System

Abstract

A system is provided.SOLUTION: The system includes a means for receiving transmitted character data, a means for analyzing the received character data by a natural language processing engine and extracting a specific keyword or emotion, a means for scoring the risk of harassment on the basis of the extracted data, a means for generating a warning message when the scoring result exceeds a fixed threshold, and a means for transmitting the warning message to a user terminal to be displayed.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] Harassment and slander are rampant in modern communication methods such as email, chat, and social media, necessitating not only mental health care for victims but also effective measures to prevent recurrence of harassment. However, currently, messages sent unintentionally by the perpetrator can have a negative impact on the recipient, and there is a lack of systems to detect this before they are sent. Therefore, there is a need to provide a means to prevent harassment from occurring by detecting potentially abusive messages in advance and encouraging appropriate corrections. [Means for solving the problem]

[0005] To solve this problem, the present invention provides a system that includes a means for receiving transmitted text data, analyzing it using a natural language processing engine, and extracting specific keywords and emotions, and a means for scoring the harassment risk level based on the extracted data. The system also includes a means for generating a warning message when the scoring result exceeds a certain threshold, and transmitting it to a user terminal for display. The system further includes a means for generating a warning message including specific suggestions for correction based on the scoring result, and a means for providing an interface that allows the user to apply the suggested corrections and resend the message when the warning message is displayed. This allows the system to warn of the harassment risk of a message before it is sent and encourage appropriate corrections, thereby preventing harassment from occurring.

[0006] "Text data to be sent" refers to text information that a user intends to send through communication means such as email, chat, or social media.

[0007] A "natural language processing engine" is a set of algorithms and software that analyzes language data and understands keywords, sentiment, and context.

[0008] "Means for extracting specific keywords and sentiment" refers to the ability to use a natural language processing engine to detect important words and phrases within text data, as well as the emotional trends of that text.

[0009] "Means for scoring the risk of harassment" refers to a function that numerically evaluates the likelihood that text constitutes harassment based on extracted keywords and emotions.

[0010] "Means for generating warning messages" refers to a function that creates an alert to prompt users to change the content of their messages when the risk of harassment exceeds a certain threshold.

[0011] "Means for sending and displaying a warning message on a user's device" refers to the function of sending the generated warning message to a user's device such as a computer or smartphone and displaying it on the screen.

[0012] "Means for generating warning messages with specific suggestions for correction" refers to the function that creates warnings for messages that pose a high risk of harassment, including specific suggestions for users on how to correct them.

[0013] "Means for providing an interface that allows the user to apply the suggested corrections and resubmit" refers to a function that, when displaying a warning message, provides a user interface that allows the user to easily correct the message and resubmit it. [Brief explanation of the drawings]

[0014] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0015] 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.

[0016] First, the terms used in the following description will be explained.

[0017] 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).

[0018] 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.

[0019] 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.

[0020] 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.

[0021] 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."

[0022] [First embodiment]

[0023] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0024] 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.

[0025] 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).

[0026] 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.

[0027] 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.

[0028] 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.

[0029] 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.

[0030] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0031] 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.

[0032] 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.

[0033] 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.

[0034] 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."

[0035] This invention is a system that checks the content of emails, chats, and social networking messages before they are sent, and detects inappropriate content or potential harassment. This system consists of a user terminal and a server, and operates as follows.

[0036] System Configuration

[0037] 1. User Device

[0038] The user terminal provides an interface for inputting character data. When the user inputs a message and clicks the send button, the character data is sent to the server.

[0039] 2. Server

[0040] The server analyzes the received text data and evaluates the risk of harassment. Specifically, it uses a natural language processing engine to extract keywords and emotions from the message and performs a scoring process based on these. If the scoring result exceeds a certain threshold, it generates a warning message and sends it to the user's device.

[0041] Program processing overview

[0042] When a user clicks the send button on a message, the device sends the message data and its metadata to the server. The server receives this data and temporarily stores it in a database. The server then analyzes the message using a natural language processing engine to extract keywords and emotions. This extracted data is used to score the harassment risk. If the score exceeds a certain threshold, the server generates a warning message and sends it to the user's device. The device then displays the warning message to the user. The user can then refer to the suggested corrections to revise the message and resend it.

[0043] Specific examples

[0044] Example 1: Sending offensive messages

[0045] 1. The user types, "You're so incompetent." and clicks the send button.

[0046] 2. The device sends this message to the server.

[0047] 3. The server receives the message and stores it in a database.

[0048] 4. The server uses a natural language processing engine to analyze the keyword "incompetent" and the overall negative sentiment.

[0049] 5. The server calculates the harassment score and determines that the score is high.

[0050] 6. The server generates a warning message stating, "This message may be considered offensive. We recommend that you correct it." and sends it to the user's device.

[0051] 7. The device displays this warning message to the user.

[0052] 8. The user corrects and resubmits, saying, "Your performance could use some improvement. Let's review the details."

[0053] Example 2: Inappropriate sending time

[0054] 1. At 2 a.m., a user types, "Please submit the materials immediately." and clicks the submit button.

[0055] 2. The device sends this message to the server.

[0056] 3. The server receives the message and stores it in a database.

[0057] 4. The server analyzes the time of sending and the message content and determines that it is an inappropriate time.

[0058] 5. The server generates a warning message stating, "This message will be sent late at night and may be offensive to the recipient. Please change the sending time or modify the message content." and sends it to the user's terminal.

[0059] 6. The device displays this warning message to the user.

[0060] 7. The user changes the sending time or corrects the message content and resends it.

[0061] In this way, the system supports safe and smooth communication by analyzing messages that users are about to send in advance, pointing out any potential harassment, and encouraging users to correct the messages.

[0062] The processing flow will be explained below.

[0063] Step 1:

[0064] The user types a message and clicks the send button.

[0065] A send trigger occurs when you enter a message on the user interface and press the send button.

[0066] Step 2:

[0067] The terminal transmits the input message data and its metadata to the server.

[0068] Data such as the message content, sender information, and scheduled sending time is transferred to the server.

[0069] Step 3:

[0070] The server temporarily stores the received message data in a database.

[0071] Temporarily stores message data and metadata so that the data can be used for subsequent processing in a secure and efficient manner.

[0072] Step 4:

[0073] The server runs a natural language processing engine to analyze the message.

[0074] Extract specific keywords and phrases from the text and then perform sentiment analysis.

[0075] Step 5:

[0076] The server scores the harassment risk based on the extracted data.

[0077] Each keyword, phrase, and emotion is evaluated to quantify the overall risk.

[0078] Step 6:

[0079] The server generates a warning message if the harassment score exceeds a certain threshold.

[0080] Create a warning message with specific suggested fixes to alert users.

[0081] Step 7:

[0082] A server-generated warning message is sent to the user terminal.

[0083] The original message is forwarded to the user's terminal along with the warning message.

[0084] Step 8:

[0085] The device will pop up a warning message to the user.

[0086] The app will show specific areas of harassment and suggest corrections, and encourage users to check and correct the content.

[0087] Step 9:

[0088] The user considers the warnings and suggestions and modifies the message.

[0089] Follow the suggested amendments or amend as you see fit.

[0090] Step 10:

[0091] The user clicks the send button again to resend the revised message.

[0092] A resend trigger occurs and the corrected message is sent to the server again.

[0093] Step 11:

[0094] The server performs a final check of the revised message and, if there are no problems, sends it to the recipient.

[0095] If the content is deemed to be acceptable, the message is officially sent to the recipient.

[0096] This process allows users to see how their message will affect the recipient before sending it, helping to prevent harassment and inappropriate messages.

[0097] Example 1

[0098] 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."

[0099] In modern communication tools such as email, chat, and social networking sites, messages sent by users can sometimes be overly aggressive or sent at inappropriate times. This can cause the recipient to feel uncomfortable or become a victim of harassment. Moreover, overly aggressive messages or messages sent at inappropriate times can disrupt smooth communication. Therefore, a system that can prevent these problems from occurring is needed.

[0100] 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.

[0101] In this invention, the server includes means for receiving text data to be sent, means for analyzing the received text data using a natural language processing engine and extracting specific keywords and emotions, means for analyzing the time of sending and identifying inappropriate messages based on the time of day, means for scoring the harassment risk level based on the extracted data, means for generating a warning message when the scoring result exceeds a certain threshold, and means for sending and displaying the warning message to the user terminal. This makes it possible to analyze in advance the content and timing of messages that users intend to send and prevent the sending of inappropriate messages.

[0102] The "means for receiving transmitted character data" is a mechanism for receiving character information transmitted by a user.

[0103] A "natural language processing engine" is software or a service that analyzes text data and extracts keywords and emotions.

[0104] "Means for extracting specific keywords and emotions" refers to a function that recognizes and acquires important words and emotional trends from received text data.

[0105] The "means for scoring the risk of harassment" is a process that quantifies the possibility of harassment based on extracted keywords and emotions.

[0106] The "means for generating a warning message" is a mechanism for creating text to warn the user based on the scoring results.

[0107] "Means for sending and displaying a warning message on a user terminal" refers to a function for delivering the generated warning message to the user's device and displaying it there.

[0108] "Means for analyzing sending times and identifying inappropriate messages based on time of day" refers to a mechanism that evaluates the time a message was sent and determines whether the content is inappropriate for that time period.

[0109] A "warning message with specific suggestions for correction" is a warning message that includes specific advice for correcting a message that has been determined to be inappropriate.

[0110] An "interface that allows a user to apply suggested corrections to a message and then resend the message" is a screen or feature that allows a user to apply suggested corrections to a message and then resend the message with the corrections.

[0111] MODE FOR CARRYING OUT THE INVENTION

[0112] The present invention is a system that analyzes messages a user is about to send in advance to detect potential harassment and inappropriate sending times. The system consists of a user terminal and a server. When a user inputs and sends a message, the message is sent to the server and analyzed using a natural language processing engine. Based on the analysis results, the system scores the risk of harassment and generates a warning message if necessary.

[0113] User Device

[0114] The user terminal provides an interface where the user can enter a message into a text entry form. When the user enters a message and clicks the send button, the terminal sends the message and associated metadata (such as the time of sending and sender information) to the server.

[0115] server

[0116] The server has the following functions:

[0117] 1. Receiving function

[0118] The server receives the text data sent from the user terminal and temporarily stores it in a database, which ensures the consistency and availability of the data.

[0119] 2. Analysis Function

[0120] The server analyzes the received message using a natural language processing engine (e.g., Google Cloud Natural Language API), extracting keywords and sentiments from the message and obtaining the data.

[0121] 3. Scoring function

[0122] The server scores the risk of harassment based on the extracted keywords and emotional data. It also analyzes the time of message sending and identifies inappropriate messages based on the time of day. If the scoring results exceed a certain threshold, a warning message is generated.

[0123] 4. Warning generation function

[0124] If the scoring result exceeds a threshold, the server generates a warning message including specific suggestions for correction, and then sends the warning message to the user terminal.

[0125] 5. Display function

[0126] The warning message sent from the server is displayed on the user's terminal, and the user can check and modify the warning message.

[0127] With the above functions, the system of the present invention can analyze messages that users are about to send in advance, point out potential harassment and inappropriate sending times, and prompt users to correct the messages, thereby supporting safe and smooth communication.

[0128] Specific examples

[0129] Example 1: Sending offensive messages

[0130] 1. The user types, "You're so incompetent." and clicks the send button.

[0131] 2. The device sends this message to the server.

[0132] 3. The server temporarily stores the received message in a database.

[0133] 4. The server uses the Google Cloud Natural Language API to extract the keyword "incompetent" and negative sentiment.

[0134] 5. The server scores the harassment risk and determines that the score exceeds the threshold.

[0135] 6. The server generates a warning message stating, "This message may be considered offensive. We recommend that you correct it." and sends it to the user's device.

[0136] 7. The device displays a warning message to the user and prompts the user to correct the message.

[0137] Example 2: Inappropriate sending time

[0138] 1. At 2 a.m., a user types, "Please submit the materials immediately." and clicks the submit button.

[0139] 2. The device sends the message and the time of sending to the server.

[0140] 3. The server temporarily stores the received message in a database.

[0141] 4. The server analyzes the sending time and determines that it is late at night.

[0142] 5. The server generates a warning message stating, "This message will be sent late at night and may be offensive to the recipient. Please change the sending time or modify the message content." and sends it to the user's terminal.

[0143] 6. The device will display a warning message to the user, prompting them to correct the sending time and message content.

[0144] Prompt Sentence Examples

[0145] "Please use a natural language processing engine to evaluate this message to determine whether it is likely to be harassment. The message content is as follows."

[0146] This system gives users the opportunity to correct inappropriate messages before they are sent, thereby promoting safe and smooth communication.

[0147] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0148] Step 1:

[0149] The user enters a message into the text entry form and clicks the send button.

[0150] Input: Text data entered by the user

[0151] Output: Text data and transmission metadata are temporarily saved on the device.

[0152] Step 2:

[0153] The device sends the message data and its metadata (e.g., sending time, sender information) to the server.

[0154] Input: User-entered text data and metadata

[0155] Output: An HTTP POST request is sent to the server

[0156] Step 3:

[0157] The message data received by the server is temporarily stored in a database.

[0158] Input: Message data and metadata received from the device.

[0159] Output: Message data and metadata stored in a database

[0160] Step 4:

[0161] The server uses a natural language processing engine (e.g., Google Cloud Natural Language API) to analyze the stored messages.

[0162] Input: Message data in the database

[0163] Output: Analysis results (keywords and sentiment data) returned by the natural language processing engine

[0164] Step 5:

[0165] The server then assigns a score to the message based on the analysis results, assessing the risk of harassment. It also analyzes the time the message was sent and identifies inappropriate messages based on the time of day.

[0166] Input: Analysis results of the natural language processing engine, message sending time

[0167] Output: Harassment score, evaluation result based on sending time

[0168] Step 6:

[0169] If the server scores above a certain threshold, it generates a warning message, such as "This message may be considered offensive. We recommend that you correct it."

[0170] Input: Harassment score, evaluation result based on submission time

[0171] Output: Warning message

[0172] Step 7:

[0173] A server-generated warning message is sent to the user terminal.

[0174] Input: warning message

[0175] Output: HTTP response to the user's device

[0176] Step 8:

[0177] The device will display a warning message to the user, allowing the user to review and modify the message.

[0178] Input: Warning message from the server

[0179] Output: A warning message that is displayed in the user interface.

[0180] The above is a description of the specific processing steps of this system.

[0181] (Application example 1)

[0182] 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."

[0183] In modern communication, especially in the workplace or when messaging between employees in brick-and-mortar stores, there is a risk of inappropriate language or harassment. Furthermore, sending messages at inappropriate times, such as late at night, can cause discomfort to the recipient. This raises concerns about a worsening work environment and reduced work efficiency. There is a need to resolve this issue and maintain healthy and comfortable communication.

[0184] 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.

[0185] In this invention, the server includes means for receiving transmitted text data, means for analyzing the received text data using a natural language processing engine and extracting specific keywords and emotions, means for scoring the harassment risk level based on the extracted data, means for generating a warning message when the scoring result exceeds a certain threshold, means for transmitting and displaying the warning message on the user terminal, and means for analyzing the time of message transmission and generating the warning message if the transmission time is inappropriate. This makes it possible to monitor both the message content and transmission time and prevent inappropriate communication before it occurs.

[0186] "Text data to be sent" refers to the text information that a user inputs as a message and intends to send.

[0187] "Means for receiving" refers to the method or mechanism by which the server receives character data sent from the user terminal.

[0188] A "natural language processing engine" is a software engine that analyzes text data and extracts meaning, emotion, and specific keywords.

[0189] "Means for analyzing and extracting" refers to methods and mechanisms for extracting specific keywords and emotional information from text data using a natural language processing engine.

[0190] "Means for scoring the risk of harassment" refers to a method or mechanism for numerically assessing the risk of harassment in text data based on extracted keywords and emotional information.

[0191] The "means for generating a warning message" refers to a method or mechanism for creating a message to warn the user based on the scoring result.

[0192] "Means for sending and displaying a warning message on a user terminal" refers to a method or mechanism for sending a generated warning message to a user's device and displaying it to the user.

[0193] The "means for analyzing the sending time of a message" refers to a method or mechanism for analyzing the time information of a message to be sent and determining whether the time is appropriate.

[0194] "Means for generating a warning message when a time is inappropriate for sending a message" refers to a method or mechanism for generating a warning message to notify the user when a time for sending a message is determined to be inappropriate.

[0195] The present invention is a system that analyzes transmitted text data, evaluates the risk of harassment, and checks the appropriateness of the transmission time. Specific embodiments will be described below.

[0196] System Configuration

[0197] (user device)

[0198] The user terminal is a device that provides an interface for inputting messages. When a user inputs a message and clicks the send button, the text data and the time of sending are sent to the server.

[0199] (server)

[0200] The server analyzes the received text data using a natural language processing engine to extract specific keywords and emotions from the text. It also scores the harassment risk based on the extracted data and generates a warning message if the scoring result exceeds a certain threshold. In addition, the server analyzes the time the message was sent and generates a separate warning message if it was sent at an inappropriate time. The generated warning message is sent to the user's device and displayed.

[0201] Hardware and software used

[0202] (Hardware)

[0203] Smartphone: Used as a user device.

[0204] (software)

[0205] Python: The language used for server-side programming.

[0206] textblob: A library used as a natural language processing engine.

[0207] Specific processing and data calculations

[0208] The server analyzes messages sent by users using a natural language processing engine (here we use textblob). This process extracts emotions and specific keywords (such as "incompetent," "stupid," and "dislike") from the text. The extracted data is then used to score the harassment risk. If the score exceeds a certain threshold, the server generates a warning message and sends it to the user's device.

[0209] The server also analyzes the time the message was sent and generates a separate warning message if it was sent at an inappropriate time, such as 2 a.m. This allows users to check not only the content of the message but also the time it was sent, preventing inappropriate communication.

[0210] Specific examples

[0211] For example, if a user types "Please submit the documents now" at 2 a.m. and clicks the send button, the system will behave as follows:

[0212] 1. The user types and sends a message.

[0213] 2. The device sends this message and the sending time to the server.

[0214] 3. The server receives the message and the time it was sent and parses it using a natural language processing engine.

[0215] 4. The server scores the harassment risk and also checks the time of transmission.

[0216] 5. Generate a warning message if it is determined to be inappropriate based on sentiment, keywords, and sending time.

[0217] 6. The server generates a warning message and sends it to the user's device.

[0218] 7. The device displays this warning message to the user.

[0219] Example prompt sentence:

[0220] "Analyze the following messages and generate a warning message if they contain harassing or inappropriate content.

[0221] Message: 'You're really incompetent.'

[0222] Sent time: '02:00'"

[0223] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0224] Step 1:

[0225] The user enters a message and clicks the send button. At this time, the message data and the current sending time are obtained. This is the user's input.

[0226] Step 2:

[0227] The device sends the message data and its metadata (such as the sending time) to the server. In this process, the device sends the message and sending time to the server as data. The input is the message data and sending time, and the output is that these data are sent to the server.

[0228] Step 3:

[0229] The server temporarily stores the received message data and sending time in a database. The input is the message data and sending time sent from the terminal, and the output is that these data are stored in the database.

[0230] Step 4:

[0231] The server analyzes the message using a natural language processing engine (textblob). During this analysis, specific keywords (such as "incompetent," "stupid," or "hate") and emotions (positive or negative) are extracted from the message. The input is message data retrieved from the database, and the output is the analyzed keywords and emotional information.

