize the rule set, means for delivering related notifications via an application installed on a smart device, means for notifying users and administrators of violation detection notifications and details, and means for generating periodic reports based on the violation detection data. This enables real-time monitoring of intra-company communications, immediate detection of compliance violations, prompt notification to target groups, and employee educational feedback.

[1521] "Internal Enterprise Communication Data" refers to the electronic data of all messages, communications, and discussions exchanged by employees within an enterprise.

[1522] "Means of collecting in real time" refers to a system that instantly collects communication data within a company and provides it for processing without delay.

[1523] "Natural language processing module" refers to software or algorithms for analyzing text data and understanding its meaning.

[1524] "Speech Recognition Module" refers to software or algorithms for converting voice data into text.

[1525] A "compliance violation detection rule set" is a set of rules for defining violation actions based on company rules and legal regulations.

[1526] "Means for issuing alerts" refers to a mechanism for immediately notifying relevant parties when a compliance violation is detected.

[1527] "Means for storing in a database" refers to a mechanism for permanently recording the analysis results and violation details and storing them for future reference and analysis.

[1528] "Means for providing educational feedback to users" refers to a mechanism for providing employees with information about compliance violations and advice for improvement.

[1529] "Means for administrators to customize rule sets" refers to a mechanism by which administrators can modify, add, or remove compliance violation detection rule sets according to the specific needs of their enterprise.

[1530] "Smart devices" refers to portable electronic devices with internet connectivity, including smartphones and tablets.

[1531] "Application" means a software program designed to perform a specific job or task.

[1532] "Means for delivering notifications" refers to a mechanism for sending real-time notifications to users and administrators via smart devices, etc.

[1533] "Means for notifying users and administrators of detected violations and providing details of those violations" refers to a mechanism for sending notifications containing detailed information to users and administrators when a violation is detected.

[1534] "Means for generating periodic reports" refers to a system for aggregating violation data over a fixed period and creating an analysis report.

[1535] In order to put the present invention into practice, a system is constructed that collects communication data within a company in real time and analyzes that data. The configuration and processing procedure of this system are described below.

[1536] System Configuration

[1537] The system consists of the following main components:

[1538] 1. Server

[1539] Data collection module: Works with internal company communication tools (chat apps and email servers) to collect data in real time.

[1540] Natural Language Processing (NLP) module: Analyzes collected text data and understands the meaning of messages. Available software includes SpaCy and Google Cloud Natural Language API.

[1541] Speech recognition module: Converts voice data into text. Uses Google Cloud Speech-to-Text API, etc.

[1542] Violation Detection Module: Compares the parsed data against a set of compliance violation detection rules.

[1543] Alerting module: Issues alerts and notifies users and administrators when violations are detected. Notification services include Firebase Cloud Messaging (FCM).

[1544] Data storage module: Analysis results and violation details are saved in a database. SQLite and PostgreSQL are available.

[1545] Report Generation Module: Generates regular reports based on violation data and provides feedback.

[1546] 2. Terminal

[1547] Smart devices (smartphones and tablets): The devices through which users access and receive notifications.

[1548] Application: Software installed on smart devices that provides real-time notifications, checks violations, and provides feedback.

[1549] Processing flow

[1550] 1. Data collection

[1551] The server's data collection module works in conjunction with each communication tool within the company to collect text and voice data in real time.

[1552] 2. Data Analysis

[1553] Natural language processing and speech recognition modules analyze the collected data and understand its meaning. For example, a message like "You're always slow at work" is identified as a negative expression.

[1554] 3. Detecting non-compliance

[1555] The server compares the parsed data against a set of compliance violation detection rules to determine whether there are any violations.

[1556] 4. Issue an alert

[1557] If a violation is detected, the alert module will be activated and send a notification to the user and administrator, such as "A compliance violation has been detected. From: Employee X, Content: 'Your work is always slow', Time: YYYY-MM-DD HH:MM."

[1558] 5. Data storage and report generation

[1559] The data storage module stores the details of violations and their analysis results in a database. Based on the stored data, the report generation module creates regular reports and provides educational feedback to employees.

