system

Abstract

Provide a system. [Solution] The generated artificial intelligence model is used to analyze the text data. means for detecting ambiguous expressions in text data; means for providing multiple interpretations of the detected ambiguous expression; means for presenting the provided interpretations to a user; A system including:

Description

[Technical Field]

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

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

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

[0004] In text-based communication, the use of punctuation and choice of particles often lead to different interpretations. This can lead to misunderstandings and miscommunication, especially in email and messaging applications. There is a need for technology that can prevent such misunderstandings and enable clearer and more accurate communication. [Means for solving the problem]

[0005] To solve this problem, a system is provided that includes the following means. First, the system includes means for analyzing text data using a generated artificial intelligence model and detecting ambiguous expressions in the text data. The system also includes means for providing multiple interpretations of the detected ambiguous expressions and means for presenting these interpretations to a user. This allows the user to recognize ambiguous expressions and obtain information to avoid misunderstandings. Furthermore, by including means for receiving text data from a user, the system analyzes the input text and can detect ambiguous expressions based on natural language processing. Such a system makes it possible to prevent misunderstandings in text-based communication.

[0006] "Generated artificial intelligence model" refers to an artificial intelligence algorithm or machine learning model that has the ability to analyze text data and detect ambiguous expressions.

[0007] "Text data" refers to any character string information entered by a user, including emails, messages, documents, and the like.

[0008] An "ambiguous expression" refers to a linguistic expression that can be interpreted differently depending on the context and placement of punctuation.

[0009] "Analysis" refers to the process of breaking down text data to understand its structure and meaning and identify ambiguous expressions.

[0010] "Detection" refers to identifying ambiguous expressions as a result of analysis.

[0011] "Interpretation" refers to giving a contextual meaning to an expression.

[0012] "Providing" refers to displaying the detected ambiguous expression and its multiple interpretations to the user.

[0013] "User" refers to an entity that uses the system to input text data and receives the analysis results.

[0014] "Punctuation" refers to symbols such as dots and commas that are used to indicate breaks or divisions in a sentence.

[0015] "Particles" refer to words used in Japanese grammar to indicate relationships between subjects, predicates, objects, etc. [Brief explanation of the drawings]

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

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

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

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

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

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

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

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

[0024] [First embodiment]

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

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

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

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

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

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

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

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

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

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

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

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

[0037] The present invention relates to a system that analyzes text data using a generated artificial intelligence model, detects ambiguous expressions in the text data, and presents multiple interpretations of the expressions to a user.

[0038] The server first loads the necessary generative AI model, specifically, a Transformer model, prepared from an existing natural language processing model such as "t5-small," which has the ability to detect ambiguous expressions and generate interpretations.

[0039] A user inputs text data from a terminal. For example, the user might input the text "I didn't agree with him at the meeting yesterday, but I agree with his opinion." This text may be interpreted differently depending on the user's intention and context.

[0040] The server then receives the text data entered by the user and analyzes it using a generative AI model. This analysis identifies ambiguous expressions in the text data. The model takes into account contextual information and punctuation placement to detect expressions that may be misleading.

[0041] The server then generates different interpretations for the detected ambiguous expressions. For example, it can provide the following interpretations for the above text:

[0042] 1. "I didn't agree with the whole meeting yesterday, but I might agree with some of his points."

[0043] 2. "I don't agree with his opinion in its entirety, but I might agree with him under certain circumstances."

[0044] The server then displays the generated interpretations on the terminal, allowing the user to refer to these interpretations and take measures to avoid misunderstandings.

[0045] As described above, the present invention provides a system that detects ambiguous expressions in text-based communication and presents their interpretations to prevent misunderstandings. By using this system, users can achieve more accurate and clearer communication.

[0046] The processing flow will be explained below.

[0047] Step 1:

[0048] The server loads the generative AI model, specifically the Transformer model "t5-small" based on natural language processing, and prepares it for analysis.

[0049] Step 2:

[0050] A user inputs text data through a terminal. For example, the user inputs "I didn't agree with him at the meeting yesterday, but I agree with his opinion."

[0051] Step 3:

[0052] The server receives text data entered by the user, which is then analyzed.

[0053] Step 4:

[0054] The server uses the generative AI model to analyze the input text data. Specifically, it performs an analysis process to detect ambiguous expressions within the text data.

[0055] Step 5:

[0056] As a result of the analysis, the server generates multiple interpretations for the detected ambiguous expressions. For example, it generates the following interpretations for the above text: "1. I did not agree with the entire meeting yesterday, but I may agree with some of his opinions." "2. I do not agree with his entire opinion, but I may agree with his opinion under certain circumstances."

[0057] Step 6:

[0058] The server transmits the generated interpretations to the user's terminal.

[0059] Step 7:

[0060] The terminal displays the received interpretation to the user, who then checks the displayed interpretation and understands the ambiguous expressions in the text data.

[0061] Step 8:

[0062] The user can refer to the proposed interpretation and correct or add to the text data as necessary. This process allows the user to choose appropriate expressions to avoid misunderstandings.

[0063] Example 1

[0064] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0065] In natural language text data exchange, ambiguous expressions can lead to misunderstandings. This misunderstanding can cause serious problems, especially in business communications and important message exchanges. Conventional systems have had difficulty automatically detecting such ambiguous expressions and providing clear interpretations to users.

[0066] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0067] In this invention, the server includes means for loading the generated AI model as a Transformer model, means for receiving user input and sending it to the server, and means for receiving analysis results from the server and displaying them to the user, thereby enabling highly accurate detection of ambiguous expressions in text data and providing the user with their interpretations.

[0068] A "generated artificial intelligence model" refers to a machine learning algorithm that has been trained on a set of data and used to perform a specific task.

[0069] "Text data" refers to a collection of information expressed as character strings, including sentences and messages written in natural language.

[0070] "Polyambiguous expressions" refer to words or phrases that can be interpreted in multiple ways depending on the context or situation.

[0071] "Interpretation" refers to supplementary explanations and possibilities that clarify the meaning contained in the text data and are provided to help users deepen their understanding.

[0072] A "Transformer model" is a type of machine learning model used in natural language processing, and refers to a model that has the ability to analyze text data taking context into account.

[0073] "Loading" refers to the operation of reading a program or data into memory and making it available for use.

[0074] "User" refers to the person who uses the system and is the entity that inputs the text data to be analyzed.

[0075] A "server" refers to a computer that provides services over a network and plays a central role in processing and managing data.

[0076] A "terminal" refers to a computer or device that is directly used by a user, and is a device that communicates with a server to send and receive data.

[0077] "Analysis results" refers to the information and data obtained after the generated artificial intelligence model analyzes the text data.

[0078] The present invention relates to a system that analyzes text data using a generated artificial intelligence model, detects ambiguous expressions in the text data, and presents multiple interpretations of the expressions to a user. An embodiment of the present invention will be described in detail below.

[0079] Hardware and Software Configuration

[0080] The server is a high-performance computer with hardware acceleration, such as a GPU. The server performs the following functions:

[0081] 1. The generated AI model is loaded as a Transformer model. At this time, a natural language processing model such as "t5-small" is downloaded from the internet and loaded into memory.

[0082] 2. Receive text data sent by the user and analyze it using the generative AI model.

[0083] 3. Detect ambiguous expressions and generate multiple interpretations for them.

[0084] 4. The generated interpretation results are sent to the user's terminal.

[0085] A terminal is a computer, mobile device, or other device that is directly operated by a user and performs the following actions:

[0086] 1. Provide an interface for users to enter text data, implemented as a web or desktop application.

[0087] 2. Send the text data entered by the user to the server.

[0088] 3. Display the analysis results received from the server to the user.

[0089] A user is someone who uses this system and performs the following operations:

[0090] 1. Enter the text data you want to analyze through the device interface. For example, enter the text "I didn't agree with him in the meeting yesterday, but I agree with his opinion."

[0091] 2. The input data is sent to the server and waits for parsing.

[0092] 3. The interpretation results sent from the server are displayed on the terminal.

[0093] Specific examples

[0094] Consider the case where the user enters the following text:

[0095] For example: "I disagreed with him in the meeting yesterday, but I agree with his opinion."

[0096] This text data is sent to the server, which analyzes it using the "t5-small" model, which generates the following interpretation:

[0097] 1. "I didn't agree with the whole meeting yesterday, but I might agree with some of his points."

[0098] 2. "I don't agree with his opinion in its entirety, but I might agree with him under certain circumstances."

[0099] These interpretations are displayed on the user's device, allowing the user to confirm the interpretation that best suits their intentions.

[0100] In this way, this system allows users to easily interpret ambiguous expressions and helps prevent misunderstandings.

[0101] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0102] Step 1:

[0103] The server loads the generative AI model.

[0104] Specific behavior:

[0105] The server downloads a Transformer model, such as "t5-small," from an internet resource and loads it into memory, using the necessary libraries, such as PyTorch and the Transformers library.

[0106] Input and Output:

[0107] Input: Model resource URL on the Internet

[0108] Output: Generative AI model loaded in memory

[0109] What happens:

[0110] The server retrieves the model file from the specified URL and loads it into memory, where it is ready to be used for text analysis.

[0111] Step 2:

[0112] The user inputs text data from the terminal.

[0113] Specific behavior:

[0114] The user inputs the text data to be analyzed using a device interface (web application or desktop application).

[0115] Input and Output:

[0116] Input: User text data (e.g., "I disagreed with him in the meeting yesterday, but I agree with his opinion.")

[0117] Output: Input data in the terminal

[0118] What happens:

[0119] The user enters the text they want to analyze in the text field and clicks the submit button. This data is stored internally and is ready for the next step.

[0120] Step 3:

[0121] The terminal transmits the input text data to the server.

[0122] Specific behavior:

[0123] The device sends the text data entered by the user to the server as an HTTP request. The data is encoded in JSON format and sent to the API endpoint using the POST method.

[0124] Input and Output:

[0125] Input: Text data in the terminal

[0126] Output: Text data sent to the server

[0127] What happens:

[0128] The terminal encodes the user's input data in JSON format and issues a POST request to the specified server endpoint.

[0129] Step 4:

[0130] The server uses a generative AI model to analyze the received text data.

[0131] Specific behavior:

[0132] The server inputs the received text data into the t5-small model and begins analysis. The model takes into account contextual information and punctuation placement to identify ambiguous expressions within the text.

[0133] Input and Output:

[0134] Input: Text data sent to the server

[0135] Output: Analysis results with ambiguous expressions identified

[0136] What happens:

[0137] The server runs the text data through a generative AI model to detect ambiguous expressions, which are then stored in memory as analysis results.

[0138] Step 5:

[0139] The server generates multiple interpretations for the detected ambiguous expression.

[0140] Specific behavior:

[0141] The server generates multiple interpretations based on the analysis results obtained from the generative AI model. For example, for expressions that can be interpreted differently, it generates sentences that explain each of their meanings.

[0142] Input and Output:

[0143] Input: Analysis results (identification of ambiguous expressions)

[0144] Output: Multiple interpretations generated

[0145] What happens:

[0146] The server uses a generative AI model to analyze the context of ambiguous expressions and generate multiple possible interpretations, which are then stored in JSON format.

[0147] Step 6:

[0148] The server transmits the generated interpretation results to the user's terminal.

[0149] Specific behavior:

[0150] The server returns the generated interpretation results to the user's device as an HTTP response. The data is again encoded in JSON format and sent.

[0151] Input and Output:

[0152] Input: Generated interpretation

[0153] Output: Interpretation results sent to the user's terminal

[0154] What happens:

[0155] The server encodes the generated interpretation results in JSON format and returns an HTTP response to the user's terminal.

[0156] Step 7:

[0157] The terminal displays the multiple interpretation results received from the server to the user.

[0158] Specific behavior:

[0159] The application or web interface used by the user displays the interpretation results received from the server in an easy-to-understand format, for example as a list of interpretations displayed on the interface for the user to review.

[0160] Input and Output:

[0161] Input: Interpretation result sent from the server

[0162] Output: The interpretation results displayed to the user

[0163] What happens:

[0164] The terminal analyzes the interpretation results received from the server and displays them on the user interface. The user can check the interpretation list and select an appropriate interpretation as needed.

[0165] As described above, the input and output for each processing step can be clearly indicated, and the system of the present invention can be implemented in a form including specific operations.

[0166] (Application example 1)

[0167] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0168] In traditional communication, ambiguous expressions often lead to misunderstandings, and accurate information transmission is particularly important in security services. However, current systems lack the ability to detect ambiguous expressions in text data and provide their interpretation. This can result in misunderstandings of important information and inappropriate decision-making.

[0169] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0170] In this invention, the server includes means for analyzing text data using the generated artificial intelligence model to detect ambiguous expressions, means for providing multiple interpretations of the detected ambiguous expressions, means for presenting the provided multiple interpretations to the user, and means for analyzing the content of a message in a security service to detect ambiguous expressions and generate a reinterpretation, and means for displaying the reinterpreted result on the user's device. This makes it possible to detect ambiguous expressions in text-based communication and present their interpretations, thereby preventing misunderstandings.

[0171] A "generated AI model" is a newly generated AI model based on an existing AI model with natural language processing capabilities, and is used to analyze and interpret ambiguous expressions.

[0172] "Text data" refers to character string information input by the user, and includes messages, sentences, and the like.

[0173] An "ambiguous expression" is an expression that can be interpreted differently depending on the context and intent, and refers to a word or phrase whose specific meaning is difficult to pin down.

[0174] "Security services" are services for maintaining the safety and confidentiality of information, particularly those for managing important communications within a company or in transactions.

[0175] A "message" is a sentence or text data that a user sends to another user, and may contain important information.

[0176] "Reinterpretation" refers to a new interpretation generated from a different angle for an input ambiguous expression, making it possible to understand it from multiple perspectives.

[0177] "User device" refers to a computer device such as a smartphone, PC, or tablet, which is the terminal through which the user receives the analysis results.

[0178] MODE FOR CARRYING OUT THE INVENTION

[0179] This invention relates to a system that uses a generated artificial intelligence model to analyze text data, detect ambiguous expressions, and present multiple interpretations to the user. This system is particularly useful in security services to prevent misunderstandings of message content. An embodiment of this invention will be described in detail below.

[0180] Hardware and software used

[0181] The following hardware and software are used to implement the present invention.

[0182] Hardware:

[0183] Server: Amazon Web Services (AWS(R)) EC2 instance

[0184] User devices: smartphone, PC, tablet

[0185] software:

[0186] Generative AI model: Hugging Face Transformer model "t5-small"

[0187] Server-side framework: Flask (based on Python®)

[0188] Front-end framework: React

[0189] Database: PostgreSQL

[0190] Program processing

[0191] Server Processing

[0192] The server runs on the Flask framework and receives text data sent by users. The received text data is analyzed using the pre-loaded Hugging Face Transformer model "t5-small." Specifically, it detects ambiguous expressions and generates multiple interpretations for those ambiguous expressions. The analysis results are sent to the user's device as multiple reinterpreted expressions. The database stores the analysis results and the user's usage history.

[0193] User Action

[0194] Users send the message they wish to have analyzed from their device to the server. They can then check the multiple interpretations provided as analysis results on their device. This allows users to prevent misunderstandings about ambiguous expressions.

[0195] Specific examples

[0196] For example, if a user sends a message from their smartphone to the server saying, "There was a lot of feedback in an important meeting, but I felt everyone should agree," the system can provide multiple interpretations, such as:

[0197] 1. "I received a lot of feedback in an important meeting, and I felt like everyone should agree with it."

[0198] 2. "I received a lot of feedback during an important meeting, and I felt like we all should agree."

[0199] Prompt Sentence Examples

[0200] If a user enters the following text and requests analysis:

[0201] text_input = "It was an important meeting and there was a lot of feedback, but everyone felt they should agree."

[0202] The server receives this input, analyzes it using a generative AI model, and generates multiple interpretations like the ones above. These results are displayed on the user's device and used as reference information to avoid misunderstandings.

[0203] This embodiment makes it possible to detect ambiguous expressions in text-based communication and present their interpretations, thereby preventing misunderstandings.

[0204] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0205] Step 1:

[0206] The user inputs the message they wish to analyze from their device (smartphone, PC, tablet, etc.). The input text data is prepared as a prompt sentence. For example, they input a message such as, "There was a lot of feedback in an important meeting, but I felt that everyone should agree."

[0207] Input: Text data entered by the user

[0208] Output: Text data as prompt sentence

[0209] Step 2:

[0210] The user's device sends the entered text data to a specified endpoint on the server, which runs on the Flask framework and receives it as an HTTP request.

[0211] Input: Text data as a prompt

[0212] Output: HTTP request sent to the server

[0213] Step 3:

[0214] The server retrieves the received text data and loads the generative AI model "t5-small." This is where the analysis of the text data begins. The server performs data analysis to detect ambiguous expressions within the text data.