[0232] Step 5:

[0233] The server scores the harassment risk based on the extracted keywords and emotional information. This scoring is calculated based on a certain algorithm (for example, the number of occurrences of specific keywords or the intensity of emotions). The input is the keywords and emotional information from the analysis results, and the output is the calculated score.

[0234] Step 6:

[0235] If the score exceeds a certain threshold, the server generates a warning message informing the user that the message may be offensive and encouraging them to correct it. The input is the calculated score, and the output is the generated warning message.

[0236] Step 7:

[0237] The server analyzes the sending time. It checks whether the sending time is an inappropriate time, such as 2 a.m. The input is the sending time retrieved from the database, and the output is the evaluation result of the appropriateness of the time (appropriate / inappropriate).

[0238] Step 8:

[0239] If the server determines that the sending time is inappropriate, it generates a separate warning message. This warning message notifies the user that the sending time is inappropriate. The input is the result of the time appropriateness evaluation, and the output is the generated warning message.

[0240] Step 9:

[0241] The server sends the generated warning message to the user terminal. The input is the generated warning message, and the output is the message sent to the user terminal.

[0242] Step 10:

[0243] The terminal displays a warning message to the user, allowing the user to receive feedback on the appropriateness of the message content and the time it was sent. The input is the warning message sent from the server, and the output is the display of the warning message to the user.

[0244] In this way, the system can properly monitor messages sent by users and prevent problems caused by the risk of harassment or inappropriate sending times.

[0245] 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.

[0246] This invention is a system that checks the content and sender's emotions of email, chat, and social networking messages before they are sent, and detects inappropriate content or possible harassment. This system consists of a user terminal, a server, and an emotion engine, and operates as follows:

[0247] System Configuration

[0248] 1. User Device

[0249] The user terminal provides an interface for inputting text data and includes an emotion engine for acquiring the user's emotion data. When the user inputs a message and clicks the send button, the text data and emotion data are transmitted to the server.

[0250] 2. Server

[0251] The server analyzes the received text data and emotional data to assess the risk of harassment. Specifically, it uses a natural language processing engine to extract keywords and emotions from the message and performs a scoring process based on these. It also integrates the user's emotional data received from the emotion engine to perform the evaluation. If the scoring result exceeds a certain threshold, it generates a warning message and sends it to the user's device.

[0252] Program processing overview

[0253] When a user clicks the send button on a message, the device sends the message data and emotion data to the server. The server receives this data and temporarily stores it in a database. The server then analyzes the message using a natural language processing engine and emotion engine to extract keywords, emotions, and the user's emotion data. This extracted data is used to score the harassment risk. If the score exceeds a certain threshold, the server generates a warning message and sends it to the user's device. The device then displays this warning message to the user. The user can then refer to the suggested corrections to revise the message and resend it.

[0254] Specific examples

[0255] Example 1: Sending offensive messages and making users angry

[0256] 1. A user types "You're so incompetent" and clicks the send button. At this time, the emotion engine detects that the user is very angry.

[0257] 2. The device sends this message and emotion data to the server.

[0258] 3. The server receives the message and emotion data and temporarily stores them in a database.

[0259] 4. The server uses a natural language processing engine to analyze the keyword "incompetent" and the overall negative sentiment, and integrates the user's anger data from the sentiment engine.

[0260] 5. The server calculates the harassment score and determines that the score is high.

[0261] 6. The server generates a warning message stating, "This message may be considered offensive. We recommend that you correct it." and sends it to the user's device.

[0262] 7. The device displays this warning message to the user.

[0263] 8. The user corrects and resubmits, saying, "Your performance could use some improvement. Let's review the details."

[0264] Example 2: Inappropriate send times and user fatigue

[0265] 1. When a user types "Please submit the materials immediately" at 2 a.m. and clicks the send button, the emotion engine detects that the user is tired.

[0266] 2. The device sends this message and emotion data to the server.

[0267] 3. The server receives the message and emotion data and temporarily stores them in a database.

[0268] 4. The server analyzes the sending time, message content, and emotional data to determine that the message was sent at an inappropriate time.

[0269] 5. The server generates a warning message stating, "This message will be sent late at night and may be offensive to the recipient. Please change the sending time or modify the message content." and sends it to the user's terminal.

[0270] 6. The device displays this warning message to the user.

[0271] 7. The user changes the sending time or corrects the message content and resends it.

[0272] In this way, this system supports safe and smooth communication by analyzing the message the user is about to send and their emotions at the time, pointing out any potential harassment and encouraging them to correct it.By taking the user's emotions into consideration, this system can provide more appropriate warnings and suggestions for correction, preventing harassment and inappropriate messages before they occur.

[0273] The processing flow will be explained below.

[0274] Step 1:

[0275] The user types a message and clicks the send button.

[0276] The user inputs a message on the user interface and presses the send button to trigger a transmission. The emotion engine obtains emotional data from the user's facial expressions, voice, etc.

[0277] Step 2:

[0278] The terminal transmits the input message data and emotion data to the server.

[0279] Metadata such as message content, sender information, emotion data, and scheduled sending time are transferred to the server.

[0280] Step 3:

[0281] The server temporarily stores the received message data and emotion data in a database.

[0282] Various data is temporarily stored in a database so that it can be used efficiently in subsequent processing.

[0283] Step 4:

[0284] The server analyzes the message using a natural language processing engine and an emotion engine.

[0285] It extracts specific keywords and phrases from text and also performs sentiment analysis, simultaneously analyzing user emotional data.

[0286] Step 5:

[0287] The server scores the harassment risk based on the extracted data.

[0288] It evaluates each keyword and phrase, the overall sentiment of the text, and user sentiment data to quantify the overall risk.

[0289] Step 6:

[0290] The server generates a warning message if the harassment score exceeds a certain threshold.

[0291] Create a warning message with specific suggestions for correction based on the harassment score.

[0292] Step 7:

[0293] A server-generated warning message is sent to the user terminal.

[0294] Along with the warning message, the original message and additional information based on emotion data are also transmitted to the user's device.

[0295] Step 8:

[0296] The device will pop up a warning message to the user.

[0297] The app will display specific areas of harassment, suggested corrections, and additional information that takes into account the user's feelings, encouraging the user to review and correct the content.

[0298] Step 9:

[0299] The user considers the warnings and suggestions and modifies the message.

[0300] Follow the suggested changes or make your own changes as you see fit, including taking into account the user's emotional feedback.

[0301] Step 10:

[0302] The user clicks the send button again to resend the revised message.

[0303] A resend trigger occurs and the corrected message is sent to the server again.

[0304] Step 11:

[0305] The server performs a final check of the revised message and, if there are no problems, sends it to the recipient.

[0306] If the corrections are deemed acceptable, the message will be officially sent to the recipient.

[0307] This series of processes allows users to check the impression their message will make on the recipient before sending it, preventing harassment and inappropriate messages. By combining it with an emotion engine, it is possible to take user emotions into account and provide more appropriate warnings and suggestions for correction.

[0308] Example 2

[0309] 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."

[0310] Messages sent via online email, chat, social media, etc. can sometimes contain inappropriate content or elements of harassment, causing trouble or upsetting experiences for some users. While there is a need for a system that can detect inappropriate messages in advance, warn senders, and encourage them to correct their messages, current systems often do not provide appropriate evaluations or warnings that fully consider user feelings. Furthermore, few systems offer specific suggestions for correction, leaving users unsure of how to correct their messages.

[0311] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: means for receiving transmitted text data and user emotion data; means for temporarily storing the received text data and emotion data in a database; means for analyzing the received text data using a natural language processing engine and extracting specific keywords and emotions; means for analyzing the received emotion data using an emotion engine and evaluating the user's emotion; means for scoring the harassment risk level based on the analysis results of the natural language processing engine and the emotion engine; means for generating a warning message when the scoring result exceeds a certain threshold; means for sending and displaying the warning message on the user terminal; and means for providing an interface that allows the user to refer to the suggested corrections and resend the message. This prevents the sending of inappropriate messages and provides the user with appropriate correction suggestions, enabling safe and smooth communication.

[0312] "Text data to be sent" refers to the text information entered by the user to be sent via email, chat, social media, etc.

[0313] "Emotion data" refers to data that indicates the user's emotional state and is information that is analyzed by the emotion engine.

[0314] "Means for temporarily storing in a database" refers to a storage device within the system for temporarily storing received character data and emotion data.

[0315] A "natural language processing engine" refers to software that analyzes text data and extracts specific keywords and emotions.

[0316] "Emotion engine" refers to software that analyzes emotional data and assesses a user's emotional state.

[0317] "Harassment risk" refers to a score that indicates the likelihood that a message constitutes harassment based on the text data and emotional data sent.

[0318] "Scoring means" refers to algorithms or programs that calculate the risk of harassment based on the analysis results of the natural language processing engine and emotion engine.

[0319] "Means for generating a warning message" refers to software or a system that automatically creates a message to warn the user when the scoring result exceeds a threshold.

[0320] "Means for sending and displaying on a user terminal" refers to a system for sending the generated warning message to a user's device and displaying it in the form of a notification or pop-up, etc.

[0321] "Means for providing an interface that allows a user to modify and resubmit a message based on the suggested modifications" refers to a system that provides a user interface that allows a user to edit and resubmit a message based on the warning message.

[0322] MODE FOR CARRYING OUT THE INVENTION

[0323] This invention is a system that analyzes the content and sender's emotions before sending messages via email, chat, social media, etc., to detect inappropriate content or potential harassment. This system consists of a user terminal, a server, and an emotion engine. Each component and its operation are described in detail below.

[0324] User Device

[0325] The user terminal provides an interface for inputting text data and includes an emotion engine for acquiring the user's emotion data. When the user inputs a message and clicks the send button, the text data and emotion data are transmitted to the server.

[0326] server

[0327] The server analyzes the received text data and emotion data to assess the risk of harassment. Specifically, it performs the following steps:

[0328] 1. Data Receipt and Storage:

[0329] The server receives the character data and emotion data sent from the user terminal.

[0330] The server temporarily stores the received data in a database.

[0331] 2. Data Analysis:

[0332] The server extracts keywords and sentiment from the message using a natural language processing engine, which performs text analysis of the message content to detect offensive keywords such as "incompetent" and the negative sentiment of the entire sentence.

[0333] The server further analyzes the received emotion data using an emotion engine to assess the user's emotional state (e.g., anger, fatigue).

[0334] 3. Scoring:

[0335] The server combines the analysis results from the natural language processing engine and the emotion engine to score the harassment risk. If offensive language and negative emotions coexist, a higher harassment score is assigned.

[0336] 4. Warning message generation:

[0337] If the server finds that the scoring result exceeds a certain threshold, it generates a warning message and prompts the user to modify the message content. For example, it generates a warning message such as, "This message may be considered offensive. We recommend that you modify it."

[0338] 5. Sending a message:

[0339] A server-generated warning message is sent to the user terminal.

[0340] Message display and editing

[0341] The device will display the received warning message to the user, allowing the user to review the warning message and correct it based on the specific correction suggestions.

[0342] If the user reflects the proposed revisions and resubmits, smoother communication is ensured.

[0343] Specific examples

[0344] Example 1: Sending offensive messages and making users angry

[0345] When a user types "You're so incompetent," and clicks the send button, the emotion engine detects that the user is very angry.

[0346] The device sends this message and emotion data to the server.

[0347] The server receives the message and emotion data and temporarily stores them in a database.

[0348] The server uses a natural language processing engine to analyze the keyword "incompetent" and overall negative sentiment, and integrates user anger data from the sentiment engine.

[0349] The server calculates the harassment score and determines that the score is high.

[0350] The server generates a warning message stating, "This message may be considered offensive. We recommend that you correct it." and sends it to the user's device.

[0351] The device will display this warning message to the user.

[0352] The user corrects the error by saying, "Your performance could use some improvement. Let's review the details." and resubmits.

[0353] Example 2: Inappropriate send times and user fatigue

[0354] If a user types "Please submit the documents immediately" at 2 a.m. and clicks the send button, the emotion engine will detect that the user is tired.

[0355] The device sends this message and emotion data to the server.

[0356] The server receives the message and emotion data and temporarily stores them in a database.

[0357] The server analyzes the sending time, message content, and emotional data to determine that the message was sent at an inappropriate time.

[0358] The server generates a warning message stating, "This message will be sent late at night and may be offensive to the recipient. Please change the sending time or modify the message content." and sends it to the user's terminal.

[0359] The device will display this warning message to the user.

[0360] The user can change the send time or modify the message content and resend it.

[0361] In this way, the system supports safe and smooth communication by analyzing the content of the message the user is about to send and their emotions at the time, pointing out any potential harassment and encouraging them to correct it. The warning message also includes specific suggestions for how to improve the message.

[0362] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0363] Step 1:

[0364] The user enters a message and clicks the send button. The user enters the text data they wish to send (e.g., "You're really incompetent."), and emotional data (e.g., angry state) is acquired at this time. The input data consists of a string of characters and emotional information.

[0365] Step 2:

[0366] The device acquires the text data and emotion data and sends them to the server. Specifically, when the user clicks the send button, the device sends both the text data (e.g., "You're really incompetent.") and emotion data (e.g., anger) to the server in a data format such as JSON.

[0367] Step 3:

[0368] The server receives the character data and emotion data and temporarily stores them in a database. Specifically, the server stores the received data in the database so that it can be used for subsequent analysis. The input is the received data, and the output is temporary storage in the database.

[0369] Step 4:

[0370] The server analyzes the text data using a natural language processing engine. Specifically, the server retrieves the text data from the database and uses the NLP engine to analyze keywords (e.g., "incompetent") and overall sentiment. The input is the text data, and the output is the analysis results.

[0371] Step 5:

[0372] The server uses an emotion engine to analyze the emotion data and evaluate the user's emotional state. Specifically, the emotion engine is used to analyze the user's emotion data in more detail, for example, to quantify the level of anger. The input is emotion data, and the output is an emotion evaluation.

[0373] Step 6:

[0374] The server integrates the results obtained from the natural language processing engine and the emotion engine to generate a score for the harassment risk. Specifically, for example, a high harassment score is generated based on the keyword "incompetent" and the user's emotion of "anger." The inputs are the analysis results and emotion assessment, and the output is a harassment score.

[0375] Step 7:

[0376] If the server scores above a certain threshold, it generates a warning message stating, "This message may be considered offensive. We recommend that you correct it." The input is the harassment score, and the output is the warning message.

[0377] Step 8:

[0378] The server sends a warning message to the user terminal. The input is the warning message and the output is the message sent to the user terminal.

[0379] Step 9:

[0380] The device displays a warning message to the user, specifically, by displaying the warning message in the user interface as a popup or notification. The input is the received warning message, and the output is the screen display.

[0381] Step 10:

[0382] The user reviews the warning message and modifies it based on the suggested corrections. For example, the user modifies the message to avoid offensive language and to say, "Your performance has room for improvement. Please review the details." The input is the warning message, and the output is the modified message.

[0383] Step 11:

[0384] The user resubmits the modified message. This time, the same processing flow is followed again to reevaluate whether the resubmitted message is appropriate. The input is the modified message, and the output is the final message transmission.

[0385] (Application example 2)

[0386] 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."

[0387] In customer support for electronic payment services, if the messages sent by the staff are inappropriate or offensive, the quality of the service may decline and customer satisfaction may be impaired. The present invention aims to solve this problem and provide a system for improving the quality of customer support and the customer experience.

[0388] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving transmitted character data, means for analyzing the received character data using a natural language processing engine and extracting specific keywords and emotions, and means for scoring the harassment risk based on the extracted data. This makes it possible to monitor customer support messages in electronic payment services and improve hospitality and customer experience.

[0389] A "user terminal" is a device for inputting and transmitting text data, and includes smartphones, personal computers, etc.

[0390] A "natural language processing engine" is software that analyzes text data and extracts specific keywords and emotions.

[0391] "Harassment risk" is a numerical representation of the likelihood that a particular message is inappropriate or offensive.

[0392] A "warning message" is a message that is generated to warn users when the risk of harassment exceeds a certain threshold.

[0393] An "electronic payment service" is a system for electronically paying for goods and services online or offline.

[0394] "Customer support" refers to services that respond to customer inquiries and problems.

[0395] "Hospitality" is a concept that refers to the quality of hospitality and treatment shown to customers.

[0396] "Customer experience" refers to the overall satisfaction and impression a customer has when using a service.

[0397] "Interface" refers to the screen and operating means that users use to interact with a system.

[0398] The present invention provides a system for monitoring customer support messages for electronic payment services to improve hospitality and customer experience. The system includes a user terminal, a server, an emotion engine, and a natural language processing engine. Specific embodiments for implementing the present invention are described below.

[0399] System Configuration

[0400] 1. User Device

[0401] The user terminal is a device for inputting and sending text data. This can be a smartphone or a PC. When the user inputs a support message and clicks the send button, the message and emotional data are sent to the server.

[0402] 2. Server

[0403] The server analyzes the received text data and emotion data to evaluate the degree of harassment risk. Specifically, the process is as follows:

[0404] Natural language processing engine: Analyzes text data and extracts specific keywords and sentiment (example of software used: spaCy).

[0405] Emotion engine: Analyzes the user's emotional data and extracts their emotional state (example of software used: IBM Watson NLP).

[0406] Scoring engine: Scores the level of harassment risk based on extracted keywords and sentiment data.

[0407] 3. Generating and sending warning messages

[0408] If the scoring result exceeds a certain threshold, the server generates a warning message and sends it to the user's device, including specific suggestions for correction.

[0409] Specific examples

[0410] 1. Inappropriate messages from support agents

[0411] 1. The support representative types "This problem is on your side" and clicks the send button.

[0412] 2. The emotion engine detects when a person is feeling stressed.

[0413] 3. The server receives the message and sentiment data and analyzes it for keywords like "the cause is on the user's side" and negative sentiment.

[0414] 4. Calculate the harassment score and determine it to be high.

[0415] 5. A warning message will be generated stating, "This message may be considered offensive. We recommend that you fix it."

[0416] 6. A warning message is sent to the user's terminal and displayed.

[0417] 7. The support person will fix it by saying, "Some of the problem is on the user's side, but we will deal with it."

[0418] Prompt Sentence Examples

[0419] The system analyzes the content and sentiment data of messages sent by customer support staff of electronic payment services, automatically detecting inappropriate content or potential harassment. If the message is potentially offensive or misleading, a warning message is generated and displayed. Example message: "The cause of this problem lies with the user."

[0420] This system will enable customer support for electronic payment services to be more professional and hospitable.

[0421] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0422] Step 1:

[0423] The user enters a message and clicks the send button. The input is text data, and emotion data is also acquired at the same time. By sending, the user device sends this data to the server.

[0424] Step 2:

[0425] The server receives the text data and emotion data from the user terminal and stores the input text data and emotion data in a temporary storage device for analysis.

[0426] Step 3:

[0427] The server uses a natural language processing engine to analyze the received text data. The input is text data, and keywords and the overall emotional state are extracted. Specifically, keyword extraction is performed using the spaCy library.

[0428] Step 4:

[0429] The server analyzes the user's emotional data using an emotion engine. The input is the emotional data, and IBM Watson NLP is used to extract the emotional state. The output is the user's specific emotional state.

[0430] Step 5:

[0431] The server integrates the data obtained from the natural language processing engine and the emotion engine. Specifically, the scoring engine evaluates the harassment risk based on the extracted keywords and emotional state, and calculates a harassment score. The input is keywords and emotion data, and the output is a harassment score.