[1560] Specific examples

[1561] Example of input prompt: "During a new project discussion, a message was sent stating, 'Your work is always slow.' Please determine whether this message is in compliance and issue an alert if necessary."

[1562] Example notification: "A compliance violation has been detected. From: Employee X, Content: 'Your work is always slow', Time: YYYY-MM-DD HH:MM"

[1563] In this way, the present invention provides real-time monitoring of corporate communication data, immediate detection of compliance violations, and prompt notification and educational feedback to target audiences.

[1564] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1565] Step 1:

[1566] The server connects with the company's internal communication tools and collects data in real time. The main inputs are employee chat messages and voice conversations. This data is sent to the server as raw data and added to a queue for further processing. The output is stored in the queue as text data or voice data.

[1567] Step 2:

[1568] The server's natural language processing (NLP) module analyzes the text data stored in the queue. The main input is the text data stored in the queue. The NLP module performs sentence tokenization (breaking down each word), part-of-speech tagging (identifying the grammatical role of each word), and sentence structure analysis (understanding the meaning of the sentence). For example, the message "Your work is always slow" is recognized as an offensive expression. The output is the analyzed text data and its semantic analysis results.

[1569] Step 3:

[1570] The server's speech recognition module converts the voice data stored in the queue into text. The main input is voice data. The speech recognition module (for example, Google Cloud Speech-to-Text API) analyzes the voice data and obtains text data. This text data is passed to the NLP module and analyzed again. The output is the text data converted from the voice data.

[1571] Step 4:

[1572] The server compares the analyzed data with a compliance violation detection rule set. The main input is the analyzed text data. The rule set defines offensive expressions and legally problematic keywords. For example, it includes keywords such as "incompetent" and "useless." This comparison determines whether or not a violation has occurred. The output is a judgment result indicating whether or not a violation has been detected.

[1573] Step 5:

[1574] If a violation is detected, the server's alert module is activated. The main input is the violation detection result. The alert module sends a notification containing the details of the violation to the user's and administrator's devices. Specifically, the notification is sent using Firebase Cloud Messaging (FCM). The output is a notification message sent to the user and administrator. For example, the notification may be in the format "A compliance violation has been detected. Sender: Employee X, Content: 'Your work is always slow', Time: YYYY-MM-DD HH:MM."

[1575] Step 6:

[1576] The server saves the details of violations and the analysis results in a database. The main inputs are the details of violations and the analysis results. This data is stored in a database system such as SQLite or PostgreSQL. The saved data is used for later analysis and audits. The output is the data recorded in the database.

[1577] Step 7:

[1578] The server's report generation module creates periodic reports based on the stored violation data. The main input is the violation data recorded in the database. The report generation module aggregates the violation data and creates an analysis report, which allows companies to provide educational feedback to employees. The output is a periodic analysis report. Specifically, a report in the format of "Violation report for the past month" is created.

[1579] Through the above processing steps, the present invention realizes real-time monitoring within a company, immediate detection and response to compliance violations, data storage, and provision of educational feedback.

[1580] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1581] The system according to the present invention monitors communication data within a company in real time and analyzes compliance violations and user sentiment. An embodiment of this system will be described below.

[1582] System Configuration

[1583] The system mainly consists of the following elements:

[1584] Server: Responsible for data collection, analysis, violation detection, emotion recognition, alert issuance, and data storage.

[1585] Terminal: A device used by users and administrators, such as a PC or smartphone.

[1586] User: An employee of a company who sends and receives messages within the system.

[1587] Administrator: The person responsible for customizing the system's rule set and emotion engine.

[1588] Program processing

[1589] 1. Data collection

[1590] The server receives data in real time from chat applications, mail servers, and voice communication systems used within the company. Depending on the format (text or voice), the data is added to a text analysis queue or a voice analysis queue.

[1591] 2. Data Analysis

[1592] The server passes the text data to a natural language processing (NLP) module, which tokenizes the received text data, tags it with parts of speech, and performs sentence structure analysis to understand the meaning and intent of the message.

[1593] The speech data is converted to text data by a speech recognition module and undergoes the same analysis as other text data: tokenization, part-of-speech tagging, and sentence structure analysis.

