system

Abstract

We provide the system. [Solution] A natural language processing means for translating input text from one language to another, A means of tokenizing and decoding text using a translation model, An output means for providing the translation result to the user, A system that includes this.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance as a response to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In translation, inaccurate information, future prediction, and emotion processing, the following problems exist in the current technology. First, conventional translation systems have difficulty fully considering cultural backgrounds and nuances, and are prone to translation errors and misinterpretations. Second, it is difficult to quickly obtain reliable information on specific topics, increasing the risk of misinformation spreading. Furthermore, when making future predictions based on past data, there are problems in selecting appropriate algorithms and processing data. Finally, there is a lack of effective means for analyzing people's emotions in real time and promoting peaceful dialogue. Solving these problems is an object of the present invention.

Means for Solving the Problems

[0005] The present invention solves the above problems by providing the following means. First, it prevents mistranslation between different languages ​​by using translation means based on natural language processing (NLP) to accurately tokenize and decode text. Second, it prevents the spread of misinformation by providing a search engine interface means and analysis means for inputting search queries on a specific topic and filtering reliable information sources. Furthermore, it provides data processing means and prediction means for analyzing past time-series data using machine learning models and predicting future trends with high accuracy. Finally, it solves the problem of emotional dialogue in real time by analyzing the content of social media posts with sentiment analysis model means and incorporating recommendation action generation means to promote peaceful dialogue based on the results. As a result, users can quickly obtain reliable information, enabling accurate translation, future predictions, and the promotion of peaceful dialogue.

[0006] "Natural language processing means" refers to technologies that analyze input text, perform tokenization and decoding, and then translate it into another language.

[0007] A "translation model" refers to an algorithm pre-trained for a specific language pair, designed to convert input text into an appropriate format and generate a translation result.

[0008] "Tokenization" refers to the process of breaking down input text into semantic units such as words and phrases.

[0009] "Decoding" refers to the process of converting tokenized data into natural language text that humans can understand.

[0010] "Output means" refers to functions that provide the user with translation and analysis results generated by the system.

[0011] "Search engine interface means" refers to a function that sends search queries entered by users to search engines on the internet and retrieves search results.

[0012] "Analysis method" refers to the process of analyzing results obtained from search engines and filtering out reliable information sources.

[0013] A "machine learning model" refers to an algorithm trained on a large amount of data, used to predict future trends from past data.

[0014] "Data processing means" refers to the process of dividing input data into training data and test data, and organizing it into a format suitable for machine learning models.

[0015] "Predictive means" refers to the ability to predict future data using trained machine learning models.

[0016] "Sentiment analysis model means" refers to algorithms and techniques used to analyze the emotional state of text data.

[0017] "Text processing means" refers to a function for converting text into a format suitable for sentiment analysis models.

[0018] The "recommended action generation method" refers to a function that proposes specific actions to promote peaceful dialogue based on the results of sentiment analysis. [Brief explanation of the drawing]

[0019] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5]It is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] It is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] It is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] It is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] It shows an emotion map to which a plurality of emotions are mapped. [Figure 10] It shows an emotion map to which a plurality of emotions are mapped. [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Example 2 when an emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when an emotion engine is combined.

Mode for Carrying Out the Invention

[0020] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0021] First, the terms used in the following description will be explained.

[0022] In the following embodiments, the signed processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Furthermore, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include CPU (Central Processing Unit), GPU (Graphics Processing Unit), GPGPU (General-Purpose computing on Graphics Processing Units), and APU (Accelerated Processing Unit).

[0023] In the following embodiments, signed RAM (Random Access Memory) is a memory that temporarily stores information and is used as work memory by the processor.

[0024] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0025] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0026] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0027] [First Embodiment]

[0028] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0029] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0030] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0031] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0032] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0033] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0034] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0035] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0036] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0037] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0038] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0039] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0040] This invention relates to a system for performing translation, filtering out inaccurate information, predicting the future, and processing emotions. Specific embodiments thereof are described below.

[0041] Translation Function Embodiment

[0042] Program processing

[0043] 1. The user enters the text they want to translate, along with the source language and the target language.

[0044] 2. The terminal receives the input information and analyzes the text using natural language processing tools.

[0045] 3. Use a translation model to tokenize the text and perform the translation.

[0046] 4. The terminal decodes the translation result and converts it into a format that humans can understand.

[0047] 5. The server provides the user with the final translation result.

[0048] Specific example

[0049] For example, if a user wants to translate the Japanese text "こんにちは" into English, the system receives "こんにちは," translates it to "Hello" based on its translation model, and provides it to the user.

[0050] Implementation of sharing correct knowledge

[0051] Program processing

[0052] 1. The user enters a question about a specific topic.

[0053] 2. The server uses a search engine interface to obtain relevant information from the internet.

[0054] 3. The server performs analysis to filter reliable information from the acquired information sources.

[0055] 4. The device provides users with filtered, reliable information.

[0056] Specific example

[0057] For example, when searching for information about "global warming," the system retrieves links to Wikipedia and government-related websites and provides them to the user.

[0058] Future predicted implementation

[0059] Program processing

[0060] 1. The user inputs historical time-series data.

[0061] 2. The server uses data processing equipment to split the input data into training data and test data.

[0062] 3. Train a machine learning model on the data and create a predictive model.

[0063] 4. The server uses the created model to predict future data.

[0064] 5. The device provides the user with the prediction results.

[0065] Specific example

[0066] For example, when a user inputs past sales data, the system uses that data to predict future sales and provides the results to the user.

[0067] A Embodiment of World Peace

[0068] Program processing

[0069] 1. The user enters the content of their social media post.

[0070] 2. The terminal uses an emotion analysis model to analyze the emotion of the input text.

[0071] 3. The terminal uses a recommended action generation mechanism based on the sentiment analysis results to create suggestions to promote peaceful dialogue.

[0072] 4. The server provides the user with the recommended actions that have been created.

[0073] Specific example

[0074] For example, in the case of a positive post like "I love this!" and a negative post like "This is so frustrating!", the system analyzes the emotions in each post, determines whether they are "positive" or "negative," and suggests actions to the user to facilitate appropriate dialogue.

[0075] These embodiments enable users to achieve highly accurate translation, access to reliable information, predict the future, and engage in peaceful, emotion-based dialogue.

[0076] The following describes the processing flow.

[0077] Translation Function Embodiment

[0078] Step 1:

[0079] The user enters the text to be translated, the source language, and the target language.

[0080] Step 2:

[0081] The terminal receives input from the user and activates natural language processing for translation.

[0082] Step 3:

[0083] The terminal tokenizes the source text and converts it into a format compatible with the language model.

[0084] Step 4:

[0085] The terminal inputs tokenized text into the translation model and performs the translation.

[0086] Step 5:

[0087] The device decodes the translation result and converts it into a natural language format that the user can understand.

[0088] Step 6:

[0089] The server provides the user with the final translation result.

[0090] Implementation of sharing correct knowledge

[0091] Step 1:

[0092] The user enters the topic they want to investigate as a query.

[0093] Step 2:

[0094] The server receives the input query and collects relevant information from the internet through the search engine interface.

[0095] Step 3:

[0096] The server filters the collected search results to identify reliable sources and extracts specific URLs.

[0097] Step 4:

[0098] The device provides users with reliable sources of information and encourages them to obtain additional information as needed.

[0099] Future predicted implementation

[0100] Step 1:

[0101] The user inputs historical time-series data.

[0102] Step 2:

[0103] The server receives the input data, preprocesses it, and splits it into training data and test data.

[0104] Step 3:

[0105] The server uses a machine learning model to create a predictive model based on the training data.

[0106] Step 4:

[0107] The server uses the created predictive model to predict future data based on the test data.

[0108] Step 5:

[0109] The device provides the user with prediction results and suggests necessary actions.

[0110] A Embodiment of World Peace

[0111] Step 1:

[0112] The user enters the content of their social media post.

[0113] Step 2:

[0114] The terminal receives the entered text and activates the sentiment analysis model.

[0115] Step 3:

[0116] The device uses an emotion analysis model to analyze the emotional state of the text.

[0117] Step 4:

[0118] Based on the results of the sentiment analysis, the device creates recommended actions to promote peaceful dialogue.

[0119] Step 5:

[0120] The server provides the user with the generated recommended actions.

[0121] This allows users to effectively translate text, obtain reliable information, predict the future, and engage in emotion-based conversations.

[0122] (Example 1)

[0123] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0124] In recent years, the importance of global communication has been steadily increasing, but handling multiple different languages ​​requires advanced translation technology. Furthermore, there is a need for systems that can quickly obtain reliable information, predict future trends, and facilitate emotion-based dialogue. To meet these needs, systems that efficiently combine highly accurate natural language processing technologies and machine learning models are essential.

[0125] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0126] In this invention, the server includes natural language processing means for translating input text from one language to another; processing means for tokenizing and decoding the text using a machine learning translation model; output means for providing the translation results to the user; search engine interface means for inputting search queries on a specific topic and obtaining relevant information; reliability evaluation means for filtering reliable information from the obtained information sources; output means for providing the filtered information to the user; machine learning prediction model means for predicting future data based on past time series data; data processing means for splitting input data into training data and test data; prediction means for predicting future data using the trained model; output means for providing the prediction results to the user; sentiment analysis means for inputting social media posts and analyzing the sentiment of the text; recommendation action generation means for promoting peaceful dialogue based on the sentiment analysis results; and output means for providing the generated recommendation actions to the user. This enables highly accurate translation, acquisition of reliable information, future prediction, and peaceful dialogue based on sentiment.

[0127] "Natural language processing means" refers to technologies that analyze input text and enable operations such as translation and sentiment analysis.

[0128] A "machine learning translation model" is a model trained using a large amount of data, and it has the ability to translate text from one language to another.

[0129] "Processing means" refers to an algorithm or program for performing a specific task, and in this invention, it performs text tokenization and decoding.

[0130] "Output means" refers to interfaces or devices for providing processing results to the user, and have functions to display translation results, search results, prediction results, etc.

[0131] A "search engine interface means" is a means of linking with a search engine for obtaining information on the internet.

[0132] A "reliability evaluation method" is a system that executes algorithms and evaluation criteria to select highly reliable information from acquired information sources.

[0133] A "machine learning prediction model" is a prediction model trained using past data, which is used to predict future data.

[0134] "Data processing means" refers to means for dividing input data into training data and test data.

[0135] A "predictive tool" is a means of using a trained model to predict future data.

[0136] "Sentiment analysis techniques" are technologies used to analyze the emotions in text and determine whether they are positive or negative.

[0137] A "recommended action generation method" is a means of generating actions and suggestions to promote peaceful dialogue based on the results of sentiment analysis.

[0138] This invention relates to a system that utilizes natural language processing technology and machine learning models to achieve translation functionality, reliable information acquisition, future prediction, and sentiment analysis and peaceful dialogue on social media. Specific embodiments thereof are described below.

[0139] Translation function

[0140] Specific Embodiments

[0141] 1. The user inputs text through a translation interface and specifies the source and target languages. This interface is implemented as a web browser or mobile application.

[0142] 2. The terminal uses the Python NLTK library to parse and tokenize the input text.

[0143] 3. The device calls the Google® Translate API or Microsoft® Translator to translate the tokenized text.

[0144] 4. The device decodes the translation result and converts it into a format that the user can understand.

[0145] 5. The server displays the final translation result on the interface and provides it to the user.

[0146] Specific example

[0147] If a user wants to translate the Japanese text "こんにちは" into English, the system receives "こんにちは," uses the Google Translate API to translate it to "Hello," and provides the result to the user.

[0148] Sharing correct knowledge

[0149] Specific Embodiments

[0150] 1. The user enters a question about a specific topic. This question is entered via a web browser or mobile application.

[0151] 2. The server uses the Google Search API and other search engine interfaces to retrieve relevant information from the internet.

[0152] 3. The server runs a reliability evaluation algorithm and filters out reliable information from the retrieved sources.

[0153] 4. The device provides filtered information to the user.

[0154] Specific example

[0155] When searching for information related to "global warming," the system uses the Google Search API to retrieve relevant information, selects the most reliable information, and provides it to the user.

