system

Abstract

We provide the system. [Solution] An input method for the user to enter text, A communication means for sending the input text to the server, The server has an analysis mechanism that performs morphological analysis, grammatical analysis, and semantic analysis on the text it receives. An error detection means and a correction generation means that identify errors in the text and generate correction proposals based on the analysis results, A means of sending the revised proposal to the terminal, A system that includes this.

Description

Technical Field

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

Background Art

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

Summary of the Invention

Problems to be Solved by the Invention

[0004] Conventionally, systems for automating the correction of errors and unnatural expressions in Japanese text have had problems of insufficient accuracy or complex and inconvenient operations. In addition, the process of detecting errors and presenting correction proposals is inefficient and often takes time and effort for users. The present invention aims to solve these drawbacks of the prior art and provide a system that can effectively detect errors in Japanese text input by users and automatically provide appropriate correction proposals.

Means for Solving the Problems

[0005] The system of the present invention first provides an input means for the user to input text. Next, it includes a communication means for sending the input text to a server. The server analyzes the received text using morphological analysis, grammatical analysis, and semantic analysis means. Based on the analysis results, the server includes error detection means and correction generation means that point out errors in the text and generate correction suggestions. These correction suggestions are sent from the server to a terminal, which displays them to the user. This allows the user to correct errors in Japanese text quickly and efficiently.

[0006] "Input means" refers to a device or interface for a user to input text.

[0007] "Communication method" refers to the internet connection and communication protocol used to send input text from the user's terminal to the server.

[0008] "Analysis means" refers to software or algorithms used to perform morphological analysis, grammatical analysis, and semantic analysis of text on a server.

[0009] Morphological analysis refers to the process of breaking down text into individual words and identifying the part of speech and conjugation of each word.

[0010] "Grammar analysis" refers to the process of detecting grammatical errors and syntactic errors within a sentence.

[0011] "Semantic analysis" refers to the process of interpreting the meaning of an entire text and evaluating the naturalness of its context and expression.

[0012] "Error detection means" refers to software or algorithms used to identify errors in text based on analysis results.

[0013] "Correction generation means" refers to software or algorithms for generating appropriate corrections for errors identified by error detection means.

[0014] "Transmission means" refers to the communication means for transmitting the generated amendment from the server to the terminal.

[0015] "Terminal" refers to electronic devices such as computers, smartphones, and tablets used by users.

[0016] "Server" refers to a computer system for receiving and analyzing the text transmitted from the user's terminal.

Brief Description of the Drawings

[0017] [Figure 1] It is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] It is a conceptual diagram showing an example of the main functions of a data processing device and a smart device according to the first embodiment. [Figure 3] It is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] It is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] It is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] It is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] It is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] It is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] It shows an emotion map to which a plurality of emotions are mapped. [Figure 10] It shows an emotion map to which a plurality of emotions are mapped. [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12]It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Example 2 when combined with an emotion engine. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when combined with an emotion engine.

Mode for Carrying Out the Invention

[0018] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0019] First, the language used in the following description will be explained.

[0020] In the following embodiments, the labeled processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0021] In the following embodiments, the labeled RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0022] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0023] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0024] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0025] [First Embodiment]

[0026] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0027] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0028] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0029] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0030] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0031] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0032] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0033] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0034] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0035] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0036] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0037] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0038] The system of the present invention is an automatic proofreading tool that assists users in accurately and naturally inputting Japanese text. In this embodiment, the specific program processing flow will be described.

[0039] First, the user uses their terminal to input Japanese text into the system's text input form. For example, suppose the user enters the following sentence:

[0040] I will be attending tomorrow's meeting, so I look forward to seeing you there.

[0041] Next, the terminal sends an HTTP request to the server using a communication method to send the entered text to the server. At this time, the text data is sent to the server.

[0042] The server begins processing the received text data for analysis. Specifically, the following three analysis methods are used:

[0043] 1. Morphological Analysis: The server uses a morphological analysis library to break down the text into individual words. For example, the input sentence is broken down into "I will be attending tomorrow's meeting, so please take care of me." The part of speech and conjugation of each word are also analyzed.

[0044] 2. Grammar Analysis: The grammar analysis engine detects grammatical errors. In the example above, it detects that the expression "sanka suru masu" is inappropriate.

[0045] 3. Semantic Analysis: A natural language understanding (NLU) model analyzes the meaning of the entire sentence and identifies unnatural phrasing. In this case, it determines that "onegaishimasu" should be changed to "onegaishimasu".

[0046] Based on the analysis results, the server uses error detection means to identify the error locations and correction generation means to generate appropriate correction suggestions. Specific correction suggestions are as follows:

[0047] Error: "surumasu" -> Corrected: "sanka shimasu"

[0048] Error: "onegaishimasu" -> Correction: "onegaishimasu"

[0049] After the revised version is generated, the server sends it back to the terminal as JSON data via the transmission method.

[0050] The terminal parses the received JSON data and displays suggested corrections to the user. For example, the user will see the following:

[0051] I will be attending tomorrow's meeting, so I look forward to seeing you there.

[0052] The user reviews the displayed correction suggestions and applies the corrections as needed. This corrects the errors in the entered Japanese text, resulting in a correct sentence.

[0053] The above describes a specific embodiment for carrying out the present invention. By using this system, users can quickly and easily correct errors in Japanese text.

[0054] The following describes the processing flow.

[0055] Step 1:

[0056] The user enters Japanese text into a text input form on their device. After entering the text, the user prepares to send the text by pressing the send button.

[0057] Step 2:

[0058] The terminal receives text data entered by the user and sends an HTTP request to the server using a communication method. This request contains text data.

[0059] Step 3:

[0060] The server parses the received HTTP request and extracts the text data from the request body. The server then passes the received text to the morphological analysis module.

[0061] Step 4:

[0062] The server's morphological analysis module breaks down text into individual words and analyzes the part of speech and conjugation of each word. This analysis includes the base form and part of speech information of each word.

[0063] Step 5:

[0064] The analyzed word information is passed to the grammar analysis module. The server's grammar analysis module identifies grammatical errors and syntactic errors in the sentence. In this process, ungrammatical verb and auxiliary verb combinations, etc., are detected.

[0065] Step 6:

[0066] Next, the server's semantic analysis module understands the overall context and meaning, detecting unnatural phrasing and misuse. This analysis evaluates the appropriateness of words and expressions.

[0067] Step 7:

[0068] Based on the results of grammatical and semantic analysis, the server identifies errors using error detection and correction generation means and generates appropriate corrections. For example, "sanka suru masu" is corrected to "sanka shimasu," and "onegai shimasu" is corrected to "onegai shimasu."

[0069] Step 8:

[0070] After the proposed correction is generated, the server formats it as JSON data. This data includes the location of the error, the original text, and the proposed correction.

[0071] Step 9:

[0072] The server sends JSON data back to the terminal via a transmission method. The terminal receives this data and processes it for display.

[0073] Step 10:

[0074] The device displays the suggested fixes to the user in a visually easy-to-understand format. The user reviews the suggested fixes and applies them as needed.

[0075] Step 11:

[0076] The user revises the text based on the suggested changes and then performs a final check. After this, the correct Japanese text can be used for other purposes.

[0077] (Example 1)

[0078] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0079] Conventional text input systems have a problem in that they have difficulty automatically detecting grammatical errors and semantic unnaturalness in Japanese text and providing accurate and natural correction suggestions. Therefore, users have to manually check and correct the text, which leads to a decrease in work efficiency.

[0080] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0081] In this invention, the server includes input means for the user to input text, communication means for transmitting the input text to the server, analysis means for performing morphological, grammatical, and semantic analysis of the text received by the server, error detection means and correction generation means for pointing out errors in the text and generating correction suggestions based on the analysis results, transmission means for transmitting the correction suggestions to the terminal, and display means for displaying the correction suggestions received by the terminal to the user. This enables the user to quickly correct errors in the input Japanese text and generate accurate and natural-sounding sentences.

[0082] A "user" is a person who inputs text into a system and utilizes the functions that the system provides.

[0083] "Input means" refers to a device or software that provides an interface for a user to input text.

[0084] "Communication means" refers to a device or software used to transmit text data entered from an input means to a server.

[0085] A "server" is a device or system that receives input text data and performs analysis on it.

[0086] "Analysis means" refers to software or algorithms used to perform morphological analysis, grammatical analysis, and semantic analysis on text received by the server.

[0087] "Error detection means" refers to a device or software that detects errors in text based on the results of analysis performed by the analysis means.

[0088] "Correction generation means" refers to a device or software that generates appropriate corrections for errors detected by error detection means.

[0089] "Transmission means" refers to a device or software used to send the generated revised draft to a terminal.

[0090] "Display means" refers to a device or software that displays the proposed revisions received by the terminal to the user.

[0091] Morphological analysis is a technique that breaks down input text into individual words and analyzes the part of speech and conjugation of each word.

[0092] "Grammar analysis" is a technique for detecting grammatical errors and syntactic errors within text.

[0093] "Semantic analysis" is a technique for understanding the meaning of an entire text and identifying unnatural phrasing or contextual problems.

[0094] A "revised version" is a suggestion for a correct or more appropriate expression, generated based on the detected error.

[0095] A "terminal" refers to a device, such as a computer or smartphone, that a user uses to access a system.

[0096] The present invention is an automated text correction tool that assists users in accurately and naturally inputting Japanese text. Specifically, the user inputs Japanese text into a text input form using a terminal. At this time, the terminal has means for sending the input text to a server.

[0097] The server has morphological, grammatical, and semantic analysis tools to analyze the received text data. First, to perform morphological analysis, it uses a morphological analysis library (e.g., MeCab or Sudachi). Morphological analysis breaks down the text into words and analyzes the part of speech and conjugation of each word. For example, if a user inputs the following sentence:

[0098] I will be attending tomorrow's meeting, so I look forward to seeing you there.

[0099] Morphological analysis breaks down this sentence into "I will be attending tomorrow's meeting, so please take care of me."

[0100] Next, the server uses a grammar analysis engine (e.g., SyntaxNet) to perform grammatical analysis. Grammar analysis detects grammatical errors and syntactic errors in the sentence. In the example above, it detects that "sanka suru masu" is grammatically incorrect.

[0101] Furthermore, to perform semantic analysis, a natural language understanding (NLU) model (e.g., BERT or GPT-3®) is used. Semantic analysis analyzes the meaning of the entire text and identifies unnatural phrasing. In this case, it is determined that "onegaishimasu" should be changed to "onegaishimasu".

[0102] Based on the analysis results, the server uses error detection means to identify the error location and correction generation means to generate appropriate correction proposals. Specific correction proposals are as follows:

[0103] Error: "surumasu" -> Corrected: "sanka shimasu"

[0104] Error: "onegaishimasu" -> Correction: "onegaishimasu"

[0105] After the proposed revisions are generated, the server sends them back to the terminal as JSON data via a transmission method. The terminal parses the received JSON data and displays the proposed revisions to the user. For example, the proposed revision might say, "I will be attending tomorrow's meeting, thank you in advance."

[0106] The user reviews the displayed correction suggestions and applies the corrections as needed. This corrects the errors in the entered Japanese text, resulting in a correct sentence.

[0107] Here are some examples of specific prompt messages:

[0108] Please identify the error in the following Japanese sentence and provide the correct sentence:

[0109] I will be attending tomorrow's meeting, so I look forward to seeing you there.

[0110] The above describes a specific embodiment for carrying out the present invention. By using this system, users can quickly and easily correct errors in Japanese text.

[0111] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0112] Step 1:

[0113] The user uses their own device to input Japanese text into the system's text input form. They use a keyboard or other input method, and the text data is recorded within the device. For example, the user might input "I will be attending tomorrow's meeting, thank you in advance."

[0114] Step 2:

[0115] The terminal sends the entered text data to the server. Specifically, it generates an HTTP request using a communication method and sends the request containing the text data to the server. The input is the text data entered by the user, and the output is the transmission of the HTTP request.

[0116] Step 3:

[0117] The server takes in the received HTTP request and extracts the text data. At this point, the input is the text data contained in the body of the HTTP request, and the output is the extracted text data.

[0118] Step 4:

[0119] The server uses a morphological analysis library (e.g., MeCab or Sudachi) to break down text data into individual words. Specifically, it starts the analysis process and passes the text data to the analysis library. The input is the extracted text data, and the output is the morphologically analyzed data (e.g., "I will be attending tomorrow's meeting, so please take care of me").

[0120] Step 5:

[0121] The server uses a grammar analysis engine (e.g., SyntaxNet) to detect grammatical and syntactic errors in the morphologically analyzed data. Specifically, it activates the grammar analysis tool and passes the analysis data. The input is the morphologically analyzed data, and the output is the identification of grammatical errors (e.g., errors such as "sanka suru masu").

[0122] Step 6:

[0123] The server uses a natural language understanding (NLU) model (e.g., BERT or GPT-3) to analyze the meaning of the entire text. Specifically, it activates a semantic analysis tool and passes the morphologically analyzed data to the NLU model. The input is the morphologically analyzed data, and the output is semantic errors or suggestions for improvement (e.g., a judgment that "onegaishimasu" should be corrected to "onegaishimasu").

[0124] Step 7:

[0125] Based on the analysis results, the server records the errors identified using error detection means and generates appropriate correction suggestions using correction generation means. Specifically, it extracts the error locations through error detection means and creates correction suggestions using correction generation means. The input is the result of grammatical and semantic analysis, and the output is the correction suggestion (e.g., "surumasu" → "sankashimasu", "onegaishimasu" → "onegaishimasu").

[0126] Step 8:

[0127] After the proposed revisions are generated, the server sends them back to the terminal as JSON data using a transmission method. Specifically, the server converts the proposed revisions to JSON format and includes them in the HTTP response before sending. The input is the proposed revisions, and the output is the transmission of JSON data.

[0128] Step 9:

[0129] The terminal parses the received JSON data and displays suggested corrections to the user. Specifically, it uses a display mechanism to visually convert the JSON data and display it on the user interface. The input is data in JSON format, and the output is a suggested correction displayed to the user (e.g., "I will be attending tomorrow's meeting, thank you in advance").

[0130] Step 10:

[0131] The user reviews the displayed suggested revisions and applies them as needed. Specifically, the user confirms the text revisions by clicking the "Apply" button. The input is the user's click operation, and the output is the text with the revisions applied.

[0132] (Application Example 1)

[0133] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0134] When ordering food delivery, users often include typos, grammatical errors, or unnatural phrasing in their special requests and order details, which can lead to misdeliveries and a poor customer experience. A system is needed to solve this problem and allow users to confirm their orders in more accurate and natural Japanese.

[0135] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0136] In this invention, the server includes input means for the user to input text, communication means for transmitting the input text to the server, analysis means for performing morphological, grammatical, and semantic analysis of the text received by the server, error detection means and correction generation means for pointing out errors in the text and generating correction suggestions based on the analysis results, transmission means for transmitting the correction suggestions to the terminal, and means featuring an automatic proofreading function for food delivery applications to correct typos, omissions, and unnatural expressions made by the user when inputting order details and special notes. This enables the user to quickly correct the content they have entered and place accurate orders.

[0137] "Input means" refers to a device or interface for a user to input text.

[0138] "Communication means" refers to a device or protocol used to send input text to a server.

[0139] "Analysis means" refers to software or hardware mechanisms that perform morphological analysis, grammatical analysis, and semantic analysis on text received by the server.

[0140] An "error detection means" is a device or program that identifies errors in text based on the analysis results.

[0141] A "correction generation means" is a device or program for generating appropriate corrections after an error has been identified.

[0142] "Transmission means" refers to a device or protocol for sending the revised version to a terminal.

[0143] The "Automatic Correction Function for Food Delivery Applications" is a feature that corrects typos, grammatical errors, and unnatural expressions when users enter order details or special notes.

[0144] The present invention provides an automated text correction tool that assists users in accurately and naturally inputting Japanese text. In this invention, the system is particularly applicable to food delivery applications.

[0145] 1. System Program

[0146] This system consists of the following main elements:

[0147] Input method: An interface for users to enter text. A text input box embedded within a food delivery application.

[0148] Communication method: An HTTP request to send the entered text to the server. Specifically, the text is sent to the server in JSON format.

[0149] Analysis methods: Functions for morphological analysis, grammatical analysis, and semantic analysis of text received by the server. Morphological analysis uses the Janome library, and grammatical and semantic analysis utilize natural language understanding (NLU) models.

[0150] Error detection and correction generation means: Based on the analysis results, errors in the text are detected and appropriate correction suggestions are generated. Examples of errors include incorrect particles and unnatural expressions.

[0151] Transmission method: An HTTP response to send JSON data containing the proposed revisions back to the terminal.

[0152] 2. Program processing description using natural language

[0153] The server receives text sent by the user through the input method. The Janome library is used to perform morphological analysis on the received text, breaking down the sentence into individual words. For example, the input sentence "I will be attending tomorrow's meeting, so thank you in advance" is broken down by morphological analysis into "tomorrow / no / kaigi / ni / participating / suru / masu / node / , / yoroshiku / onegai / shimasu".

[0154] Next, the grammar analysis engine analyzes the part of speech and conjugation of each word to detect grammatical errors. For example, it detects that the expression "sanka suru masu" (to participate) is inappropriate. Subsequently, in the semantic analysis stage, a natural language understanding (NLU) model is used to analyze the meaning of the entire sentence and identify unnatural phrasing. Here, it determines that "onegai shimasu" (please) should be changed to "onegai shimasu" (please).

[0155] Based on the analysis results, the error detection means detects errors, and the correction generation means generates appropriate correction suggestions. The server then sends the correction suggestions back to the terminal as JSON data. The terminal analyzes the received JSON data and displays the correction suggestions to the user. The user reviews the displayed correction suggestions and applies the corrections as needed.

[0156] 3. Specific examples and prompt messages

[0157] For example, a user enters the following order details:

[0158] I am allergic to shrimp, so please remove the shrimp from the dish.

[0159] When this input is sent to the server, it will suggest the following modifications:

[0160] I am allergic to shrimp, so please omit the shrimp from the dish.

[0161] Examples of prompt statements to input into a generative AI model are as follows:

[0162] Analyze the Japanese text entered by the user grammatically and semantically, and suggest appropriate corrections. Example: 'I am allergic to shrimp, so please remove the shrimp.' -> 'I am allergic to shrimp, so please remove the shrimp.'

[0163] This allows users to quickly correct their input and place orders in accurate and natural-sounding Japanese.

[0164] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0165] Step 1:

[0166] The user uses their device to enter order details and special requests into a text input form within the food delivery application. An example of the text entered is, "I am allergic to shrimp, so please omit the shrimp."

[0167] Step 2:

[0168] The terminal sends an HTTP request to the server using a communication method to send the entered text to the server. Specifically, it sends the text data to the server in JSON format.

[0169] Step 3:

[0170] The server analyzes the received text data. First, the server uses the morphological analysis library Janome to break down the text into words. For example, it breaks down "I am allergic to shrimp, so please remove the shrimp." into "Allergy / is / shrimp / , / shrimp / remove / please / please / you / please / please / remove / shrimp".

[0171] Step 4:

[0172] The server uses a grammar analysis engine to analyze the morphologically analyzed word sequence from the perspective of part of speech and conjugation, and detects grammatical errors. For example, it might determine that the unnatural expression "kudasaimasu you onegaishimasu" should be corrected to "onegaishimasu".

[0173] Step 5:

[0174] The server uses a natural language understanding (NLU) model to perform semantic analysis. Here, it analyzes the meaning of the entire sentence and identifies unnatural phrasing and expressions. For example, it might correct "onegaishimasu" to "onegaishimasu."

[0175] Step 6:

[0176] Based on the analysis results, the server uses error detection to identify errors and correction generation to generate appropriate corrections. Specifically, it generates a correction to change "onegaishimasu" to "onegaishimasu".

[0177] Step 7:

[0178] The server sends the proposed revisions back to the terminal as JSON data via a transmission method. The transmitted data includes both the original and revised text.

[0179] Step 8:

[0180] The terminal parses the received JSON data and displays suggested corrections to the user. For example, it might display "I am allergic to shrimp, so please remove the shrimp."

[0181] Step 9:

[0182] The user reviews the displayed revisions and applies them as needed. This confirms that the corrected text is finalized in the order.

[0183] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0184] The present invention is an automated proofreading tool that allows users to input Japanese text accurately and naturally, and provides feedback that takes the user's feelings into consideration. In this embodiment, the specific program processing flow will be described.