[1594] 3. Emotion Analysis

[1595] The server passes the parsed text data to an emotion engine to identify the user's emotion. This emotion engine contains algorithms that recognize the emotional tone of the text (e.g., anger, sadness, joy, etc.). For example, the message "I'm so angry" is identified as an emotion of anger.

[1596] 4. Detecting non-compliance

[1597] The server then checks the parsed data against a set of compliance detection rules, which contain keywords and phrases that identify offensive or legally questionable content, such as "useless" or "incompetent."

[1598] 5. Issuing an Alert

[1599] When a violation is detected, the server generates an alert, taking into account the emotional data. The generated alert includes information such as the content of the violating message, the emotional tone, the sender, and the time of sending. The alert is sent to the terminals of the sender of the violating message and the administrator.

[1600] 6. Data Recording

[1601] The server stores the detected violation data, analysis results, and emotional data in a database. The stored data includes the original message, analysis results, detailed violation information, and emotional tone, making it available for future analysis and audits.

[1602] 7. Providing educational feedback

[1603] On the device, users receive periodic reports and real-time notifications that allow them to learn how their behavior is non-compliant, for example, to see why their message was deemed "offensive."

[1604] 8. Customizing the ruleset

[1605] Administrators can customize the rulesets of the AI ​​modules and sentiment engines from their own devices to address the specific needs and circumstances of their companies, for example by adding new keywords that they consider "offensive."

[1606] Specific examples

[1607] Employee A sends a message in chat saying, "Your work is always slow. It really pisses me off."

[1608] The server receives the message and analyzes it with the NLP module. Based on the analysis results, the message is determined to be negative and aggressive.

[1609] The emotion engine identifies "I'm angry" as anger.

[1610] A violation is detected based on the rule set, and the server immediately issues an alert, notifying employee A and the administrator.

[1611] The message content, reason for violation, and emotional tone are stored in a database.

[1612] At a later date, Employee A will review his / her violation messages, the reasons for them, and the emotional tone of the messages through regular reports, and take appropriate action.

[1613] This specific process allows the system of the present invention to strengthen compliance and sentiment management within a company and manage risk in real time.

[1614] The processing flow will be explained below.

[1615] Step 1:

[1616] The server receives data in real time from the company's chat applications, mail servers, and voice communication systems, and depending on the format (text or voice), adds the data to a text analysis queue or a voice analysis queue.

[1617] Step 2:

[1618] The server passes the text data to a natural language processing (NLP) module, which tokenizes the received text data, tags it with parts of speech, and performs sentence structure analysis. This allows the meaning and intent of the message to be understood. For example, a message such as "You're always slow at work" is broken down into a series of words, and the role of each word is analyzed.

[1619] Step 3:

[1620] The server passes the voice data to a speech recognition module, which converts it into text data. The converted text data is then analyzed by an NLP module. For example, a voice message such as "You're always slow at work" is converted into written text, which is then analyzed.

[1621] Step 4:

[1622] The server passes the parsed text data to an emotion engine to identify the user's emotion. The emotion engine identifies the emotional tone of the text (e.g., anger, sadness, joy, etc.). For example, if the sentence "I'm so angry" is sent, it will be recognized as "anger."

[1623] Step 5:

[1624] The server compares the analyzed data and sentiment data with a set of rules to detect compliance violations. The rules include keywords and phrases that identify offensive or legally problematic content. For example, the server checks whether the message contains phrases like "useless" or "incompetent." A message like "You're always slow at work, it pisses me off" would be considered a violation.

[1625] Step 6:

[1626] When a violation is detected, the server generates an alert, taking into account emotional data. The alert includes the content of the violating message, its emotional tone (e.g., "anger"), the sender, and the time of sending. The alert is sent to the terminals of the sender of the violating message and the administrator.

[1627] Step 7:

[1628] The server stores the detected violation data, its analysis results, and emotional data in a database. The stored data includes the original message, analysis results, detailed information about the violation, and emotional tone. For example, "Your work is always slow. It makes me so angry" is stored, and the analysis results and the emotional tone of "anger" are recorded.

[1629] Step 8:

[1630] On the device, users receive periodic reports and real-time notifications, which help them learn how their actions are violating compliance. For example, users can see their violation messages, the reasons for their violations, and their emotional tone.