[0156] Future predictions

[0157] Specific Embodiments

[0158] 1. Users upload historical time-series data in CSV files or other digital formats.

[0159] 2. The server uses the Python pandas library to split the input data into training and test data.

[0160] 3. The server trains data and creates a predictive model using machine learning frameworks such as TENSORFLOW® or Scikit-learn.

[0161] 4. The server uses the trained model to predict future data.

[0162] 5. The device provides the user with prediction results in graph or table format.

[0163] Specific example

[0164] When a user inputs past sales data, the system uses a TensorFlow LSTM model based on that data to predict future sales and provides the results to the user.

[0165] World peace

[0166] Specific Embodiments

[0167] 1. The user enters the content of their social media post in text format.

[0168] 2. The terminal uses IBM Watson® or Microsoft Azure® sentiment analysis APIs to analyze the sentiment of the input text.

[0169] 3. Based on the sentiment analysis results, the device uses a recommended action generation algorithm to create suggestions to promote peaceful dialogue.

[0170] 4. The server provides the user with the recommended actions that have been created.

[0171] Specific example

[0172] If a negative post such as "This is so frustrating!" is submitted, IBM Watson will be used to determine that the sentiment is "negative" and generate and provide suggestions to the user to facilitate appropriate dialogue.

[0173] The above describes embodiments of the present invention. This system enables users to achieve highly accurate translation, obtain reliable information, predict the future, and engage in peaceful, emotion-based dialogue.

[0174] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0175] Translation Function Embodiment

[0176] Processing steps

[0177] Step 1:

[0178] The user accesses the system interface and enters the text to be translated, along with the source and target languages. For example, they might enter "こんにちは" (konnichiwa) and specify a translation from Japanese to English. This input is sent to the server in text format.

[0179] Step 2:

[0180] The terminal parses and tokenizes the input text received from the server using Python's NLTK library. The input data is "hello," and through parsing and tokenization, the internal structure of the text is understood. The output is the tokenized text.

[0181] Step 3:

[0182] The device sends tokenized text to the Google Translate API or Microsoft Translator for translation. The input is the tokenized text "こんにちは" (Konnichiwa), and the translation is performed by calling the API. The output is the translated text "Hello".

[0183] Step 4:

[0184] The terminal decodes the translation result returned from the API and converts it into a human-readable format. The input is the translated text "Hello," and the terminal performs the conversion from the encoded format to standard text format. The output is the decoded text "Hello."

[0185] Step 5:

[0186] The server provides the user with the final translation result. The input is the decoded text "Hello," which is returned to the interface for displaying it to the user. The output is the translation result displayed to the user.

[0187] Implementation of sharing correct knowledge

[0188] Processing steps

[0189] Step 1:

[0190] The user enters a question about a specific topic into the system. For example, they might enter, "Tell me about global warming." This input is sent to the server in text format.

[0191] Step 2:

[0192] The server uses the Google Search API to search for relevant information on the internet. The input is the question "Tell me about global warming," and the API is called to retrieve relevant information. The output is the retrieved search results.

[0193] Step 3:

[0194] The server executes a reliability evaluation algorithm to filter the retrieved information for the most reliable information. The input is the retrieved search results, to which the algorithm is applied for reliability evaluation. The output is the result of selecting only the most reliable information.

[0195] Step 4:

[0196] The terminal provides the user with filtered, reliable information. The input is reliable information, which is returned to the interface for displaying it to the user. The output is the reliable information displayed to the user.

[0197] Future prediction implementation

[0198] Processing steps

[0199] Step 1:

[0200] Users upload historical time-series data to the system as a CSV file. For example, they might upload sales data for the past year. This input is sent to the server in file format.

[0201] Step 2:

[0202] The server uses the Python pandas library to split the input data into training and test sets. The input is an uploaded CSV file, which is then divided into 80% training data and 20% test data. The output is the split dataset.

[0203] Step 3:

[0204] The server uses machine learning frameworks such as TensorFlow and Scikit-learn to train data and create predictive models. The input is training data, and the machine learning algorithm generates the predictive model. The output is the trained predictive model.

[0205] Step 4:

[0206] The server evaluates the performance of a model created using test data and predicts future data. The inputs are the test data and the predictive model; performance evaluation and future prediction are performed. The output is the predicted data.

[0207] Step 5:

[0208] The terminal provides prediction results to the user in graph or table format. The input is the predicted data, which is visualized and displayed on the interface. The output is the prediction result displayed to the user.

[0209] A Embodiment of World Peace

[0210] Processing steps

[0211] Step 1:

[0212] The user enters their social media post content into the system in text format. For example, they might enter "This is so frustrating!". This input is then sent to the server in text format.

[0213] Step 2:

[0214] The terminal uses IBM Watson and Microsoft Azure sentiment analysis APIs to analyze the sentiment of the input text. The input is a social media post, "This is so frustrating!", and the sentiment is analyzed by calling the sentiment analysis API. The output is the analyzed sentiment result.

[0215] Step 3:

[0216] The device applies a recommended action generation algorithm to promote peaceful dialogue based on the sentiment analysis results. The input is the sentiment analysis results, and the algorithm is used to generate an appropriate response. The output is the recommended action.

[0217] Step 4:

[0218] The server provides the user with the generated recommended actions. The input is the recommended actions, which are returned to the interface for displaying them to the user. The output is the recommended actions displayed to the user.

[0219] These specific processing steps enable users to achieve highly accurate translations, obtain reliable information, predict the future, and engage in peaceful, emotion-based dialogue.

[0220] (Application Example 1)

[0221] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0222] In many modern situations, communication between people who speak different languages, facilitating emotion-based dialogue, obtaining reliable information, and predicting the future based on diverse data are critical challenges. However, current systems lack an integrated approach to meet these requirements, particularly a lack of dialogue support incorporating real-time sentiment analysis. This makes efficient communication and appropriate decision-making difficult.

[0223] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0224] In this invention, the server includes natural language processing means for translating input text from one language to another, means for tokenizing and decoding the text using a translation model, output means for providing the translation results to the user, sentiment analysis means for analyzing the sentiment of the input text, and dialogue support means for facilitating peaceful dialogue based on the analysis results. This enables real-time sentiment analysis and dialogue support.

[0225] A "natural language processing system" is a means of analyzing input text and translating it from one language to another.

[0226] A "translation model" is a model for tokenizing text and decoding it into another language.

[0227] "Output means" refers to the means of providing translation results or analysis results to the user.

[0228] An "emotion analysis tool" is a means of analyzing the emotions in input text and identifying the type of emotion.

[0229] "Dialogue support methods" are means of creating proposals to promote peaceful dialogue based on the results of emotion analysis.

[0230] A "search engine interface" is a means of obtaining reliable information sources by entering a search query related to a specific topic.

[0231] "Analysis means" are means for filtering reliable information from acquired information sources.

[0232] A "machine learning model" is a model used to predict future trends based on past time-series data.

[0233] "Data processing means" refers to means for dividing input data into training data and test data.

[0234] A "predictive tool" is a means of predicting future data using a trained model.

[0235] This invention can be specifically implemented as follows.

[0236] First, the system receives text input from the user. This input text could be content to be translated, content requiring sentiment analysis, or a search query. Once the user enters the text, the terminal analyzes it using natural language processing tools. At this stage, the main software used includes "Transformers," a widely known natural language processing library.

[0237] Next, if translation is needed, the system uses a translation model to tokenize the text and perform the translation. The translation result is decoded and provided to the user. Open-source translation models such as "Hugging Face" are used.

[0238] Furthermore, the system uses sentiment analysis tools to analyze the sentiment of the input text. Sentiment analysis assigns specific sentiment labels (e.g., positive, negative, neutral). Based on these analysis results, dialogue support tools generate suggestions to facilitate peaceful dialogue.

[0239] When a search query is entered, the server uses a search engine interface to obtain reliable information sources and filters them using analysis tools. The reliable information is then provided to the user.

[0240] Furthermore, the system can predict future trends based on historical time-series data. The server divides the input data into training data and test data, and trains the data using a machine learning model. This allows it to predict future data and provide results to the user.

[0241] As a concrete example, suppose a security staff member detected the following statement at the scene:

[0242] “This situation is incredibly frustrating and makes me angry.”

[0243] In this case, the system analyzes the text and detects the emotion "NEGATIVE." It then suggests the following recommended action:

[0244] "Please try to engage in calm dialogue to alleviate the situation."

[0245] Examples of prompt statements include the following:

[0246] "A proposed system for emotional analysis and peaceful dialogue"

[0247] Analyze user sentiment based on their input text and provide appropriate dialogue actions.

[0248] User input: "This situation is incredibly frustrating and makes me angry."

[0249] In this way, the system provides an integrated approach to emotion analysis and dialogue support, enabling it to respond appropriately in real time.

[0250] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0251] Step 1:

[0252] The user provides input text to the terminal. The input can be text to be translated, text requiring sentiment analysis, or a search query. The terminal analyzes this input text using natural language processing. The analyzed text is then sent to the next processing step.

[0253] Step 2:

[0254] The device uses a translation model to tokenize the input text if it requires translation. Tokenization is the process of dividing each word or phrase in the text into individual tokens. The tokens are then sent to the translation model.

[0255] Step 3:

[0256] The translation model translates tokenized text into the specified language. The translation model uses a generative AI model to convert the token array into tokens in the other language. The resulting token array is decoded and converted into human-readable text. The decoded text is returned to the terminal.

[0257] Step 4:

[0258] The terminal provides the translated text to the user through an output device. The user receives the final translation result.

[0259] Step 5:

[0260] The terminal analyzes the sentiment of the input text using sentiment analysis means if the sentiment of the text should be analyzed. The sentiment analysis means analyzes the text and assigns a specific sentiment label such as positive, negative, or neutral. The analysis results and sentiment labels are identified and sent to the next processing step.

[0261] Step 6:

[0262] The device uses dialogue support mechanisms based on the analyzed emotion results to generate suggestions for peaceful dialogue. These dialogue support mechanisms create messages to alleviate negative emotions and messages to maintain a positive atmosphere for positive emotions. These suggestions are then provided to the user.

[0263] Step 7:

[0264] When a user enters a search query, the server uses a search engine interface to retrieve reliable information sources from the internet. These retrieved sources contain data corresponding to the search query.

[0265] Step 8:

[0266] The server uses analysis tools to filter reliable information from the acquired information sources. These filtering tools exclude unreliable information and extract only the reliable information. This filtered information is then provided to the user.

[0267] Step 9:

[0268] When historical time-series data is input, the server uses data processing tools to split the input data into training data and test data. The split data is then used to train a machine learning model.

[0269] Step 10:

[0270] The server uses machine learning models to predict future trends based on historical data. The predicted trend data includes estimates of future sales and demand. The prediction results are provided to the user.

[0271] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0272] This invention relates to a system for translation, filtering out inaccurate information, predicting the future, and processing emotions. Furthermore, the invention implements a form that combines this with an emotion engine that recognizes and utilizes the user's emotional information. Specific embodiments are described below.

[0273] Translation Function Embodiment

[0274] Program processing

[0275] 1. The user enters the text to be translated, the source language, and the target language.

[0276] 2. The terminal receives the input information and analyzes the text using natural language processing tools.

[0277] 3. Use a translation model to tokenize the text and perform the translation.

[0278] 4. The terminal decodes the translation result and converts it into a format that humans can understand.

[0279] 5. The server activates an emotion engine to recognize the user's emotions and performs emotion adjustments appropriate to the translation result.

[0280] 6. The server provides the user with the final translation result.

[0281] Specific example

[0282] For example, when a user wants to translate a Japanese text "こんにちは" into English, the system receives "こんにちは", translates it to "Hello" based on the translation model, and provides it in a form adapted to the user's emotional state.

[0283] Embodiments of sharing correct knowledge

[0284] Program processing

[0285] 1. The user inputs the topic to be verified as a query.

[0286] 2. The server receives the input query and collects relevant information from the Internet through the search engine interface.

[0287] 3. The server filters reliable information sources from the collected search results and extracts specific URLs.

[0288] 4. The server uses the emotion engine means to recognize the user's emotion and adjusts the reliability of the provided information according to the user's emotional state.