[0185] First, the user enters a Japanese sentence into a text input form on their device. For example, suppose the user enters the following sentence:

[0186] I will be attending tomorrow's meeting, so I look forward to seeing you there. I'm very nervous.

[0187] Next, the terminal sends an HTTP request to the server using a communication method to send the entered text to the server. At this time, the text data is sent to the server.

[0188] The server begins processing the received text data for analysis. Specifically, the following four analysis methods are used:

[0189] 1. Morphological Analysis: The server uses a morphological analysis library to break down the text into individual words. For example, the input sentence is broken down into "I will be attending tomorrow's meeting, so please be kind to me. I am very nervous." The part of speech and conjugation of each word are also analyzed.

[0190] 2. Grammar Analysis: The grammar analysis engine detects grammatical errors. In the example above, it detects that the expression "sanka suru masu" is inappropriate.

[0191] 3. Semantic Analysis: A natural language understanding (NLU) model analyzes the meaning of the entire sentence and identifies unnatural phrasing. In this case, it determines that "onegaishimasu" should be changed to "onegaishimasu".

[0192] 4. Sentiment Analysis: The server's sentiment analysis tool identifies the emotions contained in the user's text. This analysis determines that "I'm very nervous" represents the emotion of "nervousness."

[0193] Based on the analysis results, the server uses error detection and correction generation means to identify errors and generate appropriate correction suggestions. Specific correction suggestions are as follows:

[0194] Error: "surumasu" -> Corrected: "sanka shimasu"

[0195] Error: "onegaishimasu" -> Correction: "onegaishimasu"

[0196] Furthermore, the emotion engine generates user-friendly feedback based on identified emotions. For example, in response to "I'm very nervous," it might generate advice such as "Try to relax before the meeting."

[0197] After the proposed revisions and sentiment feedback are generated, the server sends them back to the terminal as JSON data via a transmission method.

[0198] The device parses the received JSON data and displays suggested revisions and sentiment feedback to the user. For example, the user might see the following:

[0199] I will be attending tomorrow's meeting, so I look forward to seeing you there. I'm very nervous.

[0200] Proposed amendment:

[0201] "I will be attending tomorrow's meeting, so I look forward to seeing you there."

[0202] Emotional feedback:

[0203] "Please relax and approach the meeting with ease."

[0204] Users review the displayed correction suggestions and sentiment feedback, and apply the corrections as needed. This corrects errors in the entered Japanese text and allows them to receive feedback that takes their feelings into consideration.

[0205] The above describes a specific embodiment for carrying out the present invention. By using this system, users can quickly and easily correct errors in Japanese text and receive emotionally sensitive feedback.

[0206] The following describes the processing flow.

[0207] Step 1:

[0208] The user enters a Japanese sentence into a text input form on their device. For example, they might enter, "I will be attending tomorrow's meeting, so I look forward to seeing you there. I'm very nervous." Once they have finished typing, they click the submit button.

[0209] Step 2:

[0210] The terminal receives the input text data and uses a communication method to send an HTTP request to the server. This request contains the text data.

[0211] Step 3:

[0212] The server parses the received HTTP request and extracts text data. The extracted text data is then passed to the morphological analysis module.

[0213] Step 4:

[0214] The server's morphological analysis module breaks down text into individual words and analyzes the part of speech and conjugation of each word. For example, the text "I will be attending tomorrow's meeting, so I look forward to seeing you. I'm very nervous." is broken down into "tomorrow / noun of / particle meeting / noun to / particle attend / verb do / auxiliary verb do / auxiliary verb so / particle, / symbol nice to see / adverb please / verb do / auxiliary verb. / symbol very / adverb nervous / noun feeling / auxiliary verb I am / auxiliary verb".

[0215] Step 5:

[0216] The analyzed word information is passed to the grammar analysis module. The server's grammar analysis module identifies grammatical errors and syntactic errors in the sentence. At this stage, the incorrect expression "sanka suru masu" (participate) is detected.

[0217] Step 6:

[0218] The results of the grammatical analysis are passed to the semantic analysis module. The server's semantic analysis module understands the context and identifies unnatural phrasing and misuse. This analysis determines that "onegaishimasu" should be corrected to "onegaishimasu".

[0219] Step 7:

[0220] Next, the server's sentiment analysis tool identifies the emotions contained in the user's text. It recognizes the emotion of tension from the expression "I'm very nervous."

[0221] Step 8:

[0222] Based on the results of grammatical analysis, semantic analysis, and sentiment analysis, the server uses error detection and correction generation means to identify errors and generate appropriate correction suggestions and sentiment feedback. For example, "participate" is corrected to "will participate," and "please" is corrected to "please." In addition, sentiment feedback such as "Please relax and enjoy the meeting" is generated for a nervous user.

[0223] Step 9:

[0224] The generated corrections and sentiment feedback are formatted in JSON format. This data includes the location of the error, the original text, the correction, and the sentiment feedback. The server sends this JSON data back to the terminal via the transmission method.

[0225] Step 10:

[0226] The device parses the received JSON data and displays suggested revisions and sentiment feedback to the user. For example, it may be displayed as follows:

[0227] I will be attending tomorrow's meeting, so I look forward to seeing you there. I'm very nervous.

[0228] Proposed amendment:

[0229] "I will be attending tomorrow's meeting, so I look forward to seeing you there."

[0230] Emotional feedback:

[0231] "Please relax and approach the meeting with ease."

[0232] Step 11:

[0233] Users review the displayed correction suggestions and sentiment feedback, and apply the corrections as needed. This corrects errors in the entered Japanese text and allows them to receive feedback that takes their feelings into consideration.

[0234] (Example 2)

[0235] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0236] When users input Japanese text, the challenge lies in accurately identifying and correcting errors, as well as providing feedback that takes the user's emotions into consideration. Conventional systems only point out errors and fail to provide emotionally sensitive feedback, thus failing to sufficiently improve user satisfaction and the naturalness of the text.

[0237] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0238] In this invention, the server includes analysis means for performing morphological, grammatical, semantic, and sentiment analysis on received text; error detection and correction generation means for pointing out errors in the text, generating suggested corrections, and providing feedback that takes the user's feelings into consideration, based on the analysis results; and transmission means for sending the suggested corrections and sentiment feedback to the terminal. This allows the user to correct errors while simultaneously receiving feedback that takes their feelings into consideration.

[0239] "Text" refers to a string of characters entered by the user, and may include Japanese sentences or words.

[0240] "Input means" refers to devices or interfaces for users to input text, including keyboards and touchscreens.

[0241] "Communication method" refers to the means by which a terminal sends text entered into it to a server, and this involves using an internet connection or a communication protocol (e.g., HTTP).

[0242] A "server" is a central processing unit that receives, analyzes, and processes text sent from a user's terminal.

[0243] "Analysis means" includes software and libraries for morphological analysis, grammatical analysis, semantic analysis, and sentiment analysis of received text.

[0244] Morphological analysis is the process of breaking down text into individual words and analyzing the part of speech and conjugation of each word.

[0245] "Grammar analysis" is the process of detecting grammatical errors and syntactic errors within a text.

[0246] "Semantic analysis" is the process of understanding the overall meaning of a text and identifying unnatural expressions.

[0247] "Sentiment analysis" is the process of identifying the user's emotions contained in text and generating appropriate feedback.

[0248] "Error detection means and correction generation means" refers to processing means for identifying errors in text based on analysis results and generating proposed corrections.

[0249] "Transmission means" refers to the means by which the server sends back the proposed revisions and sentiment feedback generated by the server to the terminal.

[0250] A "revision proposal" is a suggestion to correct errors detected by the analysis method.

[0251] "Emotional feedback" refers to providing advice and comments that take the user's emotions into consideration, based on emotion analysis.

[0252] The present invention is an automated proofreading tool that allows users to input Japanese text accurately and naturally, and provides feedback that takes into account the user's emotions. This embodiment details the specific implementation method of this system.

[0253] System Configuration

[0254] This system consists of a user terminal, a server, and various analysis tools.

[0255] 1. Terminal

[0256] This refers to a device used by a user to input text. Examples include PCs, smartphones, and tablets.

[0257] 2. Server

[0258] It is a central processing unit equipped with a high-performance processor and sufficient memory. The server receives text sent from the user and performs various analyses and processing.

[0259] 3. Analysis method

[0260] The specific software used includes the following:

[0261] Morphological analysis libraries (e.g., MeCab)

[0262] Grammar analysis engine

[0263] Natural Language Understanding (NLU) Models

[0264] Emotion analysis engine

[0265] Operation details

[0266] 1. User text input

[0267] The user enters Japanese text into an input form on their device. For example, they might enter the sentence, "I will be attending tomorrow's meeting, so thank you in advance. I'm very nervous."

[0268] 2. Sending text to the server

[0269] The entered text is sent from the terminal to the server using an HTTP request.

[0270] 3. Server-based analysis

[0271] The server processes the received text sequentially as follows:

[0272] Morphological analysis: This process breaks down text into individual words and analyzes their parts of speech and conjugations.

[0273] For example, "I will be attending tomorrow's meeting, so thank you in advance. I'm very nervous." can be broken down into "Tomorrow / no / kaigi / ni / taku / suru / masu / no / , / yoroshiku / onegai / shimasu / . / toku / kichika / shite / imasu."

[0274] Grammar analysis: Detects grammatical errors and syntactic errors in a sentence.

[0275] Example: Detect that "sanka suru masu" is grammatically incorrect.

[0276] Semantic analysis: Analyzes the meaning of the entire sentence and identifies unnatural expressions.

[0277] Example: It is decided that "onegaishimasu" should be changed to "onegaishimasu".

[0278] Sentiment analysis: Identifying the emotions contained in the text.

[0279] Example: Determine that "I am very nervous" expresses the emotion of "nervousness".

[0280] 4. Amendment and Generation of Emotion Feedback

[0281] The server points out errors based on the analysis results and generates amendments. Also, it generates feedback that takes into account the user's emotions based on emotion analysis.

[0282] Example 1: Correct the error "participate masu" to "participate shimasu".

[0283] Example 2: Correct "onegai shimasu" to "onegai shimasu".

[0284] Example 3: Generate feedback such as "Please relax and approach the meeting" for "I am very nervous".

[0285] 5. Data Return

[0286] Send the amendment and emotion feedback to the terminal as JSON-formatted data.

[0287] 6. Display to User

[0288] The terminal analyzes the received data and displays the amendment and emotion feedback to the user.

[0289] Example: Displayed as follows:

[0290] I will participate in tomorrow's meeting, so please do me a favor. I am very nervous.

[0291] Amendment:

[0292] "I will participate in tomorrow's meeting, so please do me a favor."

[0293] Emotion Feedback:

[0294] "Please attend the meeting in a relaxed manner."

[0295] Examples of prompt sentences

[0296] Examples of the set prompt sentences are shown below:

[0297] "Please input a Japanese sentence, correct the errors, and display feedback considering the sentiment. Example of the input sentence: I will attend tomorrow's meeting, so please do me a favor. I'm very nervous."

[0298] By applying this system, users can quickly and easily correct errors in Japanese text and receive feedback considering the sentiment.

[0299] The flow of the specific process in Example 2 will be described using FIG. 13.

[0300] Step 1:

[0301] This is the procedure for the user to input text.

[0302] Operation: The user inputs Japanese text into the input form on the terminal. For example, input "I will attend tomorrow's meeting, so please do me a favor. I'm very nervous."

[0303] Input: Japanese text.

[0304] Output: The text displayed on the input form on the terminal.

[0305] Step 2:

[0306] This is the procedure for the terminal to send the input text to the server.

[0307] Operation: The terminal sends the input text to the server using an HTTP request.

[0308] Input: Japanese text entered by the user.

[0309] Output: Text data sent to the server.

[0310] Step 3:

[0311] This is the procedure for performing morphological analysis on text received by the server.

[0312] Operation: The server uses a morphological analysis library to break down the text into individual words and analyze the part of speech and conjugation of each word. For example, it breaks down "I will be attending tomorrow's meeting, so thank you in advance. I'm very nervous." into "Tomorrow / no / kaigi / ni / participating / suru / masu / no / , / yoroshiku / onegai / shimasu / . / totemo / kyouki / shite / imasu".

[0313] Input: Text data sent to the server.

[0314] Output: Morphological analysis results (text data broken down into individual words).

[0315] Step 4:

[0316] This is the procedure by which the server performs grammatical analysis of the text.

[0317] Operation: The server uses a grammar analysis engine to detect grammatical errors and syntactic errors in the text. For example, it detects that the expression "sanka surumasu" (to participate) is inappropriate.

[0318] Input: Morphological analysis results.

[0319] Output: Grammar analysis results (data including information on grammatical errors and syntactic errors).

[0320] Step 5:

[0321] This is the procedure by which the server performs semantic analysis of text.

[0322] Operation: The server uses a natural language understanding (NLU) model to analyze the meaning of the entire text and identify unnatural phrasing. For example, it might determine that "onegaishimasu" should be corrected to "onegaishimasu".

[0323] Input: Grammar analysis result.

[0324] Output: Semantic analysis results (data containing information about unnatural phrasing).

[0325] Step 6:

[0326] This is the procedure for the server to perform sentiment analysis on text.

[0327] Operation: Uses an emotion analysis engine to identify the user's emotions contained in the text. For example, it determines that "I'm very nervous" represents the emotion "nervous."

[0328] Input: Semantic analysis results.

[0329] Output: Sentiment analysis results (data including user sentiment information).

[0330] Step 7:

[0331] This is a procedure in which the server performs error detection, proposes corrections, and generates sentiment feedback.

[0332] Operation: Based on the analysis results, the server identifies errors and generates appropriate correction suggestions. It also generates feedback that takes the user's emotions into account based on the sentiment analysis results. For example, it corrects "surumasu" to "shimasu" and "onegaishimasu" to "onegaishimasu". In response to "I'm very nervous," it generates feedback such as "Please relax and approach the meeting with ease."

[0333] Input: Grammatical analysis results, semantic analysis results, sentiment analysis results.

[0334] Output: Revised proposals and emotional feedback.

[0335] Step 8:

[0336] This is the procedure for the server to send suggested revisions and emotional feedback to the terminal.

[0337] Operation: The server generates JSON data containing suggested revisions and sentiment feedback, and sends it back to the terminal.

[0338] Input: Suggested revisions and emotional feedback.

[0339] Output: JSON data sent to the terminal.

[0340] Step 9:

[0341] This is the procedure for the device to display suggested fixes and emotional feedback to the user.

[0342] Operation: The device parses the received JSON data and displays suggested revisions and sentiment feedback to the user. For example, it might display: "I will be attending tomorrow's meeting, thank you in advance. I'm very nervous. Suggested revision: 'I will be attending tomorrow's meeting, thank you in advance.' Sentiment feedback: 'Please relax during the meeting.'"

[0343] Input: JSON data sent to the terminal.

[0344] Output: Suggested revisions and sentiment feedback displayed to the user.

[0345] (Application Example 2)

[0346] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0347] In virtual stores, customers often generate incorrect or unnatural sentences when entering text. Furthermore, there is a lack of means to provide feedback that is appropriate to the customer's current emotional state. This presents a challenge in improving the customer experience.

[0348] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a morphological analysis means, a grammatical analysis means, a semantic analysis means, an emotion analysis means, an error detection means and correction generation means, and a transmission means. This makes it possible not only to correct errors and unnatural expressions, but also to provide feedback based on the customer's emotional state.

[0349] A "user" is a person who uses this system to input text and sends that input to the server.

[0350] "Input means" refers to a device or interface for a user to input text.

[0351] "Communication means" refers to means, including networks and protocols, for sending input text to a server.

[0352] A "server" is a device or system that analyzes text sent by a user and generates suggested revisions and sentiment feedback.

[0353] A "morphological analysis tool" is a function that breaks down input text into words and analyzes the part of speech and conjugation of each word.

[0354] A "grammatical analysis tool" is a function that detects grammatical errors and syntactic errors in the input text.

[0355] A "semantic analysis tool" is a function that analyzes the overall meaning of the input text and identifies unnatural phrasing.

[0356] "Emotion analysis means" refers to a function that analyzes the emotions contained in the input text.

[0357] An "error detection means" is a function that identifies errors in text based on the analysis results.

[0358] A "correction generation means" is a function that generates appropriate corrections for identified errors.

[0359] "Transmission means" refers to the means of sending the generated revisions and emotional feedback to the user's device.

[0360] "Emotional feedback" refers to advice and comments provided based on the emotions contained in the user's text.

[0361] This invention is a system that corrects text entered by customers in a virtual store into accurate and natural sentences, and also provides feedback that takes into account the customer's feelings.

[0362] This system mainly consists of input means, communication means, server, morphological analysis means, grammatical analysis means, semantic analysis means, sentiment analysis means, error detection means, correction generation means, and transmission means.

[0363] First, the user enters text within the virtual store using an input device. This input device could be a smart glasses or a head-mounted display.

[0364] The text entered by the user is sent to the server via a communication method. Network communication methods such as the HTTP protocol are used for this purpose.

[0365] The server analyzes the received text. Specifically, it uses morphological analysis to break down the text into individual words and analyzes the part of speech and conjugation of each word.

[0366] Next, the grammar analysis means uses the analysis results to detect grammatical errors and syntactic errors in the input text. The semantic analysis means also analyzes the meaning of the entire text and identifies unnatural expressions.

[0367] Furthermore, the emotion analysis mechanism analyzes the emotions contained in the input text. Specifically, it uses a generative AI model to extract emotions from the text and generates corresponding feedback.

[0368] Based on the analysis results, the error detection means identifies errors in the text, and the correction generation means generates appropriate corrections and sentiment-based feedback. For example, for the text "I will be attending tomorrow's meeting, so thank you in advance. I'm very nervous," the correction is changed to "I will be attending tomorrow's meeting, so thank you in advance," and sentiment feedback such as "Please relax during the meeting" is provided.

[0369] These suggested revisions and emotional feedback are sent to the user's device via a transmission method. The user can review the submitted feedback and apply revisions as needed, thereby obtaining accurate, natural-sounding Japanese text and emotionally sensitive feedback.

[0370] For example, if the prompt is "I will be attending tomorrow's meeting, so thank you in advance. I'm very nervous," the system will respond by generating a revised version, "I will be attending tomorrow's meeting, so thank you in advance," and emotional feedback, "Please relax during the meeting."

[0371] In this way, this invention improves the accuracy and emotional consideration of text in customer service within virtual stores.

[0372] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0373] Step 1:

[0374] Users enter text within the virtual store using input devices. These input devices include smart glasses and head-mounted displays. For example, the text entered might read, "I will be attending tomorrow's meeting, so I look forward to seeing you there. I'm very nervous."

[0375] Input: Text entered by the user

[0376] Output: Raw text entered into the terminal

[0377] Step 2:

[0378] The terminal sends the entered text to the server using a communication method. This communication method typically involves network communication protocols such as HTTP.

[0379] Input: Raw text entered on the terminal

[0380] Output: Text data sent to the server

[0381] Step 3:

[0382] The server uses morphological analysis to break down the received text into individual words and analyzes the part of speech and conjugation of each word. For this purpose, it uses a morphological analysis library (e.g., Janome). For example, "I will be attending tomorrow's meeting, so thank you in advance. I'm very nervous." is analyzed as "Tomorrow / no / meeting / ni / participating / suru / masu / no / , / yoroshiku / onegai / shimasu / . / totemo / kyouki / shite / imasu".

[0383] Input: Text data sent to the server

[0384] Output: Text data broken down into words

[0385] Step 4:

[0386] The grammar analysis means detects grammatical errors and syntax errors in a sentence based on the results of morphological analysis. For example, a grammar error such as "するます" is detected.

[0387] Input: Text data decomposed into words

[0388] Output: Data with grammar errors identified

[0389] Step 5:

[0390] The semantic analysis means analyzes the overall meaning of the text using a natural language understanding model (NLU) and identifies unnatural expressions. For example, it is identified that "よろしくおねがいします" should be corrected to "お願いします".

[0391] Input: Data with grammar errors identified

[0392] Output: Data with unnatural expressions identified

[0393] Step 6:

[0394] The sentiment analysis means analyzes the sentiment contained in the input text using a generative AI model. For example, it is determined that "とても緊張しています" expresses the sentiment of "nervousness".

[0395] Input: Data with unnatural expressions identified

[0396] Output: Data with sentiment identified

[0397] Step 7:

[0398] The error detection means identifies errors in the text based on the above analysis results. For example, it is identified that "するます" is an error that should be corrected to "します".