[1631] Step 9:

[1632] Administrators can customize the rulesets for the AI ​​module and emotion engine from their devices, for example adding new keywords that are considered "aggressive" or emotional tones (such as "sadness" or "joy").

[1633] This specific processing step allows enterprises to accurately monitor and quickly respond to compliance violations and users' emotional states in real time.

[1634] Example 2

[1635] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1636] Communications within companies are increasing every day, and some of them may contain offensive language or inappropriate content, which increases the company's compliance risks. It is also difficult to grasp changes in employee emotions in real time and take appropriate action. To solve these issues, a system is needed that monitors communication data in real time, analyzes emotions, and detects compliance violations.

[1637] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1638] In this invention, the server includes means for collecting communication data within a company in real time, means for analyzing the data using a natural language processing module and a voice recognition module, means for identifying user emotions from the analyzed data using an emotion engine, means for comparing the analyzed data with a compliance violation detection rule set, means for issuing an alert including the content, emotional tone, sender, and sending time of the violation message when a compliance violation is detected, means for storing the analysis results and the violation details in a database, means for periodically providing educational feedback to users, and means for an administrator to customize the rule set. This makes it possible to monitor communications within a company in real time, detect inappropriate expressions and changes in employee emotions early, and take appropriate measures.

[1639] "Internal communications data" refers to all forms of data that take place within a company, including emails, chat messages, voice calls, video conferences, etc.

[1640] "Real-time" refers to data being processed the instant it is generated or with very little delay.

[1641] A "natural language processing module" refers to a software module that allows a computer to analyze and understand the language that humans use on a daily basis.

[1642] "Speech recognition module" refers to a software module for converting voice data into text data.

[1643] An "emotion engine" refers to a system that includes algorithms for analyzing text or speech and identifying the emotional tones (e.g., joy, anger, sadness, etc.) contained within it.

[1644] "Rule Set" refers to the list of keywords or phrases and associated rules used to determine non-compliance.

[1645] An "alert" refers to a message or warning signal that notifies a user or administrator when the system detects an abnormality or problem.

[1646] "Database" refers to a system for storing information in an organized manner and for efficiently searching and managing it.

[1647] "Educational feedback" refers to a mechanism that provides users with an evaluation of their actions and suggestions for improvement.

[1648] An "administrator" is a person who has the authority to configure and operate the system and is responsible for customizing the rule set and monitoring the system.

[1649] The system based on this invention monitors communication data within a company in real time and analyzes compliance violations and user sentiment. This system is mainly composed of three elements: a server, a terminal, and a user.

[1650] System Configuration

[1651] 1. Server:

[1652] This is the core of the system, responsible for data collection, analysis, violation detection, emotion recognition, alert generation, and data storage. The server contains the following modules:

[1653] Natural Language Processing (NLP) module: Analyzes communication data using software tools such as SpaCy and NLTK.

[1654] Speech recognition module: Converts voice data into text using a module such as Google Speech-to-Text.

[1655] Emotion engine: Identifies the emotional tone of a user's text data using tools such as IBM Watson Tone Analyzer.

[1656] Database: A storage system for storing analysis results, violation details, and emotional tones.

[1657] 2. Terminal:

[1658] These are devices used by users and administrators, such as PCs and smartphones, to receive necessary reports and notifications.

[1659] 3. User:

[1660] These are company employees who send and receive messages within this system, and communicate via chat, email, etc. in the course of their daily work.

[1661] Specific examples

[1662] Example of employee A sending a message in chat saying, "Your work is always slow. It really pisses me off."

[1663] The server receives this message and adds it to the text analysis queue.

[1664] The NLP module tokenizes the message and performs part-of-speech tagging and sentence structure analysis.

[1665] The emotion engine identifies "annoyed" as the emotion "anger."

[1666] A non-compliance rule set identifies "Your work is always slow" as an offensive expression.

[1667] The server generates an alert containing details of the violation and emotional tone, and notifies Employee A and the manager.

[1668] This message, along with the analysis results, violation details, and emotional tone, is saved in a database.