[0215] Input: Text data sent to the server

[0216] Output: Analysis results including ambiguous expressions

[0217] Step 4:

[0218] Using the generative AI model "t5-small," the server generates multiple interpretations for the detected ambiguous expressions. The model takes context into account and performs data calculations to generate multiple different interpretations.

[0219] Input: Analysis results containing ambiguous expressions

[0220] Output: Multiple interpretation results

[0221] Step 5:

[0222] The server sends the generated interpretation results to the device, where they are formatted for display on the front end.

[0223] Input: Multiple interpretation results

[0224] Output: Data sent to the front end

[0225] Step 6:

[0226] The user's device displays the received analysis results in a React-based interface, allowing the user to check multiple interpretations and choose the appropriate one.

[0227] Input: Data sent to the front end

[0228] Output: Multiple interpretation results displayed to the user

[0229] Step 7:

[0230] Users can check the analysis results and use them as reference information to prevent misunderstandings. Users can then use this information to decide on the next action to take in order to make appropriate decisions.

[0231] Input: Multiple interpretation results displayed to the user

[0232] Output: User decision

[0233] Through the above processing steps, ambiguous expressions in text data are analyzed and multiple interpretations are presented, enabling users to communicate accurately.

[0234] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0235] The present invention relates to a system that analyzes text data using a generated artificial intelligence model, detects ambiguous expressions in the text data, and presents multiple interpretations to the user. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the provided interpretations are adapted to the user's emotional state.

[0236] Hereinafter, an embodiment of the present invention will be described.

[0237] First, the server loads the necessary generative AI model and emotion engine. The generative AI model is based on natural language processing and analyzes text data and detects ambiguous expressions. The emotion engine is used to extract user emotions from text data. This enables both text analysis and emotion recognition.

[0238] A user inputs text data from a terminal. For example, the user may input the text "I didn't agree with him at the meeting yesterday, but I agree with his opinion." This text may be interpreted differently depending on the context and emotion.

[0239] The server then receives the text data entered by the user. This text data is then subjected to analysis. The server first analyzes the text data to identify ambiguous expressions. A generative AI model is used in the analysis, taking into account contextual information and punctuation placement to detect expressions that may be misleading.

[0240] Next, the server uses an emotion engine to recognize the user's emotions. Specifically, it extracts the user's emotional state from the input text data, such as positive, negative, or neutral emotions.

[0241] The server generates multiple interpretations for the detected ambiguous expressions, customizing the interpretations based on the user's emotional state. For example, the server generates the following interpretations for the above text:

[0242] 1. "I didn't agree with the whole meeting yesterday, but I might agree with some of his points." (Users express positive sentiment)

[0243] 2. "I don't agree with his opinion in general, but I might agree with him in certain circumstances." (When the user is expressing neutral or negative sentiment)

[0244] The server then sends the generated interpretations to the user's device, which then displays them to the user. By referring to these interpretations, the user can understand the ambiguous expressions in the text data and take measures to avoid misunderstandings.

[0245] In this way, the present invention provides a system that detects ambiguous expressions in text-based communication and presents their interpretation according to the emotional state, thereby enabling clearer and more accurate communication. By using this system, users can communicate more accurately and effectively.

[0246] The processing flow will be explained below.

[0247] Step 1:

[0248] The server loads the generative AI model and emotion engine. Specifically, it prepares the Transformer model "t5-small" based on natural language processing and the emotion engine for recognizing user emotions.

[0249] Step 2:

[0250] A user inputs text data through a terminal. For example, the user inputs "I didn't agree with him at the meeting yesterday, but I agree with his opinion."

[0251] Step 3:

[0252] The server receives text data entered by the user, which is then analyzed.

[0253] Step 4:

[0254] The server uses the generated AI model to analyze the input text data. Specifically, it performs an analysis process to detect ambiguous expressions within the text data.

[0255] Step 5:

[0256] The server uses an emotion engine to recognize the user's emotional state and classify the user's emotions into positive, negative, neutral, etc. based on the input text data.

[0257] Step 6:

[0258] The server generates multiple interpretations for detected ambiguous expressions, taking into account the user's emotions. For example, it provides different interpretations for positive and negative emotions.

[0259] Step 7:

[0260] The server transmits the generated interpretations to the user's terminal.

[0261] Step 8:

[0262] The terminal displays the received interpretation to the user, who then checks the displayed interpretation and understands the ambiguous expressions in the text data.

[0263] Step 9:

[0264] The user can refer to the proposed interpretation and correct or add to the text data as necessary. This process allows the user to choose appropriate expressions to avoid misunderstandings.

[0265] Example 2

[0266] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0267] Conventional text analysis systems provide a uniform interpretation of ambiguous expressions, making it difficult to provide flexible interpretations that reflect the emotional state of each individual user. Furthermore, the lack of emotion recognition functionality meant that interpretations that took the user's emotional state into account were insufficient. As a result, users were more likely to misunderstand the true meaning of the text data, hindering smooth communication.

[0268] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0269] In this invention, the server includes means for analyzing text data using the generated artificial intelligence model and detecting ambiguous expressions in the text data, means for providing multiple interpretations for the detected ambiguous expressions, means for recognizing a user's emotion, means for customizing an interpretation based on the recognized emotional state, and means for presenting the provided multiple interpretations to the user. This makes it possible to provide multiple interpretations according to the user's emotional state and reduce misunderstandings of ambiguous expressions in the text data.

[0270] A "generated artificial intelligence model" is a model that has been pre-trained to analyze text data and detect ambiguous expressions, and that uses natural language processing techniques.

[0271] "Text data" refers to character information such as sentences, documents, and chat messages entered by the user.

[0272] An "ambiguous expression" is a word or phrase that can have different meanings within a text depending on the context.

[0273] "Multiple interpretations" refers to the multiple possible meanings or connotations of an ambiguous expression.

[0274] "Means for recognizing user emotions" refers to technology or algorithms for determining a user's emotional state from text data, such as technology for extracting positive, negative, or neutral emotional states.

[0275] "Means for customizing interpretation based on emotional state" refers to a technique or algorithm that adjusts the interpretation of an ambiguous expression based on the perceived emotion of the user.

[0276] The "means for presenting the provided interpretations to the user" refers to a technique or interface for displaying the generated interpretations on the user's terminal.

[0277] The present invention is a system that uses the generated AI model and emotion engine to analyze text data and recognize user emotions, detect ambiguous expressions in the text, and provide multiple interpretations for them. This system is specifically implemented by a series of processes with the server, terminal, and user as subjects.

[0278] First, the server loads the generative AI model (e.g., OpenAI® GPT-3®) and emotion engine (e.g., IBM Watson® API), along with any necessary software libraries and configuration files, so that the system is ready for text analysis and emotion recognition.

[0279] Next, the user inputs text data from the terminal. For example, the text could be something like, "I didn't agree with him at the meeting yesterday, but I agree with his opinion." The terminal sends the input data from the user to the server as an HTTP request.

[0280] The server receives the text data sent from the device and passes it to the analysis module, which uses a generative AI model to detect ambiguous expressions in the input text data. The model takes into account contextual information and punctuation placement to identify expressions that may be misleading.

[0281] Furthermore, the server uses an emotion engine to recognize the user's emotion from the text data. The emotion engine distinguishes between positive, negative, and neutral emotional states. For example, if a user sends the text "I didn't agree with him in the meeting yesterday, but I agree with his opinion," the emotion engine recognizes the emotional state as "neutral."

[0282] The server generates multiple interpretations for the detected ambiguous expressions, taking into account the user's emotional state. For example, if the user expresses a positive emotion, the expression can be interpreted as "I didn't agree with the entire meeting yesterday, but I might agree with some of his opinions." If the user expresses a negative or neutral emotion, the expression can be interpreted as "I don't agree with his opinions overall, but I might agree with his opinions under certain circumstances."

[0283] Finally, the server sends the generated interpretations to the user's device, which displays them to the user. By referring to these interpretations, the user can more accurately understand the true meaning of the text data.

[0284] For example, the prompt the user enters is:

[0285] Generate an interpretation for the text "I disagreed with him in the meeting yesterday, but I agree with his opinion."

[0286] This system allows users to adapt the results of analysis of text data containing ambiguous expressions to their emotional state, thereby reducing misunderstandings.

[0287] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0288] System program processing flow

[0289] Step 1: The server loads the generative AI model and emotion engine

[0290] The server first loads the generative AI model and emotion engine. Specifically, it loads the libraries and configuration files required to use the generative AI model (e.g., OpenAI GPT-3) and emotion engine (e.g., IBM Watson API).

[0291] Input: Server configuration file, API key, etc.

[0292] Data processing: Initializing the library and loading the model

[0293] Output: AI model and emotion engine ready state

[0294] Specific behavior:

[0295] The server reads the configuration file and loads the API key and model paths into memory.

[0296] Initialize the generative AI model and emotion engine.

[0297] Step 2: The user inputs text data from the terminal.

[0298] The user inputs text data to be analyzed into an input field on the terminal, and this data is sent to the server.

[0299] Input: Text data that the user types into the terminal.

[0300] Data processing: None

[0301] Output: HTTP request sent from the terminal to the server

[0302] Specific behavior:

[0303] A user enters text into an input field on a web form or application and presses the submit button.

[0304] The terminal generates an HTTP request and sends the input text data to the server.

[0305] Step 3: The server receives the input text data

[0306] The server receives the text data sent from the terminal and prepares to pass it on to the next analysis step.

[0307] Input: Text data sent via the HTTP request

[0308] Data processing: Parsing HTTP requests

[0309] Output: Text data passed to the analysis module

[0310] Specific behavior:

[0311] The server receives the HTTP request and extracts the text data from it.

[0312] The extracted text data is passed to the analysis module.

[0313] Step 4: The server analyzes the text data and identifies ambiguous expressions.

[0314] The server uses a generative AI model to analyze the received text data and identify ambiguous expressions.

[0315] Input: Text data passed to the analysis module

[0316] Data Processing: Text Analysis with Generative AI Models

[0317] Output: Identification of ambiguous expressions

[0318] Specific behavior:

[0319] The server sends the text data to the generative AI model and performs the analysis.

[0320] The model takes into account contextual information and punctuation placement and returns results that detect ambiguous expressions.

[0321] Step 5: The server recognizes the user's emotion using the emotion engine.

[0322] The server uses an emotion engine to extract the user's emotion from the text data.

[0323] Input: Text data and results of identifying ambiguous expressions

[0324] Data processing: Emotion analysis using an emotion engine

[0325] Output: User's emotional state (positive, negative, neutral, etc.)

[0326] Specific behavior:

[0327] The server passes the text data to the emotion engine and performs emotion analysis.

[0328] The engine analyzes the emotional state in the text and returns the results.

[0329] Step 6: The server generates multiple interpretations and customizes them based on emotional state.

[0330] The server generates multiple interpretations based on the generative AI model and emotion recognition results, and customizes them according to the user's emotional state.

[0331] Input: Identification results of ambiguous expressions, user's emotional state

[0332] Data processing: interpretation generation and emotional state-based customization

[0333] Output: Multiple customized interpretations

[0334] Specific behavior:

[0335] The server uses a generative AI model to generate an interpretation for an ambiguous expression.

[0336] Based on the acquired emotional data, each interpretation is adapted to the emotion.

[0337] Step 7: The server sends the generated interpretation to the terminal.

[0338] The server formats the generated interpretations as an HTTP response and sends it to the user's device.

[0339] Input: Customized multiple interpretations

[0340] Data processing: Generating HTTP responses

[0341] Output: The interpreted response sent to the device

[0342] Specific behavior:

[0343] The server formats the generated interpretation into an HTTP response.

[0344] Send the interpreted response to the terminal.

[0345] Step 8: The terminal displays the interpretation to the user

[0346] The terminal displays the received interpretation to the user, who then checks the interpretation and understands the true meaning of the text data.

[0347] Input: The interpreted response sent by the server

[0348] Data processing: Display of interpretations

[0349] Output: The interpretation displayed to the user

[0350] Specific behavior:

[0351] The device receives the response from the server and updates the GUI to display the interpretation.

[0352] Provide an interface that allows users to view the interpretation.

[0353] The above are the specific processing steps of this system.

[0354] (Application example 2)

[0355] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0356] In text-based communication, ambiguous expressions in the text data entered by users can lead to misunderstandings and unpleasant interactions. In addition, the quality of communication can be reduced due to a lack of appropriate interpretation according to the user's emotional state.

[0357] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0358] In this invention, the server includes means for analyzing text data using the generated artificial intelligence model and detecting ambiguous expressions in the text data, means for providing multiple interpretations for the detected ambiguous expressions, means for presenting the provided multiple interpretations to the user, means for analyzing the user's emotions, and means for generating interpretations based on the user's emotional state. This makes it possible to accurately detect ambiguous expressions in the text data and provide interpretations that correspond to the user's emotional state, thereby reducing misunderstandings and unpleasant interactions and improving the quality of communication.

[0359] "Generated artificial intelligence model" refers to the artificial intelligence technology used to analyze text data, detect ambiguous expressions, and generate appropriate interpretations.

[0360] "Text data" refers to data expressed as character information, such as sentences or comments entered by a user.

[0361] An "ambiguous expression" refers to an expression that has different meanings depending on the context and interpretation, and may lead to misunderstanding.

[0362] "Interpretation" refers to specific explanations or proposed explanations to understand the meaning and intent of text data.

[0363] "User" refers to the entity that uses the system to input text data and have it interpreted.

[0364] "Emotion" refers to the emotional state, such as positive, negative, or neutral, contained in the text data.

[0365] "Analysis" refers to the process of analyzing text data and extracting information such as meaning and sentiment.

[0366] A "server" is the central computer of the system, and refers to the device that analyzes text data and generates interpretations.

[0367] The present invention relates to a system that uses a generated artificial intelligence model to analyze text data, detect ambiguous expressions, and present multiple interpretations to the user. Furthermore, it combines an emotion engine that recognizes the user's emotions so that the interpretations provided adapt to the user's emotional state.

[0368] First, the server loads the generated artificial intelligence model and emotion engine. The generative AI model is based on natural language processing and is used to analyze text data and detect ambiguous expressions. The emotion engine has the function of extracting user emotions from text data. This allows both text analysis and emotion recognition to be performed simultaneously.

[0369] A user inputs text data from a terminal. For example, the user may input the text, "This movie exceeded my expectations, but the ending was disappointing." This text may be interpreted differently depending on the context and emotions.

[0370] The server then receives the text data entered by the user. This text data is then analyzed. First, a generative AI model is used to analyze the text data and identify ambiguous expressions. This analysis takes into account contextual information and punctuation placement to detect expressions that may be misleading.

[0371] Next, the server uses an emotion engine to recognize the user's emotion. Specifically, it determines the user's emotional state (positive, negative, neutral, etc.) from the input text data. This emotion recognition is performed using a natural language processing library such as TextBlob.

[0372] The server then generates multiple interpretations for the detected ambiguous expressions, customizing the interpretations based on the user's emotional state. For example, the server generates the following interpretations for the above text:

[0373] 1. "The movie was good overall, but the ending was disappointing." (Users express positive emotions)

[0374] 2. "The ending of the movie was disappointing, so I didn't enjoy it overall." (Users express negative emotions)

[0375] The server sends the generated interpretations to the user's device, which then displays them to the user, allowing the user to refer to these interpretations, understand the ambiguous expressions in the text data, and take measures to avoid misunderstandings.

[0376] Example prompt sentence:

[0377] "Provide a positive interpretation of the following ambiguous text: This movie exceeded my expectations, but the ending disappointed me."

[0378] "Provide a neutral / negative interpretation of the following ambiguous text: This movie exceeded my expectations, but the ending disappointed me."

[0379] In this way, the present invention provides a system that detects ambiguous expressions in text-based communication and presents their interpretation according to the emotional state, thereby enabling clearer and more accurate communication. By using this system, users can communicate more accurately and effectively.

[0380] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0381] Step 1:

[0382] The user inputs text data using the terminal. At this time, the text data input by the user is sent to the system. The input data is in text format, and for example, a sentence such as "This movie exceeded my expectations, but the ending was disappointing" is input.

[0383] Step 2:

[0384] The server uses the generated artificial intelligence model to analyze the text data received from the user. First, it analyzes the text data based on natural language processing techniques to detect ambiguous expressions. It receives the text data as input and generates a list of ambiguous expressions as output.

[0385] Step 3:

[0386] The server uses an emotion engine to determine the user's emotional state based on the detected ambiguous expressions. Specifically, the emotion engine analyzes the text data and detects emotional states such as positive, negative, and neutral. It receives the initial text data as input and generates an emotional state as output.

[0387] Step 4:

[0388] The server generates multiple interpretations based on the detected ambiguous expressions and emotional states. In this step, a generative AI model is used, and a prompt sentence such as "Provide a positive interpretation of the following ambiguous text: This movie exceeded my expectations, but the ending was disappointing" is sent to the model. The server receives the ambiguous expressions and emotional states as input and generates a list of customized interpretations as output.