[0399] Input: Data with sentiment identified

[0400] Output: Error and correction instruction data

[0401] Step 8:

[0402] The correction generation mechanism generates appropriate correction suggestions for identified errors. It also generates user-friendly feedback based on sentiment analysis results. For example, it might generate a correction suggestion like, "I will be attending tomorrow's meeting, thank you in advance," along with emotional feedback such as, "Please relax during the meeting."

[0403] Input: Error and correction instruction data

[0404] Output: Proposed revisions and sentiment feedback data

[0405] Step 9:

[0406] The server sends the generated revisions and sentiment feedback to the user's device via a transmission method. These are then displayed on the user's device.

[0407] Input: Proposed revisions and sentiment feedback data

[0408] Output: Suggested revisions and sentiment feedback sent to the user's terminal.

[0409] Step 10:

[0410] Users review the proposed revisions and sentiment feedback, and apply the changes as needed. This ensures accurate and natural Japanese text and sentiment-sensitive feedback.

[0411] Input: Correction suggestions and sentiment feedback sent to the user's terminal.

[0412] Output: Final text with corrections and sentiment feedback applied.

[0413] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0414] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0415] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0416] [Second Embodiment]

[0417] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0418] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0419] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0420] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0421] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0422] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0423] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0424] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0425] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0426] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0427] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0428] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0429] The system of the present invention is an automatic proofreading tool that assists users in accurately and naturally inputting Japanese text. In this embodiment, the specific program processing flow will be described.

[0430] First, the user uses their terminal to input Japanese text into the system's text input form. For example, suppose the user enters the following sentence:

[0431] I will be attending tomorrow's meeting, so I look forward to seeing you there.

[0432] Next, the terminal sends an HTTP request to the server using a communication method to send the entered text to the server. At this time, the text data is sent to the server.

[0433] The server begins processing the received text data for analysis. Specifically, the following three analysis methods are used:

[0434] 1. Morphological Analysis: The server uses a morphological analysis library to break down the text into individual words. For example, the input sentence is broken down into "I will be attending tomorrow's meeting, so please take care of me." The part of speech and conjugation of each word are also analyzed.

[0435] 2. Grammar Analysis: The grammar analysis engine detects grammatical errors. In the example above, it detects that the expression "sanka suru masu" is inappropriate.

[0436] 3. Semantic Analysis: A natural language understanding (NLU) model analyzes the meaning of the entire sentence and identifies unnatural phrasing. In this case, it determines that "onegaishimasu" should be changed to "onegaishimasu".

[0437] Based on the analysis results, the server uses error detection means to identify the error locations and correction generation means to generate appropriate correction suggestions. Specific correction suggestions are as follows:

[0438] Error: "surumasu" -> Corrected: "sanka shimasu"

[0439] Error: "onegaishimasu" -> Correction: "onegaishimasu"

[0440] After the revised version is generated, the server sends it back to the terminal as JSON data via the transmission method.

[0441] The terminal parses the received JSON data and displays suggested corrections to the user. For example, the user will see the following:

[0442] I will be attending tomorrow's meeting, so I look forward to seeing you there.

[0443] The user reviews the displayed correction suggestions and applies the corrections as needed. This corrects the errors in the entered Japanese text, resulting in a correct sentence.

[0444] The above describes a specific embodiment for carrying out the present invention. By using this system, users can quickly and easily correct errors in Japanese text.

[0445] The following describes the processing flow.

[0446] Step 1:

[0447] The user enters Japanese text into a text input form on their device. After entering the text, the user prepares to send the text by pressing the send button.

[0448] Step 2:

[0449] The terminal receives text data entered by the user and sends an HTTP request to the server using a communication method. This request contains text data.

[0450] Step 3:

[0451] The server parses the received HTTP request and extracts the text data from the request body. The server then passes the received text to the morphological analysis module.

[0452] Step 4:

[0453] The server's morphological analysis module breaks down text into individual words and analyzes the part of speech and conjugation of each word. This analysis includes the base form and part of speech information of each word.

[0454] Step 5:

[0455] The analyzed word information is passed to the grammar analysis module. The server's grammar analysis module identifies grammatical errors and syntactic errors in the sentence. In this process, ungrammatical verb and auxiliary verb combinations, etc., are detected.

[0456] Step 6:

[0457] Next, the server's semantic analysis module understands the overall context and meaning, detecting unnatural phrasing and misuse. This analysis evaluates the appropriateness of words and expressions.

[0458] Step 7:

[0459] Based on the results of grammatical and semantic analysis, the server identifies errors using error detection and correction generation means and generates appropriate corrections. For example, "sanka suru masu" is corrected to "sanka shimasu," and "onegai shimasu" is corrected to "onegai shimasu."

[0460] Step 8:

[0461] After the proposed correction is generated, the server formats it as JSON data. This data includes the location of the error, the original text, and the proposed correction.

[0462] Step 9:

[0463] The server sends JSON data back to the terminal via a transmission method. The terminal receives this data and processes it for display.

[0464] Step 10:

[0465] The device displays the suggested fixes to the user in a visually easy-to-understand format. The user reviews the suggested fixes and applies them as needed.

[0466] Step 11:

[0467] The user revises the text based on the suggested changes and then performs a final check. After this, the correct Japanese text can be used for other purposes.

[0468] (Example 1)

[0469] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0470] Conventional text input systems have a problem in that they have difficulty automatically detecting grammatical errors and semantic unnaturalness in Japanese text and providing accurate and natural correction suggestions. Therefore, users have to manually check and correct the text, which leads to a decrease in work efficiency.

[0471] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0472] In this invention, the server includes input means for the user to input text, communication means for transmitting the input text to the server, analysis means for performing morphological, grammatical, and semantic analysis of the text received by the server, error detection means and correction generation means for pointing out errors in the text and generating correction suggestions based on the analysis results, transmission means for transmitting the correction suggestions to the terminal, and display means for displaying the correction suggestions received by the terminal to the user. This enables the user to quickly correct errors in the input Japanese text and generate accurate and natural-sounding sentences.

[0473] A "user" is a person who inputs text into a system and utilizes the functions that the system provides.

[0474] "Input means" refers to a device or software that provides an interface for a user to input text.

[0475] "Communication means" refers to a device or software used to transmit text data entered from an input means to a server.

[0476] A "server" is a device or system that receives input text data and performs analysis on it.

[0477] "Analysis means" refers to software or algorithms used to perform morphological analysis, grammatical analysis, and semantic analysis on text received by the server.

[0478] "Error detection means" refers to a device or software that detects errors in text based on the results of analysis performed by the analysis means.

[0479] "Correction generation means" refers to a device or software that generates appropriate corrections for errors detected by error detection means.

[0480] "Transmission means" refers to a device or software used to send the generated revised draft to a terminal.

[0481] "Display means" refers to a device or software that displays the proposed revisions received by the terminal to the user.

[0482] Morphological analysis is a technique that breaks down input text into individual words and analyzes the part of speech and conjugation of each word.

[0483] "Grammar analysis" is a technique for detecting grammatical errors and syntactic errors within text.

[0484] "Semantic analysis" is a technique for understanding the meaning of an entire text and identifying unnatural phrasing or contextual problems.

[0485] A "revised version" is a suggestion for a correct or more appropriate expression, generated based on the detected error.

[0486] A "terminal" refers to a device, such as a computer or smartphone, that a user uses to access a system.

[0487] The present invention is an automated text correction tool that assists users in accurately and naturally inputting Japanese text. Specifically, the user inputs Japanese text into a text input form using a terminal. At this time, the terminal has means for sending the input text to a server.

[0488] The server has morphological, grammatical, and semantic analysis tools to analyze the received text data. First, to perform morphological analysis, it uses a morphological analysis library (e.g., MeCab or Sudachi). Morphological analysis breaks down the text into words and analyzes the part of speech and conjugation of each word. For example, if a user inputs the following sentence:

[0489] I will be attending tomorrow's meeting, so I look forward to seeing you there.

[0490] Morphological analysis breaks down this sentence into "I will be attending tomorrow's meeting, so please take care of me."

[0491] Next, the server uses a grammar analysis engine (e.g., SyntaxNet) to perform grammatical analysis. Grammar analysis detects grammatical errors and syntactic errors in the sentence. In the example above, it detects that "sanka suru masu" is grammatically incorrect.

[0492] Furthermore, to perform semantic analysis, a natural language understanding (NLU) model (e.g., BERT or GPT-3) is used. Semantic analysis analyzes the meaning of the entire text and identifies unnatural phrasing. In this case, it is determined that "onegaishimasu" should be changed to "onegaishimasu".

[0493] Based on the analysis results, the server uses error detection means to identify the error location and correction generation means to generate appropriate correction proposals. Specific correction proposals are as follows:

[0494] Error: "surumasu" -> Corrected: "sanka shimasu"

[0495] Error: "onegaishimasu" -> Correction: "onegaishimasu"

[0496] After the proposed revisions are generated, the server sends them back to the terminal as JSON data via a transmission method. The terminal parses the received JSON data and displays the proposed revisions to the user. For example, the proposed revision might say, "I will be attending tomorrow's meeting, thank you in advance."

[0497] The user reviews the displayed correction suggestions and applies the corrections as needed. This corrects the errors in the entered Japanese text, resulting in a correct sentence.

[0498] Here are some examples of specific prompt messages:

[0499] Please identify the error in the following Japanese sentence and provide the correct sentence:

[0500] I will be attending tomorrow's meeting, so I look forward to seeing you there.

[0501] The above describes a specific embodiment for carrying out the present invention. By using this system, users can quickly and easily correct errors in Japanese text.

[0502] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0503] Step 1:

[0504] The user uses their own device to input Japanese text into the system's text input form. They use a keyboard or other input method, and the text data is recorded within the device. For example, the user might input "I will be attending tomorrow's meeting, thank you in advance."

[0505] Step 2:

[0506] The terminal sends the entered text data to the server. Specifically, it generates an HTTP request using a communication method and sends the request containing the text data to the server. The input is the text data entered by the user, and the output is the transmission of the HTTP request.

[0507] Step 3:

[0508] The server takes in the received HTTP request and extracts the text data. At this point, the input is the text data contained in the body of the HTTP request, and the output is the extracted text data.

[0509] Step 4:

[0510] The server uses a morphological analysis library (e.g., MeCab or Sudachi) to break down text data into individual words. Specifically, it starts the analysis process and passes the text data to the analysis library. The input is the extracted text data, and the output is the morphologically analyzed data (e.g., "I will be attending tomorrow's meeting, so please take care of me").

[0511] Step 5:

[0512] The server uses a grammar analysis engine (e.g., SyntaxNet) to detect grammatical and syntactic errors in the morphologically analyzed data. Specifically, it activates the grammar analysis tool and passes the analysis data. The input is the morphologically analyzed data, and the output is the identification of grammatical errors (e.g., errors such as "sanka suru masu").

[0513] Step 6:

[0514] The server uses a natural language understanding (NLU) model (e.g., BERT or GPT-3) to analyze the meaning of the entire text. Specifically, it activates a semantic analysis tool and passes the morphologically analyzed data to the NLU model. The input is the morphologically analyzed data, and the output is semantic errors or suggestions for improvement (e.g., a judgment that "onegaishimasu" should be corrected to "onegaishimasu").

[0515] Step 7:

[0516] Based on the analysis results, the server records the errors identified using error detection means and generates appropriate correction suggestions using correction generation means. Specifically, it extracts the error locations through error detection means and creates correction suggestions using correction generation means. The input is the result of grammatical and semantic analysis, and the output is the correction suggestion (e.g., "surumasu" → "sankashimasu", "onegaishimasu" → "onegaishimasu").

[0517] Step 8:

[0518] After the proposed revisions are generated, the server sends them back to the terminal as JSON data using a transmission method. Specifically, the server converts the proposed revisions to JSON format and includes them in the HTTP response before sending. The input is the proposed revisions, and the output is the transmission of JSON data.

[0519] Step 9:

[0520] The terminal parses the received JSON data and displays suggested corrections to the user. Specifically, it uses a display mechanism to visually convert the JSON data and display it on the user interface. The input is data in JSON format, and the output is a suggested correction displayed to the user (e.g., "I will be attending tomorrow's meeting, thank you in advance").

[0521] Step 10:

[0522] The user reviews the displayed suggested revisions and applies them as needed. Specifically, the user confirms the text revisions by clicking the "Apply" button. The input is the user's click operation, and the output is the text with the revisions applied.

[0523] (Application Example 1)

[0524] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0525] When ordering food delivery, users often include typos, grammatical errors, or unnatural phrasing in their special requests and order details, which can lead to misdeliveries and a poor customer experience. A system is needed to solve this problem and allow users to confirm their orders in more accurate and natural Japanese.

[0526] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0527] In this invention, the server includes input means for the user to input text, communication means for transmitting the input text to the server, analysis means for performing morphological, grammatical, and semantic analysis of the text received by the server, error detection means and correction generation means for pointing out errors in the text and generating correction suggestions based on the analysis results, transmission means for transmitting the correction suggestions to the terminal, and means featuring an automatic proofreading function for food delivery applications to correct typos, omissions, and unnatural expressions made by the user when inputting order details and special notes. This enables the user to quickly correct the content they have entered and place accurate orders.

[0528] "Input means" refers to a device or interface for a user to input text.

[0529] "Communication means" refers to a device or protocol used to send input text to a server.

[0530] "Analysis means" refers to software or hardware mechanisms that perform morphological analysis, grammatical analysis, and semantic analysis on text received by the server.

[0531] An "error detection means" is a device or program that identifies errors in text based on the analysis results.

[0532] A "correction generation means" is a device or program for generating appropriate corrections after an error has been identified.

[0533] "Transmission means" refers to a device or protocol for sending the revised version to a terminal.

[0534] The "Automatic Correction Function for Food Delivery Applications" is a feature that corrects typos, grammatical errors, and unnatural expressions when users enter order details or special notes.

[0535] The present invention provides an automated text correction tool that assists users in accurately and naturally inputting Japanese text. In this invention, the system is particularly applicable to food delivery applications.

[0536] 1. System Program

[0537] This system consists of the following main elements:

[0538] Input method: An interface for users to enter text. A text input box embedded within a food delivery application.

[0539] Communication method: An HTTP request to send the entered text to the server. Specifically, the text is sent to the server in JSON format.

[0540] Analysis methods: Functions for morphological analysis, grammatical analysis, and semantic analysis of text received by the server. Morphological analysis uses the Janome library, and grammatical and semantic analysis utilize natural language understanding (NLU) models.

[0541] Error detection and correction generation means: Based on the analysis results, errors in the text are detected and appropriate correction suggestions are generated. Examples of errors include incorrect particles and unnatural expressions.

[0542] Transmission method: An HTTP response to send JSON data containing the proposed revisions back to the terminal.

[0543] 2. Program processing description using natural language

[0544] The server receives text sent by the user through the input method. The Janome library is used to perform morphological analysis on the received text, breaking down the sentence into individual words. For example, the input sentence "I will be attending tomorrow's meeting, so thank you in advance" is broken down by morphological analysis into "tomorrow / no / kaigi / ni / participating / suru / masu / node / , / yoroshiku / onegai / shimasu".

[0545] Next, the grammar analysis engine analyzes the part of speech and conjugation of each word to detect grammatical errors. For example, it detects that the expression "sanka suru masu" (to participate) is inappropriate. Subsequently, in the semantic analysis stage, a natural language understanding (NLU) model is used to analyze the meaning of the entire sentence and identify unnatural phrasing. Here, it determines that "onegai shimasu" (please) should be changed to "onegai shimasu" (please).

[0546] Based on the analysis results, the error detection means detects errors, and the correction generation means generates appropriate correction suggestions. The server then sends the correction suggestions back to the terminal as JSON data. The terminal analyzes the received JSON data and displays the correction suggestions to the user. The user reviews the displayed correction suggestions and applies the corrections as needed.

[0547] 3. Specific examples and prompt messages

[0548] For example, a user enters the following order details:

[0549] I am allergic to shrimp, so please remove the shrimp from the dish.

[0550] When this input is sent to the server, it will suggest the following modifications:

[0551] I am allergic to shrimp, so please omit the shrimp from the dish.

[0552] Examples of prompt statements to input into a generative AI model are as follows:

[0553] Analyze the Japanese text entered by the user grammatically and semantically, and suggest appropriate corrections. Example: 'I am allergic to shrimp, so please remove the shrimp.' -> 'I am allergic to shrimp, so please remove the shrimp.'

[0554] This allows users to quickly correct their input and place orders in accurate and natural-sounding Japanese.

[0555] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0556] Step 1:

[0557] The user uses their device to enter order details and special requests into a text input form within the food delivery application. An example of the text entered is, "I am allergic to shrimp, so please omit the shrimp."

[0558] Step 2:

[0559] The terminal sends an HTTP request to the server using a communication method to send the entered text to the server. Specifically, it sends the text data to the server in JSON format.

[0560] Step 3:

[0561] The server analyzes the received text data. First, the server uses the morphological analysis library Janome to break down the text into words. For example, it breaks down "I am allergic to shrimp, so please remove the shrimp." into "Allergy / is / shrimp / , / shrimp / remove / please / please / you / please / please / remove / shrimp".

[0562] Step 4:

[0563] The server uses a grammar analysis engine to analyze the morphologically analyzed word sequence from the perspective of part of speech and conjugation, and detects grammatical errors. For example, it might determine that the unnatural expression "kudasaimasu you onegaishimasu" should be corrected to "onegaishimasu".

[0564] Step 5:

[0565] The server uses a natural language understanding (NLU) model to perform semantic analysis. Here, it analyzes the meaning of the entire sentence and identifies unnatural phrasing and expressions. For example, it might correct "onegaishimasu" to "onegaishimasu."

[0566] Step 6:

[0567] Based on the analysis results, the server uses error detection to identify errors and correction generation to generate appropriate corrections. Specifically, it generates a correction to change "onegaishimasu" to "onegaishimasu".

[0568] Step 7:

[0569] The server sends the proposed revisions back to the terminal as JSON data via a transmission method. The transmitted data includes both the original and revised text.

[0570] Step 8:

[0571] The terminal parses the received JSON data and displays suggested corrections to the user. For example, it might display "I am allergic to shrimp, so please remove the shrimp."

[0572] Step 9:

[0573] The user reviews the displayed revisions and applies them as needed. This confirms that the corrected text is finalized in the order.

[0574] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0575] The present invention is an automated proofreading tool that allows users to input Japanese text accurately and naturally, and provides feedback that takes the user's feelings into consideration. In this embodiment, the specific program processing flow will be described.

[0576] First, the user enters a Japanese sentence into a text input form on their device. For example, suppose the user enters the following sentence:

[0577] I will be attending tomorrow's meeting, so I look forward to seeing you there. I'm very nervous.

[0578] Next, the terminal sends an HTTP request to the server using a communication method to send the entered text to the server. At this time, the text data is sent to the server.

[0579] The server begins processing the received text data for analysis. Specifically, the following four analysis methods are used:

[0580] 1. Morphological Analysis: The server uses a morphological analysis library to break down the text into individual words. For example, the input sentence is broken down into "I will be attending tomorrow's meeting, so please be kind to me. I am very nervous." The part of speech and conjugation of each word are also analyzed.

[0581] 2. Grammar Analysis: The grammar analysis engine detects grammatical errors. In the example above, it detects that the expression "sanka suru masu" is inappropriate.

[0582] 3. Semantic Analysis: A natural language understanding (NLU) model analyzes the meaning of the entire sentence and identifies unnatural phrasing. In this case, it determines that "onegaishimasu" should be changed to "onegaishimasu".

[0583] 4. Sentiment Analysis: The server's sentiment analysis tool identifies the emotions contained in the user's text. This analysis determines that "I'm very nervous" represents the emotion of "nervousness."

[0584] Based on the analysis results, the server uses error detection and correction generation means to identify errors and generate appropriate correction suggestions. Specific correction suggestions are as follows:

[0585] Error: "surumasu" -> Corrected: "sanka shimasu"

[0586] Error: "onegaishimasu" -> Correction: "onegaishimasu"

[0587] Furthermore, the emotion engine generates user-friendly feedback based on identified emotions. For example, in response to "I'm very nervous," it might generate advice such as "Try to relax before the meeting."

[0588] After the proposed revisions and sentiment feedback are generated, the server sends them back to the terminal as JSON data via a transmission method.

[0589] The device parses the received JSON data and displays suggested revisions and sentiment feedback to the user. For example, the user might see the following:

[0590] I will be attending tomorrow's meeting, so I look forward to seeing you there. I'm very nervous.

[0591] Proposed amendment:

[0592] "I will be attending tomorrow's meeting, so I look forward to seeing you there."