[1669] At a later date, Employee A can review the periodic report on his / her own device and learn the reason for the violation and the emotional tone.

[1670] Prompt Sentence Examples

[1671] "Develop a system that performs employee sentiment analysis. The system collects chat data in real time, analyzes the text using natural language processing, and identifies emotional tones. It then detects compliance violations and sends alerts to managers if violations are detected. It also stores the emotional tones and violation data in a database so employees can receive feedback."

[1672] This system allows companies to monitor communication content in real time, detect and respond to inappropriate language and changes in employee sentiment early on, and provides employees with the opportunity to review and improve their own behavior through feedback.

[1673] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1674] Step 1: Collect data

[1675] The server receives data in real time from chat applications, mail servers, and voice communication systems used within the company, such as Slack, Microsoft Teams, email, and Zoom.

[1676] Input: Chat messages, emails, voice data

[1677] Output: Add data to the text analysis queue and speech analysis queue

[1678] How it works: The server sorts data received from each system into the appropriate analysis queue depending on its format (text or audio).

[1679] Step 2: Analyzing the text data

[1680] The server sends the data in the text analysis queue to a natural language processing (NLP) module, which performs tokenization, part-of-speech tagging, and sentence structure analysis of the text data.

[1681] Input: Text data

[1682] Output: Tokenized data, tagged data, sentence structure analysis results

[1683] How it works: It splits incoming text data into words and phrases, tags them with part-of-speech tags, and analyzes the structure of entire sentences.

[1684] Step 3: Analyzing the audio data

[1685] The server sends the speech data added to the speech analysis queue to a speech recognition module, which converts the speech data into text data that is then analyzed in a subsequent step.

[1686] Input: Audio data

[1687] Output: Text data

[1688] How it works: Converts speech to text and feeds it into an NLP module for further analysis.

[1689] Step 4: Sentiment Analysis

[1690] The server passes the text data analyzed by the NLP module to an emotion engine to identify the user's emotion. The emotion engine detects the emotional tone of the text (e.g., anger, sadness, joy, etc.).

[1691] Input: Parsed text data

[1692] Output: Emotional tone identification result

[1693] How it works: The emotion engine analyzes text and recognizes emotional tone, for example identifying the message "I'm so angry" as "anger."

[1694] Step 5: Detecting non-compliance

[1695] The server then compares the parsed text data against a set of compliance detection rules, which contain keywords and phrases to identify offensive or legally problematic content.

[1696] Input: Parsed text data, emotional tone

[1697] Output: Whether or not there is a violation, detailed violation information

[1698] How it works: The server checks the text data against a set of rules to see if there are any violations, such as whether it contains keywords like "useless" or "incompetent."

[1699] Step 6: Issue an alert

[1700] When a violation is detected, the server generates an alert, taking into account the emotional data. The generated alert includes information such as the content, emotional tone, sender, and time of the violating message.

[1701] Input: Violation details, emotional tone

[1702] Output: Alert message

[1703] What it does: When a violation is detected, an alert is generated to notify the sender of the offending message and an administrator. The alert includes details such as the content and emotional tone of the offending message.

[1704] Step 7: Record the data

[1705] The server stores the data on which violations were detected, along with the analysis results and emotion data, in a database for future analysis and auditing.

[1706] Input: Violation details, emotional tone, message content

[1707] Output: Saved entries in the database

[1708] How it works: Analysis results and violation information are recorded in a database and stored for future reference and verification.

[1709] Step 8: Providing educational feedback

[1710] The device displays regular reports and real-time notifications to users, allowing them to learn and improve how their behavior is out of compliance.

[1711] Input: Analysis results, violation information, emotional tone

[1712] Output: Educational feedback report, real-time notifications

[1713] Action: Generate and provide feedback to the user based on their behavior and violations, for example informing them of the specific reasons why something was deemed "offensive."

[1714] Step 9: Customizing the ruleset

[1715] Administrators can use their devices to customize the rulesets of the AI ​​modules and emotion engines to meet the specific needs and circumstances of their companies.

[1716] Input: New rule or keyword

[1717] Output: Updated ruleset

[1718] How it works: Administrators adapt the ruleset to suit their company's needs and apply it to the system, which may include adding new keywords that are considered "offensive."