[0289] 5. The terminal provides the user with reliable information sources and prompts the user to obtain additional information if necessary.

[0290] Specific example

[0291] For example, when searching for information about "global warming", the system obtains links to Wikipedia and government-related websites, and selects and provides appropriate information based on the user's emotional state.

[0292] Embodiments of future predictions

[0293] Program processing

[0294] 1. The user inputs past time-series data.

[0295] 2. The server receives the input data, preprocesses the data, and divides it into training data and test data.

[0296] 3. Use a machine learning model to train the data and create a prediction model.

[0297] 4. The server uses the created prediction model to predict future data based on the test data.

[0298] 5. The server uses the emotion engine means to adjust the prediction result based on the user's emotion information.

[0299] 6. The terminal provides the prediction result to the user and proposes the necessary actions.

[0300] Specific example

[0301] For example, when the user inputs past sales data, the system makes a future sales prediction based on that data and provides the prediction result in a form adapted to the user's emotional state.

[0302] Embodiments of world peace

[0303] Program processing

[0304] 1. The user inputs the content of the social media post.

[0305] 2. The terminal receives the input text and activates the emotion analysis model.

[0306] 3. The terminal uses the emotion analysis model to analyze the emotional state of the text.

[0307] 4. Based on the result of the emotion analysis, the terminal uses the recommended action generation means for promoting peaceful dialogue to create specific proposals.

[0308] 5. The server uses an emotion engine to adjust the optimal recommended action, taking into account the user's emotional state.

[0309] 6. The server provides the user with the generated recommended actions.

[0310] Specific example

[0311] For example, in the case of a positive post like "I love this!" and a negative post like "This is so frustrating!", the system analyzes the emotions in each post and suggests appropriate actions to facilitate dialogue based on the user's emotional state.

[0312] This enables users to achieve highly accurate translations, access to reliable information, predict the future, and engage in peaceful, emotion-based dialogue. Leveraging emotion-driven tools further enhances the user experience and increases the system's practicality.

[0313] The following describes the processing flow.

[0314] Translation Function Embodiment

[0315] Step 1:

[0316] The user enters the text to be translated, the source language, and the target language.

[0317] Step 2:

[0318] The terminal receives the input information and activates the natural language processing system.

[0319] Step 3:

[0320] The terminal tokenizes the source text and converts it into a format suitable for the translation model.

[0321] Step 4:

[0322] The terminal inputs tokenized text into the translation model and performs the translation.

[0323] Step 5:

[0324] The terminal decodes the translation result and converts it into a format that humans can understand.

[0325] Step 6:

[0326] The server activates the emotion engine and recognizes the user's emotional state.

[0327] Step 7:

[0328] The server uses the emotion information recognized by the emotion engine to adjust the translation results and adapt them to the emotions.

[0329] Step 8:

[0330] The server provides the user with the final translation result.

[0331] Implementation of sharing correct knowledge

[0332] Step 1:

[0333] The user enters the topic they want to investigate as a query.

[0334] Step 2:

[0335] The server receives the input query and collects relevant information from the internet through the search engine interface.

[0336] Step 3:

[0337] The server filters the collected search results to identify reliable sources and extracts specific URLs.

[0338] Step 4:

[0339] The server activates the emotion engine and recognizes the user's emotional state.

[0340] Step 5:

[0341] The server uses emotional information to adjust the reliability of the acquired information based on the user's emotional state.

[0342] Step 6:

[0343] The device provides users with reliable sources of information and encourages them to obtain additional information as needed.

[0344] Future predicted implementation

[0345] Step 1:

[0346] The user inputs historical time-series data.

[0347] Step 2:

[0348] The server receives the input data, preprocesses it, and splits it into training and test data.

[0349] Step 3:

[0350] The server uses a machine learning model to create a predictive model based on the training data.

[0351] Step 4:

[0352] The server uses the created predictive model to predict future data based on the test data.

[0353] Step 5:

[0354] The server activates the emotion engine and recognizes the user's emotional state.

[0355] Step 6:

[0356] The server uses the recognized emotion information to adjust the prediction results and adapt them to the user's emotions.

[0357] Step 7:

[0358] The device provides the user with prediction results and suggests necessary actions.

[0359] A Embodiment of World Peace

[0360] Step 1:

[0361] The user enters the content of their social media post.

[0362] Step 2:

[0363] The terminal receives the entered text and activates the sentiment analysis model.

[0364] Step 3:

[0365] The device uses an emotion analysis model to analyze the emotional state of the text.

[0366] Step 4:

[0367] Based on the results of the sentiment analysis, the device creates recommended actions to promote peaceful dialogue.

[0368] Step 5:

[0369] The server activates the emotion engine and recognizes the user's emotional state.

[0370] Step 6:

[0371] The server adjusts recommended actions based on emotional information, adapting to the user's emotions.

[0372] Step 7:

[0373] The server provides the user with a final recommended action.

[0374] This enables users to achieve highly accurate translations, access to reliable information, predict the future, and engage in peaceful, emotion-based dialogue. By leveraging the emotion engine, the user experience is further enhanced, and the system's practicality is increased.

[0375] (Example 2)

[0376] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0377] In today's internet environment, there is a demand for communication that transcends language barriers, the elimination of inaccurate information, future predictions, and responses based on emotions. However, conventional systems struggle to meet these multiple needs simultaneously, and a particular challenge is the lack of consideration for users' emotional information. Specifically, translation results, search results, and future predictions do not adapt to the user's emotional state, resulting in a low quality of user experience. Furthermore, without utilizing emotional information, it becomes difficult to alleviate user anxiety and stress, and to ensure the accuracy and reliability of the information provided.

[0378] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0379] In this invention, the server includes natural language processing means for translating input text from one language to another, means for tokenizing and decoding the text using a translation model, sentiment engine means for adjusting the translation result based on the user's sentiment, and output means for providing the translation result to the user. This makes it possible to provide an appropriate translation result that is in line with the user's sentiment.

[0380] Furthermore, it includes a search engine interface for inputting search queries on a specific topic and obtaining reliable information sources, an analysis means for filtering reliable information from the obtained information sources, an emotion engine means for adjusting the filtered information based on the user's emotions, and an output means for providing that information to the user. This makes it possible to provide reliable information tailored to the user's emotional state.

[0381] Furthermore, it includes a machine learning model for predicting future trends based on past time-series data, a data processing means for splitting input data into training data and test data, a prediction means for predicting future data using the trained model, an emotion engine means for adjusting the prediction results based on the user's emotions, and an output means for providing the prediction results to the user. This enables future predictions based on the user's emotional information, making it possible to provide more appropriate action suggestions.

[0382] "Natural language processing means" refers to technologies that analyze input text and convert human language into a format that machines can understand.

[0383] A "translation model" is an algorithm and computational model for converting text from one language to another.

[0384] "Tokenization" is the process of dividing input text into units of words or phrases.

[0385] "Decoding" is the process of reconstructing generated tokens and returning them to a human-understandable format as a result of translation and analysis.

[0386] An "emotional engine" is a technology that recognizes a user's emotional information and adjusts its response based on that information.

[0387] "Output means" refers to interfaces or devices used to provide text or information to users.

[0388] A "search engine interface means" is a technology for searching for information on the internet and obtaining results.

[0389] "Analysis means" refers to techniques for analyzing acquired information and extracting or filtering necessary data.

[0390] A "machine learning model" is a statistical algorithm that learns from past data and uses it to make predictions and analyses of the future.

[0391] "Data processing means" refers to technologies for pre-processing input data and converting it into the required format.

[0392] "Predictive tools" are techniques that use trained models to predict future data and trends.

[0393] This invention relates to a system that provides diverse functions using technologies such as natural language processing, emotion recognition, and machine learning. The specific forms for implementing each function are described below.

[0394] Translation Function Embodiment

[0395] In this system, the user inputs the text to be translated, along with the source and target languages. Specifically, the user inputs the text and selects the language setting through the terminal's interface. The terminal receives the input information and analyzes the text using natural language processing. Next, a translation model (e.g., a common translation API) is used to tokenize the text and perform the translation. The translation result is decoded and converted into a human-readable format. The server activates an emotion engine to recognize the user's emotions and performs emotion adjustments appropriate to the translation result. Finally, the adjusted translation result is provided to the user.

[0396] Specific example

[0397] For example, if a user wants to translate the Japanese text "こんにちは" (konnichiwa) into English, the system receives "こんにちは" as input and translates it to "Hello" based on a translation model (e.g., a general-purpose translation API). Then, using an emotion engine (e.g., a general-purpose emotion recognition API), it adjusts the translation based on the user's emotional state, such as "Hello, how can I assist you today?", and provides it to the user.

[0398] Example of a prompt

[0399] "Translate the following Japanese text 'こんにちは' into English and adjust it to suit the user's mood."

[0400] Implementation of sharing correct knowledge

[0401] In this system, the user inputs the topic they want to verify as a query. The server receives the input query and collects relevant information from the internet through a search engine interface (e.g., a general-purpose search API). It filters the retrieved search results to identify reliable sources and extracts specific URLs. Next, it uses an emotion engine to adjust the reliability of the information to match the user's emotional state. Finally, it provides reliable information to the user via their device.

[0402] Specific example

[0403] For example, when searching for information about "global warming," the system uses a general-purpose search API to collect relevant information and obtain links to Wikipedia and government websites. Then, it uses an emotion engine (e.g., a general-purpose emotion recognition API) to adapt that information to the user's emotional state and present it accordingly.

[0404] Example of a prompt

[0405] "Search for reliable information on 'global warming' and provide the most relevant information tailored to the user's emotions."

[0406] Future predicted implementation

[0407] In this system, the user inputs historical time-series data. The server receives the input data, preprocesses it, and splits it into training and test data. The software used includes, for example, general data processing libraries and machine learning libraries. Next, a machine learning model (for example, a general-purpose machine learning library) is used to train the data and create a predictive model. The prediction results are adjusted based on the user's emotional state using an emotion engine, and finally, the prediction results are provided to the user via a terminal.

[0408] Specific example

[0409] For example, when a user inputs past sales data, the system preprocesses it using a data processing library and then uses a machine learning library to predict future sales. Afterward, an emotion engine (e.g., a general-purpose emotion recognition API) is used to adapt the prediction results to the user's emotional state and provide them to the user.

[0410] Example of a prompt

[0411] "Based on past sales data, we predict future sales and adjust them according to user sentiment."

[0412] A Embodiment of World Peace

[0413] In this system, the user inputs the content of a social media post. The terminal receives the input text and activates a sentiment analysis model (e.g., a general-purpose sentiment analysis API). Using the sentiment analysis model, the emotional state of the text is analyzed, and based on the results, recommended actions to promote peaceful dialogue are generated. Next, the server uses a sentiment engine to adjust the optimal recommended actions, taking into account the user's emotional state. Finally, the generated recommended actions are provided to the user through the terminal.

[0414] Specific example

[0415] For example, in the case of a positive post like "I love this!" and a negative post like "This is so frustrating!", the system uses a sentiment analysis API to analyze the emotions in each post and then uses an emotion engine to suggest an appropriate way to interact with the user.

[0416] Example of a prompt

[0417] "Analyze the sentiment of social media posts and propose ways to promote peaceful dialogue based on user sentiment."

[0418] As a result, the system provided by the present invention can perform multi-functional responses that take into account the user's emotional information, enabling translation, elimination of inaccurate information, future prediction, and promotion of emotion-based dialogue.

[0419] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0420] Translation Function Embodiment

[0421] Program processing

[0422] Step 1:

[0423] The user inputs the text to be translated, along with the source and target languages. The user sets the text "Hello" and the source language to Japanese and the target language to English via the device's interface. Input is done using text boxes and dropdown lists.

[0424] Input: Text "Hello", Source language "Japanese", Target language "English"

[0425] Output: Input information is sent.

[0426] Step 2:

[0427] The terminal receives the input information and analyzes the text using natural language processing techniques. Here, the language of the text is confirmed, and preprocessing is performed, including analysis of words and grammar.

[0428] Input: User input information

[0429] Data processing: Text preprocessing (tokenization, stop word removal, etc.)

[0430] Output: Analysis results (tokenized text)

[0431] Step 3:

[0432] The device sends the analyzed text to a translation model for translation. The translation model uses, for example, a common translation API to convert the tokenized text into the target language.