[0389] Step 5:

[0390] The server sends the generated interpretations to the user's device, which then displays them to the user. The interpretations are presented in a format that is easy for the user to understand, and the user can use these interpretations to easily understand ambiguous expressions.

[0391] Step 6:

[0392] By referring to the presented interpretation, users can take measures to avoid misunderstandings. This reduces misunderstandings of ambiguous expressions in the text and improves the quality of communication. The user's actions and choices are the final output.

[0393] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0394] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0395] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0396] [Second embodiment]

[0397] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0398] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0399] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0400] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0401] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0402] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0403] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0404] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0405] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0406] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0407] In the smart glasses 214, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0408] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0409] The present invention relates to a system that analyzes text data using a generated artificial intelligence model, detects ambiguous expressions in the text data, and presents multiple interpretations of the expressions to a user.

[0410] The server first loads the necessary generative AI model, specifically, a Transformer model, prepared from an existing natural language processing model such as "t5-small," which has the ability to detect ambiguous expressions and generate interpretations.

[0411] A user inputs text data from a terminal. For example, the user might input the text "I didn't agree with him at the meeting yesterday, but I agree with his opinion." This text may be interpreted differently depending on the user's intention and context.

[0412] The server then receives the text data entered by the user and analyzes it using a generative AI model. This analysis identifies ambiguous expressions in the text data. The model takes into account contextual information and punctuation placement to detect expressions that may be misleading.

[0413] The server then generates different interpretations for the detected ambiguous expressions. For example, it can provide the following interpretations for the above text:

[0414] 1. "I didn't agree with the whole meeting yesterday, but I might agree with some of his points."

[0415] 2. "I don't agree with his opinion in its entirety, but I might agree with him under certain circumstances."

[0416] The server then displays the generated interpretations on the terminal, allowing the user to refer to these interpretations and take measures to avoid misunderstandings.

[0417] As described above, the present invention provides a system that detects ambiguous expressions in text-based communication and presents their interpretations to prevent misunderstandings. By using this system, users can achieve more accurate and clearer communication.

[0418] The processing flow will be explained below.

[0419] Step 1:

[0420] The server loads the generative AI model, specifically the Transformer model "t5-small" based on natural language processing, and prepares it for analysis.

[0421] Step 2:

[0422] A user inputs text data through a terminal. For example, the user inputs "I didn't agree with him at the meeting yesterday, but I agree with his opinion."

[0423] Step 3:

[0424] The server receives text data entered by the user, which is then analyzed.

[0425] Step 4:

[0426] The server uses the generative AI model to analyze the input text data. Specifically, it performs an analysis process to detect ambiguous expressions within the text data.

[0427] Step 5:

[0428] As a result of the analysis, the server generates multiple interpretations for the detected ambiguous expressions. For example, it generates the following interpretations for the above text: "1. I did not agree with the entire meeting yesterday, but I may agree with some of his opinions." "2. I do not agree with his entire opinion, but I may agree with his opinion under certain circumstances."

[0429] Step 6:

[0430] The server transmits the generated interpretations to the user's terminal.

[0431] Step 7:

[0432] The terminal displays the received interpretation to the user, who then checks the displayed interpretation and understands the ambiguous expressions in the text data.

[0433] Step 8:

[0434] The user can refer to the proposed interpretation and correct or add to the text data as necessary. This process allows the user to choose appropriate expressions to avoid misunderstandings.

[0435] Example 1

[0436] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0437] In natural language text data exchange, ambiguous expressions can lead to misunderstandings. This misunderstanding can cause serious problems, especially in business communications and important message exchanges. Conventional systems have had difficulty automatically detecting such ambiguous expressions and providing clear interpretations to users.

[0438] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0439] In this invention, the server includes means for loading the generated AI model as a Transformer model, means for receiving user input and sending it to the server, and means for receiving analysis results from the server and displaying them to the user, thereby enabling highly accurate detection of ambiguous expressions in text data and providing the user with their interpretations.

[0440] A "generated artificial intelligence model" refers to a machine learning algorithm that has been trained on a set of data and used to perform a specific task.

[0441] "Text data" refers to a collection of information expressed as character strings, including sentences and messages written in natural language.

[0442] "Polyambiguous expressions" refer to words or phrases that can be interpreted in multiple ways depending on the context or situation.

[0443] "Interpretation" refers to supplementary explanations and possibilities that clarify the meaning contained in the text data and are provided to help users deepen their understanding.

[0444] A "Transformer model" is a type of machine learning model used in natural language processing, and refers to a model that has the ability to analyze text data taking context into account.

[0445] "Loading" refers to the operation of reading a program or data into memory and making it available for use.

[0446] "User" refers to the person who uses the system and is the entity that inputs the text data to be analyzed.

[0447] A "server" refers to a computer that provides services over a network and plays a central role in processing and managing data.

[0448] A "terminal" refers to a computer or device that is directly used by a user, and is a device that communicates with a server to send and receive data.

[0449] "Analysis results" refers to the information and data obtained after the generated artificial intelligence model analyzes the text data.

[0450] The present invention relates to a system that analyzes text data using a generated artificial intelligence model, detects ambiguous expressions in the text data, and presents multiple interpretations of the expressions to a user. An embodiment of the present invention will be described in detail below.

[0451] Hardware and Software Configuration

[0452] The server is a high-performance computer with hardware acceleration, such as a GPU. The server performs the following functions:

[0453] 1. The generated AI model is loaded as a Transformer model. At this time, a natural language processing model such as "t5-small" is downloaded from the internet and loaded into memory.

[0454] 2. Receive text data sent by the user and analyze it using the generative AI model.

[0455] 3. Detect ambiguous expressions and generate multiple interpretations for them.

[0456] 4. The generated interpretation results are sent to the user's terminal.

[0457] A terminal is a computer, mobile device, or other device that is directly operated by a user and performs the following actions:

[0458] 1. Provide an interface for users to enter text data, implemented as a web or desktop application.

[0459] 2. Send the text data entered by the user to the server.

[0460] 3. Display the analysis results received from the server to the user.

[0461] A user is someone who uses this system and performs the following operations:

[0462] 1. Enter the text data you want to analyze through the device interface. For example, enter the text "I didn't agree with him in the meeting yesterday, but I agree with his opinion."

[0463] 2. The input data is sent to the server and waits for parsing.

[0464] 3. The interpretation results sent from the server are displayed on the terminal.

[0465] Specific examples

[0466] Consider the case where the user enters the following text:

[0467] For example: "I disagreed with him in the meeting yesterday, but I agree with his opinion."

[0468] This text data is sent to the server, which analyzes it using the "t5-small" model, which generates the following interpretation:

[0469] 1. "I didn't agree with the whole meeting yesterday, but I might agree with some of his points."

[0470] 2. "I don't agree with his opinion in its entirety, but I might agree with him under certain circumstances."

[0471] These interpretations are displayed on the user's device, allowing the user to confirm the interpretation that best suits their intentions.

[0472] In this way, this system allows users to easily interpret ambiguous expressions and helps prevent misunderstandings.

[0473] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0474] Step 1:

[0475] The server loads the generative AI model.

[0476] Specific behavior:

[0477] The server downloads a Transformer model, such as "t5-small," from an internet resource and loads it into memory, using the necessary libraries, such as PyTorch and the Transformers library.

[0478] Input and Output:

[0479] Input: Model resource URL on the Internet

[0480] Output: Generative AI model loaded in memory

[0481] What happens:

[0482] The server retrieves the model file from the specified URL and loads it into memory, where it is ready to be used for text analysis.

[0483] Step 2:

[0484] The user inputs text data from the terminal.

[0485] Specific behavior:

[0486] The user inputs the text data to be analyzed using a device interface (web application or desktop application).

[0487] Input and Output:

[0488] Input: User text data (e.g., "I disagreed with him in the meeting yesterday, but I agree with his opinion.")

[0489] Output: Input data in the terminal

[0490] What happens:

[0491] The user enters the text they want to analyze in the text field and clicks the submit button. This data is stored internally and is ready for the next step.

[0492] Step 3:

[0493] The terminal transmits the input text data to the server.

[0494] Specific behavior:

[0495] The device sends the text data entered by the user to the server as an HTTP request. The data is encoded in JSON format and sent to the API endpoint using the POST method.

[0496] Input and Output:

[0497] Input: Text data in the terminal

[0498] Output: Text data sent to the server

[0499] What happens:

[0500] The terminal encodes the user's input data in JSON format and issues a POST request to the specified server endpoint.

[0501] Step 4:

[0502] The server uses a generative AI model to analyze the received text data.

[0503] Specific behavior:

[0504] The server inputs the received text data into the t5-small model and begins analysis. The model takes into account contextual information and punctuation placement to identify ambiguous expressions within the text.

[0505] Input and Output:

[0506] Input: Text data sent to the server

[0507] Output: Analysis results with ambiguous expressions identified

[0508] What happens:

[0509] The server runs the text data through a generative AI model to detect ambiguous expressions, which are then stored in memory as analysis results.

[0510] Step 5:

[0511] The server generates multiple interpretations for the detected ambiguous expression.

[0512] Specific behavior:

[0513] The server generates multiple interpretations based on the analysis results obtained from the generative AI model. For example, for expressions that can be interpreted differently, it generates sentences that explain each of their meanings.

[0514] Input and Output:

[0515] Input: Analysis results (identification of ambiguous expressions)

[0516] Output: Multiple interpretations generated

[0517] What happens:

[0518] The server uses a generative AI model to analyze the context of ambiguous expressions and generate multiple possible interpretations, which are then stored in JSON format.

[0519] Step 6:

[0520] The server transmits the generated interpretation results to the user's terminal.

[0521] Specific behavior:

[0522] The server returns the generated interpretation results to the user's device as an HTTP response. The data is again encoded in JSON format and sent.

[0523] Input and Output:

[0524] Input: Generated interpretation

[0525] Output: Interpretation results sent to the user's terminal

[0526] What happens:

[0527] The server encodes the generated interpretation results in JSON format and returns an HTTP response to the user's terminal.

[0528] Step 7:

[0529] The terminal displays the multiple interpretation results received from the server to the user.

[0530] Specific behavior:

[0531] The application or web interface used by the user displays the interpretation results received from the server in an easy-to-understand format, for example as a list of interpretations displayed on the interface for the user to review.

[0532] Input and Output:

[0533] Input: Interpretation result sent from the server

[0534] Output: The interpretation results displayed to the user

[0535] What happens:

[0536] The terminal analyzes the interpretation results received from the server and displays them on the user interface. The user can check the interpretation list and select an appropriate interpretation as needed.

[0537] As described above, the input and output for each processing step can be clearly indicated, and the system of the present invention can be implemented in a form including specific operations.

[0538] (Application example 1)

[0539] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0540] In traditional communication, ambiguous expressions often lead to misunderstandings, and accurate information transmission is particularly important in security services. However, current systems lack the ability to detect ambiguous expressions in text data and provide their interpretation. This can result in misunderstandings of important information and inappropriate decision-making.

[0541] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0542] In this invention, the server includes means for analyzing text data using the generated artificial intelligence model to detect ambiguous expressions, means for providing multiple interpretations of the detected ambiguous expressions, means for presenting the provided multiple interpretations to the user, and means for analyzing the content of a message in a security service to detect ambiguous expressions and generate a reinterpretation, and means for displaying the reinterpreted result on the user's device. This makes it possible to detect ambiguous expressions in text-based communication and present their interpretations, thereby preventing misunderstandings.

[0543] A "generated AI model" is a newly generated AI model based on an existing AI model with natural language processing capabilities, and is used to analyze and interpret ambiguous expressions.

[0544] "Text data" refers to character string information input by the user, and includes messages, sentences, and the like.

[0545] An "ambiguous expression" is an expression that can be interpreted differently depending on the context and intent, and refers to a word or phrase whose specific meaning is difficult to pin down.

[0546] "Security services" are services for maintaining the safety and confidentiality of information, particularly those for managing important communications within a company or in transactions.

[0547] A "message" is a sentence or text data that a user sends to another user, and may contain important information.

[0548] "Reinterpretation" refers to a new interpretation generated from a different angle for an input ambiguous expression, making it possible to understand it from multiple perspectives.

[0549] "User device" refers to a computer device such as a smartphone, PC, or tablet, which is the terminal through which the user receives the analysis results.

[0550] MODE FOR CARRYING OUT THE INVENTION

[0551] This invention relates to a system that uses a generated artificial intelligence model to analyze text data, detect ambiguous expressions, and present multiple interpretations to the user. This system is particularly useful in security services to prevent misunderstandings of message content. An embodiment of this invention will be described in detail below.

[0552] Hardware and software used

[0553] The following hardware and software are used to implement the present invention.

[0554] Hardware:

[0555] Server: Amazon Web Services (AWS) EC2 instance

[0556] User devices: smartphone, PC, tablet

[0557] software:

[0558] Generative AI model: Hugging Face Transformer model "t5-small"

[0559] Server-side framework: Flask (Python-based)

[0560] Front-end framework: React

[0561] Database: PostgreSQL

[0562] Program processing

[0563] Server Processing

[0564] The server runs on the Flask framework and receives text data sent by users. The received text data is analyzed using the pre-loaded Hugging Face Transformer model "t5-small." Specifically, it detects ambiguous expressions and generates multiple interpretations for those ambiguous expressions. The analysis results are sent to the user's device as multiple reinterpreted expressions. The database stores the analysis results and the user's usage history.

[0565] User Action

[0566] Users send the message they wish to have analyzed from their device to the server. They can then check the multiple interpretations provided as analysis results on their device. This allows users to prevent misunderstandings about ambiguous expressions.

[0567] Specific examples

[0568] For example, if a user sends a message from their smartphone to the server saying, "There was a lot of feedback in an important meeting, but I felt everyone should agree," the system can provide multiple interpretations, such as:

[0569] 1. "I received a lot of feedback in an important meeting, and I felt like everyone should agree with it."

[0570] 2. "I received a lot of feedback during an important meeting, and I felt like we all should agree."

[0571] Prompt Sentence Examples

[0572] If a user enters the following text and requests analysis:

[0573] text_input = "It was an important meeting and there was a lot of feedback, but everyone felt they should agree."

[0574] The server receives this input, analyzes it using a generative AI model, and generates multiple interpretations like the ones above. These results are displayed on the user's device and used as reference information to avoid misunderstandings.

[0575] This embodiment makes it possible to detect ambiguous expressions in text-based communication and present their interpretations, thereby preventing misunderstandings.

[0576] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0577] Step 1:

[0578] The user inputs the message they wish to analyze from their device (smartphone, PC, tablet, etc.). The input text data is prepared as a prompt sentence. For example, they input a message such as, "There was a lot of feedback in an important meeting, but I felt that everyone should agree."

[0579] Input: Text data entered by the user

[0580] Output: Text data as prompt sentence

[0581] Step 2:

[0582] The user's device sends the entered text data to a specified endpoint on the server, which runs on the Flask framework and receives it as an HTTP request.

[0583] Input: Text data as a prompt

[0584] Output: HTTP request sent to the server

[0585] Step 3:

[0586] The server retrieves the received text data and loads the generative AI model "t5-small." This is where the analysis of the text data begins. The server performs data analysis to detect ambiguous expressions within the text data.

[0587] Input: Text data sent to the server

[0588] Output: Analysis results including ambiguous expressions

[0589] Step 4:

[0590] Using the generative AI model "t5-small," the server generates multiple interpretations for the detected ambiguous expressions. The model takes context into account and performs data calculations to generate multiple different interpretations.

[0591] Input: Analysis results containing ambiguous expressions

[0592] Output: Multiple interpretation results

[0593] Step 5:

[0594] The server sends the generated interpretation results to the device, where they are formatted for display on the front end.

[0595] Input: Multiple interpretation results

[0596] Output: Data sent to the front end

[0597] Step 6:

[0598] The user's device displays the received analysis results in a React-based interface, allowing the user to check multiple interpretations and choose the appropriate one.

[0599] Input: Data sent to the front end

[0600] Output: Multiple interpretation results displayed to the user

[0601] Step 7:

[0602] Users can check the analysis results and use them as reference information to prevent misunderstandings. Users can then use this information to decide on the next action to take in order to make appropriate decisions.

[0603] Input: Multiple interpretation results displayed to the user

[0604] Output: User decision

[0605] Through the above processing steps, ambiguous expressions in text data are analyzed and multiple interpretations are presented, enabling users to communicate accurately.

[0606] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0607] The present invention relates to a system that analyzes text data using a generated artificial intelligence model, detects ambiguous expressions in the text data, and presents multiple interpretations to the user. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the provided interpretations are adapted to the user's emotional state.

[0608] Hereinafter, an embodiment of the present invention will be described.