[0432] Step 6:

[0433] The scoring engine determines if the score exceeds a certain threshold. If so, the server generates a warning message. The input is the harassment score, and the output is the warning message. The generated warning message includes specific suggestions for correction.

[0434] Step 7:

[0435] The server sends the generated warning message to the user terminal, which displays the message and notifies the user. The input is the warning message, and the output is the displayed warning message.

[0436] Step 8:

[0437] The user checks the presented warning message, corrects the message if necessary, and resends it. Based on the warning message, the user applies the specific correction suggestions, enters a new message, and presses the resend button. The input is the corrected text data, and the output is the resent message.

[0438] This series of steps enables customer support for electronic payment services to automatically evaluate the appropriateness of sent messages and prompt necessary corrections to provide a better customer experience.

[0439] 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.

[0440] 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.

[0441] 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.

[0442] [Second embodiment]

[0443] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0444] 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.

[0445] 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).

[0446] 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.

[0447] 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.

[0448] 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).

[0449] 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.

[0450] 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.

[0451] 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.

[0452] 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.

[0453] 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.

[0454] 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."

[0455] This invention is a system that checks the content of emails, chats, and social networking messages before they are sent, and detects inappropriate content or potential harassment. This system consists of a user terminal and a server, and operates as follows.

[0456] System Configuration

[0457] 1. User Device

[0458] The user terminal provides an interface for inputting character data. When the user inputs a message and clicks the send button, the character data is sent to the server.

[0459] 2. Server

[0460] The server analyzes the received text data and evaluates the risk of harassment. Specifically, it uses a natural language processing engine to extract keywords and emotions from the message and performs a scoring process based on these. If the scoring result exceeds a certain threshold, it generates a warning message and sends it to the user's device.

[0461] Program processing overview

[0462] When a user clicks the send button on a message, the device sends the message data and its metadata to the server. The server receives this data and temporarily stores it in a database. The server then analyzes the message using a natural language processing engine to extract keywords and emotions. This extracted data is used to score the harassment risk. If the score exceeds a certain threshold, the server generates a warning message and sends it to the user's device. The device then displays the warning message to the user. The user can then refer to the suggested corrections to revise the message and resend it.

[0463] Specific examples

[0464] Example 1: Sending offensive messages

[0465] 1. The user types, "You're so incompetent." and clicks the send button.

[0466] 2. The device sends this message to the server.

[0467] 3. The server receives the message and stores it in a database.

[0468] 4. The server uses a natural language processing engine to analyze the keyword "incompetent" and the overall negative sentiment.

[0469] 5. The server calculates the harassment score and determines that the score is high.

[0470] 6. The server generates a warning message stating, "This message may be considered offensive. We recommend that you correct it." and sends it to the user's device.

[0471] 7. The device displays this warning message to the user.

[0472] 8. The user corrects and resubmits, saying, "Your performance could use some improvement. Let's review the details."

[0473] Example 2: Inappropriate sending time

[0474] 1. At 2 a.m., a user types, "Please submit the materials immediately." and clicks the submit button.

[0475] 2. The device sends this message to the server.

[0476] 3. The server receives the message and stores it in a database.

[0477] 4. The server analyzes the time of sending and the message content and determines that it is an inappropriate time.

[0478] 5. The server generates a warning message stating, "This message will be sent late at night and may be offensive to the recipient. Please change the sending time or modify the message content." and sends it to the user's terminal.

[0479] 6. The device displays this warning message to the user.

[0480] 7. The user changes the sending time or corrects the message content and resends it.

[0481] In this way, the system supports safe and smooth communication by analyzing messages that users are about to send in advance, pointing out any potential harassment, and encouraging users to correct the messages.

[0482] The processing flow will be explained below.

[0483] Step 1:

[0484] The user types a message and clicks the send button.

[0485] A send trigger occurs when you enter a message on the user interface and press the send button.

[0486] Step 2:

[0487] The terminal transmits the input message data and its metadata to the server.

[0488] Data such as the message content, sender information, and scheduled sending time is transferred to the server.

[0489] Step 3:

[0490] The server temporarily stores the received message data in a database.

[0491] Temporarily stores message data and metadata so that the data can be used for subsequent processing in a secure and efficient manner.

[0492] Step 4:

[0493] The server runs a natural language processing engine to analyze the message.

[0494] Extract specific keywords and phrases from the text and then perform sentiment analysis.

[0495] Step 5:

[0496] The server scores the harassment risk based on the extracted data.

[0497] Each keyword, phrase, and emotion is evaluated to quantify the overall risk.

[0498] Step 6:

[0499] The server generates a warning message if the harassment score exceeds a certain threshold.

[0500] Create a warning message with specific suggested fixes to alert users.

[0501] Step 7:

[0502] A server-generated warning message is sent to the user terminal.

[0503] The original message is forwarded to the user's terminal along with the warning message.

[0504] Step 8:

[0505] The device will pop up a warning message to the user.

[0506] The app will show specific areas of harassment and suggest corrections, and encourage users to check and correct the content.

[0507] Step 9:

[0508] The user considers the warnings and suggestions and modifies the message.

[0509] Follow the suggested amendments or amend as you see fit.

[0510] Step 10:

[0511] The user clicks the send button again to resend the revised message.

[0512] A resend trigger occurs and the corrected message is sent to the server again.

[0513] Step 11:

[0514] The server performs a final check of the revised message and, if there are no problems, sends it to the recipient.

[0515] If the content is deemed to be acceptable, the message is officially sent to the recipient.

[0516] This process allows users to see how their message will affect the recipient before sending it, helping to prevent harassment and inappropriate messages.

[0517] Example 1

[0518] 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."

[0519] In modern communication tools such as email, chat, and social networking sites, messages sent by users can sometimes be overly aggressive or sent at inappropriate times. This can cause the recipient to feel uncomfortable or become a victim of harassment. Moreover, overly aggressive messages or messages sent at inappropriate times can disrupt smooth communication. Therefore, a system that can prevent these problems from occurring is needed.

[0520] 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.

[0521] In this invention, the server includes means for receiving text data to be sent, means for analyzing the received text data using a natural language processing engine and extracting specific keywords and emotions, means for analyzing the time of sending and identifying inappropriate messages based on the time of day, means for scoring the harassment risk level based on the extracted data, means for generating a warning message when the scoring result exceeds a certain threshold, and means for sending and displaying the warning message to the user terminal. This makes it possible to analyze in advance the content and timing of messages that users intend to send and prevent the sending of inappropriate messages.

[0522] The "means for receiving transmitted character data" is a mechanism for receiving character information transmitted by a user.

[0523] A "natural language processing engine" is software or a service that analyzes text data and extracts keywords and emotions.

[0524] "Means for extracting specific keywords and emotions" refers to a function that recognizes and acquires important words and emotional trends from received text data.

[0525] The "means for scoring the risk of harassment" is a process that quantifies the possibility of harassment based on extracted keywords and emotions.

[0526] The "means for generating a warning message" is a mechanism for creating text to warn the user based on the scoring results.

[0527] "Means for sending and displaying a warning message on a user terminal" refers to a function for delivering the generated warning message to the user's device and displaying it there.

[0528] "Means for analyzing sending times and identifying inappropriate messages based on time of day" refers to a mechanism that evaluates the time a message was sent and determines whether the content is inappropriate for that time period.

[0529] A "warning message with specific suggestions for correction" is a warning message that includes specific advice for correcting a message that has been determined to be inappropriate.

[0530] An "interface that allows a user to apply suggested corrections to a message and then resend the message" is a screen or feature that allows a user to apply suggested corrections to a message and then resend the message with the corrections.

[0531] MODE FOR CARRYING OUT THE INVENTION

[0532] The present invention is a system that analyzes messages a user is about to send in advance to detect potential harassment and inappropriate sending times. The system consists of a user terminal and a server. When a user inputs and sends a message, the message is sent to the server and analyzed using a natural language processing engine. Based on the analysis results, the system scores the risk of harassment and generates a warning message if necessary.

[0533] User Device

[0534] The user terminal provides an interface where the user can enter a message into a text entry form. When the user enters a message and clicks the send button, the terminal sends the message and associated metadata (such as the time of sending and sender information) to the server.

[0535] server

[0536] The server has the following functions:

[0537] 1. Receiving function

[0538] The server receives the text data sent from the user terminal and temporarily stores it in a database, which ensures the consistency and availability of the data.

[0539] 2. Analysis Function

[0540] The server analyzes the received message using a natural language processing engine (e.g., Google Cloud Natural Language API), extracting keywords and sentiments from the message and obtaining the data.

[0541] 3. Scoring function

[0542] The server scores the risk of harassment based on the extracted keywords and emotional data. It also analyzes the time of message sending and identifies inappropriate messages based on the time of day. If the scoring results exceed a certain threshold, a warning message is generated.

[0543] 4. Warning generation function

[0544] If the scoring result exceeds a threshold, the server generates a warning message including specific suggestions for correction, and then sends the warning message to the user terminal.

[0545] 5. Display function

[0546] The warning message sent from the server is displayed on the user's terminal, and the user can check and modify the warning message.

[0547] With the above functions, the system of the present invention can analyze messages that users are about to send in advance, point out potential harassment and inappropriate sending times, and prompt users to correct the messages, thereby supporting safe and smooth communication.

[0548] Specific examples

[0549] Example 1: Sending offensive messages

[0550] 1. The user types, "You're so incompetent." and clicks the send button.

[0551] 2. The device sends this message to the server.

[0552] 3. The server temporarily stores the received message in a database.

[0553] 4. The server uses the Google Cloud Natural Language API to extract the keyword "incompetent" and negative sentiment.

[0554] 5. The server scores the harassment risk and determines that the score exceeds the threshold.

[0555] 6. The server generates a warning message stating, "This message may be considered offensive. We recommend that you correct it." and sends it to the user's device.

[0556] 7. The device displays a warning message to the user and prompts the user to correct the message.

[0557] Example 2: Inappropriate sending time

[0558] 1. At 2 a.m., a user types, "Please submit the materials immediately." and clicks the submit button.

[0559] 2. The device sends the message and the time of sending to the server.

[0560] 3. The server temporarily stores the received message in a database.

[0561] 4. The server analyzes the sending time and determines that it is late at night.

[0562] 5. The server generates a warning message stating, "This message will be sent late at night and may be offensive to the recipient. Please change the sending time or modify the message content." and sends it to the user's terminal.

[0563] 6. The device will display a warning message to the user, prompting them to correct the sending time and message content.

[0564] Prompt Sentence Examples

[0565] "Please use a natural language processing engine to evaluate this message to determine whether it is likely to be harassment. The message content is as follows."

[0566] This system gives users the opportunity to correct inappropriate messages before they are sent, thereby promoting safe and smooth communication.

[0567] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0568] Step 1:

[0569] The user enters a message into the text entry form and clicks the send button.

[0570] Input: Text data entered by the user

[0571] Output: Text data and transmission metadata are temporarily saved on the device.

[0572] Step 2:

[0573] The device sends the message data and its metadata (e.g., sending time, sender information) to the server.

[0574] Input: User-entered text data and metadata

[0575] Output: An HTTP POST request is sent to the server

[0576] Step 3:

[0577] The message data received by the server is temporarily stored in a database.

[0578] Input: Message data and metadata received from the device.

[0579] Output: Message data and metadata stored in a database

[0580] Step 4:

[0581] The server uses a natural language processing engine (e.g., Google Cloud Natural Language API) to analyze the stored messages.

[0582] Input: Message data in the database

[0583] Output: Analysis results (keywords and sentiment data) returned by the natural language processing engine

[0584] Step 5:

[0585] The server then assigns a score to the message based on the analysis results, assessing the risk of harassment. It also analyzes the time the message was sent and identifies inappropriate messages based on the time of day.

[0586] Input: Analysis results of the natural language processing engine, message sending time

[0587] Output: Harassment score, evaluation result based on sending time

[0588] Step 6:

[0589] If the server scores above a certain threshold, it generates a warning message, such as "This message may be considered offensive. We recommend that you correct it."

[0590] Input: Harassment score, evaluation result based on submission time

[0591] Output: Warning message

[0592] Step 7:

[0593] A server-generated warning message is sent to the user terminal.

[0594] Input: warning message

[0595] Output: HTTP response to the user's device

[0596] Step 8:

[0597] The device will display a warning message to the user, allowing the user to review and modify the message.

[0598] Input: Warning message from the server

[0599] Output: A warning message that is displayed in the user interface.

[0600] The above is a description of the specific processing steps of this system.

[0601] (Application example 1)

[0602] 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."

[0603] In modern communication, especially in the workplace or when messaging between employees in brick-and-mortar stores, there is a risk of inappropriate language or harassment. Furthermore, sending messages at inappropriate times, such as late at night, can cause discomfort to the recipient. This raises concerns about a worsening work environment and reduced work efficiency. There is a need to resolve this issue and maintain healthy and comfortable communication.

[0604] 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.

[0605] In this invention, the server includes means for receiving transmitted text data, means for analyzing the received text data using a natural language processing engine and extracting specific keywords and emotions, means for scoring the harassment risk level based on the extracted data, means for generating a warning message when the scoring result exceeds a certain threshold, means for transmitting and displaying the warning message on the user terminal, and means for analyzing the time of message transmission and generating the warning message if the transmission time is inappropriate. This makes it possible to monitor both the message content and transmission time and prevent inappropriate communication before it occurs.

[0606] "Text data to be sent" refers to the text information that a user inputs as a message and intends to send.

[0607] "Means for receiving" refers to the method or mechanism by which the server receives character data sent from the user terminal.

[0608] A "natural language processing engine" is a software engine that analyzes text data and extracts meaning, emotion, and specific keywords.

[0609] "Means for analyzing and extracting" refers to methods and mechanisms for extracting specific keywords and emotional information from text data using a natural language processing engine.

[0610] "Means for scoring the risk of harassment" refers to a method or mechanism for numerically assessing the risk of harassment in text data based on extracted keywords and emotional information.

[0611] The "means for generating a warning message" refers to a method or mechanism for creating a message to warn the user based on the scoring result.

[0612] "Means for sending and displaying a warning message on a user terminal" refers to a method or mechanism for sending a generated warning message to a user's device and displaying it to the user.

[0613] The "means for analyzing the sending time of a message" refers to a method or mechanism for analyzing the time information of a message to be sent and determining whether the time is appropriate.

[0614] "Means for generating a warning message when a time is inappropriate for sending a message" refers to a method or mechanism for generating a warning message to notify the user when a time for sending a message is determined to be inappropriate.

[0615] The present invention is a system that analyzes transmitted text data, evaluates the risk of harassment, and checks the appropriateness of the transmission time. Specific embodiments will be described below.

[0616] System Configuration

[0617] (user device)

[0618] The user terminal is a device that provides an interface for inputting messages. When a user inputs a message and clicks the send button, the text data and the time of sending are sent to the server.

[0619] (server)

[0620] The server analyzes the received text data using a natural language processing engine to extract specific keywords and emotions from the text. It also scores the harassment risk based on the extracted data and generates a warning message if the scoring result exceeds a certain threshold. In addition, the server analyzes the time the message was sent and generates a separate warning message if it was sent at an inappropriate time. The generated warning message is sent to the user's device and displayed.

[0621] Hardware and software used

[0622] (Hardware)

[0623] Smartphone: Used as a user device.

[0624] (software)

[0625] Python: The language used for server-side programming.

[0626] textblob: A library used as a natural language processing engine.

[0627] Specific processing and data calculations

[0628] The server analyzes messages sent by users using a natural language processing engine (here we use textblob). This process extracts emotions and specific keywords (such as "incompetent," "stupid," and "dislike") from the text. The extracted data is then used to score the harassment risk. If the score exceeds a certain threshold, the server generates a warning message and sends it to the user's device.

[0629] The server also analyzes the time the message was sent and generates a separate warning message if it was sent at an inappropriate time, such as 2 a.m. This allows users to check not only the content of the message but also the time it was sent, preventing inappropriate communication.

[0630] Specific examples

[0631] For example, if a user types "Please submit the documents now" at 2 a.m. and clicks the send button, the system will behave as follows:

[0632] 1. The user types and sends a message.

[0633] 2. The device sends this message and the sending time to the server.

[0634] 3. The server receives the message and the time it was sent and parses it using a natural language processing engine.

[0635] 4. The server scores the harassment risk and also checks the time of transmission.

[0636] 5. Generate a warning message if it is determined to be inappropriate based on sentiment, keywords, and sending time.

[0637] 6. The server generates a warning message and sends it to the user's device.

[0638] 7. The device displays this warning message to the user.

[0639] Example prompt sentence:

[0640] "Analyze the following messages and generate a warning message if they contain harassing or inappropriate content.

[0641] Message: 'You're really incompetent.'

[0642] Sent time: '02:00'"

[0643] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0644] Step 1:

[0645] The user enters a message and clicks the send button. At this time, the message data and the current sending time are obtained. This is the user's input.

[0646] Step 2:

[0647] The device sends the message data and its metadata (such as the sending time) to the server. In this process, the device sends the message and sending time to the server as data. The input is the message data and sending time, and the output is that these data are sent to the server.

[0648] Step 3:

[0649] The server temporarily stores the received message data and sending time in a database. The input is the message data and sending time sent from the terminal, and the output is that these data are stored in the database.

[0650] Step 4:

[0651] The server analyzes the message using a natural language processing engine (textblob). During this analysis, specific keywords (such as "incompetent," "stupid," or "hate") and emotions (positive or negative) are extracted from the message. The input is message data retrieved from the database, and the output is the analyzed keywords and emotional information.

[0652] Step 5:

[0653] The server scores the harassment risk based on the extracted keywords and emotional information. This scoring is calculated based on a certain algorithm (for example, the number of occurrences of specific keywords or the intensity of emotions). The input is the keywords and emotional information from the analysis results, and the output is the calculated score.

[0654] Step 6:

[0655] If the score exceeds a certain threshold, the server generates a warning message informing the user that the message may be offensive and encouraging them to correct it. The input is the calculated score, and the output is the generated warning message.

[0656] Step 7:

[0657] The server analyzes the sending time. It checks whether the sending time is an inappropriate time, such as 2 a.m. The input is the sending time retrieved from the database, and the output is the evaluation result of the appropriateness of the time (appropriate / inappropriate).

[0658] Step 8:

[0659] If the server determines that the sending time is inappropriate, it generates a separate warning message. This warning message notifies the user that the sending time is inappropriate. The input is the result of the time appropriateness evaluation, and the output is the generated warning message.

[0660] Step 9:

[0661] The server sends the generated warning message to the user terminal. The input is the generated warning message, and the output is the message sent to the user terminal.

[0662] Step 10:

[0663] The terminal displays a warning message to the user, allowing the user to receive feedback on the appropriateness of the message content and the time it was sent. The input is the warning message sent from the server, and the output is the display of the warning message to the user.

[0664] In this way, the system can properly monitor messages sent by users and prevent problems caused by the risk of harassment or inappropriate sending times.

[0665] 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.

[0666] This invention is a system that checks the content and sender's emotions of email, chat, and social networking messages before they are sent, and detects inappropriate content or possible harassment. This system consists of a user terminal, a server, and an emotion engine, and operates as follows:

[0667] System Configuration

[0668] 1. User Device

[0669] The user terminal provides an interface for inputting text data and includes an emotion engine for acquiring the user's emotion data. When the user inputs a message and clicks the send button, the text data and emotion data are transmitted to the server.