[1719] (Application example 2)

[1720] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1721] In recent years, inappropriate behavior and emotional changes between employees have often been the cause of trouble in corporate communications. In particular, in environments such as factories, where cooperation between employees and robots is necessary, communication is a crucial element. However, there is a lack of means to monitor this communication in real time and prevent problems before they occur. Therefore, there is a growing need for a system that can analyze all communication data, including emotions, in real time, identify compliance violations and negative emotional states, and immediately issue alerts.

[1722] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1723] In this invention, the server includes a means for collecting communication data within the company in real time, a means for analyzing the data using a natural language analysis module and a voice recognition module, and a means for comparing the analyzed data with a compliance violation detection rule set. This allows for real-time monitoring of communications between employees and robots and issuing warnings at appropriate times. The system also analyzes the user's emotional state and provides feedback based on that analysis, creating a healthier communication environment.

[1724] Definitions of important words

[1725] "Communication data" refers to data including text data and voice data generated within a company, as well as the results of sentiment analysis.

[1726] "Real-time" refers to immediate processing or analysis of data the moment it is generated or transmitted.

[1727] "Natural language analysis" refers to the process of mechanically interpreting text or audio data through tokenization, part-of-speech tagging, sentence structure analysis, etc.

[1728] "Speech recognition module" refers to a computer program or system for converting voice data into text data.

[1729] A "compliance violation detection rule set" refers to a collection of rules for detecting expressions that may violate legal or corporate standards based on specific keywords or phrases.

[1730] "Warning" refers to a notification or alert issued when non-compliance or a negative emotional state is detected.

[1731] "Database" refers to a system or structured collection of information for storing collected and analyzed data.

[1732] "Educational Feedback" refers to reports or guidance provided to users to help them learn about their communication and non-compliance behaviors.

[1733] "Customizing rule sets" refers to the act of an administrator tailoring compliance violation detection rules and sentiment analysis algorithms to the specific needs and circumstances of the company.

[1734] "User emotional state" refers to the user's emotional tone (e.g., anger, sadness, joy, etc.) derived from analyzed text data and voice data.

[1735] "Form for carrying out the invention" of the specification

[1736] System Configuration

[1737] The system according to the present invention comprises the following components:

[1738] Server: Responsible for data collection, analysis, violation detection, emotion recognition, warning issuance, and data storage. Specifically, it is equipped with a natural language analysis module, a voice recognition module, a compliance violation detection rule set, and an emotion analysis engine.

[1739] Terminal: A device used by employees and managers, such as a PC, a smartphone, or an application installed on a robot in a factory.

[1740] User: An employee of a company who sends and receives messages within the system.

[1741] Administrator: The person responsible for customizing the system's rule sets and sentiment analysis engine.

[1742] Explanation of program processing

[1743] The server uses a communications API to collect internal company communication data (text and voice data) in real time. This data is converted into text data through a natural language analysis module and a speech recognition module, and then analyzed based on the extracted tokens. In particular, the NLP (natural language processing) module performs tokenization, part-of-speech tagging, and sentence structure analysis to understand the meaning and intent of the message.

[1744] Meanwhile, the voice data is converted into text data by a speech recognition module. The analyzed text data is then passed to a sentiment analysis engine, which identifies emotional tones such as anger, sadness, joy, etc. If a user's message is "I'm so angry," the emotional tone will be recognized as "anger."

[1745] The analyzed data is compared against a set of compliance violation detection rules, which include offensive or legally questionable language. If such language is detected, the server immediately issues an alert. The alert includes information about the content, emotional tone, sender, and time of the offending message, and this information is sent to the user and administrator who detected the violation.

[1746] The data is stored in a database for permanent archival, including the original message, analysis results, violation details, and emotional tone, making it available for future analysis and audits.

[1747] At the device, users receive regular educational feedback on their communication behavior, including how they were non-compliant and their emotional state at the time. Administrators can customize the rule set and optimize the system by adding new keywords and phrases as needed.

[1748] Specific examples

[1749] For example, suppose employee A sends a message saying, "You're always slow at work. It pisses me off." In this case, the server would process it as follows:

[1750] 1. Messages are received in real time and analyzed by the NLP module.