[0609] First, the server loads the necessary generative AI model and emotion engine. The generative AI model is based on natural language processing and analyzes text data and detects ambiguous expressions. The emotion engine is used to extract user emotions from text data. This enables both text analysis and emotion recognition.

[0610] A user inputs text data from a terminal. For example, the user may input the text "I didn't agree with him at the meeting yesterday, but I agree with his opinion." This text may be interpreted differently depending on the context and emotion.

[0611] The server then receives the text data entered by the user. This text data is then subjected to analysis. The server first analyzes the text data to identify ambiguous expressions. A generative AI model is used in the analysis, taking into account contextual information and punctuation placement to detect expressions that may be misleading.

[0612] Next, the server uses an emotion engine to recognize the user's emotions. Specifically, it extracts the user's emotional state from the input text data, such as positive, negative, or neutral emotions.

[0613] The server generates multiple interpretations for the detected ambiguous expressions, customizing the interpretations based on the user's emotional state. For example, the server generates the following interpretations for the above text:

[0614] 1. "I didn't agree with the whole meeting yesterday, but I might agree with some of his points." (Users express positive sentiment)

[0615] 2. "I don't agree with his opinion in general, but I might agree with him in certain circumstances." (When the user is expressing neutral or negative sentiment)

[0616] The server then sends the generated interpretations to the user's device, which then displays them to the user. By referring to these interpretations, the user can understand the ambiguous expressions in the text data and take measures to avoid misunderstandings.

[0617] In this way, the present invention provides a system that detects ambiguous expressions in text-based communication and presents their interpretation according to the emotional state, thereby enabling clearer and more accurate communication. By using this system, users can communicate more accurately and effectively.

[0618] The processing flow will be explained below.

[0619] Step 1:

[0620] The server loads the generative AI model and emotion engine. Specifically, it prepares the Transformer model "t5-small" based on natural language processing and the emotion engine for recognizing user emotions.

[0621] Step 2:

[0622] A user inputs text data through a terminal. For example, the user inputs "I didn't agree with him at the meeting yesterday, but I agree with his opinion."

[0623] Step 3:

[0624] The server receives text data entered by the user, which is then analyzed.

[0625] Step 4:

[0626] The server uses the generated AI model to analyze the input text data. Specifically, it performs an analysis process to detect ambiguous expressions within the text data.

[0627] Step 5:

[0628] The server uses an emotion engine to recognize the user's emotional state and classify the user's emotions into positive, negative, neutral, etc. based on the input text data.

[0629] Step 6:

[0630] The server generates multiple interpretations for detected ambiguous expressions, taking into account the user's emotions. For example, it provides different interpretations for positive and negative emotions.

[0631] Step 7:

[0632] The server transmits the generated interpretations to the user's terminal.

[0633] Step 8:

[0634] The terminal displays the received interpretation to the user, who then checks the displayed interpretation and understands the ambiguous expressions in the text data.

[0635] Step 9:

[0636] The user can refer to the proposed interpretation and correct or add to the text data as necessary. This process allows the user to choose appropriate expressions to avoid misunderstandings.

[0637] Example 2

[0638] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0639] Conventional text analysis systems provide a uniform interpretation of ambiguous expressions, making it difficult to provide flexible interpretations that reflect the emotional state of each individual user. Furthermore, the lack of emotion recognition functionality meant that interpretations that took the user's emotional state into account were insufficient. As a result, users were more likely to misunderstand the true meaning of the text data, hindering smooth communication.

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

[0641] In this invention, the server includes means for analyzing text data using the generated artificial intelligence model and detecting ambiguous expressions in the text data, means for providing multiple interpretations for the detected ambiguous expressions, means for recognizing a user's emotion, means for customizing an interpretation based on the recognized emotional state, and means for presenting the provided multiple interpretations to the user. This makes it possible to provide multiple interpretations according to the user's emotional state and reduce misunderstandings of ambiguous expressions in the text data.

[0642] A "generated artificial intelligence model" is a model that has been pre-trained to analyze text data and detect ambiguous expressions, and that uses natural language processing techniques.

[0643] "Text data" refers to character information such as sentences, documents, and chat messages entered by the user.

[0644] An "ambiguous expression" is a word or phrase that can have different meanings within a text depending on the context.

[0645] "Multiple interpretations" refers to the multiple possible meanings or connotations of an ambiguous expression.

[0646] "Means for recognizing user emotions" refers to technology or algorithms for determining a user's emotional state from text data, such as technology for extracting positive, negative, or neutral emotional states.

[0647] "Means for customizing interpretation based on emotional state" refers to a technique or algorithm that adjusts the interpretation of an ambiguous expression based on the perceived emotion of the user.

[0648] The "means for presenting the provided interpretations to the user" refers to a technique or interface for displaying the generated interpretations on the user's terminal.

[0649] The present invention is a system that uses the generated AI model and emotion engine to analyze text data and recognize user emotions, detect ambiguous expressions in the text, and provide multiple interpretations for them. This system is specifically implemented by a series of processes with the server, terminal, and user as subjects.

[0650] First, the server loads the generative AI model (e.g., OpenAI GPT-3) and emotion engine (e.g., IBM Watson API), along with any necessary software libraries and configuration files, so that the system is ready for text analysis and emotion recognition.

[0651] Next, the user inputs text data from the terminal. For example, the text could be something like, "I didn't agree with him at the meeting yesterday, but I agree with his opinion." The terminal sends the input data from the user to the server as an HTTP request.

[0652] The server receives the text data sent from the device and passes it to the analysis module, which uses a generative AI model to detect ambiguous expressions in the input text data. The model takes into account contextual information and punctuation placement to identify expressions that may be misleading.

[0653] Furthermore, the server uses an emotion engine to recognize the user's emotion from the text data. The emotion engine distinguishes between positive, negative, and neutral emotional states. For example, if a user sends the text "I didn't agree with him in the meeting yesterday, but I agree with his opinion," the emotion engine recognizes the emotional state as "neutral."

[0654] The server generates multiple interpretations for the detected ambiguous expressions, taking into account the user's emotional state. For example, if the user expresses a positive emotion, the expression can be interpreted as "I didn't agree with the entire meeting yesterday, but I might agree with some of his opinions." If the user expresses a negative or neutral emotion, the expression can be interpreted as "I don't agree with his opinions overall, but I might agree with his opinions under certain circumstances."

[0655] Finally, the server sends the generated interpretations to the user's device, which displays them to the user. By referring to these interpretations, the user can more accurately understand the true meaning of the text data.

[0656] For example, the prompt the user enters is:

[0657] Generate an interpretation for the text "I disagreed with him in the meeting yesterday, but I agree with his opinion."

[0658] This system allows users to adapt the results of analysis of text data containing ambiguous expressions to their emotional state, thereby reducing misunderstandings.

[0659] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0660] System program processing flow

[0661] Step 1: The server loads the generative AI model and emotion engine

[0662] The server first loads the generative AI model and emotion engine. Specifically, it loads the libraries and configuration files required to use the generative AI model (e.g., OpenAI GPT-3) and emotion engine (e.g., IBM Watson API).

[0663] Input: Server configuration file, API key, etc.

[0664] Data processing: Initializing the library and loading the model

[0665] Output: AI model and emotion engine ready state

[0666] Specific behavior:

[0667] The server reads the configuration file and loads the API key and model paths into memory.

[0668] Initialize the generative AI model and emotion engine.

[0669] Step 2: The user inputs text data from the terminal.

[0670] The user inputs text data to be analyzed into an input field on the terminal, and this data is sent to the server.

[0671] Input: Text data that the user types into the terminal.

[0672] Data processing: None

[0673] Output: HTTP request sent from the terminal to the server

[0674] Specific behavior:

[0675] A user enters text into an input field on a web form or application and presses the submit button.

[0676] The terminal generates an HTTP request and sends the input text data to the server.

[0677] Step 3: The server receives the input text data

[0678] The server receives the text data sent from the terminal and prepares to pass it on to the next analysis step.

[0679] Input: Text data sent via the HTTP request

[0680] Data processing: Parsing HTTP requests

[0681] Output: Text data passed to the analysis module

[0682] Specific behavior:

[0683] The server receives the HTTP request and extracts the text data from it.

[0684] The extracted text data is passed to the analysis module.

[0685] Step 4: The server analyzes the text data and identifies ambiguous expressions.

[0686] The server uses a generative AI model to analyze the received text data and identify ambiguous expressions.

[0687] Input: Text data passed to the analysis module

[0688] Data Processing: Text Analysis with Generative AI Models

[0689] Output: Identification of ambiguous expressions

[0690] Specific behavior:

[0691] The server sends the text data to the generative AI model and performs the analysis.

[0692] The model takes into account contextual information and punctuation placement and returns results that detect ambiguous expressions.

[0693] Step 5: The server recognizes the user's emotion using the emotion engine.

[0694] The server uses an emotion engine to extract the user's emotion from the text data.

[0695] Input: Text data and results of identifying ambiguous expressions

[0696] Data processing: Emotion analysis using an emotion engine

[0697] Output: User's emotional state (positive, negative, neutral, etc.)

[0698] Specific behavior:

[0699] The server passes the text data to the emotion engine and performs emotion analysis.

[0700] The engine analyzes the emotional state in the text and returns the results.

[0701] Step 6: The server generates multiple interpretations and customizes them based on emotional state.

[0702] The server generates multiple interpretations based on the generative AI model and emotion recognition results, and customizes them according to the user's emotional state.

[0703] Input: Identification results of ambiguous expressions, user's emotional state

[0704] Data processing: interpretation generation and emotional state-based customization

[0705] Output: Multiple customized interpretations

[0706] Specific behavior:

[0707] The server uses a generative AI model to generate an interpretation for an ambiguous expression.

[0708] Based on the acquired emotional data, each interpretation is adapted to the emotion.

[0709] Step 7: The server sends the generated interpretation to the terminal.

[0710] The server formats the generated interpretations as an HTTP response and sends it to the user's device.

[0711] Input: Customized multiple interpretations

[0712] Data processing: Generating HTTP responses

[0713] Output: The interpreted response sent to the device

[0714] Specific behavior:

[0715] The server formats the generated interpretation into an HTTP response.

[0716] Send the interpreted response to the terminal.

[0717] Step 8: The terminal displays the interpretation to the user

[0718] The terminal displays the received interpretation to the user, who then checks the interpretation and understands the true meaning of the text data.

[0719] Input: The interpreted response sent by the server

[0720] Data processing: Display of interpretations

[0721] Output: The interpretation displayed to the user

[0722] Specific behavior:

[0723] The device receives the response from the server and updates the GUI to display the interpretation.

[0724] Provide an interface that allows users to view the interpretation.

[0725] The above are the specific processing steps of this system.

[0726] (Application example 2)

[0727] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0728] In text-based communication, ambiguous expressions in the text data entered by users can lead to misunderstandings and unpleasant interactions. In addition, the quality of communication can be reduced due to a lack of appropriate interpretation according to the user's emotional state.

[0729] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0730] In this invention, the server includes means for analyzing text data using the generated artificial intelligence model and detecting ambiguous expressions in the text data, means for providing multiple interpretations for the detected ambiguous expressions, means for presenting the provided multiple interpretations to the user, means for analyzing the user's emotions, and means for generating interpretations based on the user's emotional state. This makes it possible to accurately detect ambiguous expressions in the text data and provide interpretations that correspond to the user's emotional state, thereby reducing misunderstandings and unpleasant interactions and improving the quality of communication.

[0731] "Generated artificial intelligence model" refers to the artificial intelligence technology used to analyze text data, detect ambiguous expressions, and generate appropriate interpretations.

[0732] "Text data" refers to data expressed as character information, such as sentences or comments entered by a user.

[0733] An "ambiguous expression" refers to an expression that has different meanings depending on the context and interpretation, and may lead to misunderstanding.

[0734] "Interpretation" refers to specific explanations or proposed explanations to understand the meaning and intent of text data.

[0735] "User" refers to the entity that uses the system to input text data and have it interpreted.

[0736] "Emotion" refers to the emotional state, such as positive, negative, or neutral, contained in the text data.

[0737] "Analysis" refers to the process of analyzing text data and extracting information such as meaning and sentiment.

[0738] A "server" is the central computer of the system, and refers to the device that analyzes text data and generates interpretations.

[0739] The present invention relates to a system that uses a generated artificial intelligence model to analyze text data, detect ambiguous expressions, and present multiple interpretations to the user. Furthermore, it combines an emotion engine that recognizes the user's emotions so that the interpretations provided adapt to the user's emotional state.

[0740] First, the server loads the generated artificial intelligence model and emotion engine. The generative AI model is based on natural language processing and is used to analyze text data and detect ambiguous expressions. The emotion engine has the function of extracting user emotions from text data. This allows both text analysis and emotion recognition to be performed simultaneously.

[0741] A user inputs text data from a terminal. For example, the user may input the text, "This movie exceeded my expectations, but the ending was disappointing." This text may be interpreted differently depending on the context and emotions.

[0742] The server then receives the text data entered by the user. This text data is then analyzed. First, a generative AI model is used to analyze the text data and identify ambiguous expressions. This analysis takes into account contextual information and punctuation placement to detect expressions that may be misleading.

[0743] Next, the server uses an emotion engine to recognize the user's emotion. Specifically, it determines the user's emotional state (positive, negative, neutral, etc.) from the input text data. This emotion recognition is performed using a natural language processing library such as TextBlob.

[0744] The server then generates multiple interpretations for the detected ambiguous expressions, customizing the interpretations based on the user's emotional state. For example, the server generates the following interpretations for the above text:

[0745] 1. "The movie was good overall, but the ending was disappointing." (Users express positive emotions)

[0746] 2. "The ending of the movie was disappointing, so I didn't enjoy it overall." (Users express negative emotions)

[0747] The server sends the generated interpretations to the user's device, which then displays them to the user, allowing the user to refer to these interpretations, understand the ambiguous expressions in the text data, and take measures to avoid misunderstandings.

[0748] Example prompt sentence:

[0749] "Provide a positive interpretation of the following ambiguous text: This movie exceeded my expectations, but the ending disappointed me."

[0750] "Provide a neutral / negative interpretation of the following ambiguous text: This movie exceeded my expectations, but the ending disappointed me."

[0751] In this way, the present invention provides a system that detects ambiguous expressions in text-based communication and presents their interpretation according to the emotional state, thereby enabling clearer and more accurate communication. By using this system, users can communicate more accurately and effectively.

[0752] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0753] Step 1:

[0754] The user inputs text data using the terminal. At this time, the text data input by the user is sent to the system. The input data is in text format, and for example, a sentence such as "This movie exceeded my expectations, but the ending was disappointing" is input.

[0755] Step 2:

[0756] The server uses the generated artificial intelligence model to analyze the text data received from the user. First, it analyzes the text data based on natural language processing techniques to detect ambiguous expressions. It receives the text data as input and generates a list of ambiguous expressions as output.

[0757] Step 3:

[0758] The server uses an emotion engine to determine the user's emotional state based on the detected ambiguous expressions. Specifically, the emotion engine analyzes the text data and detects emotional states such as positive, negative, and neutral. It receives the initial text data as input and generates an emotional state as output.

[0759] Step 4:

[0760] The server generates multiple interpretations based on the detected ambiguous expressions and emotional states. In this step, a generative AI model is used, and a prompt sentence such as "Provide a positive interpretation of the following ambiguous text: This movie exceeded my expectations, but the ending was disappointing" is sent to the model. The server receives the ambiguous expressions and emotional states as input and generates a list of customized interpretations as output.

[0761] Step 5:

[0762] The server sends the generated interpretations to the user's device, which then displays them to the user. The interpretations are presented in a format that is easy for the user to understand, and the user can use these interpretations to easily understand ambiguous expressions.

[0763] Step 6:

[0764] By referring to the presented interpretation, users can take measures to avoid misunderstandings. This reduces misunderstandings of ambiguous expressions in the text and improves the quality of communication. The user's actions and choices are the final output.

[0765] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0766] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0767] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0768] [Third embodiment]

[0769] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0770] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0771] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0772] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0773] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0774] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0775] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0776] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0777] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0778] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0779] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0780] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0781] The present invention relates to a system that analyzes text data using a generated artificial intelligence model, detects ambiguous expressions in the text data, and presents multiple interpretations of the expressions to a user.

[0782] The server first loads the necessary generative AI model, specifically, a Transformer model, prepared from an existing natural language processing model such as "t5-small," which has the ability to detect ambiguous expressions and generate interpretations.

[0783] A user inputs text data from a terminal. For example, the user might input the text "I didn't agree with him at the meeting yesterday, but I agree with his opinion." This text may be interpreted differently depending on the user's intention and context.

[0784] The server then receives the text data entered by the user and analyzes it using a generative AI model. This analysis identifies ambiguous expressions in the text data. The model takes into account contextual information and punctuation placement to detect expressions that may be misleading.

[0785] The server then generates different interpretations for the detected ambiguous expressions. For example, it can provide the following interpretations for the above text:

[0786] 1. "I didn't agree with the whole meeting yesterday, but I might agree with some of his points."