[0593] Emotional feedback:

[0594] "Please relax and approach the meeting with ease."

[0595] Users review the displayed correction suggestions and sentiment feedback, and apply the corrections as needed. This corrects errors in the entered Japanese text and allows them to receive feedback that takes their feelings into consideration.

[0596] The above describes a specific embodiment for carrying out the present invention. By using this system, users can quickly and easily correct errors in Japanese text and receive emotionally sensitive feedback.

[0597] The following describes the processing flow.

[0598] Step 1:

[0599] The user enters a Japanese sentence into a text input form on their device. For example, they might enter, "I will be attending tomorrow's meeting, so I look forward to seeing you there. I'm very nervous." Once they have finished typing, they click the submit button.

[0600] Step 2:

[0601] The terminal receives the input text data and uses a communication method to send an HTTP request to the server. This request contains the text data.

[0602] Step 3:

[0603] The server parses the received HTTP request and extracts text data. The extracted text data is then passed to the morphological analysis module.

[0604] Step 4:

[0605] The server's morphological analysis module breaks down text into individual words and analyzes the part of speech and conjugation of each word. For example, the text "I will be attending tomorrow's meeting, so I look forward to seeing you. I'm very nervous." is broken down into "tomorrow / noun of / particle meeting / noun to / particle attend / verb do / auxiliary verb do / auxiliary verb so / particle, / symbol nice to see / adverb please / verb do / auxiliary verb. / symbol very / adverb nervous / noun feeling / auxiliary verb I am / auxiliary verb".

[0606] Step 5:

[0607] The analyzed word information is passed to the grammar analysis module. The server's grammar analysis module identifies grammatical errors and syntactic errors in the sentence. At this stage, the incorrect expression "sanka suru masu" (participate) is detected.

[0608] Step 6:

[0609] The results of the grammatical analysis are passed to the semantic analysis module. The server's semantic analysis module understands the context and identifies unnatural phrasing and misuse. This analysis determines that "onegaishimasu" should be corrected to "onegaishimasu".

[0610] Step 7:

[0611] Next, the server's sentiment analysis tool identifies the emotions contained in the user's text. It recognizes the emotion of tension from the expression "I'm very nervous."

[0612] Step 8:

[0613] Based on the results of grammatical analysis, semantic analysis, and sentiment analysis, the server uses error detection and correction generation means to identify errors and generate appropriate correction suggestions and sentiment feedback. For example, "participate" is corrected to "will participate," and "please" is corrected to "please." In addition, sentiment feedback such as "Please relax and enjoy the meeting" is generated for a nervous user.

[0614] Step 9:

[0615] The generated corrections and sentiment feedback are formatted in JSON format. This data includes the location of the error, the original text, the correction, and the sentiment feedback. The server sends this JSON data back to the terminal via the transmission method.

[0616] Step 10:

[0617] The device parses the received JSON data and displays suggested revisions and sentiment feedback to the user. For example, it may be displayed as follows:

[0618] I will be attending tomorrow's meeting, so I look forward to seeing you there. I'm very nervous.

[0619] Proposed amendment:

[0620] "I will be attending tomorrow's meeting, so I look forward to seeing you there."

[0621] Emotional feedback:

[0622] "Please relax and approach the meeting with ease."

[0623] Step 11:

[0624] Users review the displayed correction suggestions and sentiment feedback, and apply the corrections as needed. This corrects errors in the entered Japanese text and allows them to receive feedback that takes their feelings into consideration.

[0625] (Example 2)

[0626] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0627] When users input Japanese text, the challenge lies in accurately identifying and correcting errors, as well as providing feedback that takes the user's emotions into consideration. Conventional systems only point out errors and fail to provide emotionally sensitive feedback, thus failing to sufficiently improve user satisfaction and the naturalness of the text.

[0628] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0629] In this invention, the server includes analysis means for performing morphological, grammatical, semantic, and sentiment analysis on received text; error detection and correction generation means for pointing out errors in the text, generating suggested corrections, and providing feedback that takes the user's feelings into consideration, based on the analysis results; and transmission means for sending the suggested corrections and sentiment feedback to the terminal. This allows the user to correct errors while simultaneously receiving feedback that takes their feelings into consideration.

[0630] "Text" refers to a string of characters entered by the user, and may include Japanese sentences or words.

[0631] "Input means" refers to devices or interfaces for users to input text, including keyboards and touchscreens.

[0632] "Communication method" refers to the means by which a terminal sends text entered into it to a server, and this involves using an internet connection or a communication protocol (e.g., HTTP).

[0633] A "server" is a central processing unit that receives, analyzes, and processes text sent from a user's terminal.

[0634] "Analysis means" includes software and libraries for morphological analysis, grammatical analysis, semantic analysis, and sentiment analysis of received text.

[0635] Morphological analysis is the process of breaking down text into individual words and analyzing the part of speech and conjugation of each word.

[0636] "Grammar analysis" is the process of detecting grammatical errors and syntactic errors within a text.

[0637] "Semantic analysis" is the process of understanding the overall meaning of a text and identifying unnatural expressions.

[0638] "Sentiment analysis" is the process of identifying the user's emotions contained in text and generating appropriate feedback.

[0639] "Error detection means and correction generation means" refers to processing means for identifying errors in text based on analysis results and generating proposed corrections.

[0640] "Transmission means" refers to the means by which the server sends back the proposed revisions and sentiment feedback generated by the server to the terminal.

[0641] A "revision proposal" is a suggestion to correct errors detected by the analysis method.

[0642] "Emotional feedback" refers to providing advice and comments that take the user's emotions into consideration, based on emotion analysis.

[0643] The present invention is an automated proofreading tool that allows users to input Japanese text accurately and naturally, and provides feedback that takes into account the user's emotions. This embodiment details the specific implementation method of this system.

[0644] System Configuration

[0645] This system consists of a user terminal, a server, and various analysis tools.

[0646] 1. Terminal

[0647] This refers to a device used by a user to input text. Examples include PCs, smartphones, and tablets.

[0648] 2. Server

[0649] It is a central processing unit equipped with a high-performance processor and sufficient memory. The server receives text sent from the user and performs various analyses and processing.

[0650] 3. Analysis method

[0651] The specific software used includes the following:

[0652] Morphological analysis libraries (e.g., MeCab)

[0653] Grammar analysis engine

[0654] Natural Language Understanding (NLU) Models

[0655] Emotion analysis engine

[0656] Operation details

[0657] 1. User text input

[0658] The user enters Japanese text into an input form on their device. For example, they might enter the sentence, "I will be attending tomorrow's meeting, so thank you in advance. I'm very nervous."

[0659] 2. Sending text to the server

[0660] The entered text is sent from the terminal to the server using an HTTP request.

[0661] 3. Server-based analysis

[0662] The server processes the received text sequentially as follows:

[0663] Morphological analysis: This process breaks down text into individual words and analyzes their parts of speech and conjugations.

[0664] For example, "I will be attending tomorrow's meeting, so thank you in advance. I'm very nervous." can be broken down into "Tomorrow / no / kaigi / ni / taku / suru / masu / no / , / yoroshiku / onegai / shimasu / . / toku / kichika / shite / imasu."

[0665] Grammar analysis: Detects grammatical errors and syntactic errors in a sentence.

[0666] Example: Detect that "sanka suru masu" is grammatically incorrect.

[0667] Semantic analysis: Analyzes the meaning of the entire sentence and identifies unnatural expressions.

[0668] Example: It is decided that "onegaishimasu" should be changed to "onegaishimasu".

[0669] Sentiment analysis: Identifying the emotions contained in the text.

[0670] Example: Determine that "I am very nervous" expresses the emotion of "nervousness".

[0671] 4. Generating revised proposals and emotional feedback

[0672] The server identifies errors based on the analysis results and generates suggested corrections. It also generates feedback that takes the user's emotions into consideration based on sentiment analysis.

[0673] Example 1: Correct the error "sanka suru masu" to "sanka shimasu".

[0674] Example 2: Change "onegaishimasu" to "onegaishimasu".

[0675] Example 3: In response to "I'm very nervous," generate feedback such as "Please relax and approach the meeting with confidence."

[0676] 5. Return of data

[0677] The proposed revisions and emotional feedback will be sent to the device as data in JSON format.

[0678] 6. Display to the user

[0679] The device analyzes the received data and displays suggested revisions and sentiment feedback to the user.

[0680] Example: It will be displayed as follows:

[0681] I will be attending tomorrow's meeting, so I look forward to seeing you there. I'm very nervous.

[0682] Proposed amendment:

[0683] "I will be attending tomorrow's meeting, so I look forward to seeing you there."

[0684] Emotional feedback:

[0685] "Please relax and approach the meeting with ease."

[0686] Example of a prompt

[0687] The following is an example of a set prompt statement:

[0688] Please enter a Japanese sentence, correct any errors, and provide emotionally sensitive feedback. Example sentence: I will be attending tomorrow's meeting, so I look forward to seeing you there. I'm very nervous.

[0689] By applying this system, users can quickly and easily correct errors in Japanese text and receive emotionally sensitive feedback.

[0690] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0691] Step 1:

[0692] The user enters text.

[0693] Operation: The user enters Japanese text into an input form on the terminal. For example, they might enter, "I will be attending tomorrow's meeting, thank you in advance. I'm very nervous."

[0694] Input: Japanese text.

[0695] Output: The text displayed in the input form on the terminal.

[0696] Step 2:

[0697] This is the procedure for sending the text entered by the terminal to the server.

[0698] Operation: The terminal sends the entered text to the server using an HTTP request.

[0699] Input: Japanese text entered by the user.

[0700] Output: Text data sent to the server.

[0701] Step 3:

[0702] This is the procedure for performing morphological analysis on text received by the server.

[0703] Operation: The server uses a morphological analysis library to break down the text into individual words and analyze the part of speech and conjugation of each word. For example, it breaks down "I will be attending tomorrow's meeting, so thank you in advance. I'm very nervous." into "Tomorrow / no / kaigi / ni / participating / suru / masu / no / , / yoroshiku / onegai / shimasu / . / totemo / kyouki / shite / imasu".

[0704] Input: Text data sent to the server.

[0705] Output: Morphological analysis results (text data broken down into individual words).

[0706] Step 4:

[0707] This is the procedure by which the server performs grammatical analysis of the text.

[0708] Operation: The server uses a grammar analysis engine to detect grammatical errors and syntactic errors in the text. For example, it detects that the expression "sanka surumasu" (to participate) is inappropriate.

[0709] Input: Morphological analysis results.

[0710] Output: Grammar analysis results (data including information on grammatical errors and syntactic errors).

[0711] Step 5:

[0712] This is the procedure by which the server performs semantic analysis of text.

[0713] Operation: The server uses a natural language understanding (NLU) model to analyze the meaning of the entire text and identify unnatural phrasing. For example, it might determine that "onegaishimasu" should be corrected to "onegaishimasu".

[0714] Input: Grammar analysis result.

[0715] Output: Semantic analysis results (data containing information about unnatural phrasing).

[0716] Step 6:

[0717] This is the procedure for the server to perform sentiment analysis on text.

[0718] Operation: Uses an emotion analysis engine to identify the user's emotions contained in the text. For example, it determines that "I'm very nervous" represents the emotion "nervous."

[0719] Input: Semantic analysis results.

[0720] Output: Sentiment analysis results (data including user sentiment information).

[0721] Step 7:

[0722] This is a procedure in which the server performs error detection, proposes corrections, and generates sentiment feedback.

[0723] Operation: Based on the analysis results, the server identifies errors and generates appropriate correction suggestions. It also generates feedback that takes the user's emotions into account based on the sentiment analysis results. For example, it corrects "surumasu" to "shimasu" and "onegaishimasu" to "onegaishimasu". In response to "I'm very nervous," it generates feedback such as "Please relax and approach the meeting with ease."

[0724] Input: Grammatical analysis results, semantic analysis results, sentiment analysis results.

[0725] Output: Revised proposals and emotional feedback.

[0726] Step 8:

[0727] This is the procedure for the server to send suggested revisions and emotional feedback to the terminal.

[0728] Operation: The server generates JSON data containing suggested revisions and sentiment feedback, and sends it back to the terminal.

[0729] Input: Suggested revisions and emotional feedback.

[0730] Output: JSON data sent to the terminal.

[0731] Step 9:

[0732] This is the procedure for the device to display suggested fixes and emotional feedback to the user.

[0733] Operation: The device parses the received JSON data and displays suggested revisions and sentiment feedback to the user. For example, it might display: "I will be attending tomorrow's meeting, thank you in advance. I'm very nervous. Suggested revision: 'I will be attending tomorrow's meeting, thank you in advance.' Sentiment feedback: 'Please relax during the meeting.'"

[0734] Input: JSON data sent to the terminal.

[0735] Output: Suggested revisions and sentiment feedback displayed to the user.

[0736] (Application Example 2)

[0737] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0738] In virtual stores, customers often generate incorrect or unnatural sentences when entering text. Furthermore, there is a lack of means to provide feedback that is appropriate to the customer's current emotional state. This presents a challenge in improving the customer experience.

[0739] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a morphological analysis means, a grammatical analysis means, a semantic analysis means, an emotion analysis means, an error detection means and correction generation means, and a transmission means. This makes it possible not only to correct errors and unnatural expressions, but also to provide feedback based on the customer's emotional state.

[0740] A "user" is a person who uses this system to input text and sends that input to the server.

[0741] "Input means" refers to a device or interface for a user to input text.

[0742] "Communication means" refers to means, including networks and protocols, for sending input text to a server.

[0743] A "server" is a device or system that analyzes text sent by a user and generates suggested revisions and sentiment feedback.

[0744] A "morphological analysis tool" is a function that breaks down input text into words and analyzes the part of speech and conjugation of each word.

[0745] A "grammatical analysis tool" is a function that detects grammatical errors and syntactic errors in the input text.

[0746] A "semantic analysis tool" is a function that analyzes the overall meaning of the input text and identifies unnatural phrasing.

[0747] "Emotion analysis means" refers to a function that analyzes the emotions contained in the input text.

[0748] An "error detection means" is a function that identifies errors in text based on the analysis results.

[0749] A "correction generation means" is a function that generates appropriate corrections for identified errors.

[0750] "Transmission means" refers to the means of sending the generated revisions and emotional feedback to the user's device.

[0751] "Emotional feedback" refers to advice and comments provided based on the emotions contained in the user's text.

[0752] This invention is a system that corrects text entered by customers in a virtual store into accurate and natural sentences, and also provides feedback that takes into account the customer's feelings.

[0753] This system mainly consists of input means, communication means, server, morphological analysis means, grammatical analysis means, semantic analysis means, sentiment analysis means, error detection means, correction generation means, and transmission means.

[0754] First, the user enters text within the virtual store using an input device. This input device could be a smart glasses or a head-mounted display.

[0755] The text entered by the user is sent to the server via a communication method. Network communication methods such as the HTTP protocol are used for this purpose.

[0756] The server analyzes the received text. Specifically, it uses morphological analysis to break down the text into individual words and analyzes the part of speech and conjugation of each word.

[0757] Next, the grammar analysis means uses the analysis results to detect grammatical errors and syntactic errors in the input text. The semantic analysis means also analyzes the meaning of the entire text and identifies unnatural expressions.

[0758] Furthermore, the emotion analysis mechanism analyzes the emotions contained in the input text. Specifically, it uses a generative AI model to extract emotions from the text and generates corresponding feedback.

[0759] Based on the analysis results, the error detection means identifies errors in the text, and the correction generation means generates appropriate corrections and sentiment-based feedback. For example, for the text "I will be attending tomorrow's meeting, so thank you in advance. I'm very nervous," the correction is changed to "I will be attending tomorrow's meeting, so thank you in advance," and sentiment feedback such as "Please relax during the meeting" is provided.

[0760] These suggested revisions and emotional feedback are sent to the user's device via a transmission method. The user can review the submitted feedback and apply revisions as needed, thereby obtaining accurate, natural-sounding Japanese text and emotionally sensitive feedback.

[0761] For example, if the prompt is "I will be attending tomorrow's meeting, so thank you in advance. I'm very nervous," the system will respond by generating a revised version, "I will be attending tomorrow's meeting, so thank you in advance," and emotional feedback, "Please relax during the meeting."

[0762] In this way, this invention improves the accuracy and emotional consideration of text in customer service within virtual stores.

[0763] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0764] Step 1:

[0765] Users enter text within the virtual store using input devices. These input devices include smart glasses and head-mounted displays. For example, the text entered might read, "I will be attending tomorrow's meeting, so I look forward to seeing you there. I'm very nervous."

[0766] Input: Text entered by the user

[0767] Output: Raw text entered into the terminal

[0768] Step 2:

[0769] The terminal sends the entered text to the server using a communication method. This communication method typically involves network communication protocols such as HTTP.

[0770] Input: Raw text entered on the terminal

[0771] Output: Text data sent to the server

[0772] Step 3:

[0773] The server uses morphological analysis to break down the received text into individual words and analyzes the part of speech and conjugation of each word. For this purpose, it uses a morphological analysis library (e.g., Janome). For example, "I will be attending tomorrow's meeting, so thank you in advance. I'm very nervous." is analyzed as "Tomorrow / no / meeting / ni / participating / suru / masu / no / , / yoroshiku / onegai / shimasu / . / totemo / kyouki / shite / imasu".

[0774] Input: Text data sent to the server

[0775] Output: Text data broken down into words

[0776] Step 4:

[0777] The grammar analysis method detects grammatical errors and syntax errors in a sentence based on the results of morphological analysis. For example, a grammar error such as "するます" is detected.

[0778] Input: Text data decomposed into words

[0779] Output: Data with grammar errors identified

[0780] Step 5:

[0781] The semantic analysis method uses a natural language understanding model (NLU) to analyze the meaning of the entire text and identify unnatural expressions. For example, it is identified that "よろしくおねがいします" should be corrected to "お願いします".

[0782] Input: Data with grammar errors identified

[0783] Output: Data with unnatural expressions identified

[0784] Step 6:

[0785] The sentiment analysis method uses a generative AI model to analyze the sentiment contained in the input text. For example, it is determined that "とても緊張しています" expresses the sentiment of "nervousness".

[0786] Input: Data with unnatural expressions identified

[0787] Output: Data with sentiment identified

[0788] Step 7:

[0789] The error detection method identifies errors in the text based on the above analysis results. For example, it is identified that "するます" is an error that should be corrected to "します".

[0790] Input: Data with sentiment identified

[0791] Output: Error and correction instruction data

[0792] Step 8:

[0793] The correction generation mechanism generates appropriate correction suggestions for identified errors. It also generates user-friendly feedback based on sentiment analysis results. For example, it might generate a correction suggestion like, "I will be attending tomorrow's meeting, thank you in advance," along with emotional feedback such as, "Please relax during the meeting."

[0794] Input: Error and correction instruction data

[0795] Output: Proposed revisions and sentiment feedback data

[0796] Step 9:

[0797] The server sends the generated revisions and sentiment feedback to the user's device via a transmission method. These are then displayed on the user's device.

[0798] Input: Proposed revisions and sentiment feedback data

[0799] Output: Suggested revisions and sentiment feedback sent to the user's terminal.

[0800] Step 10:

[0801] Users review the proposed revisions and sentiment feedback, and apply the changes as needed. This ensures accurate and natural Japanese text and sentiment-sensitive feedback.

[0802] Input: Correction suggestions and sentiment feedback sent to the user's terminal.

[0803] Output: Final text with corrections and sentiment feedback applied.

[0804] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0805] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0806] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0807] [Third Embodiment]

[0808] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0809] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0810] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0811] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0812] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0813] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0814] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0815] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0816] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0817] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0818] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0819] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0820] The system of the present invention is an automatic proofreading tool that assists users in accurately and naturally inputting Japanese text. In this embodiment, the specific program processing flow will be described.

[0821] First, the user uses their terminal to input Japanese text into the system's text input form. For example, suppose the user enters the following sentence:

[0822] I will be attending tomorrow's meeting, so I look forward to seeing you there.

[0823] Next, the terminal sends an HTTP request to the server using a communication method to send the entered text to the server. At this time, the text data is sent to the server.

[0824] The server begins processing the received text data for analysis. Specifically, the following three analysis methods are used:

[0825] 1. Morphological Analysis: The server uses a morphological analysis library to break down the text into individual words. For example, the input sentence is broken down into "I will be attending tomorrow's meeting, so please take care of me." The part of speech and conjugation of each word are also analyzed.

[0826] 2. Grammar Analysis: The grammar analysis engine detects grammatical errors. In the example above, it detects that the expression "sanka suru masu" is inappropriate.