[0433] Input: Parsed text (tokens)

[0434] Data processing: Call the translation API and perform the translation.

[0435] Output: Translation result (English text "Hello")

[0436] Step 4:

[0437] The terminal receives the translation result, decodes it, and converts it into a human-readable format. Here, it reconstructs another token and formats it into the appropriate language format.

[0438] Input: English token "Hello"

[0439] Data processing: Decode and format tokens.

[0440] Output: Formatted English text "Hello"

[0441] Step 5:

[0442] The server activates the emotion engine, analyzes the translation results, and makes adjustments based on the user's emotional state. Here, sentiment analysis is performed based on the user's previous input and history, and the translation results are adjusted accordingly.

[0443] Input: Translation result "Hello"

[0444] Data processing: Analysis and adjustment using an emotion engine.

[0445] Output: Adjusted translation result: "Hello, how can I assist you today?"

[0446] Step 6:

[0447] The server sends the final translation result to the terminal, which then displays the result to the user. The user's screen displays text that has been adjusted based on their emotions.

[0448] Input: Adjusted translation result

[0449] Output: Translation results displayed to the user

[0450] ---

[0451] Implementation of sharing correct knowledge

[0452] Program processing

[0453] Step 1:

[0454] The user enters a search query related to a specific topic. For example, if they want to find information about "global warming," they would enter "global warming" in the search box.

[0455] Input: Search query "global warming"

[0456] Output: Input query

[0457] Step 2:

[0458] The server receives the input query and collects relevant information from the internet through the search engine interface. For example, it might use a general-purpose search API to retrieve information.

[0459] Input: Search query

[0460] Data processing: Call the search API and collect relevant information.

[0461] Output: Search Results

[0462] Step 3:

[0463] The server filters the collected search results to identify reliable sources. It uses algorithms to evaluate reliability, for example, to extract links to Wikipedia or official government websites.

[0464] Input: Search Results

[0465] Data processing: Reliability assessment and filtering

[0466] Output: Links to reliable sources

[0467] Step 4:

[0468] The server activates the emotion engine and adjusts the reliability of the information according to the user's emotional state. Here, the method and content of information presentation are adapted based on the user's emotional state.

[0469] Input: Links to reliable sources

[0470] Data processing: Analysis and adjustment using an emotion engine.

[0471] Output: Adjusted information

[0472] Step 5:

[0473] The device provides the user with the finalized information and prompts them to obtain additional information as needed.

[0474] Input: Adjusted information

[0475] Output: Information displayed to the user

[0476] ---

[0477] Future predicted implementation

[0478] Program processing

[0479] Step 1:

[0480] Users input historical time-series data. For example, they might upload historical sales data as a CSV file.

[0481] Input: CSV file, historical sales data

[0482] Output: Uploaded data

[0483] Step 2:

[0484] The server receives the input data, preprocesses it, and splits it into training and test data. For example, it performs data cleaning and normalization before splitting the data into training and test sets.

[0485] Input: Uploaded data

[0486] Data processing: Data cleaning, normalization, splitting

[0487] Output: Training data, Test data

[0488] Step 3:

[0489] The server uses training data to train a machine learning model. Here, for example, a general-purpose machine learning library is used to create a model for future prediction.

[0490] Input: Training data

[0491] Data processing: Model training

[0492] Output: Trained predictive model

[0493] Step 4:

[0494] The server uses a trained model to predict future data based on test data. The model is used to predict future sales data, for example.

[0495] Input: Test data

[0496] Data processing: Predicting future data

[0497] Output: Prediction result

[0498] Step 5:

[0499] The server uses an emotion engine to adjust prediction results based on the user's emotional state. For example, it might add encouraging comments to negative prediction results.

[0500] Input: Prediction result

[0501] Data processing: Analysis and adjustment using an emotion engine.

[0502] Output: Adjusted prediction results

[0503] Step 6:

[0504] The device provides the user with the final prediction results and suggests necessary actions.

[0505] Input: Adjusted prediction result

[0506] Output: Prediction results displayed to the user

[0507] ---

[0508] A Embodiment of World Peace

[0509] Program processing

[0510] Step 1:

[0511] The user enters the content of their social media post. For example, they might enter a positive post like "I love this!" in the input field.

[0512] Input: Social media post content

[0513] Output: Input text

[0514] Step 2:

[0515] The terminal receives the entered text and activates the sentiment analysis model. Here, a general-purpose sentiment analysis API is used.

[0516] Input: User's post content

[0517] Data processing: Calling the emotion analysis API

[0518] Output: Sentiment analysis results

[0519] Step 3:

[0520] The device uses an emotion analysis model to analyze the emotional state of the text. For example, it obtains positive, negative, and neutral emotion scores.

[0521] Input: User's post content

[0522] Data processing: Calculation of sentiment score

[0523] Output: Emotional state score

[0524] Step 4:

[0525] The device generates recommended actions to promote peaceful dialogue based on sentiment analysis results. For example, it might suggest "Share more!" for positive posts.

[0526] Input: Emotional state score

[0527] Data processing: Generating recommended actions

[0528] Output: Recommended Action

[0529] Step 5:

[0530] The server uses an emotion engine to adjust the optimal recommended action based on the user's emotional state. For example, it might take into account past posting history.

[0531] Input: Recommended Action

[0532] Data processing: Analysis and adjustment using an emotion engine.

[0533] Output: Optimized Recommended Actions

[0534] Step 6:

[0535] The server sends the final recommended action to the terminal, which then provides it to the user.

[0536] Input: Optimized Recommended Actions

[0537] Output: Recommended actions displayed to the user

[0538] (Application Example 2)

[0539] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0540] Conventional systems lack the ability to translate, provide information, or predict future trends that take into account user emotions, resulting in insufficient individual optimization of the user experience. Furthermore, e-commerce sites are currently unable to provide appropriate product recommendations based on user emotional states, limiting their sales promotion effectiveness.

[0541] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes natural language processing means for translating input text from one language to another, means for tokenizing and decoding the text using a translation model, sentiment analysis means for analyzing the user's sentiment information, sentiment engine means for adjusting the translation result based on the analyzed sentiment information, and output means for providing the translation result to the user. This makes it possible to provide individually optimized translation results based on the user's sentiment information.

[0542] "Natural language processing means" refers to technologies that analyze input text and perform specific processing.

[0543] A "translation model" is an algorithm that converts text from one language to another.

[0544] "Tokenization" is the process of analyzing text and dividing it into the smallest units of words or phrases.

[0545] "Decoding methods" refer to techniques for reconstructing tokenized text into natural language.

[0546] "Emotional analysis methods" refer to technologies that analyze a user's emotional information from input text and data.

[0547] An "emotional engine" is a technology that adjusts the system's response and output based on analyzed emotional information.

[0548] "Output means" refers to the technologies and devices that a system uses to display results to the user.

[0549] A "search engine interface" is a system for collecting information from the internet.

[0550] "Reliable information" refers to verified data obtained from trustworthy sources.

[0551] A "machine learning model" is an algorithm that learns patterns from past data and uses them to predict the future.

[0552] "Data processing means" refers to techniques that preprocess input data and optimize it for analysis and prediction.

[0553] A "predictive tool" is a technique that uses trained machine learning models to estimate future data.

[0554] System Overview

[0555] This invention is a system that analyzes user emotional information and optimizes translation results, information provision, future predictions, etc., based on that analysis. This system includes the following main components.

[0556] Hardware and software

[0557] 1. Hardware: Smartphones are primarily used. Smartphones provide an interface for users to input text and have a display that shows analysis results and recommendations.

[0558] 2. Software: The Python language and TextBlob library are used for sentiment analysis and natural language processing. The server implements a sentiment engine, natural language processing, data processing, and machine learning models.

[0559] Processing Overview

[0560] 1. Sentiment Analysis: The text entered by the user is analyzed using the TextBlob library to calculate an emotion score. Based on the emotion score, the user's emotional state is understood.

[0561] 2. Natural Language Processing: Natural language processing tools are used to translate text entered by the user from one language to another. The translation model performs tokenization and decoding between languages.

[0562] 3. Emotion Engine: Based on the analyzed emotional information, it adjusts the translation results and information provided. For example, if the emotional state is relaxed, it recommends relaxation-related products.

[0563] Specific example

[0564] (Specific example 1)

[0565] Suppose a user enters "I've been feeling really stressed out lately." This input text is then analyzed by the TextBlob library to calculate a negative emotion score, which is then categorized as "stress relief" by the emotion analysis tool. Next, the emotion engine is activated, and products such as stress balls or therapy sessions are recommended from the stress relief category.

[0566] (Specific example 2)

[0567] If a user enters "I'm feeling great today!", based on this positive emotion score, an aroma diffuser or relaxing tea will be recommended from the relaxation category.

[0568] Example of a prompt

[0569] This is an example where a user enters "I'm feeling overwhelmed" as their current emotional state, and relaxation or stress-relief products are recommended based on their emotional score. The product list is as follows:

[0570] Relaxation products: Aroma diffuser, Relaxing tea, Yoga mat

[0571] Stress relief products: Stress Ball, Meditation App Subscription, Therapy Sessions

[0572] If the emotional analysis score is positive, relaxation products will be recommended; if it is negative, stress relief products will be recommended.

[0573] Effects of implementation

[0574] This invention enables the provision of appropriate information and product recommendations based on the user's emotional state, thereby providing a more personalized user experience. Furthermore, it is expected to have a positive impact on sales promotion on e-commerce sites.

[0575] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0576] Step 1:

[0577] The terminal receives text input from the user. This input text reflects the user's emotional state. This text serves as the starting point for system processing. For example, the input data might be "I've been feeling really stressed out lately."

[0578] Step 2:

[0579] The terminal sends the received text to the server. The server receives the text and analyzes the sentiment information of the input text using sentiment analysis tools. It calculates the sentiment score of the text using the TextBlob library. The sentiment score is output as the calculation result. For example, a negative sentiment score may be obtained.

[0580] Step 3:

[0581] The server determines the emotional category of the text based on its emotional score. The emotional analysis tool classifies the text into the appropriate emotional category, such as "relaxation" or "stress relief." The classification result is then passed on to the next processing step. For example, if the emotional score is negative, it is classified into the "stress relief" category.

[0582] Step 4:

[0583] The server activates an emotion engine and selects the most suitable product for the user based on the analyzed emotion information. It extracts products matching the emotion category from the product list and generates a recommendation list. Random sampling or predefined rules are used to generate the recommendation list. For example, stress balls and therapy sessions might be extracted from the "stress relief" category.

[0584] Step 5:

[0585] The server sends the generated recommendation list to the terminal. The terminal displays the received recommendation list to the user. Specifically, information about the recommended products is visualized on the display device. For example, detailed information about "stress balls" and "therapy sessions" is displayed.

[0586] Step 6:

[0587] Users can make selection and purchase actions based on the displayed recommended products. The system records user selections and uses them to improve future recommendation algorithms.

[0588] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0589] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0590] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0591] [Second Embodiment]

[0592] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0593] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0594] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0595] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0596] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0597] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0598] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0599] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0600] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0601] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0602] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0603] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0604] This invention relates to a system for performing translation, filtering out inaccurate information, predicting the future, and processing emotions. Specific embodiments thereof are described below.

[0605] Translation Function Embodiment

[0606] Program processing

[0607] 1. The user enters the text they want to translate, along with the source language and the target language.

[0608] 2. The terminal receives the input information and analyzes the text using natural language processing tools.

[0609] 3. Use a translation model to tokenize the text and perform the translation.

[0610] 4. The terminal decodes the translation result and converts it into a format that humans can understand.

[0611] 5. The server provides the user with the final translation result.

[0612] Specific example

[0613] For example, if a user wants to translate the Japanese text "こんにちは" into English, the system receives "こんにちは," translates it to "Hello" based on its translation model, and provides it to the user.

[0614] Implementation of sharing correct knowledge

[0615] Program processing

[0616] 1. The user enters a question about a specific topic.

[0617] 2. The server uses a search engine interface to obtain relevant information from the internet.

[0618] 3. The server performs analysis to filter reliable information from the acquired information sources.