[0787] 2. "I don't agree with his opinion in its entirety, but I might agree with him under certain circumstances."

[0788] The server then displays the generated interpretations on the terminal, allowing the user to refer to these interpretations and take measures to avoid misunderstandings.

[0789] As described above, the present invention provides a system that detects ambiguous expressions in text-based communication and presents their interpretations to prevent misunderstandings. By using this system, users can achieve more accurate and clearer communication.

[0790] The processing flow will be explained below.

[0791] Step 1:

[0792] The server loads the generative AI model, specifically the Transformer model "t5-small" based on natural language processing, and prepares it for analysis.

[0793] Step 2:

[0794] A user inputs text data through a terminal. For example, the user inputs "I didn't agree with him at the meeting yesterday, but I agree with his opinion."

[0795] Step 3:

[0796] The server receives text data entered by the user, which is then analyzed.

[0797] Step 4:

[0798] The server uses the generative AI model to analyze the input text data. Specifically, it performs an analysis process to detect ambiguous expressions within the text data.

[0799] Step 5:

[0800] As a result of the analysis, the server generates multiple interpretations for the detected ambiguous expressions. For example, it generates the following interpretations for the above text: "1. I did not agree with the entire meeting yesterday, but I may agree with some of his opinions." "2. I do not agree with his entire opinion, but I may agree with his opinion under certain circumstances."

[0801] Step 6:

[0802] The server transmits the generated interpretations to the user's terminal.

[0803] Step 7:

[0804] The terminal displays the received interpretation to the user, who then checks the displayed interpretation and understands the ambiguous expressions in the text data.

[0805] Step 8:

[0806] The user can refer to the proposed interpretation and correct or add to the text data as necessary. This process allows the user to choose appropriate expressions to avoid misunderstandings.

[0807] Example 1

[0808] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0809] In natural language text data exchange, ambiguous expressions can lead to misunderstandings. This misunderstanding can cause serious problems, especially in business communications and important message exchanges. Conventional systems have had difficulty automatically detecting such ambiguous expressions and providing clear interpretations to users.

[0810] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0811] In this invention, the server includes means for loading the generated AI model as a Transformer model, means for receiving user input and sending it to the server, and means for receiving analysis results from the server and displaying them to the user, thereby enabling highly accurate detection of ambiguous expressions in text data and providing the user with their interpretations.

[0812] A "generated artificial intelligence model" refers to a machine learning algorithm that has been trained on a set of data and used to perform a specific task.

[0813] "Text data" refers to a collection of information expressed as character strings, including sentences and messages written in natural language.

[0814] "Polyambiguous expressions" refer to words or phrases that can be interpreted in multiple ways depending on the context or situation.

[0815] "Interpretation" refers to supplementary explanations and possibilities that clarify the meaning contained in the text data and are provided to help users deepen their understanding.

[0816] A "Transformer model" is a type of machine learning model used in natural language processing, and refers to a model that has the ability to analyze text data taking context into account.

[0817] "Loading" refers to the operation of reading a program or data into memory and making it available for use.

[0818] "User" refers to the person who uses the system and is the entity that inputs the text data to be analyzed.

[0819] A "server" refers to a computer that provides services over a network and plays a central role in processing and managing data.

[0820] A "terminal" refers to a computer or device that is directly used by a user, and is a device that communicates with a server to send and receive data.

[0821] "Analysis results" refers to the information and data obtained after the generated artificial intelligence model analyzes the text data.

[0822] The present invention relates to a system that analyzes text data using a generated artificial intelligence model, detects ambiguous expressions in the text data, and presents multiple interpretations of the expressions to a user. An embodiment of the present invention will be described in detail below.

[0823] Hardware and Software Configuration

[0824] The server is a high-performance computer with hardware acceleration, such as a GPU. The server performs the following functions:

[0825] 1. The generated AI model is loaded as a Transformer model. At this time, a natural language processing model such as "t5-small" is downloaded from the internet and loaded into memory.

[0826] 2. Receive text data sent by the user and analyze it using the generative AI model.

[0827] 3. Detect ambiguous expressions and generate multiple interpretations for them.

[0828] 4. The generated interpretation results are sent to the user's terminal.

[0829] A terminal is a computer, mobile device, or other device that is directly operated by a user and performs the following actions:

[0830] 1. Provide an interface for users to enter text data, implemented as a web or desktop application.

[0831] 2. Send the text data entered by the user to the server.

[0832] 3. Display the analysis results received from the server to the user.

[0833] A user is someone who uses this system and performs the following operations:

[0834] 1. Enter the text data you want to analyze through the device interface. For example, enter the text "I didn't agree with him in the meeting yesterday, but I agree with his opinion."

[0835] 2. The input data is sent to the server and waits for parsing.

[0836] 3. The interpretation results sent from the server are displayed on the terminal.

[0837] Specific examples

[0838] Consider the case where the user enters the following text:

[0839] For example: "I disagreed with him in the meeting yesterday, but I agree with his opinion."

[0840] This text data is sent to the server, which analyzes it using the "t5-small" model, which generates the following interpretation:

[0841] 1. "I didn't agree with the whole meeting yesterday, but I might agree with some of his points."

[0842] 2. "I don't agree with his opinion in its entirety, but I might agree with him under certain circumstances."

[0843] These interpretations are displayed on the user's device, allowing the user to confirm the interpretation that best suits their intentions.

[0844] In this way, this system allows users to easily interpret ambiguous expressions and helps prevent misunderstandings.

[0845] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0846] Step 1:

[0847] The server loads the generative AI model.

[0848] Specific behavior:

[0849] The server downloads a Transformer model, such as "t5-small," from an internet resource and loads it into memory, using the necessary libraries, such as PyTorch and the Transformers library.

[0850] Input and Output:

[0851] Input: Model resource URL on the Internet

[0852] Output: Generative AI model loaded in memory

[0853] What happens:

[0854] The server retrieves the model file from the specified URL and loads it into memory, where it is ready to be used for text analysis.

[0855] Step 2:

[0856] The user inputs text data from the terminal.

[0857] Specific behavior:

[0858] The user inputs the text data to be analyzed using a device interface (web application or desktop application).

[0859] Input and Output:

[0860] Input: User text data (e.g., "I disagreed with him in the meeting yesterday, but I agree with his opinion.")

[0861] Output: Input data in the terminal

[0862] What happens:

[0863] The user enters the text they want to analyze in the text field and clicks the submit button. This data is stored internally and is ready for the next step.

[0864] Step 3:

[0865] The terminal transmits the input text data to the server.

[0866] Specific behavior:

[0867] The device sends the text data entered by the user to the server as an HTTP request. The data is encoded in JSON format and sent to the API endpoint using the POST method.

[0868] Input and Output:

[0869] Input: Text data in the terminal

[0870] Output: Text data sent to the server

[0871] What happens:

[0872] The terminal encodes the user's input data in JSON format and issues a POST request to the specified server endpoint.

[0873] Step 4:

[0874] The server uses a generative AI model to analyze the received text data.

[0875] Specific behavior:

[0876] The server inputs the received text data into the t5-small model and begins analysis. The model takes into account contextual information and punctuation placement to identify ambiguous expressions within the text.

[0877] Input and Output:

[0878] Input: Text data sent to the server

[0879] Output: Analysis results with ambiguous expressions identified

[0880] What happens:

[0881] The server runs the text data through a generative AI model to detect ambiguous expressions, which are then stored in memory as analysis results.

[0882] Step 5:

[0883] The server generates multiple interpretations for the detected ambiguous expression.

[0884] Specific behavior:

[0885] The server generates multiple interpretations based on the analysis results obtained from the generative AI model. For example, for expressions that can be interpreted differently, it generates sentences that explain each of their meanings.

[0886] Input and Output:

[0887] Input: Analysis results (identification of ambiguous expressions)

[0888] Output: Multiple interpretations generated

[0889] What happens:

[0890] The server uses a generative AI model to analyze the context of ambiguous expressions and generate multiple possible interpretations, which are then stored in JSON format.

[0891] Step 6:

[0892] The server transmits the generated interpretation results to the user's terminal.

[0893] Specific behavior:

[0894] The server returns the generated interpretation results to the user's device as an HTTP response. The data is again encoded in JSON format and sent.

[0895] Input and Output:

[0896] Input: Generated interpretation

[0897] Output: Interpretation results sent to the user's terminal

[0898] What happens:

[0899] The server encodes the generated interpretation results in JSON format and returns an HTTP response to the user's terminal.

[0900] Step 7:

[0901] The terminal displays the multiple interpretation results received from the server to the user.

[0902] Specific behavior:

[0903] The application or web interface used by the user displays the interpretation results received from the server in an easy-to-understand format, for example as a list of interpretations displayed on the interface for the user to review.

[0904] Input and Output:

[0905] Input: Interpretation result sent from the server

[0906] Output: The interpretation results displayed to the user

[0907] What happens:

[0908] The terminal analyzes the interpretation results received from the server and displays them on the user interface. The user can check the interpretation list and select an appropriate interpretation as needed.

[0909] As described above, the input and output for each processing step can be clearly indicated, and the system of the present invention can be implemented in a form including specific operations.

[0910] (Application example 1)

[0911] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0912] In traditional communication, ambiguous expressions often lead to misunderstandings, and accurate information transmission is particularly important in security services. However, current systems lack the ability to detect ambiguous expressions in text data and provide their interpretation. This can result in misunderstandings of important information and inappropriate decision-making.

[0913] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0914] In this invention, the server includes means for analyzing text data using the generated artificial intelligence model to detect ambiguous expressions, means for providing multiple interpretations of the detected ambiguous expressions, means for presenting the provided multiple interpretations to the user, and means for analyzing the content of a message in a security service to detect ambiguous expressions and generate a reinterpretation, and means for displaying the reinterpreted result on the user's device. This makes it possible to detect ambiguous expressions in text-based communication and present their interpretations, thereby preventing misunderstandings.

[0915] A "generated AI model" is a newly generated AI model based on an existing AI model with natural language processing capabilities, and is used to analyze and interpret ambiguous expressions.

[0916] "Text data" refers to character string information input by the user, and includes messages, sentences, and the like.

[0917] An "ambiguous expression" is an expression that can be interpreted differently depending on the context and intent, and refers to a word or phrase whose specific meaning is difficult to pin down.

[0918] "Security services" are services for maintaining the safety and confidentiality of information, particularly those for managing important communications within a company or in transactions.

[0919] A "message" is a sentence or text data that a user sends to another user, and may contain important information.

[0920] "Reinterpretation" refers to a new interpretation generated from a different angle for an input ambiguous expression, making it possible to understand it from multiple perspectives.

[0921] "User device" refers to a computer device such as a smartphone, PC, or tablet, which is the terminal through which the user receives the analysis results.

[0922] MODE FOR CARRYING OUT THE INVENTION

[0923] This invention relates to a system that uses a generated artificial intelligence model to analyze text data, detect ambiguous expressions, and present multiple interpretations to the user. This system is particularly useful in security services to prevent misunderstandings of message content. An embodiment of this invention will be described in detail below.

[0924] Hardware and software used

[0925] The following hardware and software are used to implement the present invention.

[0926] Hardware:

[0927] Server: Amazon Web Services (AWS) EC2 instance

[0928] User devices: smartphone, PC, tablet

[0929] software:

[0930] Generative AI model: Hugging Face Transformer model "t5-small"

[0931] Server-side framework: Flask (Python-based)

[0932] Front-end framework: React

[0933] Database: PostgreSQL

[0934] Program processing

[0935] Server Processing

[0936] The server runs on the Flask framework and receives text data sent by users. The received text data is analyzed using the pre-loaded Hugging Face Transformer model "t5-small." Specifically, it detects ambiguous expressions and generates multiple interpretations for those ambiguous expressions. The analysis results are sent to the user's device as multiple reinterpreted expressions. The database stores the analysis results and the user's usage history.

[0937] User Action

[0938] Users send the message they wish to have analyzed from their device to the server. They can then check the multiple interpretations provided as analysis results on their device. This allows users to prevent misunderstandings about ambiguous expressions.

[0939] Specific examples

[0940] For example, if a user sends a message from their smartphone to the server saying, "There was a lot of feedback in an important meeting, but I felt everyone should agree," the system can provide multiple interpretations, such as:

[0941] 1. "I received a lot of feedback in an important meeting, and I felt like everyone should agree with it."

[0942] 2. "I received a lot of feedback during an important meeting, and I felt like we all should agree."

[0943] Prompt Sentence Examples

[0944] If a user enters the following text and requests analysis:

[0945] text_input = "It was an important meeting and there was a lot of feedback, but everyone felt they should agree."

[0946] The server receives this input, analyzes it using a generative AI model, and generates multiple interpretations like the ones above. These results are displayed on the user's device and used as reference information to avoid misunderstandings.

[0947] This embodiment makes it possible to detect ambiguous expressions in text-based communication and present their interpretations, thereby preventing misunderstandings.

[0948] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0949] Step 1:

[0950] The user inputs the message they wish to analyze from their device (smartphone, PC, tablet, etc.). The input text data is prepared as a prompt sentence. For example, they input a message such as, "There was a lot of feedback in an important meeting, but I felt that everyone should agree."

[0951] Input: Text data entered by the user

[0952] Output: Text data as prompt sentence

[0953] Step 2:

[0954] The user's device sends the entered text data to a specified endpoint on the server, which runs on the Flask framework and receives it as an HTTP request.

[0955] Input: Text data as a prompt

[0956] Output: HTTP request sent to the server

[0957] Step 3:

[0958] The server retrieves the received text data and loads the generative AI model "t5-small." This is where the analysis of the text data begins. The server performs data analysis to detect ambiguous expressions within the text data.

[0959] Input: Text data sent to the server

[0960] Output: Analysis results including ambiguous expressions

[0961] Step 4:

[0962] Using the generative AI model "t5-small," the server generates multiple interpretations for the detected ambiguous expressions. The model takes context into account and performs data calculations to generate multiple different interpretations.

[0963] Input: Analysis results containing ambiguous expressions

[0964] Output: Multiple interpretation results

[0965] Step 5:

[0966] The server sends the generated interpretation results to the device, where they are formatted for display on the front end.

[0967] Input: Multiple interpretation results

[0968] Output: Data sent to the front end

[0969] Step 6:

[0970] The user's device displays the received analysis results in a React-based interface, allowing the user to check multiple interpretations and choose the appropriate one.

[0971] Input: Data sent to the front end

[0972] Output: Multiple interpretation results displayed to the user

[0973] Step 7:

[0974] Users can check the analysis results and use them as reference information to prevent misunderstandings. Users can then use this information to decide on the next action to take in order to make appropriate decisions.

[0975] Input: Multiple interpretation results displayed to the user

[0976] Output: User decision

[0977] Through the above processing steps, ambiguous expressions in text data are analyzed and multiple interpretations are presented, enabling users to communicate accurately.

[0978] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0979] The present invention relates to a system that analyzes text data using a generated artificial intelligence model, detects ambiguous expressions in the text data, and presents multiple interpretations to the user. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the provided interpretations are adapted to the user's emotional state.

[0980] Hereinafter, an embodiment of the present invention will be described.

[0981] First, the server loads the necessary generative AI model and emotion engine. The generative AI model is based on natural language processing and analyzes text data and detects ambiguous expressions. The emotion engine is used to extract user emotions from text data. This enables both text analysis and emotion recognition.

[0982] A user inputs text data from a terminal. For example, the user may input the text "I didn't agree with him at the meeting yesterday, but I agree with his opinion." This text may be interpreted differently depending on the context and emotion.

[0983] The server then receives the text data entered by the user. This text data is then subjected to analysis. The server first analyzes the text data to identify ambiguous expressions. A generative AI model is used in the analysis, taking into account contextual information and punctuation placement to detect expressions that may be misleading.

[0984] Next, the server uses an emotion engine to recognize the user's emotions. Specifically, it extracts the user's emotional state from the input text data, such as positive, negative, or neutral emotions.

[0985] The server generates multiple interpretations for the detected ambiguous expressions, customizing the interpretations based on the user's emotional state. For example, the server generates the following interpretations for the above text:

[0986] 1. "I didn't agree with the whole meeting yesterday, but I might agree with some of his points." (Users express positive sentiment)

[0987] 2. "I don't agree with his opinion in general, but I might agree with him in certain circumstances." (When the user is expressing neutral or negative sentiment)

[0988] The server then sends the generated interpretations to the user's device, which then displays them to the user. By referring to these interpretations, the user can understand the ambiguous expressions in the text data and take measures to avoid misunderstandings.

[0989] In this way, the present invention provides a system that detects ambiguous expressions in text-based communication and presents their interpretation according to the emotional state, thereby enabling clearer and more accurate communication. By using this system, users can communicate more accurately and effectively.

[0990] The processing flow will be explained below.