[0670] 2. Server

[0671] The server analyzes the received text data and emotional data to assess the risk of harassment. Specifically, it uses a natural language processing engine to extract keywords and emotions from the message and performs a scoring process based on these. It also integrates the user's emotional data received from the emotion engine to perform the evaluation. If the scoring result exceeds a certain threshold, it generates a warning message and sends it to the user's device.

[0672] Program processing overview

[0673] When a user clicks the send button on a message, the device sends the message data and emotion data to the server. The server receives this data and temporarily stores it in a database. The server then analyzes the message using a natural language processing engine and emotion engine to extract keywords, emotions, and the user's emotion data. This extracted data is used to score the harassment risk. If the score exceeds a certain threshold, the server generates a warning message and sends it to the user's device. The device then displays this warning message to the user. The user can then refer to the suggested corrections to revise the message and resend it.

[0674] Specific examples

[0675] Example 1: Sending offensive messages and making users angry

[0676] 1. A user types "You're so incompetent" and clicks the send button. At this time, the emotion engine detects that the user is very angry.

[0677] 2. The device sends this message and emotion data to the server.

[0678] 3. The server receives the message and emotion data and temporarily stores them in a database.

[0679] 4. The server uses a natural language processing engine to analyze the keyword "incompetent" and the overall negative sentiment, and integrates the user's anger data from the sentiment engine.

[0680] 5. The server calculates the harassment score and determines that the score is high.

[0681] 6. The server generates a warning message stating, "This message may be considered offensive. We recommend that you correct it." and sends it to the user's device.

[0682] 7. The device displays this warning message to the user.

[0683] 8. The user corrects and resubmits, saying, "Your performance could use some improvement. Let's review the details."

[0684] Example 2: Inappropriate send times and user fatigue

[0685] 1. When a user types "Please submit the materials immediately" at 2 a.m. and clicks the send button, the emotion engine detects that the user is tired.

[0686] 2. The device sends this message and emotion data to the server.

[0687] 3. The server receives the message and emotion data and temporarily stores them in a database.

[0688] 4. The server analyzes the sending time, message content, and emotional data to determine that the message was sent at an inappropriate time.

[0689] 5. The server generates a warning message stating, "This message will be sent late at night and may be offensive to the recipient. Please change the sending time or modify the message content." and sends it to the user's terminal.

[0690] 6. The device displays this warning message to the user.

[0691] 7. The user changes the sending time or corrects the message content and resends it.

[0692] In this way, this system supports safe and smooth communication by analyzing the message the user is about to send and their emotions at the time, pointing out any potential harassment and encouraging them to correct it.By taking the user's emotions into consideration, this system can provide more appropriate warnings and suggestions for correction, preventing harassment and inappropriate messages before they occur.

[0693] The processing flow will be explained below.

[0694] Step 1:

[0695] The user types a message and clicks the send button.

[0696] The user inputs a message on the user interface and presses the send button to trigger a transmission. The emotion engine obtains emotional data from the user's facial expressions, voice, etc.

[0697] Step 2:

[0698] The terminal transmits the input message data and emotion data to the server.

[0699] Metadata such as message content, sender information, emotion data, and scheduled sending time are transferred to the server.

[0700] Step 3:

[0701] The server temporarily stores the received message data and emotion data in a database.

[0702] Various data is temporarily stored in a database so that it can be used efficiently in subsequent processing.

[0703] Step 4:

[0704] The server analyzes the message using a natural language processing engine and an emotion engine.

[0705] It extracts specific keywords and phrases from text and also performs sentiment analysis, simultaneously analyzing user emotional data.

[0706] Step 5:

[0707] The server scores the harassment risk based on the extracted data.

[0708] It evaluates each keyword and phrase, the overall sentiment of the text, and user sentiment data to quantify the overall risk.

[0709] Step 6:

[0710] The server generates a warning message if the harassment score exceeds a certain threshold.

[0711] Create a warning message with specific suggestions for correction based on the harassment score.

[0712] Step 7:

[0713] A server-generated warning message is sent to the user terminal.

[0714] Along with the warning message, the original message and additional information based on emotion data are also transmitted to the user's device.

[0715] Step 8:

[0716] The device will pop up a warning message to the user.

[0717] The app will display specific areas of harassment, suggested corrections, and additional information that takes into account the user's feelings, encouraging the user to review and correct the content.

[0718] Step 9:

[0719] The user considers the warnings and suggestions and modifies the message.

[0720] Follow the suggested changes or make your own changes as you see fit, including taking into account the user's emotional feedback.

[0721] Step 10:

[0722] The user clicks the send button again to resend the revised message.

[0723] A resend trigger occurs and the corrected message is sent to the server again.

[0724] Step 11:

[0725] The server performs a final check of the revised message and, if there are no problems, sends it to the recipient.

[0726] If the corrections are deemed acceptable, the message will be officially sent to the recipient.

[0727] This series of processes allows users to check the impression their message will make on the recipient before sending it, preventing harassment and inappropriate messages. By combining it with an emotion engine, it is possible to take user emotions into account and provide more appropriate warnings and suggestions for correction.

[0728] Example 2

[0729] 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."

[0730] Messages sent via online email, chat, social media, etc. can sometimes contain inappropriate content or elements of harassment, causing trouble or upsetting experiences for some users. While there is a need for a system that can detect inappropriate messages in advance, warn senders, and encourage them to correct their messages, current systems often do not provide appropriate evaluations or warnings that fully consider user feelings. Furthermore, few systems offer specific suggestions for correction, leaving users unsure of how to correct their messages.

[0731] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: means for receiving transmitted text data and user emotion data; means for temporarily storing the received text data and emotion data in a database; means for analyzing the received text data using a natural language processing engine and extracting specific keywords and emotions; means for analyzing the received emotion data using an emotion engine and evaluating the user's emotion; means for scoring the harassment risk level based on the analysis results of the natural language processing engine and the emotion engine; means for generating a warning message when the scoring result exceeds a certain threshold; means for sending and displaying the warning message on the user terminal; and means for providing an interface that allows the user to refer to the suggested corrections and resend the message. This prevents the sending of inappropriate messages and provides the user with appropriate correction suggestions, enabling safe and smooth communication.

[0732] "Text data to be sent" refers to the text information entered by the user to be sent via email, chat, social media, etc.

[0733] "Emotion data" refers to data that indicates the user's emotional state and is information that is analyzed by the emotion engine.

[0734] "Means for temporarily storing in a database" refers to a storage device within the system for temporarily storing received character data and emotion data.

[0735] A "natural language processing engine" refers to software that analyzes text data and extracts specific keywords and emotions.

[0736] "Emotion engine" refers to software that analyzes emotional data and assesses a user's emotional state.

[0737] "Harassment risk" refers to a score that indicates the likelihood that a message constitutes harassment based on the text data and emotional data sent.

[0738] "Scoring means" refers to algorithms or programs that calculate the risk of harassment based on the analysis results of the natural language processing engine and emotion engine.

[0739] "Means for generating a warning message" refers to software or a system that automatically creates a message to warn the user when the scoring result exceeds a threshold.

[0740] "Means for sending and displaying on a user terminal" refers to a system for sending the generated warning message to a user's device and displaying it in the form of a notification or pop-up, etc.

[0741] "Means for providing an interface that allows a user to modify and resubmit a message based on the suggested modifications" refers to a system that provides a user interface that allows a user to edit and resubmit a message based on the warning message.

[0742] MODE FOR CARRYING OUT THE INVENTION

[0743] This invention is a system that analyzes the content and sender's emotions before sending messages via email, chat, social media, etc., to detect inappropriate content or potential harassment. This system consists of a user terminal, a server, and an emotion engine. Each component and its operation are described in detail below.

[0744] User Device

[0745] The user terminal provides an interface for inputting text data and includes an emotion engine for acquiring the user's emotion data. When the user inputs a message and clicks the send button, the text data and emotion data are transmitted to the server.

[0746] server

[0747] The server analyzes the received text data and emotion data to assess the risk of harassment. Specifically, it performs the following steps:

[0748] 1. Data Receipt and Storage:

[0749] The server receives the character data and emotion data sent from the user terminal.

[0750] The server temporarily stores the received data in a database.

[0751] 2. Data Analysis:

[0752] The server extracts keywords and sentiment from the message using a natural language processing engine, which performs text analysis of the message content to detect offensive keywords such as "incompetent" and the negative sentiment of the entire sentence.

[0753] The server further analyzes the received emotion data using an emotion engine to assess the user's emotional state (e.g., anger, fatigue).

[0754] 3. Scoring:

[0755] The server combines the analysis results from the natural language processing engine and the emotion engine to score the harassment risk. If offensive language and negative emotions coexist, a higher harassment score is assigned.

[0756] 4. Warning message generation:

[0757] If the server finds that the scoring result exceeds a certain threshold, it generates a warning message and prompts the user to modify the message content. For example, it generates a warning message such as, "This message may be considered offensive. We recommend that you modify it."

[0758] 5. Sending a message:

[0759] A server-generated warning message is sent to the user terminal.

[0760] Message display and editing

[0761] The device will display the received warning message to the user, allowing the user to review the warning message and correct it based on the specific correction suggestions.

[0762] If the user reflects the proposed revisions and resubmits, smoother communication is ensured.

[0763] Specific examples

[0764] Example 1: Sending offensive messages and making users angry

[0765] When a user types "You're so incompetent," and clicks the send button, the emotion engine detects that the user is very angry.

[0766] The device sends this message and emotion data to the server.

[0767] The server receives the message and emotion data and temporarily stores them in a database.

[0768] The server uses a natural language processing engine to analyze the keyword "incompetent" and overall negative sentiment, and integrates user anger data from the sentiment engine.

[0769] The server calculates the harassment score and determines that the score is high.

[0770] The server generates a warning message stating, "This message may be considered offensive. We recommend that you correct it." and sends it to the user's device.

[0771] The device will display this warning message to the user.

[0772] The user corrects the error by saying, "Your performance could use some improvement. Let's review the details." and resubmits.

[0773] Example 2: Inappropriate send times and user fatigue

[0774] If a user types "Please submit the documents immediately" at 2 a.m. and clicks the send button, the emotion engine will detect that the user is tired.

[0775] The device sends this message and emotion data to the server.

[0776] The server receives the message and emotion data and temporarily stores them in a database.

[0777] The server analyzes the sending time, message content, and emotional data to determine that the message was sent at an inappropriate time.

[0778] The server generates a warning message stating, "This message will be sent late at night and may be offensive to the recipient. Please change the sending time or modify the message content." and sends it to the user's terminal.

[0779] The device will display this warning message to the user.

[0780] The user can change the send time or modify the message content and resend it.

[0781] In this way, the system supports safe and smooth communication by analyzing the content of the message the user is about to send and their emotions at the time, pointing out any potential harassment and encouraging them to correct it. The warning message also includes specific suggestions for how to improve the message.

[0782] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0783] Step 1:

[0784] The user enters a message and clicks the send button. The user enters the text data they wish to send (e.g., "You're really incompetent."), and emotional data (e.g., angry state) is acquired at this time. The input data consists of a string of characters and emotional information.

[0785] Step 2:

[0786] The device acquires the text data and emotion data and sends them to the server. Specifically, when the user clicks the send button, the device sends both the text data (e.g., "You're really incompetent.") and emotion data (e.g., anger) to the server in a data format such as JSON.

[0787] Step 3:

[0788] The server receives the character data and emotion data and temporarily stores them in a database. Specifically, the server stores the received data in the database so that it can be used for subsequent analysis. The input is the received data, and the output is temporary storage in the database.

[0789] Step 4:

[0790] The server analyzes the text data using a natural language processing engine. Specifically, the server retrieves the text data from the database and uses the NLP engine to analyze keywords (e.g., "incompetent") and overall sentiment. The input is the text data, and the output is the analysis results.

[0791] Step 5:

[0792] The server uses an emotion engine to analyze the emotion data and evaluate the user's emotional state. Specifically, the emotion engine is used to analyze the user's emotion data in more detail, for example, to quantify the level of anger. The input is emotion data, and the output is an emotion evaluation.

[0793] Step 6:

[0794] The server integrates the results obtained from the natural language processing engine and the emotion engine to generate a score for the harassment risk. Specifically, for example, a high harassment score is generated based on the keyword "incompetent" and the user's emotion of "anger." The inputs are the analysis results and emotion assessment, and the output is a harassment score.

[0795] Step 7:

[0796] If the server scores above a certain threshold, it generates a warning message stating, "This message may be considered offensive. We recommend that you correct it." The input is the harassment score, and the output is the warning message.

[0797] Step 8:

[0798] The server sends a warning message to the user terminal. The input is the warning message and the output is the message sent to the user terminal.

[0799] Step 9:

[0800] The device displays a warning message to the user, specifically, by displaying the warning message in the user interface as a popup or notification. The input is the received warning message, and the output is the screen display.

[0801] Step 10:

[0802] The user reviews the warning message and modifies it based on the suggested corrections. For example, the user modifies the message to avoid offensive language and to say, "Your performance has room for improvement. Please review the details." The input is the warning message, and the output is the modified message.

[0803] Step 11:

[0804] The user resubmits the modified message. This time, the same processing flow is followed again to reevaluate whether the resubmitted message is appropriate. The input is the modified message, and the output is the final message transmission.

[0805] (Application example 2)

[0806] 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."

[0807] In customer support for electronic payment services, if the messages sent by the staff are inappropriate or offensive, the quality of the service may decline and customer satisfaction may be impaired. The present invention aims to solve this problem and provide a system for improving the quality of customer support and the customer experience.

[0808] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving transmitted character data, means for analyzing the received character data using a natural language processing engine and extracting specific keywords and emotions, and means for scoring the harassment risk based on the extracted data. This makes it possible to monitor customer support messages in electronic payment services and improve hospitality and customer experience.

[0809] A "user terminal" is a device for inputting and transmitting text data, and includes smartphones, personal computers, etc.

[0810] A "natural language processing engine" is software that analyzes text data and extracts specific keywords and emotions.

[0811] "Harassment risk" is a numerical representation of the likelihood that a particular message is inappropriate or offensive.

[0812] A "warning message" is a message that is generated to warn users when the risk of harassment exceeds a certain threshold.

[0813] An "electronic payment service" is a system for electronically paying for goods and services online or offline.

[0814] "Customer support" refers to services that respond to customer inquiries and problems.

[0815] "Hospitality" is a concept that refers to the quality of hospitality and treatment shown to customers.

[0816] "Customer experience" refers to the overall satisfaction and impression a customer has when using a service.

[0817] "Interface" refers to the screen and operating means that users use to interact with a system.

[0818] The present invention provides a system for monitoring customer support messages for electronic payment services to improve hospitality and customer experience. The system includes a user terminal, a server, an emotion engine, and a natural language processing engine. Specific embodiments for implementing the present invention are described below.

[0819] System Configuration

[0820] 1. User Device

[0821] The user terminal is a device for inputting and sending text data. This can be a smartphone or a PC. When the user inputs a support message and clicks the send button, the message and emotional data are sent to the server.

[0822] 2. Server

[0823] The server analyzes the received text data and emotion data to evaluate the degree of harassment risk. Specifically, the process is as follows:

[0824] Natural language processing engine: Analyzes text data and extracts specific keywords and sentiment (example of software used: spaCy).

[0825] Emotion engine: Analyzes the user's emotional data and extracts their emotional state (example of software used: IBM Watson NLP).

[0826] Scoring engine: Scores the level of harassment risk based on extracted keywords and sentiment data.

[0827] 3. Generating and sending warning messages

[0828] If the scoring result exceeds a certain threshold, the server generates a warning message and sends it to the user's device, including specific suggestions for correction.

[0829] Specific examples

[0830] 1. Inappropriate messages from support agents

[0831] 1. The support representative types "This problem is on your side" and clicks the send button.

[0832] 2. The emotion engine detects when a person is feeling stressed.

[0833] 3. The server receives the message and sentiment data and analyzes it for keywords like "the cause is on the user's side" and negative sentiment.

[0834] 4. Calculate the harassment score and determine it to be high.

[0835] 5. A warning message will be generated stating, "This message may be considered offensive. We recommend that you fix it."

[0836] 6. A warning message is sent to the user's terminal and displayed.

[0837] 7. The support person will fix it by saying, "Some of the problem is on the user's side, but we will deal with it."

[0838] Prompt Sentence Examples

[0839] The system analyzes the content and sentiment data of messages sent by customer support staff of electronic payment services, automatically detecting inappropriate content or potential harassment. If the message is potentially offensive or misleading, a warning message is generated and displayed. Example message: "The cause of this problem lies with the user."

[0840] This system will enable customer support for electronic payment services to be more professional and hospitable.

[0841] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0842] Step 1:

[0843] The user enters a message and clicks the send button. The input is text data, and emotion data is also acquired at the same time. By sending, the user device sends this data to the server.

[0844] Step 2:

[0845] The server receives the text data and emotion data from the user terminal and stores the input text data and emotion data in a temporary storage device for analysis.

[0846] Step 3:

[0847] The server uses a natural language processing engine to analyze the received text data. The input is text data, and keywords and the overall emotional state are extracted. Specifically, keyword extraction is performed using the spaCy library.

[0848] Step 4:

[0849] The server analyzes the user's emotional data using an emotion engine. The input is the emotional data, and IBM Watson NLP is used to extract the emotional state. The output is the user's specific emotional state.

[0850] Step 5:

[0851] The server integrates the data obtained from the natural language processing engine and the emotion engine. Specifically, the scoring engine evaluates the harassment risk based on the extracted keywords and emotional state, and calculates a harassment score. The input is keywords and emotion data, and the output is a harassment score.

[0852] Step 6:

[0853] The scoring engine determines if the score exceeds a certain threshold. If so, the server generates a warning message. The input is the harassment score, and the output is the warning message. The generated warning message includes specific suggestions for correction.

[0854] Step 7:

[0855] The server sends the generated warning message to the user terminal, which displays the message and notifies the user. The input is the warning message, and the output is the displayed warning message.

[0856] Step 8:

[0857] The user checks the presented warning message, corrects the message if necessary, and resends it. Based on the warning message, the user applies the specific correction suggestions, enters a new message, and presses the resend button. The input is the corrected text data, and the output is the resent message.

[0858] This series of steps enables customer support for electronic payment services to automatically evaluate the appropriateness of sent messages and prompt necessary corrections to provide a better customer experience.

[0859] 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.

[0860] 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.

[0861] 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.

[0862] [Third embodiment]

[0863] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0864] 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.

[0865] 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).

[0866] 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.

[0867] 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.

[0868] 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).

[0869] 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.

[0870] 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.

[0871] 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.

[0872] 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.

[0873] 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.

[0874] 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."

[0875] This invention is a system that checks the content of emails, chats, and social networking messages before they are sent, and detects inappropriate content or potential harassment. This system consists of a user terminal and a server, and operates as follows.

[0876] System Configuration

[0877] 1. User Device

[0878] The user terminal provides an interface for inputting character data. When the user inputs a message and clicks the send button, the character data is sent to the server.

[0879] 2. Server

[0880] The server analyzes the received text data and evaluates the risk of harassment. Specifically, it uses a natural language processing engine to extract keywords and emotions from the message and performs a scoring process based on these. If the scoring result exceeds a certain threshold, it generates a warning message and sends it to the user's device.

[0881] Program processing overview

[0882] When a user clicks the send button on a message, the device sends the message data and its metadata to the server. The server receives this data and temporarily stores it in a database. The server then analyzes the message using a natural language processing engine to extract keywords and emotions. This extracted data is used to score the harassment risk. If the score exceeds a certain threshold, the server generates a warning message and sends it to the user's device. The device then displays the warning message to the user. The user can then refer to the suggested corrections to revise the message and resend it.