[0619] 4. The device provides users with filtered, reliable information.

[0620] Specific example

[0621] For example, when searching for information about "global warming," the system retrieves links to Wikipedia and government-related websites and provides them to the user.

[0622] Future predicted implementation

[0623] Program processing

[0624] 1. The user inputs historical time-series data.

[0625] 2. The server uses data processing equipment to split the input data into training data and test data.

[0626] 3. Train a machine learning model on the data and create a predictive model.

[0627] 4. The server uses the created model to predict future data.

[0628] 5. The device provides the user with the prediction results.

[0629] Specific example

[0630] For example, when a user inputs past sales data, the system uses that data to predict future sales and provides the results to the user.

[0631] A Embodiment of World Peace

[0632] Program processing

[0633] 1. The user enters the content of their social media post.

[0634] 2. The terminal uses an emotion analysis model to analyze the emotion of the input text.

[0635] 3. The terminal uses a recommended action generation mechanism based on the sentiment analysis results to create suggestions to promote peaceful dialogue.

[0636] 4. The server provides the user with the recommended actions that have been created.

[0637] Specific example

[0638] For example, in the case of a positive post like "I love this!" and a negative post like "This is so frustrating!", the system analyzes the emotions in each post, determines whether they are "positive" or "negative," and suggests actions to the user to facilitate appropriate dialogue.

[0639] These embodiments enable users to achieve highly accurate translation, access to reliable information, predict the future, and engage in peaceful, emotion-based dialogue.

[0640] The following describes the processing flow.

[0641] Translation Function Embodiment

[0642] Step 1:

[0643] The user enters the text to be translated, the source language, and the target language.

[0644] Step 2:

[0645] The terminal receives input from the user and activates natural language processing for translation.

[0646] Step 3:

[0647] The terminal tokenizes the source text and converts it into a format compatible with the language model.

[0648] Step 4:

[0649] The terminal inputs tokenized text into the translation model and performs the translation.

[0650] Step 5:

[0651] The device decodes the translation result and converts it into a natural language format that the user can understand.

[0652] Step 6:

[0653] The server provides the user with the final translation result.

[0654] Implementation of sharing correct knowledge

[0655] Step 1:

[0656] The user enters the topic they want to investigate as a query.

[0657] Step 2:

[0658] The server receives the input query and collects relevant information from the internet through the search engine interface.

[0659] Step 3:

[0660] The server filters the collected search results to identify reliable sources and extracts specific URLs.

[0661] Step 4:

[0662] The device provides users with reliable sources of information and encourages them to obtain additional information as needed.

[0663] Future predicted implementation

[0664] Step 1:

[0665] The user inputs historical time-series data.

[0666] Step 2:

[0667] The server receives the input data, preprocesses it, and splits it into training data and test data.

[0668] Step 3:

[0669] The server uses a machine learning model to create a predictive model based on the training data.

[0670] Step 4:

[0671] The server uses the created predictive model to predict future data based on the test data.

[0672] Step 5:

[0673] The device provides the user with prediction results and suggests necessary actions.

[0674] A Embodiment of World Peace

[0675] Step 1:

[0676] The user enters the content of their social media post.

[0677] Step 2:

[0678] The terminal receives the entered text and activates the sentiment analysis model.

[0679] Step 3:

[0680] The device uses an emotion analysis model to analyze the emotional state of the text.

[0681] Step 4:

[0682] Based on the results of the sentiment analysis, the device creates recommended actions to promote peaceful dialogue.

[0683] Step 5:

[0684] The server provides the user with the generated recommended actions.

[0685] This allows users to effectively translate text, obtain reliable information, predict the future, and engage in emotion-based conversations.

[0686] (Example 1)

[0687] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0688] In recent years, the importance of global communication has been steadily increasing, but handling multiple different languages ​​requires advanced translation technology. Furthermore, there is a need for systems that can quickly obtain reliable information, predict future trends, and facilitate emotion-based dialogue. To meet these needs, systems that efficiently combine highly accurate natural language processing technologies and machine learning models are essential.

[0689] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0690] In this invention, the server includes natural language processing means for translating input text from one language to another; processing means for tokenizing and decoding the text using a machine learning translation model; output means for providing the translation results to the user; search engine interface means for inputting search queries on a specific topic and obtaining relevant information; reliability evaluation means for filtering reliable information from the obtained information sources; output means for providing the filtered information to the user; machine learning prediction model means for predicting future data based on past time series data; data processing means for splitting input data into training data and test data; prediction means for predicting future data using the trained model; output means for providing the prediction results to the user; sentiment analysis means for inputting social media posts and analyzing the sentiment of the text; recommendation action generation means for promoting peaceful dialogue based on the sentiment analysis results; and output means for providing the generated recommendation actions to the user. This enables highly accurate translation, acquisition of reliable information, future prediction, and peaceful dialogue based on sentiment.

[0691] "Natural language processing means" refers to technologies that analyze input text and enable operations such as translation and sentiment analysis.

[0692] A "machine learning translation model" is a model trained using a large amount of data, and it has the ability to translate text from one language to another.

[0693] "Processing means" refers to an algorithm or program for performing a specific task, and in this invention, it performs text tokenization and decoding.

[0694] "Output means" refers to interfaces or devices for providing processing results to the user, and have functions to display translation results, search results, prediction results, etc.

[0695] A "search engine interface means" is a means of linking with a search engine for obtaining information on the internet.

[0696] A "reliability evaluation method" is a system that executes algorithms and evaluation criteria to select highly reliable information from acquired information sources.

[0697] A "machine learning prediction model" is a prediction model trained using past data, which is used to predict future data.

[0698] "Data processing means" refers to means for dividing input data into training data and test data.

[0699] A "predictive tool" is a means of using a trained model to predict future data.

[0700] "Sentiment analysis techniques" are technologies used to analyze the emotions in text and determine whether they are positive or negative.

[0701] A "recommended action generation method" is a means of generating actions and suggestions to promote peaceful dialogue based on the results of sentiment analysis.

[0702] This invention relates to a system that utilizes natural language processing technology and machine learning models to achieve translation functionality, reliable information acquisition, future prediction, and sentiment analysis and peaceful dialogue on social media. Specific embodiments thereof are described below.

[0703] Translation function

[0704] Specific Embodiments

[0705] 1. The user inputs text through a translation interface and specifies the source and target languages. This interface is implemented as a web browser or mobile application.

[0706] 2. The terminal uses the Python NLTK library to parse and tokenize the input text.

[0707] 3. The device calls the Google Translate API or Microsoft Translator to translate the tokenized text.

[0708] 4. The device decodes the translation result and converts it into a format that the user can understand.

[0709] 5. The server displays the final translation result on the interface and provides it to the user.

[0710] Specific example

[0711] If a user wants to translate the Japanese text "こんにちは" into English, the system receives "こんにちは," uses the Google Translate API to translate it to "Hello," and provides the result to the user.

[0712] Sharing correct knowledge

[0713] Specific Embodiments

[0714] 1. The user enters a question about a specific topic. This question is entered via a web browser or mobile application.

[0715] 2. The server uses the Google Search API and other search engine interfaces to retrieve relevant information from the internet.

[0716] 3. The server runs a reliability evaluation algorithm and filters out reliable information from the retrieved sources.

[0717] 4. The device provides filtered information to the user.

[0718] Specific example

[0719] When searching for information related to "global warming," the system uses the Google Search API to retrieve relevant information, selects the most reliable information, and provides it to the user.

[0720] Future predictions

[0721] Specific Embodiments

[0722] 1. Users upload historical time-series data in CSV files or other digital formats.

[0723] 2. The server uses the Python pandas library to split the input data into training and test data.

[0724] 3. The server uses machine learning frameworks such as TensorFlow and Scikit-learn to train data and create predictive models.

[0725] 4. The server uses the trained model to predict future data.

[0726] 5. The device provides the user with prediction results in graph or table format.

[0727] Specific example

[0728] When a user inputs past sales data, the system uses a TensorFlow LSTM model based on that data to predict future sales and provides the results to the user.

[0729] World peace

[0730] Specific Embodiments

[0731] 1. The user enters the content of their social media post in text format.

[0732] 2. The terminal uses IBM Watson or Microsoft Azure sentiment analysis APIs to analyze the sentiment of the input text.

[0733] 3. Based on the sentiment analysis results, the device uses a recommended action generation algorithm to create suggestions to promote peaceful dialogue.

[0734] 4. The server provides the user with the recommended actions that have been created.

[0735] Specific example

[0736] If a negative post such as "This is so frustrating!" is submitted, IBM Watson will be used to determine that the sentiment is "negative" and generate and provide suggestions to the user to facilitate appropriate dialogue.

[0737] The above describes embodiments of the present invention. This system enables users to achieve highly accurate translation, obtain reliable information, predict the future, and engage in peaceful, emotion-based dialogue.

[0738] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0739] Translation Function Embodiment

[0740] Processing steps

[0741] Step 1:

[0742] The user accesses the system interface and enters the text to be translated, along with the source and target languages. For example, they might enter "こんにちは" (konnichiwa) and specify a translation from Japanese to English. This input is sent to the server in text format.

[0743] Step 2:

[0744] The terminal parses and tokenizes the input text received from the server using Python's NLTK library. The input data is "hello," and through parsing and tokenization, the internal structure of the text is understood. The output is the tokenized text.

[0745] Step 3:

[0746] The device sends tokenized text to the Google Translate API or Microsoft Translator for translation. The input is the tokenized text "こんにちは" (Konnichiwa), and the translation is performed by calling the API. The output is the translated text "Hello".

[0747] Step 4:

[0748] The terminal decodes the translation result returned from the API and converts it into a human-readable format. The input is the translated text "Hello," and the terminal performs the conversion from the encoded format to standard text format. The output is the decoded text "Hello."

[0749] Step 5:

[0750] The server provides the user with the final translation result. The input is the decoded text "Hello," which is returned to the interface for displaying it to the user. The output is the translation result displayed to the user.

[0751] Implementation of sharing correct knowledge

[0752] Processing steps

[0753] Step 1:

[0754] The user enters a question about a specific topic into the system. For example, they might enter, "Tell me about global warming." This input is sent to the server in text format.

[0755] Step 2:

[0756] The server uses the Google Search API to search for relevant information on the internet. The input is the question "Tell me about global warming," and the API is called to retrieve relevant information. The output is the retrieved search results.

[0757] Step 3:

[0758] The server executes a reliability evaluation algorithm to filter the retrieved information for the most reliable information. The input is the retrieved search results, to which the algorithm is applied for reliability evaluation. The output is the result of selecting only the most reliable information.

[0759] Step 4:

[0760] The terminal provides the user with filtered, reliable information. The input is reliable information, which is returned to the interface for displaying it to the user. The output is the reliable information displayed to the user.

[0761] Future prediction implementation

[0762] Processing steps

[0763] Step 1:

[0764] Users upload historical time-series data to the system as a CSV file. For example, they might upload sales data for the past year. This input is sent to the server in file format.

[0765] Step 2:

[0766] The server uses the Python pandas library to split the input data into training and test sets. The input is an uploaded CSV file, which is then divided into 80% training data and 20% test data. The output is the split dataset.

[0767] Step 3:

[0768] The server uses machine learning frameworks such as TensorFlow and Scikit-learn to train data and create predictive models. The input is training data, and the machine learning algorithm generates the predictive model. The output is the trained predictive model.

[0769] Step 4:

[0770] The server evaluates the performance of a model created using test data and predicts future data. The inputs are the test data and the predictive model; performance evaluation and future prediction are performed. The output is the predicted data.

[0771] Step 5:

[0772] The terminal provides prediction results to the user in graph or table format. The input is the predicted data, which is visualized and displayed on the interface. The output is the prediction result displayed to the user.

[0773] A Embodiment of World Peace

[0774] Processing steps

[0775] Step 1:

[0776] The user enters their social media post content into the system in text format. For example, they might enter "This is so frustrating!". This input is then sent to the server in text format.

[0777] Step 2:

[0778] The terminal uses IBM Watson and Microsoft Azure sentiment analysis APIs to analyze the sentiment of the input text. The input is a social media post, "This is so frustrating!", and the sentiment is analyzed by calling the sentiment analysis API. The output is the analyzed sentiment result.