[0991] Step 1:

[0992] The server loads the generative AI model and emotion engine. Specifically, it prepares the Transformer model "t5-small" based on natural language processing and the emotion engine for recognizing user emotions.

[0993] Step 2:

[0994] A user inputs text data through a terminal. For example, the user inputs "I didn't agree with him at the meeting yesterday, but I agree with his opinion."

[0995] Step 3:

[0996] The server receives text data entered by the user, which is then analyzed.

[0997] Step 4:

[0998] The server uses the generated AI model to analyze the input text data. Specifically, it performs an analysis process to detect ambiguous expressions within the text data.

[0999] Step 5:

[1000] The server uses an emotion engine to recognize the user's emotional state and classify the user's emotions into positive, negative, neutral, etc. based on the input text data.

[1001] Step 6:

[1002] The server generates multiple interpretations for detected ambiguous expressions, taking into account the user's emotions. For example, it provides different interpretations for positive and negative emotions.

[1003] Step 7:

[1004] The server transmits the generated interpretations to the user's terminal.

[1005] Step 8:

[1006] The terminal displays the received interpretation to the user, who then checks the displayed interpretation and understands the ambiguous expressions in the text data.

[1007] Step 9:

[1008] The user can refer to the proposed interpretation and correct or add to the text data as necessary. This process allows the user to choose appropriate expressions to avoid misunderstandings.

[1009] Example 2

[1010] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1011] Conventional text analysis systems provide a uniform interpretation of ambiguous expressions, making it difficult to provide flexible interpretations that reflect the emotional state of each individual user. Furthermore, the lack of emotion recognition functionality meant that interpretations that took the user's emotional state into account were insufficient. As a result, users were more likely to misunderstand the true meaning of the text data, hindering smooth communication.

[1012] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1013] In this invention, the server includes means for analyzing text data using the generated artificial intelligence model and detecting ambiguous expressions in the text data, means for providing multiple interpretations for the detected ambiguous expressions, means for recognizing a user's emotion, means for customizing an interpretation based on the recognized emotional state, and means for presenting the provided multiple interpretations to the user. This makes it possible to provide multiple interpretations according to the user's emotional state and reduce misunderstandings of ambiguous expressions in the text data.

[1014] A "generated artificial intelligence model" is a model that has been pre-trained to analyze text data and detect ambiguous expressions, and that uses natural language processing techniques.

[1015] "Text data" refers to character information such as sentences, documents, and chat messages entered by the user.

[1016] An "ambiguous expression" is a word or phrase that can have different meanings within a text depending on the context.

[1017] "Multiple interpretations" refers to the multiple possible meanings or connotations of an ambiguous expression.

[1018] "Means for recognizing user emotions" refers to technology or algorithms for determining a user's emotional state from text data, such as technology for extracting positive, negative, or neutral emotional states.

[1019] "Means for customizing interpretation based on emotional state" refers to a technique or algorithm that adjusts the interpretation of an ambiguous expression based on the perceived emotion of the user.

[1020] The "means for presenting the provided interpretations to the user" refers to a technique or interface for displaying the generated interpretations on the user's terminal.

[1021] The present invention is a system that uses the generated AI model and emotion engine to analyze text data and recognize user emotions, detect ambiguous expressions in the text, and provide multiple interpretations for them. This system is specifically implemented by a series of processes with the server, terminal, and user as subjects.

[1022] First, the server loads the generative AI model (e.g., OpenAI GPT-3) and emotion engine (e.g., IBM Watson API), along with any necessary software libraries and configuration files, so that the system is ready for text analysis and emotion recognition.

[1023] Next, the user inputs text data from the terminal. For example, the text could be something like, "I didn't agree with him at the meeting yesterday, but I agree with his opinion." The terminal sends the input data from the user to the server as an HTTP request.

[1024] The server receives the text data sent from the device and passes it to the analysis module, which uses a generative AI model to detect ambiguous expressions in the input text data. The model takes into account contextual information and punctuation placement to identify expressions that may be misleading.

[1025] Furthermore, the server uses an emotion engine to recognize the user's emotion from the text data. The emotion engine distinguishes between positive, negative, and neutral emotional states. For example, if a user sends the text "I didn't agree with him in the meeting yesterday, but I agree with his opinion," the emotion engine recognizes the emotional state as "neutral."

[1026] The server generates multiple interpretations for the detected ambiguous expressions, taking into account the user's emotional state. For example, if the user expresses a positive emotion, the expression can be interpreted as "I didn't agree with the entire meeting yesterday, but I might agree with some of his opinions." If the user expresses a negative or neutral emotion, the expression can be interpreted as "I don't agree with his opinions overall, but I might agree with his opinions under certain circumstances."

[1027] Finally, the server sends the generated interpretations to the user's device, which displays them to the user. By referring to these interpretations, the user can more accurately understand the true meaning of the text data.

[1028] For example, the prompt the user enters is:

[1029] Generate an interpretation for the text "I disagreed with him in the meeting yesterday, but I agree with his opinion."

[1030] This system allows users to adapt the results of analysis of text data containing ambiguous expressions to their emotional state, thereby reducing misunderstandings.

[1031] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1032] System program processing flow

[1033] Step 1: The server loads the generative AI model and emotion engine

[1034] The server first loads the generative AI model and emotion engine. Specifically, it loads the libraries and configuration files required to use the generative AI model (e.g., OpenAI GPT-3) and emotion engine (e.g., IBM Watson API).

[1035] Input: Server configuration file, API key, etc.

[1036] Data processing: Initializing the library and loading the model

[1037] Output: AI model and emotion engine ready state

[1038] Specific behavior:

[1039] The server reads the configuration file and loads the API key and model paths into memory.

[1040] Initialize the generative AI model and emotion engine.

[1041] Step 2: The user inputs text data from the terminal.

[1042] The user inputs text data to be analyzed into an input field on the terminal, and this data is sent to the server.

[1043] Input: Text data that the user types into the terminal.

[1044] Data processing: None

[1045] Output: HTTP request sent from the terminal to the server

[1046] Specific behavior:

[1047] A user enters text into an input field on a web form or application and presses the submit button.

[1048] The terminal generates an HTTP request and sends the input text data to the server.

[1049] Step 3: The server receives the input text data

[1050] The server receives the text data sent from the terminal and prepares to pass it on to the next analysis step.

[1051] Input: Text data sent via the HTTP request

[1052] Data processing: Parsing HTTP requests

[1053] Output: Text data passed to the analysis module

[1054] Specific behavior:

[1055] The server receives the HTTP request and extracts the text data from it.

[1056] The extracted text data is passed to the analysis module.

[1057] Step 4: The server analyzes the text data and identifies ambiguous expressions.

[1058] The server uses a generative AI model to analyze the received text data and identify ambiguous expressions.

[1059] Input: Text data passed to the analysis module

[1060] Data Processing: Text Analysis with Generative AI Models

[1061] Output: Identification of ambiguous expressions

[1062] Specific behavior:

[1063] The server sends the text data to the generative AI model and performs the analysis.

[1064] The model takes into account contextual information and punctuation placement and returns results that detect ambiguous expressions.

[1065] Step 5: The server recognizes the user's emotion using the emotion engine.

[1066] The server uses an emotion engine to extract the user's emotion from the text data.

[1067] Input: Text data and results of identifying ambiguous expressions

[1068] Data processing: Emotion analysis using an emotion engine

[1069] Output: User's emotional state (positive, negative, neutral, etc.)

[1070] Specific behavior:

[1071] The server passes the text data to the emotion engine and performs emotion analysis.

[1072] The engine analyzes the emotional state in the text and returns the results.

[1073] Step 6: The server generates multiple interpretations and customizes them based on emotional state.

[1074] The server generates multiple interpretations based on the generative AI model and emotion recognition results, and customizes them according to the user's emotional state.

[1075] Input: Identification results of ambiguous expressions, user's emotional state

[1076] Data processing: interpretation generation and emotional state-based customization

[1077] Output: Multiple customized interpretations

[1078] Specific behavior:

[1079] The server uses a generative AI model to generate an interpretation for an ambiguous expression.

[1080] Based on the acquired emotional data, each interpretation is adapted to the emotion.

[1081] Step 7: The server sends the generated interpretation to the terminal.

[1082] The server formats the generated interpretations as an HTTP response and sends it to the user's device.

[1083] Input: Customized multiple interpretations

[1084] Data processing: Generating HTTP responses

[1085] Output: The interpreted response sent to the device

[1086] Specific behavior:

[1087] The server formats the generated interpretation into an HTTP response.

[1088] Send the interpreted response to the terminal.

[1089] Step 8: The terminal displays the interpretation to the user

[1090] The terminal displays the received interpretation to the user, who then checks the interpretation and understands the true meaning of the text data.

[1091] Input: The interpreted response sent by the server

[1092] Data processing: Display of interpretations

[1093] Output: The interpretation displayed to the user

[1094] Specific behavior:

[1095] The device receives the response from the server and updates the GUI to display the interpretation.

[1096] Provide an interface that allows users to view the interpretation.

[1097] The above are the specific processing steps of this system.

[1098] (Application example 2)

[1099] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1100] In text-based communication, ambiguous expressions in the text data entered by users can lead to misunderstandings and unpleasant interactions. In addition, the quality of communication can be reduced due to a lack of appropriate interpretation according to the user's emotional state.

[1101] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1102] In this invention, the server includes means for analyzing text data using the generated artificial intelligence model and detecting ambiguous expressions in the text data, means for providing multiple interpretations for the detected ambiguous expressions, means for presenting the provided multiple interpretations to the user, means for analyzing the user's emotions, and means for generating interpretations based on the user's emotional state. This makes it possible to accurately detect ambiguous expressions in the text data and provide interpretations that correspond to the user's emotional state, thereby reducing misunderstandings and unpleasant interactions and improving the quality of communication.

[1103] "Generated artificial intelligence model" refers to the artificial intelligence technology used to analyze text data, detect ambiguous expressions, and generate appropriate interpretations.

[1104] "Text data" refers to data expressed as character information, such as sentences or comments entered by a user.

[1105] An "ambiguous expression" refers to an expression that has different meanings depending on the context and interpretation, and may lead to misunderstanding.

[1106] "Interpretation" refers to specific explanations or proposed explanations to understand the meaning and intent of text data.

[1107] "User" refers to the entity that uses the system to input text data and have it interpreted.

[1108] "Emotion" refers to the emotional state, such as positive, negative, or neutral, contained in the text data.

[1109] "Analysis" refers to the process of analyzing text data and extracting information such as meaning and sentiment.

[1110] A "server" is the central computer of the system, and refers to the device that analyzes text data and generates interpretations.

[1111] The present invention relates to a system that uses a generated artificial intelligence model to analyze text data, detect ambiguous expressions, and present multiple interpretations to the user. Furthermore, it combines an emotion engine that recognizes the user's emotions so that the interpretations provided adapt to the user's emotional state.

[1112] First, the server loads the generated artificial intelligence model and emotion engine. The generative AI model is based on natural language processing and is used to analyze text data and detect ambiguous expressions. The emotion engine has the function of extracting user emotions from text data. This allows both text analysis and emotion recognition to be performed simultaneously.

[1113] A user inputs text data from a terminal. For example, the user may input the text, "This movie exceeded my expectations, but the ending was disappointing." This text may be interpreted differently depending on the context and emotions.

[1114] The server then receives the text data entered by the user. This text data is then analyzed. First, a generative AI model is used to analyze the text data and identify ambiguous expressions. This analysis takes into account contextual information and punctuation placement to detect expressions that may be misleading.

[1115] Next, the server uses an emotion engine to recognize the user's emotion. Specifically, it determines the user's emotional state (positive, negative, neutral, etc.) from the input text data. This emotion recognition is performed using a natural language processing library such as TextBlob.

[1116] The server then generates multiple interpretations for the detected ambiguous expressions, customizing the interpretations based on the user's emotional state. For example, the server generates the following interpretations for the above text:

[1117] 1. "The movie was good overall, but the ending was disappointing." (Users express positive emotions)

[1118] 2. "The ending of the movie was disappointing, so I didn't enjoy it overall." (Users express negative emotions)

[1119] The server sends the generated interpretations to the user's device, which then displays them to the user, allowing the user to refer to these interpretations, understand the ambiguous expressions in the text data, and take measures to avoid misunderstandings.

[1120] Example prompt sentence:

[1121] "Provide a positive interpretation of the following ambiguous text: This movie exceeded my expectations, but the ending disappointed me."

[1122] "Provide a neutral / negative interpretation of the following ambiguous text: This movie exceeded my expectations, but the ending disappointed me."

[1123] In this way, the present invention provides a system that detects ambiguous expressions in text-based communication and presents their interpretation according to the emotional state, thereby enabling clearer and more accurate communication. By using this system, users can communicate more accurately and effectively.

[1124] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1125] Step 1:

[1126] The user inputs text data using the terminal. At this time, the text data input by the user is sent to the system. The input data is in text format, and for example, a sentence such as "This movie exceeded my expectations, but the ending was disappointing" is input.

[1127] Step 2:

[1128] The server uses the generated artificial intelligence model to analyze the text data received from the user. First, it analyzes the text data based on natural language processing techniques to detect ambiguous expressions. It receives the text data as input and generates a list of ambiguous expressions as output.

[1129] Step 3:

[1130] The server uses an emotion engine to determine the user's emotional state based on the detected ambiguous expressions. Specifically, the emotion engine analyzes the text data and detects emotional states such as positive, negative, and neutral. It receives the initial text data as input and generates an emotional state as output.

[1131] Step 4:

[1132] The server generates multiple interpretations based on the detected ambiguous expressions and emotional states. In this step, a generative AI model is used, and a prompt sentence such as "Provide a positive interpretation of the following ambiguous text: This movie exceeded my expectations, but the ending was disappointing" is sent to the model. The server receives the ambiguous expressions and emotional states as input and generates a list of customized interpretations as output.

[1133] Step 5:

[1134] The server sends the generated interpretations to the user's device, which then displays them to the user. The interpretations are presented in a format that is easy for the user to understand, and the user can use these interpretations to easily understand ambiguous expressions.

[1135] Step 6:

[1136] By referring to the presented interpretation, users can take measures to avoid misunderstandings. This reduces misunderstandings of ambiguous expressions in the text and improves the quality of communication. The user's actions and choices are the final output.

[1137] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1138] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1139] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1140] [Fourth embodiment]

[1141] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1142] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1143] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1144] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1145] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1146] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1147] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1148] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1149] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1150] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1151] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1152] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1153] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1154] The present invention relates to a system that analyzes text data using a generated artificial intelligence model, detects ambiguous expressions in the text data, and presents multiple interpretations of the expressions to a user.

[1155] The server first loads the necessary generative AI model, specifically, a Transformer model, prepared from an existing natural language processing model such as "t5-small," which has the ability to detect ambiguous expressions and generate interpretations.

[1156] A user inputs text data from a terminal. For example, the user might input the text "I didn't agree with him at the meeting yesterday, but I agree with his opinion." This text may be interpreted differently depending on the user's intention and context.

[1157] The server then receives the text data entered by the user and analyzes it using a generative AI model. This analysis identifies ambiguous expressions in the text data. The model takes into account contextual information and punctuation placement to detect expressions that may be misleading.

[1158] The server then generates different interpretations for the detected ambiguous expressions. For example, it can provide the following interpretations for the above text:

[1159] 1. "I didn't agree with the whole meeting yesterday, but I might agree with some of his points."

[1160] 2. "I don't agree with his opinion in its entirety, but I might agree with him under certain circumstances."

[1161] The server then displays the generated interpretations on the terminal, allowing the user to refer to these interpretations and take measures to avoid misunderstandings.

[1162] As described above, the present invention provides a system that detects ambiguous expressions in text-based communication and presents their interpretations to prevent misunderstandings. By using this system, users can achieve more accurate and clearer communication.

[1163] The processing flow will be explained below.

[1164] Step 1:

[1165] The server loads the generative AI model, specifically the Transformer model "t5-small" based on natural language processing, and prepares it for analysis.

[1166] Step 2:

[1167] A user inputs text data through a terminal. For example, the user inputs "I didn't agree with him at the meeting yesterday, but I agree with his opinion."

[1168] Step 3:

[1169] The server receives text data entered by the user, which is then analyzed.

[1170] Step 4:

[1171] The server uses the generative AI model to analyze the input text data. Specifically, it performs an analysis process to detect ambiguous expressions within the text data.

[1172] Step 5:

[1173] As a result of the analysis, the server generates multiple interpretations for the detected ambiguous expressions. For example, it generates the following interpretations for the above text: "1. I did not agree with the entire meeting yesterday, but I may agree with some of his opinions." "2. I do not agree with his entire opinion, but I may agree with his opinion under certain circumstances."

[1174] Step 6:

[1175] The server transmits the generated interpretations to the user's terminal.

[1176] Step 7:

[1177] The terminal displays the received interpretation to the user, who then checks the displayed interpretation and understands the ambiguous expressions in the text data.