[0883] Specific examples

[0884] Example 1: Sending offensive messages

[0885] 1. The user types, "You're so incompetent." and clicks the send button.

[0886] 2. The device sends this message to the server.

[0887] 3. The server receives the message and stores it in a database.

[0888] 4. The server uses a natural language processing engine to analyze the keyword "incompetent" and the overall negative sentiment.

[0889] 5. The server calculates the harassment score and determines that the score is high.

[0890] 6. The server generates a warning message stating, "This message may be considered offensive. We recommend that you correct it." and sends it to the user's device.

[0891] 7. The device displays this warning message to the user.

[0892] 8. The user corrects and resubmits, saying, "Your performance could use some improvement. Let's review the details."

[0893] Example 2: Inappropriate sending time

[0894] 1. At 2 a.m., a user types, "Please submit the materials immediately." and clicks the submit button.

[0895] 2. The device sends this message to the server.

[0896] 3. The server receives the message and stores it in a database.

[0897] 4. The server analyzes the time of sending and the message content and determines that it is an inappropriate time.

[0898] 5. The server generates a warning message stating, "This message will be sent late at night and may be offensive to the recipient. Please change the sending time or modify the message content." and sends it to the user's terminal.

[0899] 6. The device displays this warning message to the user.

[0900] 7. The user changes the sending time or corrects the message content and resends it.

[0901] In this way, the system supports safe and smooth communication by analyzing messages that users are about to send in advance, pointing out any potential harassment, and encouraging users to correct the messages.

[0902] The processing flow will be explained below.

[0903] Step 1:

[0904] The user types a message and clicks the send button.

[0905] A send trigger occurs when you enter a message on the user interface and press the send button.

[0906] Step 2:

[0907] The terminal transmits the input message data and its metadata to the server.

[0908] Data such as the message content, sender information, and scheduled sending time is transferred to the server.

[0909] Step 3:

[0910] The server temporarily stores the received message data in a database.

[0911] Temporarily stores message data and metadata so that the data can be used for subsequent processing in a secure and efficient manner.

[0912] Step 4:

[0913] The server runs a natural language processing engine to analyze the message.

[0914] Extract specific keywords and phrases from the text and then perform sentiment analysis.

[0915] Step 5:

[0916] The server scores the harassment risk based on the extracted data.

[0917] Each keyword, phrase, and emotion is evaluated to quantify the overall risk.

[0918] Step 6:

[0919] The server generates a warning message if the harassment score exceeds a certain threshold.

[0920] Create a warning message with specific suggested fixes to alert users.

[0921] Step 7:

[0922] A server-generated warning message is sent to the user terminal.

[0923] The original message is forwarded to the user's terminal along with the warning message.

[0924] Step 8:

[0925] The device will pop up a warning message to the user.

[0926] The app will show specific areas of harassment and suggest corrections, and encourage users to check and correct the content.

[0927] Step 9:

[0928] The user considers the warnings and suggestions and modifies the message.

[0929] Follow the suggested amendments or amend as you see fit.

[0930] Step 10:

[0931] The user clicks the send button again to resend the revised message.

[0932] A resend trigger occurs and the corrected message is sent to the server again.

[0933] Step 11:

[0934] The server performs a final check of the revised message and, if there are no problems, sends it to the recipient.

[0935] If the content is deemed to be acceptable, the message is officially sent to the recipient.

[0936] This process allows users to see how their message will affect the recipient before sending it, helping to prevent harassment and inappropriate messages.

[0937] Example 1

[0938] 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."

[0939] In modern communication tools such as email, chat, and social networking sites, messages sent by users can sometimes be overly aggressive or sent at inappropriate times. This can cause the recipient to feel uncomfortable or become a victim of harassment. Moreover, overly aggressive messages or messages sent at inappropriate times can disrupt smooth communication. Therefore, a system that can prevent these problems from occurring is needed.

[0940] 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.

[0941] In this invention, the server includes means for receiving text data to be sent, means for analyzing the received text data using a natural language processing engine and extracting specific keywords and emotions, means for analyzing the time of sending and identifying inappropriate messages based on the time of day, means for scoring the harassment risk level based on the extracted data, means for generating a warning message when the scoring result exceeds a certain threshold, and means for sending and displaying the warning message to the user terminal. This makes it possible to analyze in advance the content and timing of messages that users intend to send and prevent the sending of inappropriate messages.

[0942] The "means for receiving transmitted character data" is a mechanism for receiving character information transmitted by a user.

[0943] A "natural language processing engine" is software or a service that analyzes text data and extracts keywords and emotions.

[0944] "Means for extracting specific keywords and emotions" refers to a function that recognizes and acquires important words and emotional trends from received text data.

[0945] The "means for scoring the risk of harassment" is a process that quantifies the possibility of harassment based on extracted keywords and emotions.

[0946] The "means for generating a warning message" is a mechanism for creating text to warn the user based on the scoring results.

[0947] "Means for sending and displaying a warning message on a user terminal" refers to a function for delivering the generated warning message to the user's device and displaying it there.

[0948] "Means for analyzing sending times and identifying inappropriate messages based on time of day" refers to a mechanism that evaluates the time a message was sent and determines whether the content is inappropriate for that time period.

[0949] A "warning message with specific suggestions for correction" is a warning message that includes specific advice for correcting a message that has been determined to be inappropriate.

[0950] An "interface that allows a user to apply suggested corrections to a message and then resend the message" is a screen or feature that allows a user to apply suggested corrections to a message and then resend the message with the corrections.

[0951] MODE FOR CARRYING OUT THE INVENTION

[0952] The present invention is a system that analyzes messages a user is about to send in advance to detect potential harassment and inappropriate sending times. The system consists of a user terminal and a server. When a user inputs and sends a message, the message is sent to the server and analyzed using a natural language processing engine. Based on the analysis results, the system scores the risk of harassment and generates a warning message if necessary.

[0953] User Device

[0954] The user terminal provides an interface where the user can enter a message into a text entry form. When the user enters a message and clicks the send button, the terminal sends the message and associated metadata (such as the time of sending and sender information) to the server.

[0955] server

[0956] The server has the following functions:

[0957] 1. Receiving function

[0958] The server receives the text data sent from the user terminal and temporarily stores it in a database, which ensures the consistency and availability of the data.

[0959] 2. Analysis Function

[0960] The server analyzes the received message using a natural language processing engine (e.g., Google Cloud Natural Language API), extracting keywords and sentiments from the message and obtaining the data.

[0961] 3. Scoring function

[0962] The server scores the risk of harassment based on the extracted keywords and emotional data. It also analyzes the time of message sending and identifies inappropriate messages based on the time of day. If the scoring results exceed a certain threshold, a warning message is generated.

[0963] 4. Warning generation function

[0964] If the scoring result exceeds a threshold, the server generates a warning message including specific suggestions for correction, and then sends the warning message to the user terminal.

[0965] 5. Display function

[0966] The warning message sent from the server is displayed on the user's terminal, and the user can check and modify the warning message.

[0967] With the above functions, the system of the present invention can analyze messages that users are about to send in advance, point out potential harassment and inappropriate sending times, and prompt users to correct the messages, thereby supporting safe and smooth communication.

[0968] Specific examples

[0969] Example 1: Sending offensive messages

[0970] 1. The user types, "You're so incompetent." and clicks the send button.

[0971] 2. The device sends this message to the server.

[0972] 3. The server temporarily stores the received message in a database.

[0973] 4. The server uses the Google Cloud Natural Language API to extract the keyword "incompetent" and negative sentiment.

[0974] 5. The server scores the harassment risk and determines that the score exceeds the threshold.

[0975] 6. The server generates a warning message stating, "This message may be considered offensive. We recommend that you correct it." and sends it to the user's device.

[0976] 7. The device displays a warning message to the user and prompts the user to correct the message.

[0977] Example 2: Inappropriate sending time

[0978] 1. At 2 a.m., a user types, "Please submit the materials immediately." and clicks the submit button.

[0979] 2. The device sends the message and the time of sending to the server.

[0980] 3. The server temporarily stores the received message in a database.

[0981] 4. The server analyzes the sending time and determines that it is late at night.

[0982] 5. The server generates a warning message stating, "This message will be sent late at night and may be offensive to the recipient. Please change the sending time or modify the message content." and sends it to the user's terminal.

[0983] 6. The device will display a warning message to the user, prompting them to correct the sending time and message content.

[0984] Prompt Sentence Examples

[0985] "Please use a natural language processing engine to evaluate this message to determine whether it is likely to be harassment. The message content is as follows."

[0986] This system gives users the opportunity to correct inappropriate messages before they are sent, thereby promoting safe and smooth communication.

[0987] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0988] Step 1:

[0989] The user enters a message into the text entry form and clicks the send button.

[0990] Input: Text data entered by the user

[0991] Output: Text data and transmission metadata are temporarily saved on the device.

[0992] Step 2:

[0993] The device sends the message data and its metadata (e.g., sending time, sender information) to the server.

[0994] Input: User-entered text data and metadata

[0995] Output: An HTTP POST request is sent to the server

[0996] Step 3:

[0997] The message data received by the server is temporarily stored in a database.

[0998] Input: Message data and metadata received from the device.

[0999] Output: Message data and metadata stored in a database

[1000] Step 4:

[1001] The server uses a natural language processing engine (e.g., Google Cloud Natural Language API) to analyze the stored messages.

[1002] Input: Message data in the database

[1003] Output: Analysis results (keywords and sentiment data) returned by the natural language processing engine

[1004] Step 5:

[1005] The server then assigns a score to the message based on the analysis results, assessing the risk of harassment. It also analyzes the time the message was sent and identifies inappropriate messages based on the time of day.

[1006] Input: Analysis results of the natural language processing engine, message sending time

[1007] Output: Harassment score, evaluation result based on sending time

[1008] Step 6:

[1009] If the server scores above a certain threshold, it generates a warning message, such as "This message may be considered offensive. We recommend that you correct it."

[1010] Input: Harassment score, evaluation result based on submission time

[1011] Output: Warning message

[1012] Step 7:

[1013] A server-generated warning message is sent to the user terminal.

[1014] Input: warning message

[1015] Output: HTTP response to the user's device

[1016] Step 8:

[1017] The device will display a warning message to the user, allowing the user to review and modify the message.

[1018] Input: Warning message from the server

[1019] Output: A warning message that is displayed in the user interface.

[1020] The above is a description of the specific processing steps of this system.

[1021] (Application example 1)

[1022] 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."

[1023] In modern communication, especially in the workplace or when messaging between employees in brick-and-mortar stores, there is a risk of inappropriate language or harassment. Furthermore, sending messages at inappropriate times, such as late at night, can cause discomfort to the recipient. This raises concerns about a worsening work environment and reduced work efficiency. There is a need to resolve this issue and maintain healthy and comfortable communication.

[1024] 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.

[1025] In this invention, the server includes means for receiving transmitted text data, means for analyzing the received text data using a natural language processing engine and extracting specific keywords and emotions, means for scoring the harassment risk level based on the extracted data, means for generating a warning message when the scoring result exceeds a certain threshold, means for transmitting and displaying the warning message on the user terminal, and means for analyzing the time of message transmission and generating the warning message if the transmission time is inappropriate. This makes it possible to monitor both the message content and transmission time and prevent inappropriate communication before it occurs.

[1026] "Text data to be sent" refers to the text information that a user inputs as a message and intends to send.

[1027] "Means for receiving" refers to the method or mechanism by which the server receives character data sent from the user terminal.

[1028] A "natural language processing engine" is a software engine that analyzes text data and extracts meaning, emotion, and specific keywords.

[1029] "Means for analyzing and extracting" refers to methods and mechanisms for extracting specific keywords and emotional information from text data using a natural language processing engine.

[1030] "Means for scoring the risk of harassment" refers to a method or mechanism for numerically assessing the risk of harassment in text data based on extracted keywords and emotional information.

[1031] The "means for generating a warning message" refers to a method or mechanism for creating a message to warn the user based on the scoring result.

[1032] "Means for sending and displaying a warning message on a user terminal" refers to a method or mechanism for sending a generated warning message to a user's device and displaying it to the user.

[1033] The "means for analyzing the sending time of a message" refers to a method or mechanism for analyzing the time information of a message to be sent and determining whether the time is appropriate.

[1034] "Means for generating a warning message when a time is inappropriate for sending a message" refers to a method or mechanism for generating a warning message to notify the user when a time for sending a message is determined to be inappropriate.

[1035] The present invention is a system that analyzes transmitted text data, evaluates the risk of harassment, and checks the appropriateness of the transmission time. Specific embodiments will be described below.

[1036] System Configuration

[1037] (user device)

[1038] The user terminal is a device that provides an interface for inputting messages. When a user inputs a message and clicks the send button, the text data and the time of sending are sent to the server.

[1039] (server)

[1040] The server analyzes the received text data using a natural language processing engine to extract specific keywords and emotions from the text. It also scores the harassment risk based on the extracted data and generates a warning message if the scoring result exceeds a certain threshold. In addition, the server analyzes the time the message was sent and generates a separate warning message if it was sent at an inappropriate time. The generated warning message is sent to the user's device and displayed.

[1041] Hardware and software used

[1042] (Hardware)

[1043] Smartphone: Used as a user device.

[1044] (software)

[1045] Python: The language used for server-side programming.

[1046] textblob: A library used as a natural language processing engine.

[1047] Specific processing and data calculations

[1048] The server analyzes messages sent by users using a natural language processing engine (here we use textblob). This process extracts emotions and specific keywords (such as "incompetent," "stupid," and "dislike") from the text. The extracted data is then used to score the harassment risk. If the score exceeds a certain threshold, the server generates a warning message and sends it to the user's device.

[1049] The server also analyzes the time the message was sent and generates a separate warning message if it was sent at an inappropriate time, such as 2 a.m. This allows users to check not only the content of the message but also the time it was sent, preventing inappropriate communication.

[1050] Specific examples

[1051] For example, if a user types "Please submit the documents now" at 2 a.m. and clicks the send button, the system will behave as follows:

[1052] 1. The user types and sends a message.

[1053] 2. The device sends this message and the sending time to the server.

[1054] 3. The server receives the message and the time it was sent and parses it using a natural language processing engine.

[1055] 4. The server scores the harassment risk and also checks the time of transmission.

[1056] 5. Generate a warning message if it is determined to be inappropriate based on sentiment, keywords, and sending time.

[1057] 6. The server generates a warning message and sends it to the user's device.

[1058] 7. The device displays this warning message to the user.

[1059] Example prompt sentence:

[1060] "Analyze the following messages and generate a warning message if they contain harassing or inappropriate content.

[1061] Message: 'You're really incompetent.'

[1062] Sent time: '02:00'"

[1063] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1064] Step 1:

[1065] The user enters a message and clicks the send button. At this time, the message data and the current sending time are obtained. This is the user's input.

[1066] Step 2:

[1067] The device sends the message data and its metadata (such as the sending time) to the server. In this process, the device sends the message and sending time to the server as data. The input is the message data and sending time, and the output is that these data are sent to the server.

[1068] Step 3:

[1069] The server temporarily stores the received message data and sending time in a database. The input is the message data and sending time sent from the terminal, and the output is that these data are stored in the database.

[1070] Step 4:

[1071] The server analyzes the message using a natural language processing engine (textblob). During this analysis, specific keywords (such as "incompetent," "stupid," or "hate") and emotions (positive or negative) are extracted from the message. The input is message data retrieved from the database, and the output is the analyzed keywords and emotional information.

[1072] Step 5:

[1073] The server scores the harassment risk based on the extracted keywords and emotional information. This scoring is calculated based on a certain algorithm (for example, the number of occurrences of specific keywords or the intensity of emotions). The input is the keywords and emotional information from the analysis results, and the output is the calculated score.

[1074] Step 6:

[1075] If the score exceeds a certain threshold, the server generates a warning message informing the user that the message may be offensive and encouraging them to correct it. The input is the calculated score, and the output is the generated warning message.

[1076] Step 7:

[1077] The server analyzes the sending time. It checks whether the sending time is an inappropriate time, such as 2 a.m. The input is the sending time retrieved from the database, and the output is the evaluation result of the appropriateness of the time (appropriate / inappropriate).

[1078] Step 8:

[1079] If the server determines that the sending time is inappropriate, it generates a separate warning message. This warning message notifies the user that the sending time is inappropriate. The input is the result of the time appropriateness evaluation, and the output is the generated warning message.

[1080] Step 9:

[1081] The server sends the generated warning message to the user terminal. The input is the generated warning message, and the output is the message sent to the user terminal.

[1082] Step 10:

[1083] The terminal displays a warning message to the user, allowing the user to receive feedback on the appropriateness of the message content and the time it was sent. The input is the warning message sent from the server, and the output is the display of the warning message to the user.

[1084] In this way, the system can properly monitor messages sent by users and prevent problems caused by the risk of harassment or inappropriate sending times.

[1085] 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.

[1086] This invention is a system that checks the content and sender's emotions of email, chat, and social networking messages before they are sent, and detects inappropriate content or possible harassment. This system consists of a user terminal, a server, and an emotion engine, and operates as follows:

[1087] System Configuration

[1088] 1. User Device

[1089] The user terminal provides an interface for inputting text data and includes an emotion engine for acquiring the user's emotion data. When the user inputs a message and clicks the send button, the text data and emotion data are transmitted to the server.

[1090] 2. Server

[1091] The server analyzes the received text data and emotional data to assess the risk of harassment. Specifically, it uses a natural language processing engine to extract keywords and emotions from the message and performs a scoring process based on these. It also integrates the user's emotional data received from the emotion engine to perform the evaluation. If the scoring result exceeds a certain threshold, it generates a warning message and sends it to the user's device.

[1092] Program processing overview

[1093] When a user clicks the send button on a message, the device sends the message data and emotion data to the server. The server receives this data and temporarily stores it in a database. The server then analyzes the message using a natural language processing engine and emotion engine to extract keywords, emotions, and the user's emotion data. This extracted data is used to score the harassment risk. If the score exceeds a certain threshold, the server generates a warning message and sends it to the user's device. The device then displays this warning message to the user. The user can then refer to the suggested corrections to revise the message and resend it.

[1094] Specific examples

[1095] Example 1: Sending offensive messages and making users angry

[1096] 1. A user types "You're so incompetent" and clicks the send button. At this time, the emotion engine detects that the user is very angry.

[1097] 2. The device sends this message and emotion data to the server.

[1098] 3. The server receives the message and emotion data and temporarily stores them in a database.

[1099] 4. The server uses a natural language processing engine to analyze the keyword "incompetent" and the overall negative sentiment, and integrates the user's anger data from the sentiment engine.

[1100] 5. The server calculates the harassment score and determines that the score is high.

[1101] 6. The server generates a warning message stating, "This message may be considered offensive. We recommend that you correct it." and sends it to the user's device.

[1102] 7. The device displays this warning message to the user.

[1103] 8. The user corrects and resubmits, saying, "Your performance could use some improvement. Let's review the details."

[1104] Example 2: Inappropriate send times and user fatigue

[1105] 1. When a user types "Please submit the materials immediately" at 2 a.m. and clicks the send button, the emotion engine detects that the user is tired.

[1106] 2. The device sends this message and emotion data to the server.

[1107] 3. The server receives the message and emotion data and temporarily stores them in a database.

[1108] 4. The server analyzes the sending time, message content, and emotional data to determine that the message was sent at an inappropriate time.