[0779] Step 3:

[0780] The device applies a recommended action generation algorithm to promote peaceful dialogue based on the sentiment analysis results. The input is the sentiment analysis results, and the algorithm is used to generate an appropriate response. The output is the recommended action.

[0781] Step 4:

[0782] The server provides the user with the generated recommended actions. The input is the recommended actions, which are returned to the interface for displaying them to the user. The output is the recommended actions displayed to the user.

[0783] These specific processing steps enable users to achieve highly accurate translations, obtain reliable information, predict the future, and engage in peaceful, emotion-based dialogue.

[0784] (Application Example 1)

[0785] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0786] In many modern situations, communication between people who speak different languages, facilitating emotion-based dialogue, obtaining reliable information, and predicting the future based on diverse data are critical challenges. However, current systems lack an integrated approach to meet these requirements, particularly a lack of dialogue support incorporating real-time sentiment analysis. This makes efficient communication and appropriate decision-making difficult.

[0787] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0788] In this invention, the server includes natural language processing means for translating input text from one language to another, means for tokenizing and decoding the text using a translation model, output means for providing the translation results to the user, sentiment analysis means for analyzing the sentiment of the input text, and dialogue support means for facilitating peaceful dialogue based on the analysis results. This enables real-time sentiment analysis and dialogue support.

[0789] A "natural language processing system" is a means of analyzing input text and translating it from one language to another.

[0790] A "translation model" is a model for tokenizing text and decoding it into another language.

[0791] "Output means" refers to the means of providing translation results or analysis results to the user.

[0792] An "emotion analysis tool" is a means of analyzing the emotions in input text and identifying the type of emotion.

[0793] "Dialogue support methods" are means of creating proposals to promote peaceful dialogue based on the results of emotion analysis.

[0794] A "search engine interface" is a means of obtaining reliable information sources by entering a search query related to a specific topic.

[0795] "Analysis means" are means for filtering reliable information from acquired information sources.

[0796] A "machine learning model" is a model used to predict future trends based on past time-series data.

[0797] "Data processing means" refers to means for dividing input data into training data and test data.

[0798] A "predictive tool" is a means of predicting future data using a trained model.

[0799] This invention can be specifically implemented as follows.

[0800] First, the system receives text input from the user. This input text could be content to be translated, content requiring sentiment analysis, or a search query. Once the user enters the text, the terminal analyzes it using natural language processing tools. At this stage, the main software used includes "Transformers," a widely known natural language processing library.

[0801] Next, if translation is needed, the system uses a translation model to tokenize the text and perform the translation. The translation result is decoded and provided to the user. Open-source translation models such as "Hugging Face" are used.

[0802] Furthermore, the system uses sentiment analysis tools to analyze the sentiment of the input text. Sentiment analysis assigns specific sentiment labels (e.g., positive, negative, neutral). Based on these analysis results, dialogue support tools generate suggestions to facilitate peaceful dialogue.

[0803] When a search query is entered, the server uses a search engine interface to obtain reliable information sources and filters them using analysis tools. The reliable information is then provided to the user.

[0804] Furthermore, the system can predict future trends based on historical time-series data. The server divides the input data into training data and test data, and trains the data using a machine learning model. This allows it to predict future data and provide results to the user.

[0805] As a concrete example, suppose a security staff member detected the following statement at the scene:

[0806] “This situation is incredibly frustrating and makes me angry.”

[0807] In this case, the system analyzes the text and detects the emotion "NEGATIVE." It then suggests the following recommended action:

[0808] "Please try to engage in calm dialogue to alleviate the situation."

[0809] Examples of prompt statements include the following:

[0810] "A proposed system for emotional analysis and peaceful dialogue"

[0811] Analyze user sentiment based on their input text and provide appropriate dialogue actions.

[0812] User input: "This situation is incredibly frustrating and makes me angry."

[0813] In this way, the system provides an integrated approach to emotion analysis and dialogue support, enabling it to respond appropriately in real time.

[0814] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0815] Step 1:

[0816] The user provides input text to the terminal. The input can be text to be translated, text requiring sentiment analysis, or a search query. The terminal analyzes this input text using natural language processing. The analyzed text is then sent to the next processing step.

[0817] Step 2:

[0818] The device uses a translation model to tokenize the input text if it requires translation. Tokenization is the process of dividing each word or phrase in the text into individual tokens. The tokens are then sent to the translation model.

[0819] Step 3:

[0820] The translation model translates tokenized text into the specified language. The translation model uses a generative AI model to convert the token array into tokens in the other language. The resulting token array is decoded and converted into human-readable text. The decoded text is returned to the terminal.

[0821] Step 4:

[0822] The terminal provides the translated text to the user through an output device. The user receives the final translation result.

[0823] Step 5:

[0824] The terminal analyzes the sentiment of the input text using sentiment analysis means if the sentiment of the text should be analyzed. The sentiment analysis means analyzes the text and assigns a specific sentiment label such as positive, negative, or neutral. The analysis results and sentiment labels are identified and sent to the next processing step.

[0825] Step 6:

[0826] The device uses dialogue support mechanisms based on the analyzed emotion results to generate suggestions for peaceful dialogue. These dialogue support mechanisms create messages to alleviate negative emotions and messages to maintain a positive atmosphere for positive emotions. These suggestions are then provided to the user.

[0827] Step 7:

[0828] When a user enters a search query, the server uses a search engine interface to retrieve reliable information sources from the internet. These retrieved sources contain data corresponding to the search query.

[0829] Step 8:

[0830] The server uses analysis tools to filter reliable information from the acquired information sources. These filtering tools exclude unreliable information and extract only the reliable information. This filtered information is then provided to the user.

[0831] Step 9:

[0832] When historical time-series data is input, the server uses data processing tools to split the input data into training data and test data. The split data is then used to train a machine learning model.

[0833] Step 10:

[0834] The server uses machine learning models to predict future trends based on historical data. The predicted trend data includes estimates of future sales and demand. The prediction results are provided to the user.

[0835] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0836] This invention relates to a system for translation, filtering out inaccurate information, predicting the future, and processing emotions. Furthermore, the invention implements a form that combines this with an emotion engine that recognizes and utilizes the user's emotional information. Specific embodiments are described below.

[0837] Translation Function Embodiment

[0838] Program processing

[0839] 1. The user enters the text to be translated, the source language, and the target language.

[0840] 2. The terminal receives the input information and analyzes the text using natural language processing tools.

[0841] 3. Use a translation model to tokenize the text and perform the translation.

[0842] 4. The terminal decodes the translation result and converts it into a format that humans can understand.

[0843] 5. The server activates an emotion engine to recognize the user's emotions and performs emotion adjustments appropriate to the translation result.

[0844] 6. The server provides the user with the final translation result.

[0845] Specific example

[0846] For example, if a user wants to translate the Japanese text "こんにちは" (konnichiwa) into English, the system receives "こんにちは," translates it to "Hello" based on its translation model, and then provides it in a form that is adapted to the user's emotional state.

[0847] Implementation of sharing correct knowledge

[0848] Program processing

[0849] 1. The user enters the topic they want to verify as a query.

[0850] 2. The server receives the input query and collects relevant information from the internet through the search engine interface.

[0851] 3. The server filters the collected search results to identify reliable sources and extracts specific URLs.

[0852] 4. The server uses an emotion engine to recognize the user's emotions and adjusts the reliability of the information it provides according to the user's emotional state.

[0853] 5. The device provides users with reliable sources of information and encourages them to obtain additional information as needed.

[0854] Specific example

[0855] For example, when searching for information about "global warming," the system retrieves links from Wikipedia and government-related websites, and then selects and provides appropriate information based on the user's emotional state.

[0856] Future predicted implementation

[0857] Program processing

[0858] 1. The user inputs historical time-series data.

[0859] 2. The server receives the input data, preprocesses it, and splits it into training data and test data.

[0860] 3. Train a machine learning model on the data and create a predictive model.

[0861] 4. The server uses the created predictive model to predict future data based on the test data.

[0862] 5. The server uses an emotion engine to adjust prediction results based on the user's emotional information.

[0863] 6. The device provides the user with prediction results and suggests necessary actions.

[0864] Specific example

[0865] For example, when a user inputs past sales data, the system uses that data to predict future sales and provides the prediction results in a way that is tailored to the user's emotional state.

[0866] A Embodiment of World Peace

[0867] Program processing

[0868] 1. The user enters the content of their social media post.

[0869] 2. The terminal receives the entered text and activates the sentiment analysis model.

[0870] 3. The device uses an emotion analysis model to analyze the emotional state of the text.

[0871] 4. Based on the results of the sentiment analysis, the terminal creates specific suggestions using a recommended action generation mechanism to promote peaceful dialogue.

[0872] 5. The server uses an emotion engine to adjust the optimal recommended action, taking into account the user's emotional state.

[0873] 6. The server provides the user with the generated recommended actions.

[0874] Specific example

[0875] For example, in the case of a positive post like "I love this!" and a negative post like "This is so frustrating!", the system analyzes the emotions in each post and suggests appropriate actions to facilitate dialogue based on the user's emotional state.

[0876] This enables users to achieve highly accurate translations, access to reliable information, predict the future, and engage in peaceful, emotion-based dialogue. Leveraging emotion-driven tools further enhances the user experience and increases the system's practicality.

[0877] The following describes the processing flow.

[0878] Translation Function Embodiment

[0879] Step 1:

[0880] The user enters the text to be translated, the source language, and the target language.

[0881] Step 2:

[0882] The terminal receives the input information and activates the natural language processing system.

[0883] Step 3:

[0884] The terminal tokenizes the source text and converts it into a format suitable for the translation model.

[0885] Step 4:

[0886] The terminal inputs tokenized text into the translation model and performs the translation.

[0887] Step 5:

[0888] The terminal decodes the translation result and converts it into a format that humans can understand.

[0889] Step 6:

[0890] The server activates the emotion engine and recognizes the user's emotional state.

[0891] Step 7:

[0892] The server uses the emotion information recognized by the emotion engine to adjust the translation results and adapt them to the emotions.

[0893] Step 8:

[0894] The server provides the user with the final translation result.

[0895] Implementation of sharing correct knowledge

[0896] Step 1:

[0897] The user enters the topic they want to investigate as a query.

[0898] Step 2:

[0899] The server receives the input query and collects relevant information from the internet through the search engine interface.

[0900] Step 3:

[0901] The server filters the collected search results to identify reliable sources and extracts specific URLs.

[0902] Step 4:

[0903] The server activates the emotion engine and recognizes the user's emotional state.

[0904] Step 5:

[0905] The server uses emotional information to adjust the reliability of the acquired information based on the user's emotional state.

[0906] Step 6:

[0907] The device provides users with reliable sources of information and encourages them to obtain additional information as needed.

[0908] Future predicted implementation

[0909] Step 1:

[0910] The user inputs historical time-series data.

[0911] Step 2:

[0912] The server receives the input data, preprocesses it, and splits it into training and test data.

[0913] Step 3:

[0914] The server uses a machine learning model to create a predictive model based on the training data.

[0915] Step 4:

[0916] The server uses the created predictive model to predict future data based on the test data.

[0917] Step 5:

[0918] The server activates the emotion engine and recognizes the user's emotional state.

[0919] Step 6:

[0920] The server uses the recognized emotion information to adjust the prediction results and adapt them to the user's emotions.

[0921] Step 7:

[0922] The device provides the user with prediction results and suggests necessary actions.

[0923] A Embodiment of World Peace

[0924] Step 1:

[0925] The user enters the content of their social media post.

[0926] Step 2:

[0927] The terminal receives the entered text and activates the sentiment analysis model.

[0928] Step 3:

[0929] The device uses an emotion analysis model to analyze the emotional state of the text.

[0930] Step 4:

[0931] Based on the results of the sentiment analysis, the device creates recommended actions to promote peaceful dialogue.

[0932] Step 5:

[0933] The server activates the emotion engine and recognizes the user's emotional state.

[0934] Step 6:

[0935] The server adjusts recommended actions based on emotional information, adapting to the user's emotions.

[0936] Step 7:

[0937] The server provides the user with a final recommended action.