[1178] Step 8:

[1179] The user can refer to the proposed interpretation and correct or add to the text data as necessary. This process allows the user to choose appropriate expressions to avoid misunderstandings.

[1180] Example 1

[1181] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1182] In natural language text data exchange, ambiguous expressions can lead to misunderstandings. This misunderstanding can cause serious problems, especially in business communications and important message exchanges. Conventional systems have had difficulty automatically detecting such ambiguous expressions and providing clear interpretations to users.

[1183] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1184] In this invention, the server includes means for loading the generated AI model as a Transformer model, means for receiving user input and sending it to the server, and means for receiving analysis results from the server and displaying them to the user, thereby enabling highly accurate detection of ambiguous expressions in text data and providing the user with their interpretations.

[1185] A "generated artificial intelligence model" refers to a machine learning algorithm that has been trained on a set of data and used to perform a specific task.

[1186] "Text data" refers to a collection of information expressed as character strings, including sentences and messages written in natural language.

[1187] "Polyambiguous expressions" refer to words or phrases that can be interpreted in multiple ways depending on the context or situation.

[1188] "Interpretation" refers to supplementary explanations and possibilities that clarify the meaning contained in the text data and are provided to help users deepen their understanding.

[1189] A "Transformer model" is a type of machine learning model used in natural language processing, and refers to a model that has the ability to analyze text data taking context into account.

[1190] "Loading" refers to the operation of reading a program or data into memory and making it available for use.

[1191] "User" refers to the person who uses the system and is the entity that inputs the text data to be analyzed.

[1192] A "server" refers to a computer that provides services over a network and plays a central role in processing and managing data.

[1193] A "terminal" refers to a computer or device that is directly used by a user, and is a device that communicates with a server to send and receive data.

[1194] "Analysis results" refers to the information and data obtained after the generated artificial intelligence model analyzes the text data.

[1195] The present invention relates to a system that analyzes text data using a generated artificial intelligence model, detects ambiguous expressions in the text data, and presents multiple interpretations of the expressions to a user. An embodiment of the present invention will be described in detail below.

[1196] Hardware and Software Configuration

[1197] The server is a high-performance computer with hardware acceleration, such as a GPU. The server performs the following functions:

[1198] 1. The generated AI model is loaded as a Transformer model. At this time, a natural language processing model such as "t5-small" is downloaded from the internet and loaded into memory.

[1199] 2. Receive text data sent by the user and analyze it using the generative AI model.

[1200] 3. Detect ambiguous expressions and generate multiple interpretations for them.

[1201] 4. The generated interpretation results are sent to the user's terminal.

[1202] A terminal is a computer, mobile device, or other device that is directly operated by a user and performs the following actions:

[1203] 1. Provide an interface for users to enter text data, implemented as a web or desktop application.

[1204] 2. Send the text data entered by the user to the server.

[1205] 3. Display the analysis results received from the server to the user.

[1206] A user is someone who uses this system and performs the following operations:

[1207] 1. Enter the text data you want to analyze through the device interface. For example, enter the text "I didn't agree with him in the meeting yesterday, but I agree with his opinion."

[1208] 2. The input data is sent to the server and waits for parsing.

[1209] 3. The interpretation results sent from the server are displayed on the terminal.

[1210] Specific examples

[1211] Consider the case where the user enters the following text:

[1212] For example: "I disagreed with him in the meeting yesterday, but I agree with his opinion."

[1213] This text data is sent to the server, which analyzes it using the "t5-small" model, which generates the following interpretation:

[1214] 1. "I didn't agree with the whole meeting yesterday, but I might agree with some of his points."

[1215] 2. "I don't agree with his opinion in its entirety, but I might agree with him under certain circumstances."

[1216] These interpretations are displayed on the user's device, allowing the user to confirm the interpretation that best suits their intentions.

[1217] In this way, this system allows users to easily interpret ambiguous expressions and helps prevent misunderstandings.

[1218] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1219] Step 1:

[1220] The server loads the generative AI model.

[1221] Specific behavior:

[1222] The server downloads a Transformer model, such as "t5-small," from an internet resource and loads it into memory, using the necessary libraries, such as PyTorch and the Transformers library.

[1223] Input and Output:

[1224] Input: Model resource URL on the Internet

[1225] Output: Generative AI model loaded in memory

[1226] What happens:

[1227] The server retrieves the model file from the specified URL and loads it into memory, where it is ready to be used for text analysis.

[1228] Step 2:

[1229] The user inputs text data from the terminal.

[1230] Specific behavior:

[1231] The user inputs the text data to be analyzed using a device interface (web application or desktop application).

[1232] Input and Output:

[1233] Input: User text data (e.g., "I disagreed with him in the meeting yesterday, but I agree with his opinion.")

[1234] Output: Input data in the terminal

[1235] What happens:

[1236] The user enters the text they want to analyze in the text field and clicks the submit button. This data is stored internally and is ready for the next step.

[1237] Step 3:

[1238] The terminal transmits the input text data to the server.

[1239] Specific behavior:

[1240] The device sends the text data entered by the user to the server as an HTTP request. The data is encoded in JSON format and sent to the API endpoint using the POST method.

[1241] Input and Output:

[1242] Input: Text data in the terminal

[1243] Output: Text data sent to the server

[1244] What happens:

[1245] The terminal encodes the user's input data in JSON format and issues a POST request to the specified server endpoint.

[1246] Step 4:

[1247] The server uses a generative AI model to analyze the received text data.

[1248] Specific behavior:

[1249] The server inputs the received text data into the t5-small model and begins analysis. The model takes into account contextual information and punctuation placement to identify ambiguous expressions within the text.

[1250] Input and Output:

[1251] Input: Text data sent to the server

[1252] Output: Analysis results with ambiguous expressions identified

[1253] What happens:

[1254] The server runs the text data through a generative AI model to detect ambiguous expressions, which are then stored in memory as analysis results.

[1255] Step 5:

[1256] The server generates multiple interpretations for the detected ambiguous expression.

[1257] Specific behavior:

[1258] The server generates multiple interpretations based on the analysis results obtained from the generative AI model. For example, for expressions that can be interpreted differently, it generates sentences that explain each of their meanings.

[1259] Input and Output:

[1260] Input: Analysis results (identification of ambiguous expressions)

[1261] Output: Multiple interpretations generated

[1262] What happens:

[1263] The server uses a generative AI model to analyze the context of ambiguous expressions and generate multiple possible interpretations, which are then stored in JSON format.

[1264] Step 6:

[1265] The server transmits the generated interpretation results to the user's terminal.

[1266] Specific behavior:

[1267] The server returns the generated interpretation results to the user's device as an HTTP response. The data is again encoded in JSON format and sent.

[1268] Input and Output:

[1269] Input: Generated interpretation

[1270] Output: Interpretation results sent to the user's terminal

[1271] What happens:

[1272] The server encodes the generated interpretation results in JSON format and returns an HTTP response to the user's terminal.

[1273] Step 7:

[1274] The terminal displays the multiple interpretation results received from the server to the user.

[1275] Specific behavior:

[1276] The application or web interface used by the user displays the interpretation results received from the server in an easy-to-understand format, for example as a list of interpretations displayed on the interface for the user to review.

[1277] Input and Output:

[1278] Input: Interpretation result sent from the server

[1279] Output: The interpretation results displayed to the user

[1280] What happens:

[1281] The terminal analyzes the interpretation results received from the server and displays them on the user interface. The user can check the interpretation list and select an appropriate interpretation as needed.

[1282] As described above, the input and output for each processing step can be clearly indicated, and the system of the present invention can be implemented in a form including specific operations.

[1283] (Application example 1)

[1284] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1285] In traditional communication, ambiguous expressions often lead to misunderstandings, and accurate information transmission is particularly important in security services. However, current systems lack the ability to detect ambiguous expressions in text data and provide their interpretation. This can result in misunderstandings of important information and inappropriate decision-making.

[1286] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1287] In this invention, the server includes means for analyzing text data using the generated artificial intelligence model to detect ambiguous expressions, means for providing multiple interpretations of the detected ambiguous expressions, means for presenting the provided multiple interpretations to the user, and means for analyzing the content of a message in a security service to detect ambiguous expressions and generate a reinterpretation, and means for displaying the reinterpreted result on the user's device. This makes it possible to detect ambiguous expressions in text-based communication and present their interpretations, thereby preventing misunderstandings.

[1288] A "generated AI model" is a newly generated AI model based on an existing AI model with natural language processing capabilities, and is used to analyze and interpret ambiguous expressions.

[1289] "Text data" refers to character string information input by the user, and includes messages, sentences, and the like.

[1290] An "ambiguous expression" is an expression that can be interpreted differently depending on the context and intent, and refers to a word or phrase whose specific meaning is difficult to pin down.

[1291] "Security services" are services for maintaining the safety and confidentiality of information, particularly those for managing important communications within a company or in transactions.

[1292] A "message" is a sentence or text data that a user sends to another user, and may contain important information.

[1293] "Reinterpretation" refers to a new interpretation generated from a different angle for an input ambiguous expression, making it possible to understand it from multiple perspectives.

[1294] "User device" refers to a computer device such as a smartphone, PC, or tablet, which is the terminal through which the user receives the analysis results.

[1295] MODE FOR CARRYING OUT THE INVENTION

[1296] This invention relates to a system that uses a generated artificial intelligence model to analyze text data, detect ambiguous expressions, and present multiple interpretations to the user. This system is particularly useful in security services to prevent misunderstandings of message content. An embodiment of this invention will be described in detail below.

[1297] Hardware and software used

[1298] The following hardware and software are used to implement the present invention.

[1299] Hardware:

[1300] Server: Amazon Web Services (AWS) EC2 instance

[1301] User devices: smartphone, PC, tablet

[1302] software:

[1303] Generative AI model: Hugging Face Transformer model "t5-small"

[1304] Server-side framework: Flask (Python-based)

[1305] Front-end framework: React

[1306] Database: PostgreSQL

[1307] Program processing

[1308] Server Processing

[1309] The server runs on the Flask framework and receives text data sent by users. The received text data is analyzed using the pre-loaded Hugging Face Transformer model "t5-small." Specifically, it detects ambiguous expressions and generates multiple interpretations for those ambiguous expressions. The analysis results are sent to the user's device as multiple reinterpreted expressions. The database stores the analysis results and the user's usage history.

[1310] User Action

[1311] Users send the message they wish to have analyzed from their device to the server. They can then check the multiple interpretations provided as analysis results on their device. This allows users to prevent misunderstandings about ambiguous expressions.

[1312] Specific examples

[1313] For example, if a user sends a message from their smartphone to the server saying, "There was a lot of feedback in an important meeting, but I felt everyone should agree," the system can provide multiple interpretations, such as:

[1314] 1. "I received a lot of feedback in an important meeting, and I felt like everyone should agree with it."

[1315] 2. "I received a lot of feedback during an important meeting, and I felt like we all should agree."

[1316] Prompt Sentence Examples

[1317] If a user enters the following text and requests analysis:

[1318] text_input = "It was an important meeting and there was a lot of feedback, but everyone felt they should agree."

[1319] The server receives this input, analyzes it using a generative AI model, and generates multiple interpretations like the ones above. These results are displayed on the user's device and used as reference information to avoid misunderstandings.

[1320] This embodiment makes it possible to detect ambiguous expressions in text-based communication and present their interpretations, thereby preventing misunderstandings.

[1321] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1322] Step 1:

[1323] The user inputs the message they wish to analyze from their device (smartphone, PC, tablet, etc.). The input text data is prepared as a prompt sentence. For example, they input a message such as, "There was a lot of feedback in an important meeting, but I felt that everyone should agree."

[1324] Input: Text data entered by the user

[1325] Output: Text data as prompt sentence

[1326] Step 2:

[1327] The user's device sends the entered text data to a specified endpoint on the server, which runs on the Flask framework and receives it as an HTTP request.

[1328] Input: Text data as a prompt

[1329] Output: HTTP request sent to the server

[1330] Step 3:

[1331] The server retrieves the received text data and loads the generative AI model "t5-small." This is where the analysis of the text data begins. The server performs data analysis to detect ambiguous expressions within the text data.

[1332] Input: Text data sent to the server

[1333] Output: Analysis results including ambiguous expressions

[1334] Step 4:

[1335] Using the generative AI model "t5-small," the server generates multiple interpretations for the detected ambiguous expressions. The model takes context into account and performs data calculations to generate multiple different interpretations.

[1336] Input: Analysis results containing ambiguous expressions

[1337] Output: Multiple interpretation results

[1338] Step 5:

[1339] The server sends the generated interpretation results to the device, where they are formatted for display on the front end.

[1340] Input: Multiple interpretation results

[1341] Output: Data sent to the front end

[1342] Step 6:

[1343] The user's device displays the received analysis results in a React-based interface, allowing the user to check multiple interpretations and choose the appropriate one.

[1344] Input: Data sent to the front end

[1345] Output: Multiple interpretation results displayed to the user

[1346] Step 7:

[1347] Users can check the analysis results and use them as reference information to prevent misunderstandings. Users can then use this information to decide on the next action to take in order to make appropriate decisions.

[1348] Input: Multiple interpretation results displayed to the user

[1349] Output: User decision

[1350] Through the above processing steps, ambiguous expressions in text data are analyzed and multiple interpretations are presented, enabling users to communicate accurately.

[1351] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1352] The present invention relates to a system that analyzes text data using a generated artificial intelligence model, detects ambiguous expressions in the text data, and presents multiple interpretations to the user. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, the provided interpretations are adapted to the user's emotional state.

[1353] Hereinafter, an embodiment of the present invention will be described.

[1354] First, the server loads the necessary generative AI model and emotion engine. The generative AI model is based on natural language processing and analyzes text data and detects ambiguous expressions. The emotion engine is used to extract user emotions from text data. This enables both text analysis and emotion recognition.

[1355] A user inputs text data from a terminal. For example, the user may input the text "I didn't agree with him at the meeting yesterday, but I agree with his opinion." This text may be interpreted differently depending on the context and emotion.

[1356] The server then receives the text data entered by the user. This text data is then subjected to analysis. The server first analyzes the text data to identify ambiguous expressions. A generative AI model is used in the analysis, taking into account contextual information and punctuation placement to detect expressions that may be misleading.

[1357] Next, the server uses an emotion engine to recognize the user's emotions. Specifically, it extracts the user's emotional state from the input text data, such as positive, negative, or neutral emotions.

[1358] The server generates multiple interpretations for the detected ambiguous expressions, customizing the interpretations based on the user's emotional state. For example, the server generates the following interpretations for the above text:

[1359] 1. "I didn't agree with the whole meeting yesterday, but I might agree with some of his points." (Users express positive sentiment)

[1360] 2. "I don't agree with his opinion in general, but I might agree with him in certain circumstances." (When the user is expressing neutral or negative sentiment)

[1361] The server then sends the generated interpretations to the user's device, which then displays them to the user. By referring to these interpretations, the user can understand the ambiguous expressions in the text data and take measures to avoid misunderstandings.

[1362] In this way, the present invention provides a system that detects ambiguous expressions in text-based communication and presents their interpretation according to the emotional state, thereby enabling clearer and more accurate communication. By using this system, users can communicate more accurately and effectively.

[1363] The processing flow will be explained below.

[1364] Step 1:

[1365] The server loads the generative AI model and emotion engine. Specifically, it prepares the Transformer model "t5-small" based on natural language processing and the emotion engine for recognizing user emotions.

[1366] Step 2:

[1367] A user inputs text data through a terminal. For example, the user inputs "I didn't agree with him at the meeting yesterday, but I agree with his opinion."

[1368] Step 3:

[1369] The server receives text data entered by the user, which is then analyzed.

[1370] Step 4:

[1371] The server uses the generated AI model to analyze the input text data. Specifically, it performs an analysis process to detect ambiguous expressions within the text data.

[1372] Step 5:

[1373] The server uses an emotion engine to recognize the user's emotional state and classify the user's emotions into positive, negative, neutral, etc. based on the input text data.

[1374] Step 6:

[1375] The server generates multiple interpretations for detected ambiguous expressions, taking into account the user's emotions. For example, it provides different interpretations for positive and negative emotions.

[1376] Step 7:

[1377] The server transmits the generated interpretations to the user's terminal.

[1378] Step 8:

[1379] The terminal displays the received interpretation to the user, who then checks the displayed interpretation and understands the ambiguous expressions in the text data.

[1380] Step 9:

[1381] The user can refer to the proposed interpretation and correct or add to the text data as necessary. This process allows the user to choose appropriate expressions to avoid misunderstandings.

[1382] Example 2

[1383] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1384] Conventional text analysis systems provide a uniform interpretation of ambiguous expressions, making it difficult to provide flexible interpretations that reflect the emotional state of each individual user. Furthermore, the lack of emotion recognition functionality meant that interpretations that took the user's emotional state into account were insufficient. As a result, users were more likely to misunderstand the true meaning of the text data, hindering smooth communication.