[1109] 5. The server generates a warning message stating, "This message will be sent late at night and may be offensive to the recipient. Please change the sending time or modify the message content." and sends it to the user's terminal.

[1110] 6. The device displays this warning message to the user.

[1111] 7. The user changes the sending time or corrects the message content and resends it.

[1112] In this way, this system supports safe and smooth communication by analyzing the message the user is about to send and their emotions at the time, pointing out any potential harassment and encouraging them to correct it.By taking the user's emotions into consideration, this system can provide more appropriate warnings and suggestions for correction, preventing harassment and inappropriate messages before they occur.

[1113] The processing flow will be explained below.

[1114] Step 1:

[1115] The user types a message and clicks the send button.

[1116] The user inputs a message on the user interface and presses the send button to trigger a transmission. The emotion engine obtains emotional data from the user's facial expressions, voice, etc.

[1117] Step 2:

[1118] The terminal transmits the input message data and emotion data to the server.

[1119] Metadata such as message content, sender information, emotion data, and scheduled sending time are transferred to the server.

[1120] Step 3:

[1121] The server temporarily stores the received message data and emotion data in a database.

[1122] Various data is temporarily stored in a database so that it can be used efficiently in subsequent processing.

[1123] Step 4:

[1124] The server analyzes the message using a natural language processing engine and an emotion engine.

[1125] It extracts specific keywords and phrases from text and also performs sentiment analysis, simultaneously analyzing user emotional data.

[1126] Step 5:

[1127] The server scores the harassment risk based on the extracted data.

[1128] It evaluates each keyword and phrase, the overall sentiment of the text, and user sentiment data to quantify the overall risk.

[1129] Step 6:

[1130] The server generates a warning message if the harassment score exceeds a certain threshold.

[1131] Create a warning message with specific suggestions for correction based on the harassment score.

[1132] Step 7:

[1133] A server-generated warning message is sent to the user terminal.

[1134] Along with the warning message, the original message and additional information based on emotion data are also transmitted to the user's device.

[1135] Step 8:

[1136] The device will pop up a warning message to the user.

[1137] The app will display specific areas of harassment, suggested corrections, and additional information that takes into account the user's feelings, encouraging the user to review and correct the content.

[1138] Step 9:

[1139] The user considers the warnings and suggestions and modifies the message.

[1140] Follow the suggested changes or make your own changes as you see fit, including taking into account the user's emotional feedback.

[1141] Step 10:

[1142] The user clicks the send button again to resend the revised message.

[1143] A resend trigger occurs and the corrected message is sent to the server again.

[1144] Step 11:

[1145] The server performs a final check of the revised message and, if there are no problems, sends it to the recipient.

[1146] If the corrections are deemed acceptable, the message will be officially sent to the recipient.

[1147] This series of processes allows users to check the impression their message will make on the recipient before sending it, preventing harassment and inappropriate messages. By combining it with an emotion engine, it is possible to take user emotions into account and provide more appropriate warnings and suggestions for correction.

[1148] Example 2

[1149] 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."

[1150] Messages sent via online email, chat, social media, etc. can sometimes contain inappropriate content or elements of harassment, causing trouble or upsetting experiences for some users. While there is a need for a system that can detect inappropriate messages in advance, warn senders, and encourage them to correct their messages, current systems often do not provide appropriate evaluations or warnings that fully consider user feelings. Furthermore, few systems offer specific suggestions for correction, leaving users unsure of how to correct their messages.

[1151] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: means for receiving transmitted text data and user emotion data; means for temporarily storing the received text data and emotion data in a database; means for analyzing the received text data using a natural language processing engine and extracting specific keywords and emotions; means for analyzing the received emotion data using an emotion engine and evaluating the user's emotion; means for scoring the harassment risk level based on the analysis results of the natural language processing engine and the emotion engine; means for generating a warning message when the scoring result exceeds a certain threshold; means for sending and displaying the warning message on the user terminal; and means for providing an interface that allows the user to refer to the suggested corrections and resend the message. This prevents the sending of inappropriate messages and provides the user with appropriate correction suggestions, enabling safe and smooth communication.

[1152] "Text data to be sent" refers to the text information entered by the user to be sent via email, chat, social media, etc.

[1153] "Emotion data" refers to data that indicates the user's emotional state and is information that is analyzed by the emotion engine.

[1154] "Means for temporarily storing in a database" refers to a storage device within the system for temporarily storing received character data and emotion data.

[1155] A "natural language processing engine" refers to software that analyzes text data and extracts specific keywords and emotions.

[1156] "Emotion engine" refers to software that analyzes emotional data and assesses a user's emotional state.

[1157] "Harassment risk" refers to a score that indicates the likelihood that a message constitutes harassment based on the text data and emotional data sent.

[1158] "Scoring means" refers to algorithms or programs that calculate the risk of harassment based on the analysis results of the natural language processing engine and emotion engine.

[1159] "Means for generating a warning message" refers to software or a system that automatically creates a message to warn the user when the scoring result exceeds a threshold.

[1160] "Means for sending and displaying on a user terminal" refers to a system for sending the generated warning message to a user's device and displaying it in the form of a notification or pop-up, etc.

[1161] "Means for providing an interface that allows a user to modify and resubmit a message based on the suggested modifications" refers to a system that provides a user interface that allows a user to edit and resubmit a message based on the warning message.

[1162] MODE FOR CARRYING OUT THE INVENTION

[1163] This invention is a system that analyzes the content and sender's emotions before sending messages via email, chat, social media, etc., to detect inappropriate content or potential harassment. This system consists of a user terminal, a server, and an emotion engine. Each component and its operation are described in detail below.

[1164] User Device

[1165] The user terminal provides an interface for inputting text data and includes an emotion engine for acquiring the user's emotion data. When the user inputs a message and clicks the send button, the text data and emotion data are transmitted to the server.

[1166] server

[1167] The server analyzes the received text data and emotion data to assess the risk of harassment. Specifically, it performs the following steps:

[1168] 1. Data Receipt and Storage:

[1169] The server receives the character data and emotion data sent from the user terminal.

[1170] The server temporarily stores the received data in a database.

[1171] 2. Data Analysis:

[1172] The server extracts keywords and sentiment from the message using a natural language processing engine, which performs text analysis of the message content to detect offensive keywords such as "incompetent" and the negative sentiment of the entire sentence.

[1173] The server further analyzes the received emotion data using an emotion engine to assess the user's emotional state (e.g., anger, fatigue).

[1174] 3. Scoring:

[1175] The server combines the analysis results from the natural language processing engine and the emotion engine to score the harassment risk. If offensive language and negative emotions coexist, a higher harassment score is assigned.

[1176] 4. Warning message generation:

[1177] If the server finds that the scoring result exceeds a certain threshold, it generates a warning message and prompts the user to modify the message content. For example, it generates a warning message such as, "This message may be considered offensive. We recommend that you modify it."

[1178] 5. Sending a message:

[1179] A server-generated warning message is sent to the user terminal.

[1180] Message display and editing

[1181] The device will display the received warning message to the user, allowing the user to review the warning message and correct it based on the specific correction suggestions.

[1182] If the user reflects the proposed revisions and resubmits, smoother communication is ensured.

[1183] Specific examples

[1184] Example 1: Sending offensive messages and making users angry

[1185] When a user types "You're so incompetent," and clicks the send button, the emotion engine detects that the user is very angry.

[1186] The device sends this message and emotion data to the server.

[1187] The server receives the message and emotion data and temporarily stores them in a database.

[1188] The server uses a natural language processing engine to analyze the keyword "incompetent" and overall negative sentiment, and integrates user anger data from the sentiment engine.

[1189] The server calculates the harassment score and determines that the score is high.

[1190] The server generates a warning message stating, "This message may be considered offensive. We recommend that you correct it." and sends it to the user's device.

[1191] The device will display this warning message to the user.

[1192] The user corrects the error by saying, "Your performance could use some improvement. Let's review the details." and resubmits.

[1193] Example 2: Inappropriate send times and user fatigue

[1194] If a user types "Please submit the documents immediately" at 2 a.m. and clicks the send button, the emotion engine will detect that the user is tired.

[1195] The device sends this message and emotion data to the server.

[1196] The server receives the message and emotion data and temporarily stores them in a database.

[1197] The server analyzes the sending time, message content, and emotional data to determine that the message was sent at an inappropriate time.

[1198] The server generates a warning message stating, "This message will be sent late at night and may be offensive to the recipient. Please change the sending time or modify the message content." and sends it to the user's terminal.

[1199] The device will display this warning message to the user.

[1200] The user can change the send time or modify the message content and resend it.

[1201] In this way, the system supports safe and smooth communication by analyzing the content of the message the user is about to send and their emotions at the time, pointing out any potential harassment and encouraging them to correct it. The warning message also includes specific suggestions for how to improve the message.

[1202] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1203] Step 1:

[1204] The user enters a message and clicks the send button. The user enters the text data they wish to send (e.g., "You're really incompetent."), and emotional data (e.g., angry state) is acquired at this time. The input data consists of a string of characters and emotional information.

[1205] Step 2:

[1206] The device acquires the text data and emotion data and sends them to the server. Specifically, when the user clicks the send button, the device sends both the text data (e.g., "You're really incompetent.") and emotion data (e.g., anger) to the server in a data format such as JSON.

[1207] Step 3:

[1208] The server receives the character data and emotion data and temporarily stores them in a database. Specifically, the server stores the received data in the database so that it can be used for subsequent analysis. The input is the received data, and the output is temporary storage in the database.

[1209] Step 4:

[1210] The server analyzes the text data using a natural language processing engine. Specifically, the server retrieves the text data from the database and uses the NLP engine to analyze keywords (e.g., "incompetent") and overall sentiment. The input is the text data, and the output is the analysis results.

[1211] Step 5:

[1212] The server uses an emotion engine to analyze the emotion data and evaluate the user's emotional state. Specifically, the emotion engine is used to analyze the user's emotion data in more detail, for example, to quantify the level of anger. The input is emotion data, and the output is an emotion evaluation.

[1213] Step 6:

[1214] The server integrates the results obtained from the natural language processing engine and the emotion engine to generate a score for the harassment risk. Specifically, for example, a high harassment score is generated based on the keyword "incompetent" and the user's emotion of "anger." The inputs are the analysis results and emotion assessment, and the output is a harassment score.

[1215] Step 7:

[1216] If the server scores above a certain threshold, it generates a warning message stating, "This message may be considered offensive. We recommend that you correct it." The input is the harassment score, and the output is the warning message.

[1217] Step 8:

[1218] The server sends a warning message to the user terminal. The input is the warning message and the output is the message sent to the user terminal.

[1219] Step 9:

[1220] The device displays a warning message to the user, specifically, by displaying the warning message in the user interface as a popup or notification. The input is the received warning message, and the output is the screen display.

[1221] Step 10:

[1222] The user reviews the warning message and modifies it based on the suggested corrections. For example, the user modifies the message to avoid offensive language and to say, "Your performance has room for improvement. Please review the details." The input is the warning message, and the output is the modified message.

[1223] Step 11:

[1224] The user resubmits the modified message. This time, the same processing flow is followed again to reevaluate whether the resubmitted message is appropriate. The input is the modified message, and the output is the final message transmission.

[1225] (Application example 2)

[1226] 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."

[1227] In customer support for electronic payment services, if the messages sent by the staff are inappropriate or offensive, the quality of the service may decline and customer satisfaction may be impaired. The present invention aims to solve this problem and provide a system for improving the quality of customer support and the customer experience.

[1228] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving transmitted character data, means for analyzing the received character data using a natural language processing engine and extracting specific keywords and emotions, and means for scoring the harassment risk based on the extracted data. This makes it possible to monitor customer support messages in electronic payment services and improve hospitality and customer experience.

[1229] A "user terminal" is a device for inputting and transmitting text data, and includes smartphones, personal computers, etc.

[1230] A "natural language processing engine" is software that analyzes text data and extracts specific keywords and emotions.

[1231] "Harassment risk" is a numerical representation of the likelihood that a particular message is inappropriate or offensive.

[1232] A "warning message" is a message that is generated to warn users when the risk of harassment exceeds a certain threshold.

[1233] An "electronic payment service" is a system for electronically paying for goods and services online or offline.

[1234] "Customer support" refers to services that respond to customer inquiries and problems.

[1235] "Hospitality" is a concept that refers to the quality of hospitality and treatment shown to customers.

[1236] "Customer experience" refers to the overall satisfaction and impression a customer has when using a service.

[1237] "Interface" refers to the screen and operating means that users use to interact with a system.

[1238] The present invention provides a system for monitoring customer support messages for electronic payment services to improve hospitality and customer experience. The system includes a user terminal, a server, an emotion engine, and a natural language processing engine. Specific embodiments for implementing the present invention are described below.

[1239] System Configuration

[1240] 1. User Device

[1241] The user terminal is a device for inputting and sending text data. This can be a smartphone or a PC. When the user inputs a support message and clicks the send button, the message and emotional data are sent to the server.

[1242] 2. Server

[1243] The server analyzes the received text data and emotion data to evaluate the degree of harassment risk. Specifically, the process is as follows:

[1244] Natural language processing engine: Analyzes text data and extracts specific keywords and sentiment (example of software used: spaCy).

[1245] Emotion engine: Analyzes the user's emotional data and extracts their emotional state (example of software used: IBM Watson NLP).

[1246] Scoring engine: Scores the level of harassment risk based on extracted keywords and sentiment data.

[1247] 3. Generating and sending warning messages

[1248] If the scoring result exceeds a certain threshold, the server generates a warning message and sends it to the user's device, including specific suggestions for correction.

[1249] Specific examples

[1250] 1. Inappropriate messages from support agents

[1251] 1. The support representative types "This problem is on your side" and clicks the send button.

[1252] 2. The emotion engine detects when a person is feeling stressed.

[1253] 3. The server receives the message and sentiment data and analyzes it for keywords like "the cause is on the user's side" and negative sentiment.

[1254] 4. Calculate the harassment score and determine it to be high.

[1255] 5. A warning message will be generated stating, "This message may be considered offensive. We recommend that you fix it."

[1256] 6. A warning message is sent to the user's terminal and displayed.

[1257] 7. The support person will fix it by saying, "Some of the problem is on the user's side, but we will deal with it."

[1258] Prompt Sentence Examples

[1259] The system analyzes the content and sentiment data of messages sent by customer support staff of electronic payment services, automatically detecting inappropriate content or potential harassment. If the message is potentially offensive or misleading, a warning message is generated and displayed. Example message: "The cause of this problem lies with the user."

[1260] This system will enable customer support for electronic payment services to be more professional and hospitable.

[1261] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1262] Step 1:

[1263] The user enters a message and clicks the send button. The input is text data, and emotion data is also acquired at the same time. By sending, the user device sends this data to the server.

[1264] Step 2:

[1265] The server receives the text data and emotion data from the user terminal and stores the input text data and emotion data in a temporary storage device for analysis.

[1266] Step 3:

[1267] The server uses a natural language processing engine to analyze the received text data. The input is text data, and keywords and the overall emotional state are extracted. Specifically, keyword extraction is performed using the spaCy library.

[1268] Step 4:

[1269] The server analyzes the user's emotional data using an emotion engine. The input is the emotional data, and IBM Watson NLP is used to extract the emotional state. The output is the user's specific emotional state.

[1270] Step 5:

[1271] The server integrates the data obtained from the natural language processing engine and the emotion engine. Specifically, the scoring engine evaluates the harassment risk based on the extracted keywords and emotional state, and calculates a harassment score. The input is keywords and emotion data, and the output is a harassment score.

[1272] Step 6:

[1273] The scoring engine determines if the score exceeds a certain threshold. If so, the server generates a warning message. The input is the harassment score, and the output is the warning message. The generated warning message includes specific suggestions for correction.

[1274] Step 7:

[1275] The server sends the generated warning message to the user terminal, which displays the message and notifies the user. The input is the warning message, and the output is the displayed warning message.

[1276] Step 8:

[1277] The user checks the presented warning message, corrects the message if necessary, and resends it. Based on the warning message, the user applies the specific correction suggestions, enters a new message, and presses the resend button. The input is the corrected text data, and the output is the resent message.

[1278] This series of steps enables customer support for electronic payment services to automatically evaluate the appropriateness of sent messages and prompt necessary corrections to provide a better customer experience.

[1279] 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.

[1280] 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.

[1281] 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.

[1282] [Fourth embodiment]

[1283] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1284] 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.

[1285] 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).

[1286] 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.

[1287] 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.

[1288] 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).

[1289] 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.

[1290] 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.

[1291] 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.

[1292] 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.

[1293] 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.

[1294] 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.

[1295] 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."

[1296] This invention is a system that checks the content of emails, chats, and social networking messages before they are sent, and detects inappropriate content or potential harassment. This system consists of a user terminal and a server, and operates as follows.

[1297] System Configuration

[1298] 1. User Device

[1299] The user terminal provides an interface for inputting character data. When the user inputs a message and clicks the send button, the character data is sent to the server.

[1300] 2. Server

[1301] The server analyzes the received text data and evaluates the risk of harassment. Specifically, it uses a natural language processing engine to extract keywords and emotions from the message and performs a scoring process based on these. If the scoring result exceeds a certain threshold, it generates a warning message and sends it to the user's device.

[1302] Program processing overview

[1303] When a user clicks the send button on a message, the device sends the message data and its metadata to the server. The server receives this data and temporarily stores it in a database. The server then analyzes the message using a natural language processing engine to extract keywords and emotions. This extracted data is used to score the harassment risk. If the score exceeds a certain threshold, the server generates a warning message and sends it to the user's device. The device then displays the warning message to the user. The user can then refer to the suggested corrections to revise the message and resend it.

[1304] Specific examples

[1305] Example 1: Sending offensive messages

[1306] 1. The user types, "You're so incompetent." and clicks the send button.

[1307] 2. The device sends this message to the server.

[1308] 3. The server receives the message and stores it in a database.

[1309] 4. The server uses a natural language processing engine to analyze the keyword "incompetent" and the overall negative sentiment.

[1310] 5. The server calculates the harassment score and determines that the score is high.

[1311] 6. The server generates a warning message stating, "This message may be considered offensive. We recommend that you correct it." and sends it to the user's device.

[1312] 7. The device displays this warning message to the user.

[1313] 8. The user corrects and resubmits, saying, "Your performance could use some improvement. Let's review the details."

[1314] Example 2: Inappropriate sending time

[1315] 1. At 2 a.m., a user types, "Please submit the materials immediately." and clicks the submit button.

[1316] 2. The device sends this message to the server.

[1317] 3. The server receives the message and stores it in a database.

[1318] 4. The server analyzes the time of sending and the message content and determines that it is an inappropriate time.

[1319] 5. The server generates a warning message stating, "This message will be sent late at night and may be offensive to the recipient. Please change the sending time or modify the message content." and sends it to the user's terminal.

[1320] 6. The device displays this warning message to the user.

[1321] 7. The user changes the sending time or corrects the message content and resends it.

[1322] In this way, the system supports safe and smooth communication by analyzing messages that users are about to send in advance, pointing out any potential harassment, and encouraging users to correct the messages.

[1323] The processing flow will be explained below.

[1324] Step 1:

[1325] The user types a message and clicks the send button.

[1326] A send trigger occurs when you enter a message on the user interface and press the send button.

[1327] Step 2:

[1328] The terminal transmits the input message data and its metadata to the server.

[1329] Data such as the message content, sender information, and scheduled sending time is transferred to the server.

[1330] Step 3:

[1331] The server temporarily stores the received message data in a database.