[1751] 2. The sentiment analysis engine identifies the phrase "I'm angry" as anger.

[1752] 3. Based on the rule set, this message is detected as non-compliant.

[1753] 4. The server immediately issues an alert and notifies Employee A and the administrator.

[1754] Prompt Sentence Examples

[1755] Monitor chat logs between Employee A and Employee B in real time to detect compliance violations and sentiment changes. Start your analysis with the following chat logs:

[1756] "Employee A: You're always slow at work. It really pisses me off."

[1757] The system monitors communication between employees and robots in real time and provides immediate and appropriate feedback to support safe and efficient operations within the factory.

[1758] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1759] Program processing flow

[1760] Step 1:

[1761] The server collects communication data (text data and voice data) within the company in real time. Specifically, it uses a communication API to obtain messages and call content and adds them to a data queue. The input data is raw text or voice data, and the output is data added to the analysis queue.

[1762] Step 2:

[1763] The server passes the collected text data to a natural language analysis module, which tokenizes the data, tags it with parts of speech, and analyzes sentence structure. The input data is text data, and the output is tokenized analyzed data. The analyzed data is used to understand the meaning and intent of the message.

[1764] Step 3:

[1765] The server passes the collected voice data to a voice recognition module and converts it into text data. The voice recognition module analyzes the voice data and converts it into text. The input data is voice data and the output is text data. The converted text data is further analyzed by a natural language analysis module.

[1766] Step 4:

[1767] The server passes the parsed text data to the sentiment analysis engine, which analyzes the emotional tone of the text to identify the user's emotional state. The input data is the parsed text data, and the output is the identified emotional tone. For example, the message "I'm so angry" is recognized as "Anger."

[1768] Step 5:

[1769] The server compares the analyzed data against a set of compliance violation detection rules. The rule set contains keywords and phrases that determine whether a violation has occurred. The input data is the analyzed and sentiment analysis data, and the output is a determination of whether or not there is a compliance violation. If a violation is detected, information about it is generated.

[1770] Step 6:

[1771] The server immediately issues a warning if a violation is detected. The warning includes the content, emotional tone, sender, and time of sending of the violating message. The input data is the compliance violation judgment result, and the output is the generation and notification of a warning message. The notification is sent to the user who detected the violation and the administrator.

[1772] Step 7:

[1773] The server stores the analysis results and violation details in a database. The stored data includes the original message, analysis results, violation details, and emotional tone. The input data is the analyzed and violation detection data, and the output is storage in the database.

[1774] Step 8:

[1775] The device periodically provides educational feedback to the user, including how the user violated compliance and their emotional state at the time. The input data are the analysis results stored in the database, and the output is a report of the educational feedback.

[1776] Step 9:

[1777] Administrators use a terminal to customize the rule set. They can add new keywords and phrases. The input is the administrator's configuration information, and the output is an updated rule set. This allows the system to adapt to the specific needs and circumstances of the company.

[1778] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1779] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1780] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1781] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1782] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1783] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1784] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1785] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1786] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1787] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1788] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1789] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1790] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1791] 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.

[1792] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1793] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1794] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1795] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1796] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1797] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1798] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1799] The following is further disclosed regarding the above embodiment.

[1800] (Claim 1)

[1801] A means of collecting communication data within a company in real time,

[1802] means for analyzing the data using a natural language processing module and a voice recognition module;

[1803] means for matching the parsed data with a set of compliance violation detection rules;

[1804] A means of alerting when non-compliance is detected; and

[1805] A means for storing the analysis results and violation details in a database;

[1806] a means for providing educational feedback to the user on a periodic basis;

[1807] A means for administrators to customize the ruleset;

[1808] A system including:

[1809] (Claim 2)

[1810] 10. The system of claim 1, wherein the data collected in real time includes both text data and audio data.

[1811] (Claim 3)

[1812] 10. The system of claim 1, wherein the alert is sent to a user who detected the violation and to an administrator.