[1385] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1386] In this invention, the server includes means for analyzing text data using the generated artificial intelligence model and detecting ambiguous expressions in the text data, means for providing multiple interpretations for the detected ambiguous expressions, means for recognizing a user's emotion, means for customizing an interpretation based on the recognized emotional state, and means for presenting the provided multiple interpretations to the user. This makes it possible to provide multiple interpretations according to the user's emotional state and reduce misunderstandings of ambiguous expressions in the text data.

[1387] A "generated artificial intelligence model" is a model that has been pre-trained to analyze text data and detect ambiguous expressions, and that uses natural language processing techniques.

[1388] "Text data" refers to character information such as sentences, documents, and chat messages entered by the user.

[1389] An "ambiguous expression" is a word or phrase that can have different meanings within a text depending on the context.

[1390] "Multiple interpretations" refers to the multiple possible meanings or connotations of an ambiguous expression.

[1391] "Means for recognizing user emotions" refers to technology or algorithms for determining a user's emotional state from text data, such as technology for extracting positive, negative, or neutral emotional states.

[1392] "Means for customizing interpretation based on emotional state" refers to a technique or algorithm that adjusts the interpretation of an ambiguous expression based on the perceived emotion of the user.

[1393] The "means for presenting the provided interpretations to the user" refers to a technique or interface for displaying the generated interpretations on the user's terminal.

[1394] The present invention is a system that uses the generated AI model and emotion engine to analyze text data and recognize user emotions, detect ambiguous expressions in the text, and provide multiple interpretations for them. This system is specifically implemented by a series of processes with the server, terminal, and user as subjects.

[1395] First, the server loads the generative AI model (e.g., OpenAI GPT-3) and emotion engine (e.g., IBM Watson API), along with any necessary software libraries and configuration files, so that the system is ready for text analysis and emotion recognition.

[1396] Next, the user inputs text data from the terminal. For example, the text could be something like, "I didn't agree with him at the meeting yesterday, but I agree with his opinion." The terminal sends the input data from the user to the server as an HTTP request.

[1397] The server receives the text data sent from the device and passes it to the analysis module, which uses a generative AI model to detect ambiguous expressions in the input text data. The model takes into account contextual information and punctuation placement to identify expressions that may be misleading.

[1398] Furthermore, the server uses an emotion engine to recognize the user's emotion from the text data. The emotion engine distinguishes between positive, negative, and neutral emotional states. For example, if a user sends the text "I didn't agree with him in the meeting yesterday, but I agree with his opinion," the emotion engine recognizes the emotional state as "neutral."

[1399] The server generates multiple interpretations for the detected ambiguous expressions, taking into account the user's emotional state. For example, if the user expresses a positive emotion, the expression can be interpreted as "I didn't agree with the entire meeting yesterday, but I might agree with some of his opinions." If the user expresses a negative or neutral emotion, the expression can be interpreted as "I don't agree with his opinions overall, but I might agree with his opinions under certain circumstances."

[1400] Finally, the server sends the generated interpretations to the user's device, which displays them to the user. By referring to these interpretations, the user can more accurately understand the true meaning of the text data.

[1401] For example, the prompt the user enters is:

[1402] Generate an interpretation for the text "I disagreed with him in the meeting yesterday, but I agree with his opinion."

[1403] This system allows users to adapt the results of analysis of text data containing ambiguous expressions to their emotional state, thereby reducing misunderstandings.

[1404] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1405] System program processing flow

[1406] Step 1: The server loads the generative AI model and emotion engine

[1407] The server first loads the generative AI model and emotion engine. Specifically, it loads the libraries and configuration files required to use the generative AI model (e.g., OpenAI GPT-3) and emotion engine (e.g., IBM Watson API).

[1408] Input: Server configuration file, API key, etc.

[1409] Data processing: Initializing the library and loading the model

[1410] Output: AI model and emotion engine ready state

[1411] Specific behavior:

[1412] The server reads the configuration file and loads the API key and model paths into memory.

[1413] Initialize the generative AI model and emotion engine.

[1414] Step 2: The user inputs text data from the terminal.

[1415] The user inputs text data to be analyzed into an input field on the terminal, and this data is sent to the server.

[1416] Input: Text data that the user types into the terminal.

[1417] Data processing: None

[1418] Output: HTTP request sent from the terminal to the server

[1419] Specific behavior:

[1420] A user enters text into an input field on a web form or application and presses the submit button.

[1421] The terminal generates an HTTP request and sends the input text data to the server.

[1422] Step 3: The server receives the input text data

[1423] The server receives the text data sent from the terminal and prepares to pass it on to the next analysis step.

[1424] Input: Text data sent via the HTTP request

[1425] Data processing: Parsing HTTP requests

[1426] Output: Text data passed to the analysis module

[1427] Specific behavior:

[1428] The server receives the HTTP request and extracts the text data from it.

[1429] The extracted text data is passed to the analysis module.

[1430] Step 4: The server analyzes the text data and identifies ambiguous expressions.

[1431] The server uses a generative AI model to analyze the received text data and identify ambiguous expressions.

[1432] Input: Text data passed to the analysis module

[1433] Data Processing: Text Analysis with Generative AI Models

[1434] Output: Identification of ambiguous expressions

[1435] Specific behavior:

[1436] The server sends the text data to the generative AI model and performs the analysis.

[1437] The model takes into account contextual information and punctuation placement and returns results that detect ambiguous expressions.

[1438] Step 5: The server recognizes the user's emotion using the emotion engine.

[1439] The server uses an emotion engine to extract the user's emotion from the text data.

[1440] Input: Text data and results of identifying ambiguous expressions

[1441] Data processing: Emotion analysis using an emotion engine

[1442] Output: User's emotional state (positive, negative, neutral, etc.)

[1443] Specific behavior:

[1444] The server passes the text data to the emotion engine and performs emotion analysis.

[1445] The engine analyzes the emotional state in the text and returns the results.

[1446] Step 6: The server generates multiple interpretations and customizes them based on emotional state.

[1447] The server generates multiple interpretations based on the generative AI model and emotion recognition results, and customizes them according to the user's emotional state.

[1448] Input: Identification results of ambiguous expressions, user's emotional state

[1449] Data processing: interpretation generation and emotional state-based customization

[1450] Output: Multiple customized interpretations

[1451] Specific behavior:

[1452] The server uses a generative AI model to generate an interpretation for an ambiguous expression.

[1453] Based on the acquired emotional data, each interpretation is adapted to the emotion.

[1454] Step 7: The server sends the generated interpretation to the terminal.

[1455] The server formats the generated interpretations as an HTTP response and sends it to the user's device.

[1456] Input: Customized multiple interpretations

[1457] Data processing: Generating HTTP responses

[1458] Output: The interpreted response sent to the device

[1459] Specific behavior:

[1460] The server formats the generated interpretation into an HTTP response.

[1461] Send the interpreted response to the terminal.

[1462] Step 8: The terminal displays the interpretation to the user

[1463] The terminal displays the received interpretation to the user, who then checks the interpretation and understands the true meaning of the text data.

[1464] Input: The interpreted response sent by the server

[1465] Data processing: Display of interpretations

[1466] Output: The interpretation displayed to the user

[1467] Specific behavior:

[1468] The device receives the response from the server and updates the GUI to display the interpretation.

[1469] Provide an interface that allows users to view the interpretation.

[1470] The above are the specific processing steps of this system.

[1471] (Application example 2)

[1472] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1473] In text-based communication, ambiguous expressions in the text data entered by users can lead to misunderstandings and unpleasant interactions. In addition, the quality of communication can be reduced due to a lack of appropriate interpretation according to the user's emotional state.

[1474] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1475] In this invention, the server includes means for analyzing text data using the generated artificial intelligence model and detecting ambiguous expressions in the text data, means for providing multiple interpretations for the detected ambiguous expressions, means for presenting the provided multiple interpretations to the user, means for analyzing the user's emotions, and means for generating interpretations based on the user's emotional state. This makes it possible to accurately detect ambiguous expressions in the text data and provide interpretations that correspond to the user's emotional state, thereby reducing misunderstandings and unpleasant interactions and improving the quality of communication.

[1476] "Generated artificial intelligence model" refers to the artificial intelligence technology used to analyze text data, detect ambiguous expressions, and generate appropriate interpretations.

[1477] "Text data" refers to data expressed as character information, such as sentences or comments entered by a user.

[1478] An "ambiguous expression" refers to an expression that has different meanings depending on the context and interpretation, and may lead to misunderstanding.

[1479] "Interpretation" refers to specific explanations or proposed explanations to understand the meaning and intent of text data.

[1480] "User" refers to the entity that uses the system to input text data and have it interpreted.

[1481] "Emotion" refers to the emotional state, such as positive, negative, or neutral, contained in the text data.

[1482] "Analysis" refers to the process of analyzing text data and extracting information such as meaning and sentiment.

[1483] A "server" is the central computer of the system, and refers to the device that analyzes text data and generates interpretations.

[1484] The present invention relates to a system that uses a generated artificial intelligence model to analyze text data, detect ambiguous expressions, and present multiple interpretations to the user. Furthermore, it combines an emotion engine that recognizes the user's emotions so that the interpretations provided adapt to the user's emotional state.

[1485] First, the server loads the generated artificial intelligence model and emotion engine. The generative AI model is based on natural language processing and is used to analyze text data and detect ambiguous expressions. The emotion engine has the function of extracting user emotions from text data. This allows both text analysis and emotion recognition to be performed simultaneously.

[1486] A user inputs text data from a terminal. For example, the user may input the text, "This movie exceeded my expectations, but the ending was disappointing." This text may be interpreted differently depending on the context and emotions.

[1487] The server then receives the text data entered by the user. This text data is then analyzed. First, a generative AI model is used to analyze the text data and identify ambiguous expressions. This analysis takes into account contextual information and punctuation placement to detect expressions that may be misleading.

[1488] Next, the server uses an emotion engine to recognize the user's emotion. Specifically, it determines the user's emotional state (positive, negative, neutral, etc.) from the input text data. This emotion recognition is performed using a natural language processing library such as TextBlob.

[1489] The server then generates multiple interpretations for the detected ambiguous expressions, customizing the interpretations based on the user's emotional state. For example, the server generates the following interpretations for the above text:

[1490] 1. "The movie was good overall, but the ending was disappointing." (Users express positive emotions)

[1491] 2. "The ending of the movie was disappointing, so I didn't enjoy it overall." (Users express negative emotions)

[1492] The server sends the generated interpretations to the user's device, which then displays them to the user, allowing the user to refer to these interpretations, understand the ambiguous expressions in the text data, and take measures to avoid misunderstandings.

[1493] Example prompt sentence:

[1494] "Provide a positive interpretation of the following ambiguous text: This movie exceeded my expectations, but the ending disappointed me."

[1495] "Provide a neutral / negative interpretation of the following ambiguous text: This movie exceeded my expectations, but the ending disappointed me."

[1496] In this way, the present invention provides a system that detects ambiguous expressions in text-based communication and presents their interpretation according to the emotional state, thereby enabling clearer and more accurate communication. By using this system, users can communicate more accurately and effectively.

[1497] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1498] Step 1:

[1499] The user inputs text data using the terminal. At this time, the text data input by the user is sent to the system. The input data is in text format, and for example, a sentence such as "This movie exceeded my expectations, but the ending was disappointing" is input.

[1500] Step 2:

[1501] The server uses the generated artificial intelligence model to analyze the text data received from the user. First, it analyzes the text data based on natural language processing techniques to detect ambiguous expressions. It receives the text data as input and generates a list of ambiguous expressions as output.

[1502] Step 3:

[1503] The server uses an emotion engine to determine the user's emotional state based on the detected ambiguous expressions. Specifically, the emotion engine analyzes the text data and detects emotional states such as positive, negative, and neutral. It receives the initial text data as input and generates an emotional state as output.

[1504] Step 4:

[1505] The server generates multiple interpretations based on the detected ambiguous expressions and emotional states. In this step, a generative AI model is used, and a prompt sentence such as "Provide a positive interpretation of the following ambiguous text: This movie exceeded my expectations, but the ending was disappointing" is sent to the model. The server receives the ambiguous expressions and emotional states as input and generates a list of customized interpretations as output.

[1506] Step 5:

[1507] The server sends the generated interpretations to the user's device, which then displays them to the user. The interpretations are presented in a format that is easy for the user to understand, and the user can use these interpretations to easily understand ambiguous expressions.

[1508] Step 6:

[1509] By referring to the presented interpretation, users can take measures to avoid misunderstandings. This reduces misunderstandings of ambiguous expressions in the text and improves the quality of communication. The user's actions and choices are the final output.

[1510] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1511] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1512] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1513] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1514] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1515] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1516] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1517] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1518] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1519] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1520] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1521] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1522] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1523] 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.

[1524] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1525] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1526] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1527] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1528] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1529] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1530] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1531] The following is further disclosed regarding the above embodiment.

[1532] (Claim 1)

[1533] The generated artificial intelligence model is used to analyze the text data.

[1534] means for detecting ambiguous expressions in text data;

[1535] means for providing multiple interpretations of the detected ambiguous expression;

[1536] means for presenting the provided interpretations to a user;

[1537] A system including:

[1538] (Claim 2)

[1539] 10. The system of claim 1, further comprising: means for receiving text data from a user.

[1540] (Claim 3)

[1541] 10. The system of claim 1, wherein the generated artificial intelligence model includes means for detecting ambiguous expressions based on natural language processing.

[1542] "Example 1"

[1543] (Claim 1)

[1544] The generated artificial intelligence model is used to analyze the text data.

[1545] means for detecting ambiguous expressions in text data;

[1546] means for providing multiple interpretations of the detected ambiguous expression;

[1547] means for presenting the provided interpretations to a user;

[1548] means for loading the generated artificial intelligence model as a Transformer model;

[1549] means for receiving user input and transmitting it to a server;

[1550] means for receiving the analysis results from the server and displaying them to the user;

[1551] A system including:

[1552] (Claim 2)

[1553] 10. The system of claim 1, further comprising: means for receiving text data from a user.

[1554] (Claim 3)

[1555] 10. The system of claim 1, wherein the generated artificial intelligence model includes means for detecting ambiguous expressions based on natural language processing.

[1556] "Application Example 1"

[1557] (Claim 1)

[1558] a means for analyzing text data using the generated artificial intelligence model and detecting ambiguous expressions;

[1559] means for providing multiple interpretations of the detected ambiguous expression;

[1560] means for presenting the provided interpretations to a user;

[1561] In a security service, a means for analyzing the content of a message to detect ambiguous expressions and generate reinterpretations;

[1562] means for displaying the reinterpreted results on the user's device;

[1563] A system including:

[1564] (Claim 2)

[1565] 10. The system of claim 1, further comprising: means for receiving text data from a user.

[1566] (Claim 3)

[1567] 10. The system of claim 1, wherein the generated artificial intelligence model includes means for detecting ambiguous expressions based on natural language processing.

[1568] "Example 2: Combining Emotion Engines"

[1569] (Claim 1)

[1570] The generated artificial intelligence model is used to analyze the text data.

[1571] means for detecting ambiguous expressions in text data;

[1572] means for providing multiple interpretations of the detected ambiguous expression;

[1573] means for recognizing a user's emotion;

[1574] a means of customizing interpretations based on perceived emotional states;

[1575] means for presenting the provided interpretations to a user;

[1576] A system including:

[1577] (Claim 2)

[1578] 10. The system of claim 1, further comprising: means for receiving text data from a user.

[1579] (Claim 3)

[1580] 10. The system of claim 1, wherein the generated artificial intelligence model includes means for detecting ambiguous expressions based on natural language processing.

[1581] "Application example 2 when combining emotion engines"

[1582] (Claim 1)

[1583] The generated artificial intelligence model is used to analyze the text data.

[1584] means for detecting ambiguous expressions in text data;

[1585] means for providing multiple interpretations of the detected ambiguous expression;

[1586] means for presenting the provided interpretations to a user;

[1587] means for analyzing user emotions;

[1588] means for generating an interpretation based on the emotional state;

[1589] A system including:

[1590] (Claim 2)

[1591] 10. The system of claim 1, further comprising: means for receiving text data from a user.

[1592] (Claim 3)

[1593] 10. The system of claim 1, wherein the generated artificial intelligence model includes means for detecting ambiguous expressions based on natural language processing. [Explanation of symbols]

[1594] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. The generated artificial intelligence model is used to analyze the text data. means for detecting ambiguous expressions in text data; a means for providing multiple interpretations of the detected ambiguous expression; means for presenting the provided interpretations to a user; A system including:

2. The system of claim 1 further comprising means for receiving text data from a user.

3. The system of claim 1 , wherein the generated artificial intelligence model includes means for detecting ambiguous expressions based on natural language processing.

Images