[0827] 3. Semantic Analysis: A natural language understanding (NLU) model analyzes the meaning of the entire sentence and identifies unnatural phrasing. In this case, it determines that "onegaishimasu" should be changed to "onegaishimasu".

[0828] Based on the analysis results, the server uses error detection means to identify the error locations and correction generation means to generate appropriate correction suggestions. Specific correction suggestions are as follows:

[0829] Error: "surumasu" -> Corrected: "sanka shimasu"

[0830] Error: "onegaishimasu" -> Correction: "onegaishimasu"

[0831] After the revised version is generated, the server sends it back to the terminal as JSON data via the transmission method.

[0832] The terminal parses the received JSON data and displays suggested corrections to the user. For example, the user will see the following:

[0833] I will be attending tomorrow's meeting, so I look forward to seeing you there.

[0834] The user reviews the displayed correction suggestions and applies the corrections as needed. This corrects the errors in the entered Japanese text, resulting in a correct sentence.

[0835] The above describes a specific embodiment for carrying out the present invention. By using this system, users can quickly and easily correct errors in Japanese text.

[0836] The following describes the processing flow.

[0837] Step 1:

[0838] The user enters Japanese text into a text input form on their device. After entering the text, the user prepares to send the text by pressing the send button.

[0839] Step 2:

[0840] The terminal receives text data entered by the user and sends an HTTP request to the server using a communication method. This request contains text data.

[0841] Step 3:

[0842] The server parses the received HTTP request and extracts the text data from the request body. The server then passes the received text to the morphological analysis module.

[0843] Step 4:

[0844] The server's morphological analysis module breaks down text into individual words and analyzes the part of speech and conjugation of each word. This analysis includes the base form and part of speech information of each word.

[0845] Step 5:

[0846] The analyzed word information is passed to the grammar analysis module. The server's grammar analysis module identifies grammatical errors and syntactic errors in the sentence. In this process, ungrammatical verb and auxiliary verb combinations, etc., are detected.

[0847] Step 6:

[0848] Next, the server's semantic analysis module understands the overall context and meaning, detecting unnatural phrasing and misuse. This analysis evaluates the appropriateness of words and expressions.

[0849] Step 7:

[0850] Based on the results of grammatical and semantic analysis, the server identifies errors using error detection and correction generation means and generates appropriate corrections. For example, "sanka suru masu" is corrected to "sanka shimasu," and "onegai shimasu" is corrected to "onegai shimasu."

[0851] Step 8:

[0852] After the proposed correction is generated, the server formats it as JSON data. This data includes the location of the error, the original text, and the proposed correction.

[0853] Step 9:

[0854] The server sends JSON data back to the terminal via a transmission method. The terminal receives this data and processes it for display.

[0855] Step 10:

[0856] The device displays the suggested fixes to the user in a visually easy-to-understand format. The user reviews the suggested fixes and applies them as needed.

[0857] Step 11:

[0858] The user revises the text based on the suggested changes and then performs a final check. After this, the correct Japanese text can be used for other purposes.

[0859] (Example 1)

[0860] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0861] Conventional text input systems have a problem in that they have difficulty automatically detecting grammatical errors and semantic unnaturalness in Japanese text and providing accurate and natural correction suggestions. Therefore, users have to manually check and correct the text, which leads to a decrease in work efficiency.

[0862] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0863] In this invention, the server includes input means for the user to input text, communication means for transmitting the input text to the server, analysis means for performing morphological, grammatical, and semantic analysis of the text received by the server, error detection means and correction generation means for pointing out errors in the text and generating correction suggestions based on the analysis results, transmission means for transmitting the correction suggestions to the terminal, and display means for displaying the correction suggestions received by the terminal to the user. This enables the user to quickly correct errors in the input Japanese text and generate accurate and natural-sounding sentences.

[0864] A "user" is a person who inputs text into a system and utilizes the functions that the system provides.

[0865] "Input means" refers to a device or software that provides an interface for a user to input text.

[0866] "Communication means" refers to a device or software used to transmit text data entered from an input means to a server.

[0867] A "server" is a device or system that receives input text data and performs analysis on it.

[0868] "Analysis means" refers to software or algorithms used to perform morphological analysis, grammatical analysis, and semantic analysis on text received by the server.

[0869] "Error detection means" refers to a device or software that detects errors in text based on the results of analysis performed by the analysis means.

[0870] "Correction generation means" refers to a device or software that generates appropriate corrections for errors detected by error detection means.

[0871] "Transmission means" refers to a device or software used to send the generated revised draft to a terminal.

[0872] "Display means" refers to a device or software that displays the proposed revisions received by the terminal to the user.

[0873] Morphological analysis is a technique that breaks down input text into individual words and analyzes the part of speech and conjugation of each word.

[0874] "Grammar analysis" is a technique for detecting grammatical errors and syntactic errors within text.

[0875] "Semantic analysis" is a technique for understanding the meaning of an entire text and identifying unnatural phrasing or contextual problems.

[0876] A "revised version" is a suggestion for a correct or more appropriate expression, generated based on the detected error.

[0877] A "terminal" refers to a device, such as a computer or smartphone, that a user uses to access a system.

[0878] The present invention is an automated text correction tool that assists users in accurately and naturally inputting Japanese text. Specifically, the user inputs Japanese text into a text input form using a terminal. At this time, the terminal has means for sending the input text to a server.

[0879] The server has morphological, grammatical, and semantic analysis tools to analyze the received text data. First, to perform morphological analysis, it uses a morphological analysis library (e.g., MeCab or Sudachi). Morphological analysis breaks down the text into words and analyzes the part of speech and conjugation of each word. For example, if a user inputs the following sentence:

[0880] I will be attending tomorrow's meeting, so I look forward to seeing you there.

[0881] Morphological analysis breaks down this sentence into "I will be attending tomorrow's meeting, so please take care of me."

[0882] Next, the server uses a grammar analysis engine (e.g., SyntaxNet) to perform grammatical analysis. Grammar analysis detects grammatical errors and syntactic errors in the sentence. In the example above, it detects that "sanka suru masu" is grammatically incorrect.

[0883] Furthermore, to perform semantic analysis, a natural language understanding (NLU) model (e.g., BERT or GPT-3) is used. Semantic analysis analyzes the meaning of the entire text and identifies unnatural phrasing. In this case, it is determined that "onegaishimasu" should be changed to "onegaishimasu".

[0884] Based on the analysis results, the server uses error detection means to identify the error location and correction generation means to generate appropriate correction proposals. Specific correction proposals are as follows:

[0885] Error: "surumasu" -> Corrected: "sanka shimasu"

[0886] Error: "onegaishimasu" -> Correction: "onegaishimasu"

[0887] After the proposed revisions are generated, the server sends them back to the terminal as JSON data via a transmission method. The terminal parses the received JSON data and displays the proposed revisions to the user. For example, the proposed revision might say, "I will be attending tomorrow's meeting, thank you in advance."

[0888] The user reviews the displayed correction suggestions and applies the corrections as needed. This corrects the errors in the entered Japanese text, resulting in a correct sentence.

[0889] Here are some examples of specific prompt messages:

[0890] Please identify the error in the following Japanese sentence and provide the correct sentence:

[0891] I will be attending tomorrow's meeting, so I look forward to seeing you there.

[0892] The above describes a specific embodiment for carrying out the present invention. By using this system, users can quickly and easily correct errors in Japanese text.

[0893] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0894] Step 1:

[0895] The user uses their own device to input Japanese text into the system's text input form. They use a keyboard or other input method, and the text data is recorded within the device. For example, the user might input "I will be attending tomorrow's meeting, thank you in advance."

[0896] Step 2:

[0897] The terminal sends the entered text data to the server. Specifically, it generates an HTTP request using a communication method and sends the request containing the text data to the server. The input is the text data entered by the user, and the output is the transmission of the HTTP request.

[0898] Step 3:

[0899] The server takes in the received HTTP request and extracts the text data. At this point, the input is the text data contained in the body of the HTTP request, and the output is the extracted text data.

[0900] Step 4:

[0901] The server uses a morphological analysis library (e.g., MeCab or Sudachi) to break down text data into individual words. Specifically, it starts the analysis process and passes the text data to the analysis library. The input is the extracted text data, and the output is the morphologically analyzed data (e.g., "I will be attending tomorrow's meeting, so please take care of me").

[0902] Step 5:

[0903] The server uses a grammar analysis engine (e.g., SyntaxNet) to detect grammatical and syntactic errors in the morphologically analyzed data. Specifically, it activates the grammar analysis tool and passes the analysis data. The input is the morphologically analyzed data, and the output is the identification of grammatical errors (e.g., errors such as "sanka suru masu").

[0904] Step 6:

[0905] The server uses a natural language understanding (NLU) model (e.g., BERT or GPT-3) to analyze the meaning of the entire text. Specifically, it activates a semantic analysis tool and passes the morphologically analyzed data to the NLU model. The input is the morphologically analyzed data, and the output is semantic errors or suggestions for improvement (e.g., a judgment that "onegaishimasu" should be corrected to "onegaishimasu").

[0906] Step 7:

[0907] Based on the analysis results, the server records the errors identified using error detection means and generates appropriate correction suggestions using correction generation means. Specifically, it extracts the error locations through error detection means and creates correction suggestions using correction generation means. The input is the result of grammatical and semantic analysis, and the output is the correction suggestion (e.g., "surumasu" → "sankashimasu", "onegaishimasu" → "onegaishimasu").

[0908] Step 8:

[0909] After the proposed revisions are generated, the server sends them back to the terminal as JSON data using a transmission method. Specifically, the server converts the proposed revisions to JSON format and includes them in the HTTP response before sending. The input is the proposed revisions, and the output is the transmission of JSON data.

[0910] Step 9:

[0911] The terminal parses the received JSON data and displays suggested corrections to the user. Specifically, it uses a display mechanism to visually convert the JSON data and display it on the user interface. The input is data in JSON format, and the output is a suggested correction displayed to the user (e.g., "I will be attending tomorrow's meeting, thank you in advance").

[0912] Step 10:

[0913] The user reviews the displayed suggested revisions and applies them as needed. Specifically, the user confirms the text revisions by clicking the "Apply" button. The input is the user's click operation, and the output is the text with the revisions applied.

[0914] (Application Example 1)

[0915] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0916] When ordering food delivery, users often include typos, grammatical errors, or unnatural phrasing in their special requests and order details, which can lead to misdeliveries and a poor customer experience. A system is needed to solve this problem and allow users to confirm their orders in more accurate and natural Japanese.

[0917] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0918] In this invention, the server includes input means for the user to input text, communication means for transmitting the input text to the server, analysis means for performing morphological, grammatical, and semantic analysis of the text received by the server, error detection means and correction generation means for pointing out errors in the text and generating correction suggestions based on the analysis results, transmission means for transmitting the correction suggestions to the terminal, and means featuring an automatic proofreading function for food delivery applications to correct typos, omissions, and unnatural expressions made by the user when inputting order details and special notes. This enables the user to quickly correct the content they have entered and place accurate orders.

[0919] "Input means" refers to a device or interface for a user to input text.

[0920] "Communication means" refers to a device or protocol used to send input text to a server.

[0921] "Analysis means" refers to software or hardware mechanisms that perform morphological analysis, grammatical analysis, and semantic analysis on text received by the server.

[0922] An "error detection means" is a device or program that identifies errors in text based on the analysis results.

[0923] A "correction generation means" is a device or program for generating appropriate corrections after an error has been identified.

[0924] "Transmission means" refers to a device or protocol for sending the revised version to a terminal.

[0925] The "Automatic Correction Function for Food Delivery Applications" is a feature that corrects typos, grammatical errors, and unnatural expressions when users enter order details or special notes.

[0926] The present invention provides an automated text correction tool that assists users in accurately and naturally inputting Japanese text. In this invention, the system is particularly applicable to food delivery applications.

[0927] 1. System Program

[0928] This system consists of the following main elements:

[0929] Input method: An interface for users to enter text. A text input box embedded within a food delivery application.

[0930] Communication method: An HTTP request to send the entered text to the server. Specifically, the text is sent to the server in JSON format.

[0931] Analysis methods: Functions for morphological analysis, grammatical analysis, and semantic analysis of text received by the server. Morphological analysis uses the Janome library, and grammatical and semantic analysis utilize natural language understanding (NLU) models.

[0932] Error detection and correction generation means: Based on the analysis results, errors in the text are detected and appropriate correction suggestions are generated. Examples of errors include incorrect particles and unnatural expressions.

[0933] Transmission method: An HTTP response to send JSON data containing the proposed revisions back to the terminal.

[0934] 2. Program processing description using natural language

[0935] The server receives text sent by the user through the input method. The Janome library is used to perform morphological analysis on the received text, breaking down the sentence into individual words. For example, the input sentence "I will be attending tomorrow's meeting, so thank you in advance" is broken down by morphological analysis into "tomorrow / no / kaigi / ni / participating / suru / masu / node / , / yoroshiku / onegai / shimasu".

[0936] Next, the grammar analysis engine analyzes the part of speech and conjugation of each word to detect grammatical errors. For example, it detects that the expression "sanka suru masu" (to participate) is inappropriate. Subsequently, in the semantic analysis stage, a natural language understanding (NLU) model is used to analyze the meaning of the entire sentence and identify unnatural phrasing. Here, it determines that "onegai shimasu" (please) should be changed to "onegai shimasu" (please).

[0937] Based on the analysis results, the error detection means detects errors, and the correction generation means generates appropriate correction suggestions. The server then sends the correction suggestions back to the terminal as JSON data. The terminal analyzes the received JSON data and displays the correction suggestions to the user. The user reviews the displayed correction suggestions and applies the corrections as needed.

[0938] 3. Specific examples and prompt messages

[0939] For example, a user enters the following order details:

[0940] I am allergic to shrimp, so please remove the shrimp from the dish.

[0941] When this input is sent to the server, it will suggest the following modifications:

[0942] I am allergic to shrimp, so please omit the shrimp from the dish.

[0943] Examples of prompt statements to input into a generative AI model are as follows:

[0944] Analyze the Japanese text entered by the user grammatically and semantically, and suggest appropriate corrections. Example: 'I am allergic to shrimp, so please remove the shrimp.' -> 'I am allergic to shrimp, so please remove the shrimp.'

[0945] This allows users to quickly correct their input and place orders in accurate and natural-sounding Japanese.

[0946] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0947] Step 1:

[0948] The user uses their device to enter order details and special requests into a text input form within the food delivery application. An example of the text entered is, "I am allergic to shrimp, so please omit the shrimp."

[0949] Step 2:

[0950] The terminal sends an HTTP request to the server using a communication method to send the entered text to the server. Specifically, it sends the text data to the server in JSON format.

[0951] Step 3:

[0952] The server analyzes the received text data. First, the server uses the morphological analysis library Janome to break down the text into words. For example, it breaks down "I am allergic to shrimp, so please remove the shrimp." into "Allergy / is / shrimp / , / shrimp / remove / please / please / you / please / please / remove / shrimp".

[0953] Step 4:

[0954] The server uses a grammar analysis engine to analyze the morphologically analyzed word sequence from the perspective of part of speech and conjugation, and detects grammatical errors. For example, it might determine that the unnatural expression "kudasaimasu you onegaishimasu" should be corrected to "onegaishimasu".

[0955] Step 5:

[0956] The server uses a natural language understanding (NLU) model to perform semantic analysis. Here, it analyzes the meaning of the entire sentence and identifies unnatural phrasing and expressions. For example, it might correct "onegaishimasu" to "onegaishimasu."

[0957] Step 6:

[0958] Based on the analysis results, the server uses error detection to identify errors and correction generation to generate appropriate corrections. Specifically, it generates a correction to change "onegaishimasu" to "onegaishimasu".

[0959] Step 7:

[0960] The server sends the proposed revisions back to the terminal as JSON data via a transmission method. The transmitted data includes both the original and revised text.

[0961] Step 8:

[0962] The terminal parses the received JSON data and displays suggested corrections to the user. For example, it might display "I am allergic to shrimp, so please remove the shrimp."

[0963] Step 9:

[0964] The user reviews the displayed revisions and applies them as needed. This confirms that the corrected text is finalized in the order.

[0965] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0966] The present invention is an automated proofreading tool that allows users to input Japanese text accurately and naturally, and provides feedback that takes the user's feelings into consideration. In this embodiment, the specific program processing flow will be described.

[0967] First, the user enters a Japanese sentence into a text input form on their device. For example, suppose the user enters the following sentence:

[0968] I will be attending tomorrow's meeting, so I look forward to seeing you there. I'm very nervous.

[0969] Next, the terminal sends an HTTP request to the server using a communication method to send the entered text to the server. At this time, the text data is sent to the server.

[0970] The server begins processing the received text data for analysis. Specifically, the following four analysis methods are used:

[0971] 1. Morphological Analysis: The server uses a morphological analysis library to break down the text into individual words. For example, the input sentence is broken down into "I will be attending tomorrow's meeting, so please be kind to me. I am very nervous." The part of speech and conjugation of each word are also analyzed.

[0972] 2. Grammar Analysis: The grammar analysis engine detects grammatical errors. In the example above, it detects that the expression "sanka suru masu" is inappropriate.

[0973] 3. Semantic Analysis: A natural language understanding (NLU) model analyzes the meaning of the entire sentence and identifies unnatural phrasing. In this case, it determines that "onegaishimasu" should be changed to "onegaishimasu".

[0974] 4. Sentiment Analysis: The server's sentiment analysis tool identifies the emotions contained in the user's text. This analysis determines that "I'm very nervous" represents the emotion of "nervousness."

[0975] Based on the analysis results, the server uses error detection and correction generation means to identify errors and generate appropriate correction suggestions. Specific correction suggestions are as follows:

[0976] Error: "surumasu" -> Corrected: "sanka shimasu"

[0977] Error: "onegaishimasu" -> Correction: "onegaishimasu"

[0978] Furthermore, the emotion engine generates user-friendly feedback based on identified emotions. For example, in response to "I'm very nervous," it might generate advice such as "Try to relax before the meeting."

[0979] After the proposed revisions and sentiment feedback are generated, the server sends them back to the terminal as JSON data via a transmission method.

[0980] The device parses the received JSON data and displays suggested revisions and sentiment feedback to the user. For example, the user might see the following:

[0981] I will be attending tomorrow's meeting, so I look forward to seeing you there. I'm very nervous.

[0982] Proposed amendment:

[0983] "I will be attending tomorrow's meeting, so I look forward to seeing you there."

[0984] Emotional feedback:

[0985] "Please relax and approach the meeting with ease."

[0986] Users review the displayed correction suggestions and sentiment feedback, and apply the corrections as needed. This corrects errors in the entered Japanese text and allows them to receive feedback that takes their feelings into consideration.

[0987] The above describes a specific embodiment for carrying out the present invention. By using this system, users can quickly and easily correct errors in Japanese text and receive emotionally sensitive feedback.

[0988] The following describes the processing flow.

[0989] Step 1:

[0990] The user enters a Japanese sentence into a text input form on their device. For example, they might enter, "I will be attending tomorrow's meeting, so I look forward to seeing you there. I'm very nervous." Once they have finished typing, they click the submit button.

[0991] Step 2:

[0992] The terminal receives the input text data and uses a communication method to send an HTTP request to the server. This request contains the text data.

[0993] Step 3:

[0994] The server parses the received HTTP request and extracts text data. The extracted text data is then passed to the morphological analysis module.

[0995] Step 4:

[0996] The server's morphological analysis module breaks down text into individual words and analyzes the part of speech and conjugation of each word. For example, the text "I will be attending tomorrow's meeting, so I look forward to seeing you. I'm very nervous." is broken down into "tomorrow / noun of / particle meeting / noun to / particle attend / verb do / auxiliary verb do / auxiliary verb so / particle, / symbol nice to see / adverb please / verb do / auxiliary verb. / symbol very / adverb nervous / noun feeling / auxiliary verb I am / auxiliary verb".

[0997] Step 5:

[0998] The analyzed word information is passed to the grammar analysis module. The server's grammar analysis module identifies grammatical errors and syntactic errors in the sentence. At this stage, the incorrect expression "sanka suru masu" (participate) is detected.

[0999] Step 6:

[1000] The results of the grammatical analysis are passed to the semantic analysis module. The server's semantic analysis module understands the context and identifies unnatural phrasing and misuse. This analysis determines that "onegaishimasu" should be corrected to "onegaishimasu".