[1332] Temporarily stores message data and metadata so that the data can be used for subsequent processing in a secure and efficient manner.

[1333] Step 4:

[1334] The server runs a natural language processing engine to analyze the message.

[1335] Extract specific keywords and phrases from the text and then perform sentiment analysis.

[1336] Step 5:

[1337] The server scores the harassment risk based on the extracted data.

[1338] Each keyword, phrase, and emotion is evaluated to quantify the overall risk.

[1339] Step 6:

[1340] The server generates a warning message if the harassment score exceeds a certain threshold.

[1341] Create a warning message with specific suggested fixes to alert users.

[1342] Step 7:

[1343] A server-generated warning message is sent to the user terminal.

[1344] The original message is forwarded to the user's terminal along with the warning message.

[1345] Step 8:

[1346] The device will pop up a warning message to the user.

[1347] The app will show specific areas of harassment and suggest corrections, and encourage users to check and correct the content.

[1348] Step 9:

[1349] The user considers the warnings and suggestions and modifies the message.

[1350] Follow the suggested amendments or amend as you see fit.

[1351] Step 10:

[1352] The user clicks the send button again to resend the revised message.

[1353] A resend trigger occurs and the corrected message is sent to the server again.

[1354] Step 11:

[1355] The server performs a final check of the revised message and, if there are no problems, sends it to the recipient.

[1356] If the content is deemed to be acceptable, the message is officially sent to the recipient.

[1357] This process allows users to see how their message will affect the recipient before sending it, helping to prevent harassment and inappropriate messages.

[1358] Example 1

[1359] 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."

[1360] In modern communication tools such as email, chat, and social networking sites, messages sent by users can sometimes be overly aggressive or sent at inappropriate times. This can cause the recipient to feel uncomfortable or become a victim of harassment. Moreover, overly aggressive messages or messages sent at inappropriate times can disrupt smooth communication. Therefore, a system that can prevent these problems from occurring is needed.

[1361] 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.

[1362] In this invention, the server includes means for receiving text data to be sent, means for analyzing the received text data using a natural language processing engine and extracting specific keywords and emotions, means for analyzing the time of sending and identifying inappropriate messages based on the time of day, means for scoring the harassment risk level based on the extracted data, means for generating a warning message when the scoring result exceeds a certain threshold, and means for sending and displaying the warning message to the user terminal. This makes it possible to analyze in advance the content and timing of messages that users intend to send and prevent the sending of inappropriate messages.

[1363] The "means for receiving transmitted character data" is a mechanism for receiving character information transmitted by a user.

[1364] A "natural language processing engine" is software or a service that analyzes text data and extracts keywords and emotions.

[1365] "Means for extracting specific keywords and emotions" refers to a function that recognizes and acquires important words and emotional trends from received text data.

[1366] The "means for scoring the risk of harassment" is a process that quantifies the possibility of harassment based on extracted keywords and emotions.

[1367] The "means for generating a warning message" is a mechanism for creating text to warn the user based on the scoring results.

[1368] "Means for sending and displaying a warning message on a user terminal" refers to a function for delivering the generated warning message to the user's device and displaying it there.

[1369] "Means for analyzing sending times and identifying inappropriate messages based on time of day" refers to a mechanism that evaluates the time a message was sent and determines whether the content is inappropriate for that time period.

[1370] A "warning message with specific suggestions for correction" is a warning message that includes specific advice for correcting a message that has been determined to be inappropriate.

[1371] An "interface that allows a user to apply suggested corrections to a message and then resend the message" is a screen or feature that allows a user to apply suggested corrections to a message and then resend the message with the corrections.

[1372] MODE FOR CARRYING OUT THE INVENTION

[1373] The present invention is a system that analyzes messages a user is about to send in advance to detect potential harassment and inappropriate sending times. The system consists of a user terminal and a server. When a user inputs and sends a message, the message is sent to the server and analyzed using a natural language processing engine. Based on the analysis results, the system scores the risk of harassment and generates a warning message if necessary.

[1374] User Device

[1375] The user terminal provides an interface where the user can enter a message into a text entry form. When the user enters a message and clicks the send button, the terminal sends the message and associated metadata (such as the time of sending and sender information) to the server.

[1376] server

[1377] The server has the following functions:

[1378] 1. Receiving function

[1379] The server receives the text data sent from the user terminal and temporarily stores it in a database, which ensures the consistency and availability of the data.

[1380] 2. Analysis Function

[1381] The server analyzes the received message using a natural language processing engine (e.g., Google Cloud Natural Language API), extracting keywords and sentiments from the message and obtaining the data.

[1382] 3. Scoring function

[1383] The server scores the risk of harassment based on the extracted keywords and emotional data. It also analyzes the time of message sending and identifies inappropriate messages based on the time of day. If the scoring results exceed a certain threshold, a warning message is generated.

[1384] 4. Warning generation function

[1385] If the scoring result exceeds a threshold, the server generates a warning message including specific suggestions for correction, and then sends the warning message to the user terminal.

[1386] 5. Display function

[1387] The warning message sent from the server is displayed on the user's terminal, and the user can check and modify the warning message.

[1388] With the above functions, the system of the present invention can analyze messages that users are about to send in advance, point out potential harassment and inappropriate sending times, and prompt users to correct the messages, thereby supporting safe and smooth communication.

[1389] Specific examples

[1390] Example 1: Sending offensive messages

[1391] 1. The user types, "You're so incompetent." and clicks the send button.

[1392] 2. The device sends this message to the server.

[1393] 3. The server temporarily stores the received message in a database.

[1394] 4. The server uses the Google Cloud Natural Language API to extract the keyword "incompetent" and negative sentiment.

[1395] 5. The server scores the harassment risk and determines that the score exceeds the threshold.

[1396] 6. The server generates a warning message stating, "This message may be considered offensive. We recommend that you correct it." and sends it to the user's device.

[1397] 7. The device displays a warning message to the user and prompts the user to correct the message.

[1398] Example 2: Inappropriate sending time

[1399] 1. At 2 a.m., a user types, "Please submit the materials immediately." and clicks the submit button.

[1400] 2. The device sends the message and the time of sending to the server.

[1401] 3. The server temporarily stores the received message in a database.

[1402] 4. The server analyzes the sending time and determines that it is late at night.

[1403] 5. The server generates a warning message stating, "This message will be sent late at night and may be offensive to the recipient. Please change the sending time or modify the message content." and sends it to the user's terminal.

[1404] 6. The device will display a warning message to the user, prompting them to correct the sending time and message content.

[1405] Prompt Sentence Examples

[1406] "Please use a natural language processing engine to evaluate this message to determine whether it is likely to be harassment. The message content is as follows."

[1407] This system gives users the opportunity to correct inappropriate messages before they are sent, thereby promoting safe and smooth communication.

[1408] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1409] Step 1:

[1410] The user enters a message into the text entry form and clicks the send button.

[1411] Input: Text data entered by the user

[1412] Output: Text data and transmission metadata are temporarily saved on the device.

[1413] Step 2:

[1414] The device sends the message data and its metadata (e.g., sending time, sender information) to the server.

[1415] Input: User-entered text data and metadata

[1416] Output: An HTTP POST request is sent to the server

[1417] Step 3:

[1418] The message data received by the server is temporarily stored in a database.

[1419] Input: Message data and metadata received from the device.

[1420] Output: Message data and metadata stored in a database

[1421] Step 4:

[1422] The server uses a natural language processing engine (e.g., Google Cloud Natural Language API) to analyze the stored messages.

[1423] Input: Message data in the database

[1424] Output: Analysis results (keywords and sentiment data) returned by the natural language processing engine

[1425] Step 5:

[1426] The server then assigns a score to the message based on the analysis results, assessing the risk of harassment. It also analyzes the time the message was sent and identifies inappropriate messages based on the time of day.

[1427] Input: Analysis results of the natural language processing engine, message sending time

[1428] Output: Harassment score, evaluation result based on sending time

[1429] Step 6:

[1430] If the server scores above a certain threshold, it generates a warning message, such as "This message may be considered offensive. We recommend that you correct it."

[1431] Input: Harassment score, evaluation result based on submission time

[1432] Output: Warning message

[1433] Step 7:

[1434] A server-generated warning message is sent to the user terminal.

[1435] Input: warning message

[1436] Output: HTTP response to the user's device

[1437] Step 8:

[1438] The device will display a warning message to the user, allowing the user to review and modify the message.

[1439] Input: Warning message from the server

[1440] Output: A warning message that is displayed in the user interface.

[1441] The above is a description of the specific processing steps of this system.

[1442] (Application example 1)

[1443] 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."

[1444] In modern communication, especially in the workplace or when messaging between employees in brick-and-mortar stores, there is a risk of inappropriate language or harassment. Furthermore, sending messages at inappropriate times, such as late at night, can cause discomfort to the recipient. This raises concerns about a worsening work environment and reduced work efficiency. There is a need to resolve this issue and maintain healthy and comfortable communication.

[1445] 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.

[1446] In this invention, the server includes means for receiving transmitted text data, means for analyzing the received text data using a natural language processing engine and extracting specific keywords and emotions, means for scoring the harassment risk level based on the extracted data, means for generating a warning message when the scoring result exceeds a certain threshold, means for transmitting and displaying the warning message on the user terminal, and means for analyzing the time of message transmission and generating the warning message if the transmission time is inappropriate. This makes it possible to monitor both the message content and transmission time and prevent inappropriate communication before it occurs.

[1447] "Text data to be sent" refers to the text information that a user inputs as a message and intends to send.

[1448] "Means for receiving" refers to the method or mechanism by which the server receives character data sent from the user terminal.

[1449] A "natural language processing engine" is a software engine that analyzes text data and extracts meaning, emotion, and specific keywords.

[1450] "Means for analyzing and extracting" refers to methods and mechanisms for extracting specific keywords and emotional information from text data using a natural language processing engine.

[1451] "Means for scoring the risk of harassment" refers to a method or mechanism for numerically assessing the risk of harassment in text data based on extracted keywords and emotional information.

[1452] The "means for generating a warning message" refers to a method or mechanism for creating a message to warn the user based on the scoring result.

[1453] "Means for sending and displaying a warning message on a user terminal" refers to a method or mechanism for sending a generated warning message to a user's device and displaying it to the user.

[1454] The "means for analyzing the sending time of a message" refers to a method or mechanism for analyzing the time information of a message to be sent and determining whether the time is appropriate.

[1455] "Means for generating a warning message when a time is inappropriate for sending a message" refers to a method or mechanism for generating a warning message to notify the user when a time for sending a message is determined to be inappropriate.

[1456] The present invention is a system that analyzes transmitted text data, evaluates the risk of harassment, and checks the appropriateness of the transmission time. Specific embodiments will be described below.

[1457] System Configuration

[1458] (user device)

[1459] The user terminal is a device that provides an interface for inputting messages. When a user inputs a message and clicks the send button, the text data and the time of sending are sent to the server.

[1460] (server)

[1461] The server analyzes the received text data using a natural language processing engine to extract specific keywords and emotions from the text. It also scores the harassment risk based on the extracted data and generates a warning message if the scoring result exceeds a certain threshold. In addition, the server analyzes the time the message was sent and generates a separate warning message if it was sent at an inappropriate time. The generated warning message is sent to the user's device and displayed.

[1462] Hardware and software used

[1463] (Hardware)

[1464] Smartphone: Used as a user device.

[1465] (software)

[1466] Python: The language used for server-side programming.

[1467] textblob: A library used as a natural language processing engine.

[1468] Specific processing and data calculations

[1469] The server analyzes messages sent by users using a natural language processing engine (here we use textblob). This process extracts emotions and specific keywords (such as "incompetent," "stupid," and "dislike") from the text. The extracted data is then used to score the harassment risk. If the score exceeds a certain threshold, the server generates a warning message and sends it to the user's device.

[1470] The server also analyzes the time the message was sent and generates a separate warning message if it was sent at an inappropriate time, such as 2 a.m. This allows users to check not only the content of the message but also the time it was sent, preventing inappropriate communication.

[1471] Specific examples

[1472] For example, if a user types "Please submit the documents now" at 2 a.m. and clicks the send button, the system will behave as follows:

[1473] 1. The user types and sends a message.

[1474] 2. The device sends this message and the sending time to the server.

[1475] 3. The server receives the message and the time it was sent and parses it using a natural language processing engine.

[1476] 4. The server scores the harassment risk and also checks the time of transmission.

[1477] 5. Generate a warning message if it is determined to be inappropriate based on sentiment, keywords, and sending time.

[1478] 6. The server generates a warning message and sends it to the user's device.

[1479] 7. The device displays this warning message to the user.

[1480] Example prompt sentence:

[1481] "Analyze the following messages and generate a warning message if they contain harassing or inappropriate content.

[1482] Message: 'You're really incompetent.'

[1483] Sent time: '02:00'"

[1484] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1485] Step 1:

[1486] The user enters a message and clicks the send button. At this time, the message data and the current sending time are obtained. This is the user's input.

[1487] Step 2:

[1488] The device sends the message data and its metadata (such as the sending time) to the server. In this process, the device sends the message and sending time to the server as data. The input is the message data and sending time, and the output is that these data are sent to the server.

[1489] Step 3:

[1490] The server temporarily stores the received message data and sending time in a database. The input is the message data and sending time sent from the terminal, and the output is that these data are stored in the database.

[1491] Step 4:

[1492] The server analyzes the message using a natural language processing engine (textblob). During this analysis, specific keywords (such as "incompetent," "stupid," or "hate") and emotions (positive or negative) are extracted from the message. The input is message data retrieved from the database, and the output is the analyzed keywords and emotional information.

[1493] Step 5:

[1494] The server scores the harassment risk based on the extracted keywords and emotional information. This scoring is calculated based on a certain algorithm (for example, the number of occurrences of specific keywords or the intensity of emotions). The input is the keywords and emotional information from the analysis results, and the output is the calculated score.

[1495] Step 6:

[1496] If the score exceeds a certain threshold, the server generates a warning message informing the user that the message may be offensive and encouraging them to correct it. The input is the calculated score, and the output is the generated warning message.

[1497] Step 7:

[1498] The server analyzes the sending time. It checks whether the sending time is an inappropriate time, such as 2 a.m. The input is the sending time retrieved from the database, and the output is the evaluation result of the appropriateness of the time (appropriate / inappropriate).

[1499] Step 8:

[1500] If the server determines that the sending time is inappropriate, it generates a separate warning message. This warning message notifies the user that the sending time is inappropriate. The input is the result of the time appropriateness evaluation, and the output is the generated warning message.

[1501] Step 9:

[1502] The server sends the generated warning message to the user terminal. The input is the generated warning message, and the output is the message sent to the user terminal.

[1503] Step 10:

[1504] The terminal displays a warning message to the user, allowing the user to receive feedback on the appropriateness of the message content and the time it was sent. The input is the warning message sent from the server, and the output is the display of the warning message to the user.

[1505] In this way, the system can properly monitor messages sent by users and prevent problems caused by the risk of harassment or inappropriate sending times.

[1506] 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.

[1507] This invention is a system that checks the content and sender's emotions of email, chat, and social networking messages before they are sent, and detects inappropriate content or possible harassment. This system consists of a user terminal, a server, and an emotion engine, and operates as follows:

[1508] System Configuration

[1509] 1. User Device

[1510] The user terminal provides an interface for inputting text data and includes an emotion engine for acquiring the user's emotion data. When the user inputs a message and clicks the send button, the text data and emotion data are transmitted to the server.

[1511] 2. Server

[1512] The server analyzes the received text data and emotional data to assess the risk of harassment. Specifically, it uses a natural language processing engine to extract keywords and emotions from the message and performs a scoring process based on these. It also integrates the user's emotional data received from the emotion engine to perform the evaluation. If the scoring result exceeds a certain threshold, it generates a warning message and sends it to the user's device.

[1513] Program processing overview

[1514] When a user clicks the send button on a message, the device sends the message data and emotion data to the server. The server receives this data and temporarily stores it in a database. The server then analyzes the message using a natural language processing engine and emotion engine to extract keywords, emotions, and the user's emotion data. This extracted data is used to score the harassment risk. If the score exceeds a certain threshold, the server generates a warning message and sends it to the user's device. The device then displays this warning message to the user. The user can then refer to the suggested corrections to revise the message and resend it.

[1515] Specific examples

[1516] Example 1: Sending offensive messages and making users angry

[1517] 1. A user types "You're so incompetent" and clicks the send button. At this time, the emotion engine detects that the user is very angry.

[1518] 2. The device sends this message and emotion data to the server.

[1519] 3. The server receives the message and emotion data and temporarily stores them in a database.

[1520] 4. The server uses a natural language processing engine to analyze the keyword "incompetent" and the overall negative sentiment, and integrates the user's anger data from the sentiment engine.

[1521] 5. The server calculates the harassment score and determines that the score is high.

[1522] 6. The server generates a warning message stating, "This message may be considered offensive. We recommend that you correct it." and sends it to the user's device.

[1523] 7. The device displays this warning message to the user.

[1524] 8. The user corrects and resubmits, saying, "Your performance could use some improvement. Let's review the details."

[1525] Example 2: Inappropriate send times and user fatigue

[1526] 1. When a user types "Please submit the materials immediately" at 2 a.m. and clicks the send button, the emotion engine detects that the user is tired.

[1527] 2. The device sends this message and emotion data to the server.

[1528] 3. The server receives the message and emotion data and temporarily stores them in a database.

[1529] 4. The server analyzes the sending time, message content, and emotional data to determine that the message was sent at an inappropriate time.

[1530] 5. The server generates a warning message stating, "This message will be sent late at night and may be offensive to the recipient. Please change the sending time or modify the message content." and sends it to the user's terminal.

[1531] 6. The device displays this warning message to the user.

[1532] 7. The user changes the sending time or corrects the message content and resends it.

[1533] In this way, this system supports safe and smooth communication by analyzing the message the user is about to send and their emotions at the time, pointing out any potential harassment and encouraging them to correct it.By taking the user's emotions into consideration, this system can provide more appropriate warnings and suggestions for correction, preventing harassment and inappropriate messages before they occur.

[1534] The processing flow will be explained below.

[1535] Step 1:

[1536] The user types a message and clicks the send button.

[1537] The user inputs a message on the user interface and presses the send button to trigger a transmission. The emotion engine obtains emotional data from the user's facial expressions, voice, etc.

[1538] Step 2:

[1539] The terminal transmits the input message data and emotion data to the server.

[1540] Metadata such as message content, sender information, emotion data, and scheduled sending time are transferred to the server.

[1541] Step 3:

[1542] The server temporarily stores the received message data and emotion data in a database.

[1543] Various data is temporarily stored in a database so that it can be used efficiently in subsequent processing.

[1544] Step 4:

[1545] The server analyzes the message using a natural language processing engine and an emotion engine.

[1546] It extracts specific keywords and phrases from text and also performs sentiment analysis, simultaneously analyzing user emotional data.

[1547] Step 5:

[1548] The server scores the harassment risk based on the extracted data.

[1549] It evaluates each keyword and phrase, the overall sentiment of the text, and user sentiment data to quantify the overall risk.

[1550] Step 6:

[1551] The server generates a warning message if the harassment score exceeds a certain threshold.

[1552] Create a warning message with specific suggestions for correction based on the harassment score.

[1553] Step 7:

[1554] A server-generated warning message is sent to the user terminal.

[1555] Along with the warning message, the original message and additional information based on emotion data are also transmitted to the user's device.

[1556] Step 8:

[1557] The device will pop up a warning message to the user.