[1813] "Example 1"

[1814] (Claim 1)

[1815] A means of collecting communication data within a company in real time,

[1816] means for analyzing the data using a natural language processing module and a voice recognition module;

[1817] means for matching the analyzed data with a set of non-compliance detection rules;

[1818] A means of issuing notifications when non-compliance is detected; and

[1819] a means for storing the analysis results and the details of the violation in a storage device;

[1820] a means of providing educational feedback to users on a regular basis;

[1821] A means for administrators to modify the ruleset;

[1822] Real-time breach detection notification and analysis visualization;

[1823] ...

[1824] A system including:

[1825] (Claim 2)

[1826] 10. The system of claim 1, wherein the data collected in real time includes both text data and audio data.

[1827] (Claim 3)

[1828] 2. The system according to claim 1, wherein the notification is sent to the user who detected the violation and to an administrator.

[1829] "Application Example 1"

[1830] (Claim 1)

[1831] A means of collecting communication data within a company in real time,

[1832] means for analyzing the data using a natural language processing module and a voice recognition module;

[1833] means for matching the parsed data with a set of compliance violation detection rules;

[1834] A means of alerting when non-compliance is detected; and

[1835] A means for storing the analysis results and violation details in a database;

[1836] a means for providing educational feedback to the user on a periodic basis;

[1837] A means for administrators to customize the ruleset;

[1838] means for delivering the relevant notifications via an application installed on the smart device;

[1839] a means for notifying users and administrators of violation detection notifications and their details;

[1840] means for generating periodic reports based on the violation detection data;

[1841] A system including:

[1842] (Claim 2)

[1843] 10. The system of claim 1, wherein the data collected in real time includes both text data and audio data.

[1844] (Claim 3)

[1845] 10. The system of claim 1, wherein the alert is sent to a user who detected the violation and to an administrator.

[1846] "Example 2: Combining Emotion Engines"

[1847] (Claim 1)

[1848] A means of collecting communication data within a company in real time,

[1849] means for analyzing the data using a natural language processing module and a voice recognition module;

[1850] A means for identifying the user's emotion using an emotion engine based on the analyzed data;

[1851] means for matching the parsed data with a set of compliance violation detection rules;

[1852] A means of issuing an alert when a compliance violation is detected, including the content, emotional tone, sender, and time of the non-compliant message;

[1853] A means for storing the analysis results and violation details in a database;

[1854] a means for providing educational feedback to the user on a periodic basis;

[1855] A means for administrators to customize the ruleset;

[1856] A system including:

[1857] (Claim 2)

[1858] 10. The system of claim 1, wherein the data collected in real time includes both text data and audio data.

[1859] (Claim 3)

[1860] 10. The system of claim 1, wherein the alert is sent to a user who detected the violation and to an administrator.

[1861] "Application example 2 when combining emotion engines"

[1862] Rewritten claims

[1863] (Claim 1)

[1864] A means of collecting communication data within a company in real time,

[1865] means for analyzing the data using a natural language analysis module and a voice recognition module;

[1866] means for matching the parsed data with a set of compliance violation detection rules;

[1867] A means of issuing an alert when non-compliance is detected; and

[1868] A means for storing the analysis results and violation details in a database;

[1869] a means of providing educational feedback to users on a regular basis;

[1870] A means for administrators to customize the ruleset;

[1871] means for analyzing the emotional state of the user and providing feedback based on the emotion;

[1872] A system including:

[1873] (Claim 2)

[1874] 10. The system of claim 1, wherein the data collected in real time includes text data, voice data, and results of sentiment analysis.

[1875] (Claim 3)

[1876] 2. The system of claim 1, wherein the warning is notified to the user who detected the violation and to an administrator, and further notifies the emotional state of the violation message. [Explanation of symbols]

[1877] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of collecting communication data within a company in real time, means for analyzing the data using a natural language processing module and a voice recognition module; means for matching the parsed data with a set of compliance violation detection rules; A means of alerting when non-compliance is detected; and A means for storing the analysis results and violation details in a database; a means for providing educational feedback to the user on a periodic basis; A means for administrators to customize the ruleset; A system including:

2. 10. The system of claim 1, wherein the real-time collected data includes both text data and audio data.

3. The system of claim 1 , wherein the alert notifies the user who detected the violation and an administrator.

Images