[1001] Step 7:

[1002] Next, the server's sentiment analysis tool identifies the emotions contained in the user's text. It recognizes the emotion of tension from the expression "I'm very nervous."

[1003] Step 8:

[1004] Based on the results of grammatical analysis, semantic analysis, and sentiment analysis, the server uses error detection and correction generation means to identify errors and generate appropriate correction suggestions and sentiment feedback. For example, "participate" is corrected to "will participate," and "please" is corrected to "please." In addition, sentiment feedback such as "Please relax and enjoy the meeting" is generated for a nervous user.

[1005] Step 9:

[1006] The generated corrections and sentiment feedback are formatted in JSON format. This data includes the location of the error, the original text, the correction, and the sentiment feedback. The server sends this JSON data back to the terminal via the transmission method.

[1007] Step 10:

[1008] The device parses the received JSON data and displays suggested revisions and sentiment feedback to the user. For example, it may be displayed as follows:

[1009] I will be attending tomorrow's meeting, so I look forward to seeing you there. I'm very nervous.

[1010] Proposed amendment:

[1011] "I will be attending tomorrow's meeting, so I look forward to seeing you there."

[1012] Emotional feedback:

[1013] "Please relax and approach the meeting with ease."

[1014] Step 11:

[1015] Users review the displayed correction suggestions and sentiment feedback, and apply the corrections as needed. This corrects errors in the entered Japanese text and allows them to receive feedback that takes their feelings into consideration.

[1016] (Example 2)

[1017] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1018] When users input Japanese text, the challenge lies in accurately identifying and correcting errors, as well as providing feedback that takes the user's emotions into consideration. Conventional systems only point out errors and fail to provide emotionally sensitive feedback, thus failing to sufficiently improve user satisfaction and the naturalness of the text.

[1019] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1020] In this invention, the server includes analysis means for performing morphological, grammatical, semantic, and sentiment analysis on received text; error detection and correction generation means for pointing out errors in the text, generating suggested corrections, and providing feedback that takes the user's feelings into consideration, based on the analysis results; and transmission means for sending the suggested corrections and sentiment feedback to the terminal. This allows the user to correct errors while simultaneously receiving feedback that takes their feelings into consideration.

[1021] "Text" refers to a string of characters entered by the user, and may include Japanese sentences or words.

[1022] "Input means" refers to devices or interfaces for users to input text, including keyboards and touchscreens.

[1023] "Communication method" refers to the means by which a terminal sends text entered into it to a server, and this involves using an internet connection or a communication protocol (e.g., HTTP).

[1024] A "server" is a central processing unit that receives, analyzes, and processes text sent from a user's terminal.

[1025] "Analysis means" includes software and libraries for morphological analysis, grammatical analysis, semantic analysis, and sentiment analysis of received text.

[1026] Morphological analysis is the process of breaking down text into individual words and analyzing the part of speech and conjugation of each word.

[1027] "Grammar analysis" is the process of detecting grammatical errors and syntactic errors within a text.

[1028] "Semantic analysis" is the process of understanding the overall meaning of a text and identifying unnatural expressions.

[1029] "Sentiment analysis" is the process of identifying the user's emotions contained in text and generating appropriate feedback.

[1030] "Error detection means and correction generation means" refers to processing means for identifying errors in text based on analysis results and generating proposed corrections.

[1031] "Transmission means" refers to the means by which the server sends back the proposed revisions and sentiment feedback generated by the server to the terminal.

[1032] A "revision proposal" is a suggestion to correct errors detected by the analysis method.

[1033] "Emotional feedback" refers to providing advice and comments that take the user's emotions into consideration, based on emotion analysis.

[1034] The present invention is an automated proofreading tool that allows users to input Japanese text accurately and naturally, and provides feedback that takes into account the user's emotions. This embodiment details the specific implementation method of this system.

[1035] System Configuration

[1036] This system consists of a user terminal, a server, and various analysis tools.

[1037] 1. Terminal

[1038] This refers to a device used by a user to input text. Examples include PCs, smartphones, and tablets.

[1039] 2. Server

[1040] It is a central processing unit equipped with a high-performance processor and sufficient memory. The server receives text sent from the user and performs various analyses and processing.

[1041] 3. Analysis method

[1042] The specific software used includes the following:

[1043] Morphological analysis libraries (e.g., MeCab)

[1044] Grammar analysis engine

[1045] Natural Language Understanding (NLU) Models

[1046] Emotion analysis engine

[1047] Operation details

[1048] 1. User text input

[1049] The user enters Japanese text into an input form on their device. For example, they might enter the sentence, "I will be attending tomorrow's meeting, so thank you in advance. I'm very nervous."

[1050] 2. Sending text to the server

[1051] The entered text is sent from the terminal to the server using an HTTP request.

[1052] 3. Server-based analysis

[1053] The server processes the received text sequentially as follows:

[1054] Morphological analysis: This process breaks down text into individual words and analyzes their parts of speech and conjugations.

[1055] For example, "I will be attending tomorrow's meeting, so thank you in advance. I'm very nervous." can be broken down into "Tomorrow / no / kaigi / ni / taku / suru / masu / no / , / yoroshiku / onegai / shimasu / . / toku / kichika / shite / imasu."

[1056] Grammar analysis: Detects grammatical errors and syntactic errors in a sentence.

[1057] Example: Detect that "sanka suru masu" is grammatically incorrect.

[1058] Semantic analysis: Analyzes the meaning of the entire sentence and identifies unnatural expressions.

[1059] Example: It is decided that "onegaishimasu" should be changed to "onegaishimasu".

[1060] Sentiment analysis: Identifying the emotions contained in the text.

[1061] Example: Determine that "I am very nervous" expresses the emotion of "nervousness".

[1062] 4. Generating revised proposals and emotional feedback

[1063] The server identifies errors based on the analysis results and generates suggested corrections. It also generates feedback that takes the user's emotions into consideration based on sentiment analysis.

[1064] Example 1: Correct the error "sanka suru masu" to "sanka shimasu".

[1065] Example 2: Change "onegaishimasu" to "onegaishimasu".

[1066] Example 3: In response to "I'm very nervous," generate feedback such as "Please relax and approach the meeting with confidence."

[1067] 5. Return of data

[1068] The proposed revisions and emotional feedback will be sent to the device as data in JSON format.

[1069] 6. Display to the user

[1070] The device analyzes the received data and displays suggested revisions and sentiment feedback to the user.

[1071] Example: It will be displayed as follows:

[1072] I will be attending tomorrow's meeting, so I look forward to seeing you there. I'm very nervous.

[1073] Proposed amendment:

[1074] "I will be attending tomorrow's meeting, so I look forward to seeing you there."

[1075] Emotional feedback:

[1076] "Please relax and approach the meeting with ease."

[1077] Example of a prompt

[1078] The following is an example of a set prompt statement:

[1079] Please enter a Japanese sentence, correct any errors, and provide emotionally sensitive feedback. Example sentence: I will be attending tomorrow's meeting, so I look forward to seeing you there. I'm very nervous.

[1080] By applying this system, users can quickly and easily correct errors in Japanese text and receive emotionally sensitive feedback.

[1081] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1082] Step 1:

[1083] The user enters text.

[1084] Operation: The user enters Japanese text into an input form on the terminal. For example, they might enter, "I will be attending tomorrow's meeting, thank you in advance. I'm very nervous."

[1085] Input: Japanese text.

[1086] Output: The text displayed in the input form on the terminal.

[1087] Step 2:

[1088] This is the procedure for sending the text entered by the terminal to the server.

[1089] Operation: The terminal sends the entered text to the server using an HTTP request.

[1090] Input: Japanese text entered by the user.

[1091] Output: Text data sent to the server.

[1092] Step 3:

[1093] This is the procedure for performing morphological analysis on text received by the server.

[1094] Operation: The server uses a morphological analysis library to break down the text into individual words and analyze the part of speech and conjugation of each word. For example, it breaks down "I will be attending tomorrow's meeting, so thank you in advance. I'm very nervous." into "Tomorrow / no / kaigi / ni / participating / suru / masu / no / , / yoroshiku / onegai / shimasu / . / totemo / kyouki / shite / imasu".

[1095] Input: Text data sent to the server.

[1096] Output: Morphological analysis results (text data broken down into individual words).

[1097] Step 4:

[1098] This is the procedure by which the server performs grammatical analysis of the text.

[1099] Operation: The server uses a grammar analysis engine to detect grammatical errors and syntactic errors in the text. For example, it detects that the expression "sanka surumasu" (to participate) is inappropriate.

[1100] Input: Morphological analysis results.

[1101] Output: Grammar analysis results (data including information on grammatical errors and syntactic errors).

[1102] Step 5:

[1103] This is the procedure by which the server performs semantic analysis of text.

[1104] Operation: The server uses a natural language understanding (NLU) model to analyze the meaning of the entire text and identify unnatural phrasing. For example, it might determine that "onegaishimasu" should be corrected to "onegaishimasu".

[1105] Input: Grammar analysis result.

[1106] Output: Semantic analysis results (data containing information about unnatural phrasing).

[1107] Step 6:

[1108] This is the procedure for the server to perform sentiment analysis on text.

[1109] Operation: Uses an emotion analysis engine to identify the user's emotions contained in the text. For example, it determines that "I'm very nervous" represents the emotion "nervous."

[1110] Input: Semantic analysis results.

[1111] Output: Sentiment analysis results (data including user sentiment information).

[1112] Step 7:

[1113] This is a procedure in which the server performs error detection, proposes corrections, and generates sentiment feedback.

[1114] Operation: Based on the analysis results, the server identifies errors and generates appropriate correction suggestions. It also generates feedback that takes the user's emotions into account based on the sentiment analysis results. For example, it corrects "surumasu" to "shimasu" and "onegaishimasu" to "onegaishimasu". In response to "I'm very nervous," it generates feedback such as "Please relax and approach the meeting with ease."

[1115] Input: Grammatical analysis results, semantic analysis results, sentiment analysis results.

[1116] Output: Revised proposals and emotional feedback.

[1117] Step 8:

[1118] This is the procedure for the server to send suggested revisions and emotional feedback to the terminal.

[1119] Operation: The server generates JSON data containing suggested revisions and sentiment feedback, and sends it back to the terminal.

[1120] Input: Suggested revisions and emotional feedback.

[1121] Output: JSON data sent to the terminal.

[1122] Step 9:

[1123] This is the procedure for the device to display suggested fixes and emotional feedback to the user.

[1124] Operation: The device parses the received JSON data and displays suggested revisions and sentiment feedback to the user. For example, it might display: "I will be attending tomorrow's meeting, thank you in advance. I'm very nervous. Suggested revision: 'I will be attending tomorrow's meeting, thank you in advance.' Sentiment feedback: 'Please relax during the meeting.'"

[1125] Input: JSON data sent to the terminal.

[1126] Output: Suggested revisions and sentiment feedback displayed to the user.

[1127] (Application Example 2)

[1128] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1129] In virtual stores, customers often generate incorrect or unnatural sentences when entering text. Furthermore, there is a lack of means to provide feedback that is appropriate to the customer's current emotional state. This presents a challenge in improving the customer experience.

[1130] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a morphological analysis means, a grammatical analysis means, a semantic analysis means, an emotion analysis means, an error detection means and correction generation means, and a transmission means. This makes it possible not only to correct errors and unnatural expressions, but also to provide feedback based on the customer's emotional state.

[1131] A "user" is a person who uses this system to input text and sends that input to the server.

[1132] "Input means" refers to a device or interface for a user to input text.

[1133] "Communication means" refers to means, including networks and protocols, for sending input text to a server.

[1134] A "server" is a device or system that analyzes text sent by a user and generates suggested revisions and sentiment feedback.

[1135] A "morphological analysis tool" is a function that breaks down input text into words and analyzes the part of speech and conjugation of each word.

[1136] A "grammatical analysis tool" is a function that detects grammatical errors and syntactic errors in the input text.

[1137] A "semantic analysis tool" is a function that analyzes the overall meaning of the input text and identifies unnatural phrasing.

[1138] "Emotion analysis means" refers to a function that analyzes the emotions contained in the input text.

[1139] An "error detection means" is a function that identifies errors in text based on the analysis results.

[1140] A "correction generation means" is a function that generates appropriate corrections for identified errors.

[1141] "Transmission means" refers to the means of sending the generated revisions and emotional feedback to the user's device.

[1142] "Emotional feedback" refers to advice and comments provided based on the emotions contained in the user's text.

[1143] This invention is a system that corrects text entered by customers in a virtual store into accurate and natural sentences, and also provides feedback that takes into account the customer's feelings.

[1144] This system mainly consists of input means, communication means, server, morphological analysis means, grammatical analysis means, semantic analysis means, sentiment analysis means, error detection means, correction generation means, and transmission means.

[1145] First, the user enters text within the virtual store using an input device. This input device could be a smart glasses or a head-mounted display.

[1146] The text entered by the user is sent to the server via a communication method. Network communication methods such as the HTTP protocol are used for this purpose.

[1147] The server analyzes the received text. Specifically, it uses morphological analysis to break down the text into individual words and analyzes the part of speech and conjugation of each word.

[1148] Next, the grammar analysis means uses the analysis results to detect grammatical errors and syntactic errors in the input text. The semantic analysis means also analyzes the meaning of the entire text and identifies unnatural expressions.

[1149] Furthermore, the emotion analysis mechanism analyzes the emotions contained in the input text. Specifically, it uses a generative AI model to extract emotions from the text and generates corresponding feedback.

[1150] Based on the analysis results, the error detection means identifies errors in the text, and the correction generation means generates appropriate corrections and sentiment-based feedback. For example, for the text "I will be attending tomorrow's meeting, so thank you in advance. I'm very nervous," the correction is changed to "I will be attending tomorrow's meeting, so thank you in advance," and sentiment feedback such as "Please relax during the meeting" is provided.

[1151] These suggested revisions and emotional feedback are sent to the user's device via a transmission method. The user can review the submitted feedback and apply revisions as needed, thereby obtaining accurate, natural-sounding Japanese text and emotionally sensitive feedback.

[1152] For example, if the prompt is "I will be attending tomorrow's meeting, so thank you in advance. I'm very nervous," the system will respond by generating a revised version, "I will be attending tomorrow's meeting, so thank you in advance," and emotional feedback, "Please relax during the meeting."

[1153] In this way, this invention improves the accuracy and emotional consideration of text in customer service within virtual stores.

[1154] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1155] Step 1:

[1156] Users enter text within the virtual store using input devices. These input devices include smart glasses and head-mounted displays. For example, the text entered might read, "I will be attending tomorrow's meeting, so I look forward to seeing you there. I'm very nervous."

[1157] Input: Text entered by the user

[1158] Output: Raw text entered into the terminal

[1159] Step 2:

[1160] The terminal sends the entered text to the server using a communication method. This communication method typically involves network communication protocols such as HTTP.

[1161] Input: Raw text entered on the terminal

[1162] Output: Text data sent to the server

[1163] Step 3:

[1164] The server uses morphological analysis to break down the received text into individual words and analyzes the part of speech and conjugation of each word. For this purpose, it uses a morphological analysis library (e.g., Janome). For example, "I will be attending tomorrow's meeting, so thank you in advance. I'm very nervous." is analyzed as "Tomorrow / no / meeting / ni / participating / suru / masu / no / , / yoroshiku / onegai / shimasu / . / totemo / kyouki / shite / imasu".

[1165] Input: Text data sent to the server

[1166] Output: Text data broken down into words

[1167] Step 4:

[1168] The grammatical analysis means detects grammatical errors and syntax errors in the text based on the results of morphological analysis. For example, a grammatical error such as "するます" is detected.

[1169] Input: Text data decomposed into words

[1170] Output: Data with grammatical errors identified

[1171] Step 5:

[1172] The semantic analysis means analyzes the overall meaning of the text using a natural language understanding model (NLU) and identifies unnatural expressions. For example, it is identified that "よろしくおねがいします" should be corrected to "お願いします".

[1173] Input: Data with grammatical errors identified

[1174] Output: Data with unnatural expressions identified

[1175] Step 6:

[1176] The sentiment analysis means analyzes the sentiment contained in the input text using a generative AI model. For example, it is determined that "とても緊張しています" expresses the sentiment of "nervousness".

[1177] Input: Data with unnatural expressions identified

[1178] Output: Data with sentiment identified

[1179] Step 7:

[1180] The error detection means identifies errors in the text based on the above analysis results. For example, it is identified that "するます" is an error that should be corrected to "します".

[1181] Input: Data with sentiment identified

[1182] Output: Error and correction instruction data

[1183] Step 8:

[1184] The correction generation mechanism generates appropriate correction suggestions for identified errors. It also generates user-friendly feedback based on sentiment analysis results. For example, it might generate a correction suggestion like, "I will be attending tomorrow's meeting, thank you in advance," along with emotional feedback such as, "Please relax during the meeting."

[1185] Input: Error and correction instruction data

[1186] Output: Proposed revisions and sentiment feedback data

[1187] Step 9:

[1188] The server sends the generated revisions and sentiment feedback to the user's device via a transmission method. These are then displayed on the user's device.

[1189] Input: Proposed revisions and sentiment feedback data

[1190] Output: Suggested revisions and sentiment feedback sent to the user's terminal.

[1191] Step 10:

[1192] Users review the proposed revisions and sentiment feedback, and apply the changes as needed. This ensures accurate and natural Japanese text and sentiment-sensitive feedback.

[1193] Input: Correction suggestions and sentiment feedback sent to the user's terminal.

[1194] Output: Final text with corrections and sentiment feedback applied.

[1195] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1196] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1197] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[1198] [Fourth Embodiment]

[1199] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1200] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1201] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1202] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[1203] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1204] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1205] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1206] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[1207] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1208] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1209] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1210] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1211] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1212] The system of the present invention is an automatic proofreading tool that assists users in accurately and naturally inputting Japanese text. In this embodiment, the specific program processing flow will be described.

[1213] First, the user uses their terminal to input Japanese text into the system's text input form. For example, suppose the user enters the following sentence:

[1214] I will be attending tomorrow's meeting, so I look forward to seeing you there.

[1215] Next, the terminal sends an HTTP request to the server using a communication method to send the entered text to the server. At this time, the text data is sent to the server.

[1216] The server begins processing the received text data for analysis. Specifically, the following three analysis methods are used:

[1217] 1. Morphological Analysis: The server uses a morphological analysis library to break down the text into individual words. For example, the input sentence is broken down into "I will be attending tomorrow's meeting, so please take care of me." The part of speech and conjugation of each word are also analyzed.

[1218] 2. Grammar Analysis: The grammar analysis engine detects grammatical errors. In the example above, it detects that the expression "sanka suru masu" is inappropriate.

[1219] 3. Semantic Analysis: A natural language understanding (NLU) model analyzes the meaning of the entire sentence and identifies unnatural phrasing. In this case, it determines that "onegaishimasu" should be changed to "onegaishimasu".

[1220] Based on the analysis results, the server uses error detection means to identify the error locations and correction generation means to generate appropriate correction suggestions. Specific correction suggestions are as follows:

[1221] Error: "surumasu" -> Corrected: "sanka shimasu"

[1222] Error: "onegaishimasu" -> Correction: "onegaishimasu"

[1223] After the revised version is generated, the server sends it back to the terminal as JSON data via the transmission method.

[1224] The terminal parses the received JSON data and displays suggested corrections to the user. For example, the user will see the following:

[1225] I will be attending tomorrow's meeting, so I look forward to seeing you there.

[1226] The user reviews the displayed correction suggestions and applies the corrections as needed. This corrects the errors in the entered Japanese text, resulting in a correct sentence.

[1227] The above describes a specific embodiment for carrying out the present invention. By using this system, users can quickly and easily correct errors in Japanese text.

[1228] The following describes the processing flow.

[1229] Step 1:

[1230] The user enters Japanese text into a text input form on their device. After entering the text, the user prepares to send the text by pressing the send button.

[1231] Step 2:

[1232] The terminal receives text data entered by the user and sends an HTTP request to the server using a communication method. This request contains text data.

[1233] Step 3:

[1234] The server parses the received HTTP request and extracts the text data from the request body. The server then passes the received text to the morphological analysis module.

[1235] Step 4:

[1236] The server's morphological analysis module breaks down text into individual words and analyzes the part of speech and conjugation of each word. This analysis includes the base form and part of speech information of each word.

[1237] Step 5:

[1238] The analyzed word information is passed to the grammar analysis module. The server's grammar analysis module identifies grammatical errors and syntactic errors in the sentence. In this process, ungrammatical verb and auxiliary verb combinations, etc., are detected.