[1558] The app will display specific areas of harassment, suggested corrections, and additional information that takes into account the user's feelings, encouraging the user to review and correct the content.

[1559] Step 9:

[1560] The user considers the warnings and suggestions and modifies the message.

[1561] Follow the suggested changes or make your own changes as you see fit, including taking into account the user's emotional feedback.

[1562] Step 10:

[1563] The user clicks the send button again to resend the revised message.

[1564] A resend trigger occurs and the corrected message is sent to the server again.

[1565] Step 11:

[1566] The server performs a final check of the revised message and, if there are no problems, sends it to the recipient.

[1567] If the corrections are deemed acceptable, the message will be officially sent to the recipient.

[1568] This series of processes allows users to check the impression their message will make on the recipient before sending it, preventing harassment and inappropriate messages. By combining it with an emotion engine, it is possible to take user emotions into account and provide more appropriate warnings and suggestions for correction.

[1569] Example 2

[1570] 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."

[1571] Messages sent via online email, chat, social media, etc. can sometimes contain inappropriate content or elements of harassment, causing trouble or upsetting experiences for some users. While there is a need for a system that can detect inappropriate messages in advance, warn senders, and encourage them to correct their messages, current systems often do not provide appropriate evaluations or warnings that fully consider user feelings. Furthermore, few systems offer specific suggestions for correction, leaving users unsure of how to correct their messages.

[1572] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: means for receiving transmitted text data and user emotion data; means for temporarily storing the received text data and emotion data in a database; means for analyzing the received text data using a natural language processing engine and extracting specific keywords and emotions; means for analyzing the received emotion data using an emotion engine and evaluating the user's emotion; means for scoring the harassment risk level based on the analysis results of the natural language processing engine and the emotion engine; means for generating a warning message when the scoring result exceeds a certain threshold; means for sending and displaying the warning message on the user terminal; and means for providing an interface that allows the user to refer to the suggested corrections and resend the message. This prevents the sending of inappropriate messages and provides the user with appropriate correction suggestions, enabling safe and smooth communication.

[1573] "Text data to be sent" refers to the text information entered by the user to be sent via email, chat, social media, etc.

[1574] "Emotion data" refers to data that indicates the user's emotional state and is information that is analyzed by the emotion engine.

[1575] "Means for temporarily storing in a database" refers to a storage device within the system for temporarily storing received character data and emotion data.

[1576] A "natural language processing engine" refers to software that analyzes text data and extracts specific keywords and emotions.

[1577] "Emotion engine" refers to software that analyzes emotional data and assesses a user's emotional state.

[1578] "Harassment risk" refers to a score that indicates the likelihood that a message constitutes harassment based on the text data and emotional data sent.

[1579] "Scoring means" refers to algorithms or programs that calculate the risk of harassment based on the analysis results of the natural language processing engine and emotion engine.

[1580] "Means for generating a warning message" refers to software or a system that automatically creates a message to warn the user when the scoring result exceeds a threshold.

[1581] "Means for sending and displaying on a user terminal" refers to a system for sending the generated warning message to a user's device and displaying it in the form of a notification or pop-up, etc.

[1582] "Means for providing an interface that allows a user to modify and resubmit a message based on the suggested modifications" refers to a system that provides a user interface that allows a user to edit and resubmit a message based on the warning message.

[1583] MODE FOR CARRYING OUT THE INVENTION

[1584] This invention is a system that analyzes the content and sender's emotions before sending messages via email, chat, social media, etc., to detect inappropriate content or potential harassment. This system consists of a user terminal, a server, and an emotion engine. Each component and its operation are described in detail below.

[1585] User Device

[1586] The user terminal provides an interface for inputting text data and includes an emotion engine for acquiring the user's emotion data. When the user inputs a message and clicks the send button, the text data and emotion data are transmitted to the server.

[1587] server

[1588] The server analyzes the received text data and emotion data to assess the risk of harassment. Specifically, it performs the following steps:

[1589] 1. Data Receipt and Storage:

[1590] The server receives the character data and emotion data sent from the user terminal.

[1591] The server temporarily stores the received data in a database.

[1592] 2. Data Analysis:

[1593] The server extracts keywords and sentiment from the message using a natural language processing engine, which performs text analysis of the message content to detect offensive keywords such as "incompetent" and the negative sentiment of the entire sentence.

[1594] The server further analyzes the received emotion data using an emotion engine to assess the user's emotional state (e.g., anger, fatigue).

[1595] 3. Scoring:

[1596] The server combines the analysis results from the natural language processing engine and the emotion engine to score the harassment risk. If offensive language and negative emotions coexist, a higher harassment score is assigned.

[1597] 4. Warning message generation:

[1598] If the server finds that the scoring result exceeds a certain threshold, it generates a warning message and prompts the user to modify the message content. For example, it generates a warning message such as, "This message may be considered offensive. We recommend that you modify it."

[1599] 5. Sending a message:

[1600] A server-generated warning message is sent to the user terminal.

[1601] Message display and editing

[1602] The device will display the received warning message to the user, allowing the user to review the warning message and correct it based on the specific correction suggestions.

[1603] If the user reflects the proposed revisions and resubmits, smoother communication is ensured.

[1604] Specific examples

[1605] Example 1: Sending offensive messages and making users angry

[1606] When a user types "You're so incompetent," and clicks the send button, the emotion engine detects that the user is very angry.

[1607] The device sends this message and emotion data to the server.

[1608] The server receives the message and emotion data and temporarily stores them in a database.

[1609] The server uses a natural language processing engine to analyze the keyword "incompetent" and overall negative sentiment, and integrates user anger data from the sentiment engine.

[1610] The server calculates the harassment score and determines that the score is high.

[1611] The server generates a warning message stating, "This message may be considered offensive. We recommend that you correct it." and sends it to the user's device.

[1612] The device will display this warning message to the user.

[1613] The user corrects the error by saying, "Your performance could use some improvement. Let's review the details." and resubmits.

[1614] Example 2: Inappropriate send times and user fatigue

[1615] If a user types "Please submit the documents immediately" at 2 a.m. and clicks the send button, the emotion engine will detect that the user is tired.

[1616] The device sends this message and emotion data to the server.

[1617] The server receives the message and emotion data and temporarily stores them in a database.

[1618] The server analyzes the sending time, message content, and emotional data to determine that the message was sent at an inappropriate time.

[1619] The server generates a warning message stating, "This message will be sent late at night and may be offensive to the recipient. Please change the sending time or modify the message content." and sends it to the user's terminal.

[1620] The device will display this warning message to the user.

[1621] The user can change the send time or modify the message content and resend it.

[1622] In this way, the system supports safe and smooth communication by analyzing the content of the message the user is about to send and their emotions at the time, pointing out any potential harassment and encouraging them to correct it. The warning message also includes specific suggestions for how to improve the message.

[1623] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1624] Step 1:

[1625] The user enters a message and clicks the send button. The user enters the text data they wish to send (e.g., "You're really incompetent."), and emotional data (e.g., angry state) is acquired at this time. The input data consists of a string of characters and emotional information.

[1626] Step 2:

[1627] The device acquires the text data and emotion data and sends them to the server. Specifically, when the user clicks the send button, the device sends both the text data (e.g., "You're really incompetent.") and emotion data (e.g., anger) to the server in a data format such as JSON.

[1628] Step 3:

[1629] The server receives the character data and emotion data and temporarily stores them in a database. Specifically, the server stores the received data in the database so that it can be used for subsequent analysis. The input is the received data, and the output is temporary storage in the database.

[1630] Step 4:

[1631] The server analyzes the text data using a natural language processing engine. Specifically, the server retrieves the text data from the database and uses the NLP engine to analyze keywords (e.g., "incompetent") and overall sentiment. The input is the text data, and the output is the analysis results.

[1632] Step 5:

[1633] The server uses an emotion engine to analyze the emotion data and evaluate the user's emotional state. Specifically, the emotion engine is used to analyze the user's emotion data in more detail, for example, to quantify the level of anger. The input is emotion data, and the output is an emotion evaluation.

[1634] Step 6:

[1635] The server integrates the results obtained from the natural language processing engine and the emotion engine to generate a score for the harassment risk. Specifically, for example, a high harassment score is generated based on the keyword "incompetent" and the user's emotion of "anger." The inputs are the analysis results and emotion assessment, and the output is a harassment score.

[1636] Step 7:

[1637] If the server scores above a certain threshold, it generates a warning message stating, "This message may be considered offensive. We recommend that you correct it." The input is the harassment score, and the output is the warning message.

[1638] Step 8:

[1639] The server sends a warning message to the user terminal. The input is the warning message and the output is the message sent to the user terminal.

[1640] Step 9:

[1641] The device displays a warning message to the user, specifically, by displaying the warning message in the user interface as a popup or notification. The input is the received warning message, and the output is the screen display.

[1642] Step 10:

[1643] The user reviews the warning message and modifies it based on the suggested corrections. For example, the user modifies the message to avoid offensive language and to say, "Your performance has room for improvement. Please review the details." The input is the warning message, and the output is the modified message.

[1644] Step 11:

[1645] The user resubmits the modified message. This time, the same processing flow is followed again to reevaluate whether the resubmitted message is appropriate. The input is the modified message, and the output is the final message transmission.

[1646] (Application example 2)

[1647] 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."

[1648] In customer support for electronic payment services, if the messages sent by the staff are inappropriate or offensive, the quality of the service may decline and customer satisfaction may be impaired. The present invention aims to solve this problem and provide a system for improving the quality of customer support and the customer experience.

[1649] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving transmitted character data, means for analyzing the received character data using a natural language processing engine and extracting specific keywords and emotions, and means for scoring the harassment risk based on the extracted data. This makes it possible to monitor customer support messages in electronic payment services and improve hospitality and customer experience.

[1650] A "user terminal" is a device for inputting and transmitting text data, and includes smartphones, personal computers, etc.

[1651] A "natural language processing engine" is software that analyzes text data and extracts specific keywords and emotions.

[1652] "Harassment risk" is a numerical representation of the likelihood that a particular message is inappropriate or offensive.

[1653] A "warning message" is a message that is generated to warn users when the risk of harassment exceeds a certain threshold.

[1654] An "electronic payment service" is a system for electronically paying for goods and services online or offline.

[1655] "Customer support" refers to services that respond to customer inquiries and problems.

[1656] "Hospitality" is a concept that refers to the quality of hospitality and treatment shown to customers.

[1657] "Customer experience" refers to the overall satisfaction and impression a customer has when using a service.

[1658] "Interface" refers to the screen and operating means that users use to interact with a system.

[1659] The present invention provides a system for monitoring customer support messages for electronic payment services to improve hospitality and customer experience. The system includes a user terminal, a server, an emotion engine, and a natural language processing engine. Specific embodiments for implementing the present invention are described below.

[1660] System Configuration

[1661] 1. User Device

[1662] The user terminal is a device for inputting and sending text data. This can be a smartphone or a PC. When the user inputs a support message and clicks the send button, the message and emotional data are sent to the server.

[1663] 2. Server

[1664] The server analyzes the received text data and emotion data to evaluate the degree of harassment risk. Specifically, the process is as follows:

[1665] Natural language processing engine: Analyzes text data and extracts specific keywords and sentiment (example of software used: spaCy).

[1666] Emotion engine: Analyzes the user's emotional data and extracts their emotional state (example of software used: IBM Watson NLP).

[1667] Scoring engine: Scores the level of harassment risk based on extracted keywords and sentiment data.

[1668] 3. Generating and sending warning messages

[1669] If the scoring result exceeds a certain threshold, the server generates a warning message and sends it to the user's device, including specific suggestions for correction.

[1670] Specific examples

[1671] 1. Inappropriate messages from support agents

[1672] 1. The support representative types "This problem is on your side" and clicks the send button.

[1673] 2. The emotion engine detects when a person is feeling stressed.

[1674] 3. The server receives the message and sentiment data and analyzes it for keywords like "the cause is on the user's side" and negative sentiment.

[1675] 4. Calculate the harassment score and determine it to be high.

[1676] 5. A warning message will be generated stating, "This message may be considered offensive. We recommend that you fix it."

[1677] 6. A warning message is sent to the user's terminal and displayed.

[1678] 7. The support person will fix it by saying, "Some of the problem is on the user's side, but we will deal with it."

[1679] Prompt Sentence Examples

[1680] The system analyzes the content and sentiment data of messages sent by customer support staff of electronic payment services, automatically detecting inappropriate content or potential harassment. If the message is potentially offensive or misleading, a warning message is generated and displayed. Example message: "The cause of this problem lies with the user."

[1681] This system will enable customer support for electronic payment services to be more professional and hospitable.

[1682] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1683] Step 1:

[1684] The user enters a message and clicks the send button. The input is text data, and emotion data is also acquired at the same time. By sending, the user device sends this data to the server.

[1685] Step 2:

[1686] The server receives the text data and emotion data from the user terminal and stores the input text data and emotion data in a temporary storage device for analysis.

[1687] Step 3:

[1688] The server uses a natural language processing engine to analyze the received text data. The input is text data, and keywords and the overall emotional state are extracted. Specifically, keyword extraction is performed using the spaCy library.

[1689] Step 4:

[1690] The server analyzes the user's emotional data using an emotion engine. The input is the emotional data, and IBM Watson NLP is used to extract the emotional state. The output is the user's specific emotional state.

[1691] Step 5:

[1692] The server integrates the data obtained from the natural language processing engine and the emotion engine. Specifically, the scoring engine evaluates the harassment risk based on the extracted keywords and emotional state, and calculates a harassment score. The input is keywords and emotion data, and the output is a harassment score.

[1693] Step 6:

[1694] The scoring engine determines if the score exceeds a certain threshold. If so, the server generates a warning message. The input is the harassment score, and the output is the warning message. The generated warning message includes specific suggestions for correction.

[1695] Step 7:

[1696] The server sends the generated warning message to the user terminal, which displays the message and notifies the user. The input is the warning message, and the output is the displayed warning message.

[1697] Step 8:

[1698] The user checks the presented warning message, corrects the message if necessary, and resends it. Based on the warning message, the user applies the specific correction suggestions, enters a new message, and presses the resend button. The input is the corrected text data, and the output is the resent message.

[1699] This series of steps enables customer support for electronic payment services to automatically evaluate the appropriateness of sent messages and prompt necessary corrections to provide a better customer experience.

[1700] 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.

[1701] 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.

[1702] 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.

[1703] 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.

[1704] 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.

[1705] 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.

[1706] 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).

[1707] 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.

[1708] 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."

[1709] 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.

[1710] 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).

[1711] 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.

[1712] 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.

[1713] 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.

[1714] 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.

[1715] 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.

[1716] 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.

[1717] 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.

[1718] 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.

[1719] 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.

[1720] 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.

[1721] The following is further disclosed regarding the above embodiment.

[1722] (Claim 1)

[1723] means for receiving transmitted character data;

[1724] A means of analyzing the received text data using a natural language processing engine and extracting specific keywords and emotions;

[1725] A means for scoring the risk of harassment based on the extracted data;

[1726] means for generating a warning message when the scoring result exceeds a certain threshold;

[1727] A system including means for transmitting and displaying a warning message to a user terminal.

[1728] (Claim 2)

[1729] 10. The system of claim 1, further comprising: means for generating a warning message including specific suggestions for correction based on the scoring result.

[1730] (Claim 3)

[1731] 10. The system of claim 1, further comprising: means for providing an interface upon displaying the warning message that allows a user to apply suggested corrections and resubmit.

[1732] "Example 1"

[1733] (Claim 1)

[1734] means for receiving transmitted character data;

[1735] A means of analyzing the received text data using a natural language processing engine and extracting specific keywords and emotions;

[1736] A means for scoring the risk of harassment based on the extracted data;

[1737] means for generating a warning message when the scoring result exceeds a certain threshold;

[1738] means for transmitting and displaying a warning message on a user terminal;

[1739] means for analyzing the sending times and identifying inappropriate messages based on time of day;

[1740] A system including:

[1741] (Claim 2)

[1742] 10. The system of claim 1, further comprising: means for generating a warning message including specific suggestions for correction based on the scoring result.

[1743] (Claim 3)

[1744] 10. The system of claim 1, further comprising: means for providing an interface upon displaying the warning message that allows a user to apply suggested corrections and resubmit.

[1745] "Application Example 1"

[1746] (Claim 1)

[1747] means for receiving transmitted character data;

[1748] A means of analyzing the received text data using a natural language processing engine and extracting specific keywords and emotions;

[1749] A means for scoring the risk of harassment based on the extracted data;

[1750] means for generating a warning message when the scoring result exceeds a certain threshold;

[1751] means for transmitting and displaying a warning message on a user terminal;

[1752] The system includes means for analyzing the time of message transmission and generating a warning message if the transmission time is inappropriate.

[1753] (Claim 2)

[1754] 10. The system of claim 1, further comprising: means for generating a warning message including specific suggestions for correction based on the scoring result.

[1755] (Claim 3)

[1756] 10. The system of claim 1, further comprising: means for providing an interface upon displaying the warning message that allows a user to apply suggested corrections and resubmit.

[1757] "Example 2: Combining Emotion Engines"

[1758] (Claim 1)

[1759] means for receiving the transmitted character data and user emotion data;

[1760] a means for temporarily storing the received character data and emotion data in a database;

[1761] A means of analyzing the received text data using a natural language processing engine and extracting specific keywords and emotions;

[1762] a means for analyzing the received emotion data by an emotion engine and evaluating the emotion of the user;

[1763] A means for scoring the risk of harassment based on the analysis results of the natural language processing engine and the emotion engine;

[1764] means for generating a warning message when the scoring result exceeds a certain threshold;

[1765] means for transmitting and displaying a warning message on a user terminal;

[1766] providing an interface that allows the user to modify and resubmit the message using the suggested modifications; and

[1767] A system including:

[1768] (Claim 2)

[1769] 10. The system of claim 1, further comprising: means for generating a warning message including specific suggestions for correction based on the scoring result.

[1770] (Claim 3)

[1771] 10. The system of claim 1, further comprising: means for providing an interface upon displaying the warning message that allows a user to apply suggested corrections and resubmit.

[1772] "Application example 2 when combining emotion engines"

[1773] (Claim 1)

[1774] means for receiving transmitted character data;

[1775] A means of analyzing the received text data using a natural language processing engine and extracting specific keywords and emotions;

[1776] A means for scoring the risk of harassment based on the extracted data;

[1777] means for generating a warning message when the scoring result exceeds a certain threshold;

[1778] means for transmitting and displaying a warning message on a user terminal;

[1779] A system that monitors customer support messages in electronic payment services and includes measures to improve hospitality and customer experience.

[1780] (Claim 2)

[1781] 10. The system of claim 1, further comprising: means for generating a warning message including specific suggestions for correction based on the scoring result.

[1782] (Claim 3)

[1783] 10. The system of claim 1, further comprising: means for providing an interface upon displaying the warning message that allows a user to apply suggested corrections and resubmit. [Explanation of symbols]

[1784] 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. means for receiving transmitted character data; A means of analyzing the received text data using a natural language processing engine and extracting specific keywords and emotions; A means for scoring the risk of harassment based on the extracted data; means for generating a warning message when the scoring result exceeds a certain threshold; A system including means for transmitting and displaying a warning message to a user terminal.

2. The system of claim 1 , further comprising: means for generating a warning message including specific suggestions for correction based on the scoring results.

3. 10. The system of claim 1, further comprising means for providing an interface that allows a user to apply suggested corrections and resubmit when displaying the warning message.

Images