[0938] This enables users to achieve highly accurate translations, access to reliable information, predict the future, and engage in peaceful, emotion-based dialogue. By leveraging the emotion engine, the user experience is further enhanced, and the system's practicality is increased.

[0939] (Example 2)

[0940] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0941] In today's internet environment, there is a demand for communication that transcends language barriers, the elimination of inaccurate information, future predictions, and responses based on emotions. However, conventional systems struggle to meet these multiple needs simultaneously, and a particular challenge is the lack of consideration for users' emotional information. Specifically, translation results, search results, and future predictions do not adapt to the user's emotional state, resulting in a low quality of user experience. Furthermore, without utilizing emotional information, it becomes difficult to alleviate user anxiety and stress, and to ensure the accuracy and reliability of the information provided.

[0942] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0943] In this invention, the server includes natural language processing means for translating input text from one language to another, means for tokenizing and decoding the text using a translation model, sentiment engine means for adjusting the translation result based on the user's sentiment, and output means for providing the translation result to the user. This makes it possible to provide an appropriate translation result that is in line with the user's sentiment.

[0944] Furthermore, it includes a search engine interface for inputting search queries on a specific topic and obtaining reliable information sources, an analysis means for filtering reliable information from the obtained information sources, an emotion engine means for adjusting the filtered information based on the user's emotions, and an output means for providing that information to the user. This makes it possible to provide reliable information tailored to the user's emotional state.

[0945] Furthermore, it includes a machine learning model for predicting future trends based on past time-series data, a data processing means for splitting input data into training data and test data, a prediction means for predicting future data using the trained model, an emotion engine means for adjusting the prediction results based on the user's emotions, and an output means for providing the prediction results to the user. This enables future predictions based on the user's emotional information, making it possible to provide more appropriate action suggestions.

[0946] "Natural language processing means" refers to technologies that analyze input text and convert human language into a format that machines can understand.

[0947] A "translation model" is an algorithm and computational model for converting text from one language to another.

[0948] "Tokenization" is the process of dividing input text into units of words or phrases.

[0949] "Decoding" is the process of reconstructing generated tokens and returning them to a human-understandable format as a result of translation and analysis.

[0950] An "emotional engine" is a technology that recognizes a user's emotional information and adjusts its response based on that information.

[0951] "Output means" refers to interfaces or devices used to provide text or information to users.

[0952] A "search engine interface means" is a technology for searching for information on the internet and obtaining results.

[0953] "Analysis means" refers to techniques for analyzing acquired information and extracting or filtering necessary data.

[0954] A "machine learning model" is a statistical algorithm that learns from past data and uses it to make predictions and analyses of the future.

[0955] "Data processing means" refers to technologies for pre-processing input data and converting it into the required format.

[0956] "Predictive tools" are techniques that use trained models to predict future data and trends.

[0957] This invention relates to a system that provides diverse functions using technologies such as natural language processing, emotion recognition, and machine learning. The specific forms for implementing each function are described below.

[0958] Translation Function Embodiment

[0959] In this system, the user inputs the text to be translated, along with the source and target languages. Specifically, the user inputs the text and selects the language setting through the terminal's interface. The terminal receives the input information and analyzes the text using natural language processing. Next, a translation model (e.g., a common translation API) is used to tokenize the text and perform the translation. The translation result is decoded and converted into a human-readable format. The server activates an emotion engine to recognize the user's emotions and performs emotion adjustments appropriate to the translation result. Finally, the adjusted translation result is provided to the user.

[0960] Specific example

[0961] For example, if a user wants to translate the Japanese text "こんにちは" (konnichiwa) into English, the system receives "こんにちは" as input and translates it to "Hello" based on a translation model (e.g., a general-purpose translation API). Then, using an emotion engine (e.g., a general-purpose emotion recognition API), it adjusts the translation based on the user's emotional state, such as "Hello, how can I assist you today?", and provides it to the user.

[0962] Example of a prompt

[0963] "Translate the following Japanese text 'こんにちは' into English and adjust it to suit the user's mood."

[0964] Implementation of sharing correct knowledge

[0965] In this system, the user inputs the topic they want to verify as a query. The server receives the input query and collects relevant information from the internet through a search engine interface (e.g., a general-purpose search API). It filters the retrieved search results to identify reliable sources and extracts specific URLs. Next, it uses an emotion engine to adjust the reliability of the information to match the user's emotional state. Finally, it provides reliable information to the user via their device.

[0966] Specific example

[0967] For example, when searching for information about "global warming," the system uses a general-purpose search API to collect relevant information and obtain links to Wikipedia and government websites. Then, it uses an emotion engine (e.g., a general-purpose emotion recognition API) to adapt that information to the user's emotional state and present it accordingly.

[0968] Example of a prompt

[0969] "Search for reliable information on 'global warming' and provide the most relevant information tailored to the user's emotions."

[0970] Future predicted implementation

[0971] In this system, the user inputs historical time-series data. The server receives the input data, preprocesses it, and splits it into training and test data. The software used includes, for example, general data processing libraries and machine learning libraries. Next, a machine learning model (for example, a general-purpose machine learning library) is used to train the data and create a predictive model. The prediction results are adjusted based on the user's emotional state using an emotion engine, and finally, the prediction results are provided to the user via a terminal.

[0972] Specific example

[0973] For example, when a user inputs past sales data, the system preprocesses it using a data processing library and then uses a machine learning library to predict future sales. Afterward, an emotion engine (e.g., a general-purpose emotion recognition API) is used to adapt the prediction results to the user's emotional state and provide them to the user.

[0974] Example of a prompt

[0975] "Based on past sales data, we predict future sales and adjust them according to user sentiment."

[0976] A Embodiment of World Peace

[0977] In this system, the user inputs the content of a social media post. The terminal receives the input text and activates a sentiment analysis model (e.g., a general-purpose sentiment analysis API). Using the sentiment analysis model, the emotional state of the text is analyzed, and based on the results, recommended actions to promote peaceful dialogue are generated. Next, the server uses a sentiment engine to adjust the optimal recommended actions, taking into account the user's emotional state. Finally, the generated recommended actions are provided to the user through the terminal.

[0978] Specific example

[0979] For example, in the case of a positive post like "I love this!" and a negative post like "This is so frustrating!", the system uses a sentiment analysis API to analyze the emotions in each post and then uses an emotion engine to suggest an appropriate way to interact with the user.

[0980] Example of a prompt

[0981] "Analyze the sentiment of social media posts and propose ways to promote peaceful dialogue based on user sentiment."

[0982] As a result, the system provided by the present invention can perform multi-functional responses that take into account the user's emotional information, enabling translation, elimination of inaccurate information, future prediction, and promotion of emotion-based dialogue.

[0983] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0984] Translation Function Embodiment

[0985] Program processing

[0986] Step 1:

[0987] The user inputs the text to be translated, along with the source and target languages. The user sets the text "Hello" and the source language to Japanese and the target language to English via the device's interface. Input is done using text boxes and dropdown lists.

[0988] Input: Text "Hello", Source language "Japanese", Target language "English"

[0989] Output: Input information is sent.

[0990] Step 2:

[0991] The terminal receives the input information and analyzes the text using natural language processing techniques. Here, the language of the text is confirmed, and preprocessing is performed, including analysis of words and grammar.

[0992] Input: User input information

[0993] Data processing: Text preprocessing (tokenization, stop word removal, etc.)

[0994] Output: Analysis results (tokenized text)

[0995] Step 3:

[0996] The device sends the analyzed text to a translation model for translation. The translation model uses, for example, a common translation API to convert the tokenized text into the target language.

[0997] Input: Parsed text (tokens)

[0998] Data processing: Call the translation API and perform the translation.

[0999] Output: Translation result (English text "Hello")

[1000] Step 4:

[1001] The terminal receives the translation result, decodes it, and converts it into a human-readable format. Here, it reconstructs another token and formats it into the appropriate language format.

[1002] Input: English token "Hello"

[1003] Data processing: Decode and format tokens.

[1004] Output: Formatted English text "Hello"

[1005] Step 5:

[1006] The server activates the emotion engine, analyzes the translation results, and makes adjustments based on the user's emotional state. Here, sentiment analysis is performed based on the user's previous input and history, and the translation results are adjusted accordingly.

[1007] Input: Translation result "Hello"

[1008] Data processing: Analysis and adjustment using an emotion engine.

[1009] Output: Adjusted translation result: "Hello, how can I assist you today?"

[1010] Step 6:

[1011] The server sends the final translation result to the terminal, which then displays the result to the user. The user's screen displays text that has been adjusted based on their emotions.

[1012] Input: Adjusted translation result

[1013] Output: Translation results displayed to the user

[1014] ---

[1015] Implementation of sharing correct knowledge

[1016] Program processing

[1017] Step 1:

[1018] The user enters a search query related to a specific topic. For example, if they want to find information about "global warming," they would enter "global warming" in the search box.

[1019] Input: Search query "global warming"

[1020] Output: Input query

[1021] Step 2:

[1022] The server receives the input query and collects relevant information from the internet through the search engine interface. For example, it might use a general-purpose search API to retrieve information.

[1023] Input: Search query

[1024] Data processing: Call the search API and collect relevant information.

[1025] Output: Search Results

[1026] Step 3:

[1027] The server filters the collected search results to identify reliable sources. It uses algorithms to evaluate reliability, for example, to extract links to Wikipedia or official government websites.

[1028] Input: Search Results

[1029] Data processing: Reliability assessment and filtering

[1030] Output: Links to reliable sources

[1031] Step 4:

[1032] The server activates the emotion engine and adjusts the reliability of the information according to the user's emotional state. Here, the method and content of information presentation are adapted based on the user's emotional state.

[1033] Input: Links to reliable sources

[1034] Data processing: Analysis and adjustment using an emotion engine.

[1035] Output: Adjusted information

[1036] Step 5:

[1037] The device provides the user with the finalized information and prompts them to obtain additional information as needed.

[1038] Input: Adjusted information

[1039] Output: Information displayed to the user

[1040] ---

[1041] Future predicted implementation

[1042] Program processing

[1043] Step 1:

[1044] Users input historical time-series data. For example, they might upload historical sales data as a CSV file.

[1045] Input: CSV file, historical sales data

[1046] Output: Uploaded data

[1047] Step 2:

[1048] The server receives the input data, preprocesses it, and splits it into training and test data. For example, it performs data cleaning and normalization before splitting the data into training and test sets.

[1049] Input: Uploaded data

[1050] Data processing: Data cleaning, normalization, splitting

[1051] Output: Training data, Test data

[1052] Step 3:

[1053] The server uses training data to train a machine learning model. Here, for example, a general-purpose machine learning library is used to create a model for future prediction.

[1054] Input: Training data

[1055] Data processing: Model training

[1056] Output: Trained predictive model

[1057] Step 4:

[1058] The server uses a trained model to predict future data based on test data. The model is used to predict future sales data, for example.

[1059] Input: Test data

[1060] Data processing: Predicting future data

[1061] Output: Prediction result

[1062] Step 5:

[1063] The server uses an emotion engine to adjust prediction results based on the user's emotional state. For example, it might add encouraging comments to negative prediction results.

[1064] Input: Prediction result

[1065] Data processing: Analysis and adjustment using an emotion engine.

[1066] Output: Adjusted prediction results

[1067] Step 6:

[1068] The device provides the user with the final prediction results and suggests necessary actions.

[1069] Input: Adjusted prediction result

[1070] Output: Prediction results displayed to the user

[1071] ---

[1072] A Embodiment of World Peace

[1073] Program processing

[1074] Step 1:

[1075] The user enters the content of their social media post. For example, they might enter a positive post like "I love this!" in the input field.

[1076] Input: Social media post content

[1077] Output: Input text

[1078] Step 2:

[1079] The terminal receives the entered text and activates the sentiment analysis model. Here, a general-purpose sentiment analysis API is used.

[1080] Input: User's post content

[1081] Data processing: Calling the emotion analysis API

[1082] Output: Sentiment analysis results

[1083] Step 3:

[1084] The device uses an emotion analysis model to analyze the emotional state of the text. For example, it obtains positive, negative, and neutral emotion scores.