[1239] Step 6:

[1240] Next, the server's semantic analysis module understands the overall context and meaning, detecting unnatural phrasing and misuse. This analysis evaluates the appropriateness of words and expressions.

[1241] Step 7:

[1242] Based on the results of grammatical and semantic analysis, the server identifies errors using error detection and correction generation means and generates appropriate corrections. For example, "sanka suru masu" is corrected to "sanka shimasu," and "onegai shimasu" is corrected to "onegai shimasu."

[1243] Step 8:

[1244] After the proposed correction is generated, the server formats it as JSON data. This data includes the location of the error, the original text, and the proposed correction.

[1245] Step 9:

[1246] The server sends JSON data back to the terminal via a transmission method. The terminal receives this data and processes it for display.

[1247] Step 10:

[1248] The device displays the suggested fixes to the user in a visually easy-to-understand format. The user reviews the suggested fixes and applies them as needed.

[1249] Step 11:

[1250] The user revises the text based on the suggested changes and then performs a final check. After this, the correct Japanese text can be used for other purposes.

[1251] (Example 1)

[1252] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1253] Conventional text input systems have a problem in that they have difficulty automatically detecting grammatical errors and semantic unnaturalness in Japanese text and providing accurate and natural correction suggestions. Therefore, users have to manually check and correct the text, which leads to a decrease in work efficiency.

[1254] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1255] In this invention, the server includes input means for the user to input text, communication means for transmitting the input text to the server, analysis means for performing morphological, grammatical, and semantic analysis of the text received by the server, error detection means and correction generation means for pointing out errors in the text and generating correction suggestions based on the analysis results, transmission means for transmitting the correction suggestions to the terminal, and display means for displaying the correction suggestions received by the terminal to the user. This enables the user to quickly correct errors in the input Japanese text and generate accurate and natural-sounding sentences.

[1256] A "user" is a person who inputs text into a system and utilizes the functions that the system provides.

[1257] "Input means" refers to a device or software that provides an interface for a user to input text.

[1258] "Communication means" refers to a device or software used to transmit text data entered from an input means to a server.

[1259] A "server" is a device or system that receives input text data and performs analysis on it.

[1260] "Analysis means" refers to software or algorithms used to perform morphological analysis, grammatical analysis, and semantic analysis on text received by the server.

[1261] "Error detection means" refers to a device or software that detects errors in text based on the results of analysis performed by the analysis means.

[1262] "Correction generation means" refers to a device or software that generates appropriate corrections for errors detected by error detection means.

[1263] "Transmission means" refers to a device or software used to send the generated revised draft to a terminal.

[1264] "Display means" refers to a device or software that displays the proposed revisions received by the terminal to the user.

[1265] Morphological analysis is a technique that breaks down input text into individual words and analyzes the part of speech and conjugation of each word.

[1266] "Grammar analysis" is a technique for detecting grammatical errors and syntactic errors within text.

[1267] "Semantic analysis" is a technique for understanding the meaning of an entire text and identifying unnatural phrasing or contextual problems.

[1268] A "revised version" is a suggestion for a correct or more appropriate expression, generated based on the detected error.

[1269] A "terminal" refers to a device, such as a computer or smartphone, that a user uses to access a system.

[1270] The present invention is an automated text correction tool that assists users in accurately and naturally inputting Japanese text. Specifically, the user inputs Japanese text into a text input form using a terminal. At this time, the terminal has means for sending the input text to a server.

[1271] The server has morphological, grammatical, and semantic analysis tools to analyze the received text data. First, to perform morphological analysis, it uses a morphological analysis library (e.g., MeCab or Sudachi). Morphological analysis breaks down the text into words and analyzes the part of speech and conjugation of each word. For example, if a user inputs the following sentence:

[1272] I will be attending tomorrow's meeting, so I look forward to seeing you there.

[1273] Morphological analysis breaks down this sentence into "I will be attending tomorrow's meeting, so please take care of me."

[1274] Next, the server uses a grammar analysis engine (e.g., SyntaxNet) to perform grammatical analysis. Grammar analysis detects grammatical errors and syntactic errors in the sentence. In the example above, it detects that "sanka suru masu" is grammatically incorrect.

[1275] Furthermore, to perform semantic analysis, a natural language understanding (NLU) model (e.g., BERT or GPT-3) is used. Semantic analysis analyzes the meaning of the entire text and identifies unnatural phrasing. In this case, it is determined that "onegaishimasu" should be changed to "onegaishimasu".

[1276] Based on the analysis results, the server uses error detection means to identify the error location and correction generation means to generate appropriate correction proposals. Specific correction proposals are as follows:

[1277] Error: "surumasu" -> Corrected: "sanka shimasu"

[1278] Error: "onegaishimasu" -> Correction: "onegaishimasu"

[1279] After the proposed revisions are generated, the server sends them back to the terminal as JSON data via a transmission method. The terminal parses the received JSON data and displays the proposed revisions to the user. For example, the proposed revision might say, "I will be attending tomorrow's meeting, thank you in advance."

[1280] The user reviews the displayed correction suggestions and applies the corrections as needed. This corrects the errors in the entered Japanese text, resulting in a correct sentence.

[1281] Here are some examples of specific prompt messages:

[1282] Please identify the error in the following Japanese sentence and provide the correct sentence:

[1283] I will be attending tomorrow's meeting, so I look forward to seeing you there.

[1284] The above describes a specific embodiment for carrying out the present invention. By using this system, users can quickly and easily correct errors in Japanese text.

[1285] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1286] Step 1:

[1287] The user uses their own device to input Japanese text into the system's text input form. They use a keyboard or other input method, and the text data is recorded within the device. For example, the user might input "I will be attending tomorrow's meeting, thank you in advance."

[1288] Step 2:

[1289] The terminal sends the entered text data to the server. Specifically, it generates an HTTP request using a communication method and sends the request containing the text data to the server. The input is the text data entered by the user, and the output is the transmission of the HTTP request.

[1290] Step 3:

[1291] The server takes in the received HTTP request and extracts the text data. At this point, the input is the text data contained in the body of the HTTP request, and the output is the extracted text data.

[1292] Step 4:

[1293] The server uses a morphological analysis library (e.g., MeCab or Sudachi) to break down text data into individual words. Specifically, it starts the analysis process and passes the text data to the analysis library. The input is the extracted text data, and the output is the morphologically analyzed data (e.g., "I will be attending tomorrow's meeting, so please take care of me").

[1294] Step 5:

[1295] The server uses a grammar analysis engine (e.g., SyntaxNet) to detect grammatical and syntactic errors in the morphologically analyzed data. Specifically, it activates the grammar analysis tool and passes the analysis data. The input is the morphologically analyzed data, and the output is the identification of grammatical errors (e.g., errors such as "sanka suru masu").

[1296] Step 6:

[1297] The server uses a natural language understanding (NLU) model (e.g., BERT or GPT-3) to analyze the meaning of the entire text. Specifically, it activates a semantic analysis tool and passes the morphologically analyzed data to the NLU model. The input is the morphologically analyzed data, and the output is semantic errors or suggestions for improvement (e.g., a judgment that "onegaishimasu" should be corrected to "onegaishimasu").

[1298] Step 7:

[1299] Based on the analysis results, the server records the errors identified using error detection means and generates appropriate correction suggestions using correction generation means. Specifically, it extracts the error locations through error detection means and creates correction suggestions using correction generation means. The input is the result of grammatical and semantic analysis, and the output is the correction suggestion (e.g., "surumasu" → "sankashimasu", "onegaishimasu" → "onegaishimasu").

[1300] Step 8:

[1301] After the proposed revisions are generated, the server sends them back to the terminal as JSON data using a transmission method. Specifically, the server converts the proposed revisions to JSON format and includes them in the HTTP response before sending. The input is the proposed revisions, and the output is the transmission of JSON data.

[1302] Step 9:

[1303] The terminal parses the received JSON data and displays suggested corrections to the user. Specifically, it uses a display mechanism to visually convert the JSON data and display it on the user interface. The input is data in JSON format, and the output is a suggested correction displayed to the user (e.g., "I will be attending tomorrow's meeting, thank you in advance").

[1304] Step 10:

[1305] The user reviews the displayed suggested revisions and applies them as needed. Specifically, the user confirms the text revisions by clicking the "Apply" button. The input is the user's click operation, and the output is the text with the revisions applied.

[1306] (Application Example 1)

[1307] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1308] When ordering food delivery, users often include typos, grammatical errors, or unnatural phrasing in their special requests and order details, which can lead to misdeliveries and a poor customer experience. A system is needed to solve this problem and allow users to confirm their orders in more accurate and natural Japanese.

[1309] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1310] In this invention, the server includes input means for the user to input text, communication means for transmitting the input text to the server, analysis means for performing morphological, grammatical, and semantic analysis of the text received by the server, error detection means and correction generation means for pointing out errors in the text and generating correction suggestions based on the analysis results, transmission means for transmitting the correction suggestions to the terminal, and means featuring an automatic proofreading function for food delivery applications to correct typos, omissions, and unnatural expressions made by the user when inputting order details and special notes. This enables the user to quickly correct the content they have entered and place accurate orders.

[1311] "Input means" refers to a device or interface for a user to input text.

[1312] "Communication means" refers to a device or protocol used to send input text to a server.

[1313] "Analysis means" refers to software or hardware mechanisms that perform morphological analysis, grammatical analysis, and semantic analysis on text received by the server.

[1314] An "error detection means" is a device or program that identifies errors in text based on the analysis results.

[1315] A "correction generation means" is a device or program for generating appropriate corrections after an error has been identified.

[1316] "Transmission means" refers to a device or protocol for sending the revised version to a terminal.

[1317] The "Automatic Correction Function for Food Delivery Applications" is a feature that corrects typos, grammatical errors, and unnatural expressions when users enter order details or special notes.

[1318] The present invention provides an automated text correction tool that assists users in accurately and naturally inputting Japanese text. In this invention, the system is particularly applicable to food delivery applications.

[1319] 1. System Program

[1320] This system consists of the following main elements:

[1321] Input method: An interface for users to enter text. A text input box embedded within a food delivery application.

[1322] Communication method: An HTTP request to send the entered text to the server. Specifically, the text is sent to the server in JSON format.

[1323] Analysis methods: Functions for morphological analysis, grammatical analysis, and semantic analysis of text received by the server. Morphological analysis uses the Janome library, and grammatical and semantic analysis utilize natural language understanding (NLU) models.

[1324] Error detection and correction generation means: Based on the analysis results, errors in the text are detected and appropriate correction suggestions are generated. Examples of errors include incorrect particles and unnatural expressions.

[1325] Transmission method: An HTTP response to send JSON data containing the proposed revisions back to the terminal.

[1326] 2. Program processing description using natural language

[1327] The server receives text sent by the user through the input method. The Janome library is used to perform morphological analysis on the received text, breaking down the sentence into individual words. For example, the input sentence "I will be attending tomorrow's meeting, so thank you in advance" is broken down by morphological analysis into "tomorrow / no / kaigi / ni / participating / suru / masu / node / , / yoroshiku / onegai / shimasu".

[1328] Next, the grammar analysis engine analyzes the part of speech and conjugation of each word to detect grammatical errors. For example, it detects that the expression "sanka suru masu" (to participate) is inappropriate. Subsequently, in the semantic analysis stage, a natural language understanding (NLU) model is used to analyze the meaning of the entire sentence and identify unnatural phrasing. Here, it determines that "onegai shimasu" (please) should be changed to "onegai shimasu" (please).

[1329] Based on the analysis results, the error detection means detects errors, and the correction generation means generates appropriate correction suggestions. The server then sends the correction suggestions back to the terminal as JSON data. The terminal analyzes the received JSON data and displays the correction suggestions to the user. The user reviews the displayed correction suggestions and applies the corrections as needed.

[1330] 3. Specific examples and prompt messages

[1331] For example, a user enters the following order details:

[1332] I am allergic to shrimp, so please remove the shrimp from the dish.

[1333] When this input is sent to the server, it will suggest the following modifications:

[1334] I am allergic to shrimp, so please omit the shrimp from the dish.

[1335] Examples of prompt statements to input into a generative AI model are as follows:

[1336] Analyze the Japanese text entered by the user grammatically and semantically, and suggest appropriate corrections. Example: 'I am allergic to shrimp, so please remove the shrimp.' -> 'I am allergic to shrimp, so please remove the shrimp.'

[1337] This allows users to quickly correct their input and place orders in accurate and natural-sounding Japanese.

[1338] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1339] Step 1:

[1340] The user uses their device to enter order details and special requests into a text input form within the food delivery application. An example of the text entered is, "I am allergic to shrimp, so please omit the shrimp."

[1341] Step 2:

[1342] The terminal sends an HTTP request to the server using a communication method to send the entered text to the server. Specifically, it sends the text data to the server in JSON format.

[1343] Step 3:

[1344] The server analyzes the received text data. First, the server uses the morphological analysis library Janome to break down the text into words. For example, it breaks down "I am allergic to shrimp, so please remove the shrimp." into "Allergy / is / shrimp / , / shrimp / remove / please / please / you / please / please / remove / shrimp".

[1345] Step 4:

[1346] The server uses a grammar analysis engine to analyze the morphologically analyzed word sequence from the perspective of part of speech and conjugation, and detects grammatical errors. For example, it might determine that the unnatural expression "kudasaimasu you onegaishimasu" should be corrected to "onegaishimasu".

[1347] Step 5:

[1348] The server uses a natural language understanding (NLU) model to perform semantic analysis. Here, it analyzes the meaning of the entire sentence and identifies unnatural phrasing and expressions. For example, it might correct "onegaishimasu" to "onegaishimasu."

[1349] Step 6:

[1350] Based on the analysis results, the server uses error detection to identify errors and correction generation to generate appropriate corrections. Specifically, it generates a correction to change "onegaishimasu" to "onegaishimasu".

[1351] Step 7:

[1352] The server sends the proposed revisions back to the terminal as JSON data via a transmission method. The transmitted data includes both the original and revised text.

[1353] Step 8:

[1354] The terminal parses the received JSON data and displays suggested corrections to the user. For example, it might display "I am allergic to shrimp, so please remove the shrimp."

[1355] Step 9:

[1356] The user reviews the displayed revisions and applies them as needed. This confirms that the corrected text is finalized in the order.

[1357] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1358] The present invention is an automated proofreading tool that allows users to input Japanese text accurately and naturally, and provides feedback that takes the user's feelings into consideration. In this embodiment, the specific program processing flow will be described.

[1359] First, the user enters a Japanese sentence into a text input form on their device. For example, suppose the user enters the following sentence:

[1360] I will be attending tomorrow's meeting, so I look forward to seeing you there. I'm very nervous.

[1361] Next, the terminal sends an HTTP request to the server using a communication method to send the entered text to the server. At this time, the text data is sent to the server.

[1362] The server begins processing the received text data for analysis. Specifically, the following four analysis methods are used:

[1363] 1. Morphological Analysis: The server uses a morphological analysis library to break down the text into individual words. For example, the input sentence is broken down into "I will be attending tomorrow's meeting, so please be kind to me. I am very nervous." The part of speech and conjugation of each word are also analyzed.

[1364] 2. Grammar Analysis: The grammar analysis engine detects grammatical errors. In the example above, it detects that the expression "sanka suru masu" is inappropriate.

[1365] 3. Semantic Analysis: A natural language understanding (NLU) model analyzes the meaning of the entire sentence and identifies unnatural phrasing. In this case, it determines that "onegaishimasu" should be changed to "onegaishimasu".

[1366] 4. Sentiment Analysis: The server's sentiment analysis tool identifies the emotions contained in the user's text. This analysis determines that "I'm very nervous" represents the emotion of "nervousness."

[1367] Based on the analysis results, the server uses error detection and correction generation means to identify errors and generate appropriate correction suggestions. Specific correction suggestions are as follows:

[1368] Error: "surumasu" -> Corrected: "sanka shimasu"

[1369] Error: "onegaishimasu" -> Correction: "onegaishimasu"

[1370] Furthermore, the emotion engine generates user-friendly feedback based on identified emotions. For example, in response to "I'm very nervous," it might generate advice such as "Try to relax before the meeting."

[1371] After the proposed revisions and sentiment feedback are generated, the server sends them back to the terminal as JSON data via a transmission method.

[1372] The device parses the received JSON data and displays suggested revisions and sentiment feedback to the user. For example, the user might see the following:

[1373] I will be attending tomorrow's meeting, so I look forward to seeing you there. I'm very nervous.

[1374] Proposed amendment:

[1375] "I will be attending tomorrow's meeting, so I look forward to seeing you there."

[1376] Emotional feedback:

[1377] "Please relax and approach the meeting with ease."

[1378] Users review the displayed correction suggestions and sentiment feedback, and apply the corrections as needed. This corrects errors in the entered Japanese text and allows them to receive feedback that takes their feelings into consideration.

[1379] The above describes a specific embodiment for carrying out the present invention. By using this system, users can quickly and easily correct errors in Japanese text and receive emotionally sensitive feedback.

[1380] The following describes the processing flow.

[1381] Step 1:

[1382] The user enters a Japanese sentence into a text input form on their device. For example, they might enter, "I will be attending tomorrow's meeting, so I look forward to seeing you there. I'm very nervous." Once they have finished typing, they click the submit button.

[1383] Step 2:

[1384] The terminal receives the input text data and uses a communication method to send an HTTP request to the server. This request contains the text data.

[1385] Step 3:

[1386] The server parses the received HTTP request and extracts text data. The extracted text data is then passed to the morphological analysis module.

[1387] Step 4:

[1388] The server's morphological analysis module breaks down text into individual words and analyzes the part of speech and conjugation of each word. For example, the text "I will be attending tomorrow's meeting, so I look forward to seeing you. I'm very nervous." is broken down into "tomorrow / noun of / particle meeting / noun to / particle attend / verb do / auxiliary verb do / auxiliary verb so / particle, / symbol nice to see / adverb please / verb do / auxiliary verb. / symbol very / adverb nervous / noun feeling / auxiliary verb I am / auxiliary verb".

[1389] Step 5:

[1390] The analyzed word information is passed to the grammar analysis module. The server's grammar analysis module identifies grammatical errors and syntactic errors in the sentence. At this stage, the incorrect expression "sanka suru masu" (participate) is detected.

[1391] Step 6:

[1392] The results of the grammatical analysis are passed to the semantic analysis module. The server's semantic analysis module understands the context and identifies unnatural phrasing and misuse. This analysis determines that "onegaishimasu" should be corrected to "onegaishimasu".

[1393] Step 7:

[1394] Next, the server's sentiment analysis tool identifies the emotions contained in the user's text. It recognizes the emotion of tension from the expression "I'm very nervous."

[1395] Step 8:

[1396] Based on the results of grammatical analysis, semantic analysis, and sentiment analysis, the server uses error detection and correction generation means to identify errors and generate appropriate correction suggestions and sentiment feedback. For example, "participate" is corrected to "will participate," and "please" is corrected to "please." In addition, sentiment feedback such as "Please relax and enjoy the meeting" is generated for a nervous user.

[1397] Step 9:

[1398] The generated corrections and sentiment feedback are formatted in JSON format. This data includes the location of the error, the original text, the correction, and the sentiment feedback. The server sends this JSON data back to the terminal via the transmission method.

[1399] Step 10:

[1400] The device parses the received JSON data and displays suggested revisions and sentiment feedback to the user. For example, it may be displayed as follows:

[1401] I will be attending tomorrow's meeting, so I look forward to seeing you there. I'm very nervous.

[1402] Proposed amendment:

[1403] "I will be attending tomorrow's meeting, so I look forward to seeing you there."

[1404] Emotional feedback:

[1405] "Please relax and approach the meeting with ease."

[1406] Step 11:

[1407] Users review the displayed correction suggestions and sentiment feedback, and apply the corrections as needed. This corrects errors in the entered Japanese text and allows them to receive feedback that takes their feelings into consideration.

[1408] (Example 2)

[1409] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1410] When users input Japanese text, the challenge lies in accurately identifying and correcting errors, as well as providing feedback that takes the user's emotions into consideration. Conventional systems only point out errors and fail to provide emotionally sensitive feedback, thus failing to sufficiently improve user satisfaction and the naturalness of the text.

[1411] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1412] In this invention, the server includes analysis means for performing morphological, grammatical, semantic, and sentiment analysis on received text; error detection and correction generation means for pointing out errors in the text, generating suggested corrections, and providing feedback that takes the user's feelings into consideration, based on the analysis results; and transmission means for sending the suggested corrections and sentiment feedback to the terminal. This allows the user to correct errors while simultaneously receiving feedback that takes their feelings into consideration.

[1413] "Text" refers to a string of characters entered by the user, and may include Japanese sentences or words.

[1414] "Input means" refers to devices or interfaces for users to input text, including keyboards and touchscreens.

[1415] "Communication method" refers to the means by which a terminal sends text entered into it to a server, and this involves using an internet connection or a communication protocol (e.g., HTTP).

[1416] A "server" is a central processing unit that receives, analyzes, and processes text sent from a user's terminal.

[1417] "Analysis means" includes software and libraries for morphological analysis, grammatical analysis, semantic analysis, and sentiment analysis of received text.

[1418] Morphological analysis is the process of breaking down text into individual words and analyzing the part of speech and conjugation of each word.

[1419] "Grammar analysis" is the process of detecting grammatical errors and syntactic errors within a text.

[1420] "Semantic analysis" is the process of understanding the overall meaning of a text and identifying unnatural expressions.

[1421] "Sentiment analysis" is the process of identifying the user's emotions contained in text and generating appropriate feedback.

[1422] "Error detection means and correction generation means" refers to processing means for identifying errors in text based on analysis results and generating proposed corrections.

[1423] "Transmission means" refers to the means by which the server sends back the proposed revisions and sentiment feedback generated by the server to the terminal.

[1424] A "revision proposal" is a suggestion to correct errors detected by the analysis method.

[1425] "Emotional feedback" refers to providing advice and comments that take the user's emotions into consideration, based on emotion analysis.

[1426] The present invention is an automated proofreading tool that allows users to input Japanese text accurately and naturally, and provides feedback that takes into account the user's emotions. This embodiment details the specific implementation method of this system.

[1427] System Configuration

[1428] This system consists of a user terminal, a server, and various analysis tools.

[1429] 1. Terminal

[1430] This refers to a device used by a user to input text. Examples include PCs, smartphones, and tablets.

[1431] 2. Server

[1432] It is a central processing unit equipped with a high-performance processor and sufficient memory. The server receives text sent from the user and performs various analyses and processing.

[1433] 3. Analysis method

[1434] The specific software used includes the following:

[1435] Morphological analysis libraries (e.g., MeCab)

[1436] Grammar analysis engine

[1437] Natural Language Understanding (NLU) Models

[1438] Emotion analysis engine

[1439] Operation details

[1440] 1. User text input

[1441] The user enters Japanese text into an input form on their device. For example, they might enter the sentence, "I will be attending tomorrow's meeting, so thank you in advance. I'm very nervous."

[1442] 2. Sending text to the server

[1443] The entered text is sent from the terminal to the server using an HTTP request.

[1444] 3. Server-based analysis

[1445] The server processes the received text sequentially as follows:

[1446] Morphological analysis: This process breaks down text into individual words and analyzes their parts of speech and conjugations.

[1447] For example, "I will be attending tomorrow's meeting, so thank you in advance. I'm very nervous." can be broken down into "Tomorrow / no / kaigi / ni / taku / suru / masu / no / , / yoroshiku / onegai / shimasu / . / toku / kichika / shite / imasu."

[1448] Grammar analysis: Detects grammatical errors and syntactic errors in a sentence.

[1449] Example: Detect that "sanka suru masu" is grammatically incorrect.

[1450] Semantic analysis: Analyzes the meaning of the entire sentence and identifies unnatural expressions.

[1451] Example: It is decided that "onegaishimasu" should be changed to "onegaishimasu".

[1452] Sentiment analysis: Identifying the emotions contained in the text.

[1453] Example: Determine that "I am very nervous" expresses the emotion of "nervousness".

[1454] 4. Generating revised proposals and emotional feedback

[1455] The server identifies errors based on the analysis results and generates suggested corrections. It also generates feedback that takes the user's emotions into consideration based on sentiment analysis.

[1456] Example 1: Correct the error "sanka suru masu" to "sanka shimasu".

[1457] Example 2: Change "onegaishimasu" to "onegaishimasu".

[1458] Example 3: In response to "I'm very nervous," generate feedback such as "Please relax and approach the meeting with confidence."

[1459] 5. Return of data

[1460] The proposed revisions and emotional feedback will be sent to the device as data in JSON format.

[1461] 6. Display to the user

[1462] The device analyzes the received data and displays suggested revisions and sentiment feedback to the user.

[1463] Example: It will be displayed as follows:

[1464] I will be attending tomorrow's meeting, so I look forward to seeing you there. I'm very nervous.

[1465] Proposed amendment:

[1466] "I will be attending tomorrow's meeting, so I look forward to seeing you there."

[1467] Emotional feedback:

[1468] "Please relax and approach the meeting with ease."

[1469] Example of a prompt

[1470] The following is an example of a set prompt statement:

[1471] Please enter a Japanese sentence, correct any errors, and provide emotionally sensitive feedback. Example sentence: I will be attending tomorrow's meeting, so I look forward to seeing you there. I'm very nervous.

[1472] By applying this system, users can quickly and easily correct errors in Japanese text and receive emotionally sensitive feedback.

[1473] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1474] Step 1:

[1475] The user enters text.

[1476] Operation: The user enters Japanese text into an input form on the terminal. For example, they might enter, "I will be attending tomorrow's meeting, thank you in advance. I'm very nervous."

[1477] Input: Japanese text.

[1478] Output: The text displayed in the input form on the terminal.

[1479] Step 2:

[1480] This is the procedure for sending the text entered by the terminal to the server.

[1481] Operation: The terminal sends the entered text to the server using an HTTP request.

[1482] Input: Japanese text entered by the user.

[1483] Output: Text data sent to the server.

[1484] Step 3:

[1485] This is the procedure for performing morphological analysis on text received by the server.

[1486] Operation: The server uses a morphological analysis library to break down the text into individual words and analyze the part of speech and conjugation of each word. For example, it breaks down "I will be attending tomorrow's meeting, so thank you in advance. I'm very nervous." into "Tomorrow / no / kaigi / ni / participating / suru / masu / no / , / yoroshiku / onegai / shimasu / . / totemo / kyouki / shite / imasu".

[1487] Input: Text data sent to the server.

[1488] Output: Morphological analysis results (text data broken down into individual words).

[1489] Step 4:

[1490] This is the procedure by which the server performs grammatical analysis of the text.

[1491] Operation: The server uses a grammar analysis engine to detect grammatical errors and syntactic errors in the text. For example, it detects that the expression "sanka surumasu" (to participate) is inappropriate.

[1492] Input: Morphological analysis results.

[1493] Output: Grammar analysis results (data including information on grammatical errors and syntactic errors).

[1494] Step 5:

[1495] This is the procedure by which the server performs semantic analysis of text.

[1496] Operation: The server uses a natural language understanding (NLU) model to analyze the meaning of the entire text and identify unnatural phrasing. For example, it might determine that "onegaishimasu" should be corrected to "onegaishimasu".

[1497] Input: Grammar analysis result.

[1498] Output: Semantic analysis results (data containing information about unnatural phrasing).

[1499] Step 6:

[1500] This is the procedure for the server to perform sentiment analysis on text.

[1501] Operation: Uses an emotion analysis engine to identify the user's emotions contained in the text. For example, it determines that "I'm very nervous" represents the emotion "nervous."

[1502] Input: Semantic analysis results.

[1503] Output: Sentiment analysis results (data including user sentiment information).

[1504] Step 7:

[1505] This is a procedure in which the server performs error detection, proposes corrections, and generates sentiment feedback.

[1506] Operation: Based on the analysis results, the server identifies errors and generates appropriate correction suggestions. It also generates feedback that takes the user's emotions into account based on the sentiment analysis results. For example, it corrects "surumasu" to "shimasu" and "onegaishimasu" to "onegaishimasu". In response to "I'm very nervous," it generates feedback such as "Please relax and approach the meeting with ease."

[1507] Input: Grammatical analysis results, semantic analysis results, sentiment analysis results.

[1508] Output: Revised proposals and emotional feedback.

[1509] Step 8:

[1510] This is the procedure for the server to send suggested revisions and emotional feedback to the terminal.

[1511] Operation: The server generates JSON data containing suggested revisions and sentiment feedback, and sends it back to the terminal.

[1512] Input: Suggested revisions and emotional feedback.

[1513] Output: JSON data sent to the terminal.

[1514] Step 9:

[1515] This is the procedure for the device to display suggested fixes and emotional feedback to the user.

[1516] Operation: The device parses the received JSON data and displays suggested revisions and sentiment feedback to the user. For example, it might display: "I will be attending tomorrow's meeting, thank you in advance. I'm very nervous. Suggested revision: 'I will be attending tomorrow's meeting, thank you in advance.' Sentiment feedback: 'Please relax during the meeting.'"

[1517] Input: JSON data sent to the terminal.

[1518] Output: Suggested revisions and sentiment feedback displayed to the user.

[1519] (Application Example 2)

[1520] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1521] In virtual stores, customers often generate incorrect or unnatural sentences when entering text. Furthermore, there is a lack of means to provide feedback that is appropriate to the customer's current emotional state. This presents a challenge in improving the customer experience.

[1522] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a morphological analysis means, a grammatical analysis means, a semantic analysis means, an emotion analysis means, an error detection means and correction generation means, and a transmission means. This makes it possible not only to correct errors and unnatural expressions, but also to provide feedback based on the customer's emotional state.

[1523] A "user" is a person who uses this system to input text and sends that input to the server.

[1524] "Input means" refers to a device or interface for a user to input text.

[1525] "Communication means" refers to means, including networks and protocols, for sending input text to a server.

[1526] A "server" is a device or system that analyzes text sent by a user and generates suggested revisions and sentiment feedback.

[1527] A "morphological analysis tool" is a function that breaks down input text into words and analyzes the part of speech and conjugation of each word.

[1528] A "grammatical analysis tool" is a function that detects grammatical errors and syntactic errors in the input text.

[1529] A "semantic analysis tool" is a function that analyzes the overall meaning of the input text and identifies unnatural phrasing.

[1530] "Emotion analysis means" refers to a function that analyzes the emotions contained in the input text.

[1531] An "error detection means" is a function that identifies errors in text based on the analysis results.

[1532] A "correction generation means" is a function that generates appropriate corrections for identified errors.

[1533] "Transmission means" refers to the means of sending the generated revisions and emotional feedback to the user's device.

[1534] "Emotional feedback" refers to advice and comments provided based on the emotions contained in the user's text.

[1535] This invention is a system that corrects text entered by customers in a virtual store into accurate and natural sentences, and also provides feedback that takes into account the customer's feelings.

[1536] This system mainly consists of input means, communication means, server, morphological analysis means, grammatical analysis means, semantic analysis means, sentiment analysis means, error detection means, correction generation means, and transmission means.

[1537] First, the user enters text within the virtual store using an input device. This input device could be a smart glasses or a head-mounted display.

[1538] The text entered by the user is sent to the server via a communication method. Network communication methods such as the HTTP protocol are used for this purpose.

[1539] The server analyzes the received text. Specifically, it uses morphological analysis to break down the text into individual words and analyzes the part of speech and conjugation of each word.

[1540] Next, the grammar analysis means uses the analysis results to detect grammatical errors and syntactic errors in the input text. The semantic analysis means also analyzes the meaning of the entire text and identifies unnatural expressions.

[1541] Furthermore, the emotion analysis mechanism analyzes the emotions contained in the input text. Specifically, it uses a generative AI model to extract emotions from the text and generates corresponding feedback.

[1542] Based on the analysis results, the error detection means identifies errors in the text, and the correction generation means generates appropriate corrections and sentiment-based feedback. For example, for the text "I will be attending tomorrow's meeting, so thank you in advance. I'm very nervous," the correction is changed to "I will be attending tomorrow's meeting, so thank you in advance," and sentiment feedback such as "Please relax during the meeting" is provided.

[1543] These suggested revisions and emotional feedback are sent to the user's device via a transmission method. The user can review the submitted feedback and apply revisions as needed, thereby obtaining accurate, natural-sounding Japanese text and emotionally sensitive feedback.

[1544] For example, if the prompt is "I will be attending tomorrow's meeting, so thank you in advance. I'm very nervous," the system will respond by generating a revised version, "I will be attending tomorrow's meeting, so thank you in advance," and emotional feedback, "Please relax during the meeting."

[1545] In this way, this invention improves the accuracy and emotional consideration of text in customer service within virtual stores.

[1546] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1547] Step 1:

[1548] Users enter text within the virtual store using input devices. These input devices include smart glasses and head-mounted displays. For example, the text entered might read, "I will be attending tomorrow's meeting, so I look forward to seeing you there. I'm very nervous."

[1549] Input: Text entered by the user

[1550] Output: Raw text entered into the terminal

[1551] Step 2:

[1552] The terminal sends the entered text to the server using a communication method. This communication method typically involves network communication protocols such as HTTP.

[1553] Input: Raw text entered on the terminal

[1554] Output: Text data sent to the server

[1555] Step 3:

[1556] The server uses morphological analysis to break down the received text into individual words and analyzes the part of speech and conjugation of each word. For this purpose, it uses a morphological analysis library (e.g., Janome). For example, "I will be attending tomorrow's meeting, so thank you in advance. I'm very nervous." is analyzed as "Tomorrow / no / meeting / ni / participating / suru / masu / no / , / yoroshiku / onegai / shimasu / . / totemo / kyouki / shite / imasu".

[1557] Input: Text data sent to the server

[1558] Output: Text data broken down into words

[1559] Step 4:

[1560] The grammar analysis means detects grammatical errors and syntax errors in a sentence based on the results of morphological analysis. For example, a grammar error such as "するます" is detected.

[1561] Input: Text data decomposed into words

[1562] Output: Data with grammar errors identified

[1563] Step 5:

[1564] The semantic analysis means analyzes the overall meaning of the text using a natural language understanding model (NLU) and identifies unnatural expressions. For example, it is identified that "よろしくおねがいします" should be corrected to "お願いします".

[1565] Input: Data with grammar errors identified

[1566] Output: Data with unnatural expressions identified

[1567] Step 6:

[1568] The sentiment analysis means analyzes the sentiment contained in the input text using a generative AI model. For example, it is determined that "とても緊張しています" expresses the sentiment of "nervousness".

[1569] Input: Data with unnatural expressions identified

[1570] Output: Data with sentiment identified

[1571] Step 7:

[1572] The error detection means identifies errors in the text based on the above analysis results. For example, it is identified that "するます" is an error that should be corrected to "します".

[1573] Input: Data with sentiment identified

[1574] Output: Error and correction instruction data

[1575] Step 8:

[1576] The correction generation mechanism generates appropriate correction suggestions for identified errors. It also generates user-friendly feedback based on sentiment analysis results. For example, it might generate a correction suggestion like, "I will be attending tomorrow's meeting, thank you in advance," along with emotional feedback such as, "Please relax during the meeting."

[1577] Input: Error and correction instruction data

[1578] Output: Proposed revisions and sentiment feedback data

[1579] Step 9:

[1580] The server sends the generated revisions and sentiment feedback to the user's device via a transmission method. These are then displayed on the user's device.

[1581] Input: Proposed revisions and sentiment feedback data

[1582] Output: Suggested revisions and sentiment feedback sent to the user's terminal.

[1583] Step 10:

[1584] Users review the proposed revisions and sentiment feedback, and apply the changes as needed. This ensures accurate and natural Japanese text and sentiment-sensitive feedback.

[1585] Input: Correction suggestions and sentiment feedback sent to the user's terminal.

[1586] Output: Final text with corrections and sentiment feedback applied.

[1587] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1588] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1589] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[1590] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1591] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[1592] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[1593] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[1594] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[1595] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[1596] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[1597] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[1598] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[1599] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[1600] 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.

[1601] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[1602] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[1603] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[1604] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[1605] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[1606] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[1607] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted as being incorporated by reference.

[1608] The following is further disclosed regarding the embodiments described above.

[1609] (Claim 1)

[1610] An input method for the user to enter text,

[1611] A communication means for sending the input text to the server,

[1612] The server has an analysis mechanism that performs morphological analysis, grammatical analysis, and semantic analysis on the text it receives.

[1613] An error detection means and a correction generation means that identify errors in the text and generate correction proposals based on the analysis results,

[1614] A means of sending the revised proposal to the terminal,

[1615] A system that includes this.

[1616] (Claim 2)

[1617] The system according to claim 1, wherein the morphological analysis means breaks down text into words and analyzes the parts of speech and conjugations.

[1618] (Claim 3)

[1619] The system according to claim 1, wherein the grammatical analysis means detects grammatical errors or syntactic errors in a sentence.

[1620] "Example 1"

[1621] (Claim 1)

[1622] An input method for the user to enter text,

[1623] A communication means for sending the input text to the server,

[1624] The server has an analysis mechanism that performs morphological analysis, grammatical analysis, and semantic analysis on the text it receives.

[1625] An error detection means and a correction generation means that identify errors in the text and generate correction proposals based on the analysis results,

[1626] A means of sending the revised proposal to the terminal,

[1627] A display means for displaying the proposed corrections received by the terminal to the user,

[1628] A system that includes this.

[1629] (Claim 2)

[1630] The system according to claim 1, wherein the morphological analysis means breaks down text into words and analyzes the parts of speech and conjugations.

[1631] (Claim 3)

[1632] The system according to claim 1, wherein the grammatical analysis means detects grammatical errors or syntactic errors in a sentence.

[1633] "Application Example 1"

[1634] (Claim 1)

[1635] An input method for the user to enter text,

[1636] A communication means for sending the input text to the server,

[1637] The server has an analysis mechanism that performs morphological analysis, grammatical analysis, and semantic analysis on the text it receives.

[1638] An error detection means and a correction generation means that identify errors in the text and generate correction proposals based on the analysis results,

[1639] A means of sending the revised proposal to the terminal,

[1640] A method featuring an automated proofreading function for food delivery applications to correct typos, grammatical errors, and unnatural expressions in the input of order details and special notes by the user,

[1641] A system that includes this.

[1642] (Claim 2)

[1643] The system according to claim 1, wherein the morphological analysis means breaks down text into words and analyzes the parts of speech and conjugations.

[1644] (Claim 3)

[1645] The system according to claim 1, wherein the grammatical analysis means detects grammatical errors or syntactic errors in a sentence.

[1646] "Example 2 of combining an emotion engine"

[1647] (Claim 1)

[1648] An input method for the user to enter text,

[1649] A communication means for sending the input text to the server,

[1650] The server performs morphological analysis, grammatical analysis, semantic analysis, and sentiment analysis on the text it receives.

[1651] An error detection means and a correction generation means that, based on the analysis results, identifies errors in the text, generates correction suggestions, and provides feedback that takes into account the user's feelings.

[1652] A means for sending revised proposals and emotional feedback to a terminal,

[1653] A system that includes this.

[1654] (Claim 2)

[1655] The system according to claim 1, wherein the morphological analysis means breaks down text into words and analyzes the parts of speech and conjugations.

[1656] (Claim 3)

[1657] The system according to claim 1, wherein the grammatical analysis means detects grammatical errors or syntactic errors in a sentence.

[1658] (Claim 4)

[1659] The system according to claim 1, wherein the emotion analysis means identifies the user's emotions contained in the text and generates corresponding feedback.

[1660] "Application example 2 when combining with an emotional engine"

[1661] (Claim 1)

[1662] An input method for the user to enter text,

[1663] A communication means for sending the input text to the server,

[1664] The server performs morphological analysis, grammatical analysis, semantic analysis, and sentiment analysis on the text it receives.

[1665] An error detection means and a correction generation means that, based on the analysis results, identifies errors in the text and generates correction suggestions and sentiment-based feedback,

[1666] A means for sending revised proposals and emotional feedback to a terminal,

[1667] A system that includes this.

[1668] (Claim 2)

[1669] The system according to claim 1, wherein the morphological analysis means breaks down text into words and analyzes the parts of speech and conjugations.

[1670] (Claim 3)

[1671] The system according to claim 1, wherein the grammatical analysis means detects grammatical errors or syntactic errors in a sentence. [Explanation of Symbols]

[1672] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. An input method for the user to enter text, A communication means for sending the input text to the server, The server has an analysis mechanism that performs morphological analysis, grammatical analysis, and semantic analysis on the text it receives. An error detection means and a correction generation means that identify errors in the text and generate correction proposals based on the analysis results, A means of sending the revised proposal to the terminal, A system that includes this.

2. The system according to claim 1, wherein the morphological analysis means breaks down text into words and analyzes the parts of speech and conjugations.

3. The system according to claim 1, wherein the grammatical analysis means detects grammatical errors or syntactic errors in a sentence.

Images