[1085] Input: User's post content

[1086] Data processing: Calculation of sentiment score

[1087] Output: Emotional state score

[1088] Step 4:

[1089] The device generates recommended actions to promote peaceful dialogue based on sentiment analysis results. For example, it might suggest "Share more!" for positive posts.

[1090] Input: Emotional state score

[1091] Data processing: Generating recommended actions

[1092] Output: Recommended Action

[1093] Step 5:

[1094] The server uses an emotion engine to adjust the optimal recommended action based on the user's emotional state. For example, it might take into account past posting history.

[1095] Input: Recommended Action

[1096] Data processing: Analysis and adjustment using an emotion engine.

[1097] Output: Optimized Recommended Actions

[1098] Step 6:

[1099] The server sends the final recommended action to the terminal, which then provides it to the user.

[1100] Input: Optimized Recommended Actions

[1101] Output: Recommended actions displayed to the user

[1102] (Application Example 2)

[1103] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[1104] Conventional systems lack the ability to translate, provide information, or predict future trends that take into account user emotions, resulting in insufficient individual optimization of the user experience. Furthermore, e-commerce sites are currently unable to provide appropriate product recommendations based on user emotional states, limiting their sales promotion effectiveness.

[1105] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes natural language processing means for translating input text from one language to another, means for tokenizing and decoding the text using a translation model, sentiment analysis means for analyzing the user's sentiment information, sentiment engine means for adjusting the translation result based on the analyzed sentiment information, and output means for providing the translation result to the user. This makes it possible to provide individually optimized translation results based on the user's sentiment information.

[1106] "Natural language processing means" refers to technologies that analyze input text and perform specific processing.

[1107] A "translation model" is an algorithm that converts text from one language to another.

[1108] "Tokenization" is the process of analyzing text and dividing it into the smallest units of words or phrases.

[1109] "Decoding methods" refer to techniques for reconstructing tokenized text into natural language.

[1110] "Emotional analysis methods" refer to technologies that analyze a user's emotional information from input text and data.

[1111] An "emotional engine" is a technology that adjusts the system's response and output based on analyzed emotional information.

[1112] "Output means" refers to the technologies and devices that a system uses to display results to the user.

[1113] A "search engine interface" is a system for collecting information from the internet.

[1114] "Reliable information" refers to verified data obtained from trustworthy sources.

[1115] A "machine learning model" is an algorithm that learns patterns from past data and uses them to predict the future.

[1116] "Data processing means" refers to techniques that preprocess input data and optimize it for analysis and prediction.

[1117] A "predictive tool" is a technique that uses trained machine learning models to estimate future data.

[1118] System Overview

[1119] This invention is a system that analyzes user emotional information and optimizes translation results, information provision, future predictions, etc., based on that analysis. This system includes the following main components.

[1120] Hardware and software

[1121] 1. Hardware: Smartphones are primarily used. Smartphones provide an interface for users to input text and have a display that shows analysis results and recommendations.

[1122] 2. Software: The Python language and TextBlob library are used for sentiment analysis and natural language processing. The server implements a sentiment engine, natural language processing, data processing, and machine learning models.

[1123] Processing Overview

[1124] 1. Sentiment Analysis: The text entered by the user is analyzed using the TextBlob library to calculate an emotion score. Based on the emotion score, the user's emotional state is understood.

[1125] 2. Natural Language Processing: Natural language processing tools are used to translate text entered by the user from one language to another. The translation model performs tokenization and decoding between languages.

[1126] 3. Emotion Engine: Based on the analyzed emotional information, it adjusts the translation results and information provided. For example, if the emotional state is relaxed, it recommends relaxation-related products.

[1127] Specific example

[1128] (Specific example 1)

[1129] Suppose a user enters "I've been feeling really stressed out lately." This input text is then analyzed by the TextBlob library to calculate a negative emotion score, which is then categorized as "stress relief" by the emotion analysis tool. Next, the emotion engine is activated, and products such as stress balls or therapy sessions are recommended from the stress relief category.

[1130] (Specific example 2)

[1131] If a user enters "I'm feeling great today!", based on this positive emotion score, an aroma diffuser or relaxing tea will be recommended from the relaxation category.

[1132] Example of a prompt

[1133] This is an example where a user enters "I'm feeling overwhelmed" as their current emotional state, and relaxation or stress-relief products are recommended based on their emotional score. The product list is as follows:

[1134] Relaxation products: Aroma diffuser, Relaxing tea, Yoga mat

[1135] Stress relief products: Stress Ball, Meditation App Subscription, Therapy Sessions

[1136] If the emotional analysis score is positive, relaxation products will be recommended; if it is negative, stress relief products will be recommended.

[1137] Effects of implementation

[1138] This invention enables the provision of appropriate information and product recommendations based on the user's emotional state, thereby providing a more personalized user experience. Furthermore, it is expected to have a positive impact on sales promotion on e-commerce sites.

[1139] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1140] Step 1:

[1141] The terminal receives text input from the user. This input text reflects the user's emotional state. This text serves as the starting point for system processing. For example, the input data might be "I've been feeling really stressed out lately."

[1142] Step 2:

[1143] The terminal sends the received text to the server. The server receives the text and analyzes the sentiment information of the input text using sentiment analysis tools. It calculates the sentiment score of the text using the TextBlob library. The sentiment score is output as the calculation result. For example, a negative sentiment score may be obtained.

[1144] Step 3:

[1145] The server determines the emotional category of the text based on its emotional score. The emotional analysis tool classifies the text into the appropriate emotional category, such as "relaxation" or "stress relief." The classification result is then passed on to the next processing step. For example, if the emotional score is negative, it is classified into the "stress relief" category.

[1146] Step 4:

[1147] The server activates an emotion engine and selects the most suitable product for the user based on the analyzed emotion information. It extracts products matching the emotion category from the product list and generates a recommendation list. Random sampling or predefined rules are used to generate the recommendation list. For example, stress balls and therapy sessions might be extracted from the "stress relief" category.

[1148] Step 5:

[1149] The server sends the generated recommendation list to the terminal. The terminal displays the received recommendation list to the user. Specifically, information about the recommended products is visualized on the display device. For example, detailed information about "stress balls" and "therapy sessions" is displayed.

[1150] Step 6:

[1151] Users can make selection and purchase actions based on the displayed recommended products. The system records user selections and uses them to improve future recommendation algorithms.

[1152] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1153] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1154] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[1155] [Third Embodiment]

[1156] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[1157] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[1158] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1159] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[1160] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1161] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1162] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1163] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1164] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1165] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1166] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1167] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[1168] This invention relates to a system for performing translation, filtering out inaccurate information, predicting the future, and processing emotions. Specific embodiments thereof are described below.

[1169] Translation Function Embodiment

[1170] Program processing

[1171] 1. The user enters the text they want to translate, along with the source language and the target language.

[1172] 2. The terminal receives the input information and analyzes the text using natural language processing tools.

[1173] 3. Use a translation model to tokenize the text and perform the translation.

[1174] 4. The terminal decodes the translation result and converts it into a format that humans can understand.

[1175] 5. The server provides the user with the final translation result.

[1176] Specific example

[1177] For example, if a user wants to translate the Japanese text "こんにちは" into English, the system receives "こんにちは," translates it to "Hello" based on its translation model, and provides it to the user.

[1178] Implementation of sharing correct knowledge

[1179] Program processing

[1180] 1. The user enters a question about a specific topic.

[1181] 2. The server uses a search engine interface to obtain relevant information from the internet.

[1182] 3. The server performs analysis to filter reliable information from the acquired information sources.

[1183] 4. The device provides users with filtered, reliable information.

[1184] Specific example

[1185] For example, when searching for information about "global warming," the system retrieves links to Wikipedia and government-related websites and provides them to the user.

[1186] Future predicted implementation

[1187] Program processing

[1188] 1. The user inputs historical time-series data.

[1189] 2. The server uses data processing equipment to split the input data into training data and test data.

[1190] 3. Train a machine learning model on the data and create a predictive model.

[1191] 4. The server uses the created model to predict future data.

[1192] 5. The device provides the user with the prediction results.

[1193] Specific example

[1194] For example, when a user inputs past sales data, the system uses that data to predict future sales and provides the results to the user.

[1195] A Embodiment of World Peace

[1196] Program processing

[1197] 1. The user enters the content of their social media post.

[1198] 2. The terminal uses an emotion analysis model to analyze the emotion of the input text.

[1199] 3. The terminal uses a recommended action generation mechanism based on the sentiment analysis results to create suggestions to promote peaceful dialogue.

[1200] 4. The server provides the user with the recommended actions that have been created.

[1201] Specific example

[1202] For example, in the case of a positive post like "I love this!" and a negative post like "This is so frustrating!", the system analyzes the emotions in each post, determines whether they are "positive" or "negative," and suggests actions to the user to facilitate appropriate dialogue.

[1203] These embodiments enable users to achieve highly accurate translation, access to reliable information, predict the future, and engage in peaceful, emotion-based dialogue.

[1204] The following describes the processing flow.

[1205] Translation Function Embodiment

[1206] Step 1:

[1207] The user enters the text to be translated, the source language, and the target language.

[1208] Step 2:

[1209] The terminal receives input from the user and activates natural language processing for translation.

[1210] Step 3:

[1211] The terminal tokenizes the source text and converts it into a format compatible with the language model.

[1212] Step 4:

[1213] The terminal inputs tokenized text into the translation model and performs the translation.

[1214] Step 5:

[1215] The device decodes the translation result and converts it into a natural language format that the user can understand.

[1216] Step 6:

[1217] The server provides the user with the final translation result.

[1218] Implementation of sharing correct knowledge

[1219] Step 1:

[1220] The user enters the topic they want to investigate as a query.

[1221] Step 2:

[1222] The server receives the input query and collects relevant information from the internet through the search engine interface.

[1223] Step 3:

[1224] The server filters the collected search results to identify reliable sources and extracts specific URLs.

[1225] Step 4:

[1226] The device provides users with reliable sources of information and encourages them to obtain additional information as needed.

[1227] Future predicted implementation

[1228] Step 1:

[1229] The user inputs historical time-series data.

[1230] Step 2:

[1231] The server receives the input data, preprocesses it, and splits it into training data and test data.

[1232] Step 3:

[1233] The server uses a machine learning model to create a predictive model based on the training data.

[1234] Step 4:

[1235] The server uses the created predictive model to predict future data based on the test data.

[1236] Step 5:

[1237] The device provides the user with prediction results and suggests necessary actions.

[1238] A Embodiment of World Peace

[1239] Step 1:

[1240] The user enters the content of their social media post.

[1241] Step 2:

[1242] The terminal receives the entered text and activates the sentiment analysis model.

[1243] Step 3:

[1244] The device uses an emotion analysis model to analyze the emotional state of the text.

[1245] Step 4:

[1246] Based on the results of the sentiment analysis, the device creates recommended actions to promote peaceful dialogue.

[1247] Step 5:

[1248] The server provides the user with the generated recommended actions.

[1249] This allows users to effectively translate text, obtain reliable information, predict the future, and engage in emotion-based conversations.

[1250] (Example 1)

[1251] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1252] In recent years, the importance of global communication has been steadily increasing, but handling multiple different languages ​​requires advanced translation technology. Furthermore, there is a need for systems that can quickly obtain reliable information, predict future trends, and facilitate emotion-based dialogue. To meet these needs, systems that efficiently combine highly accurate natural language processing technologies and machine learning models are essential.

[1253] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1254] In this invention, the server includes natural language processing means for translating input text from one language to another; processing means for tokenizing and decoding the text using a machine learning translation model; output means for providing the translation results to the user; search engine interface means for inputting search queries on a specific topic and obtaining relevant information; reliability evaluation means for filtering reliable information from the obtained information sources; output means for providing the filtered information to the user; machine learning prediction model means for predicting future data based on past time series data; data processing means for splitting input data into training data and test data; prediction means for predicting future data using the trained model; output means for providing the prediction results to the user; sentiment analysis means for inputting social media posts and analyzing the sentiment of the text; recommendation action generation means for promoting peaceful dialogue based on the sentiment analysis results; and output means for providing the generated recommendation actions to the user. This enables highly accurate translation, acquisition of reliable information, future prediction, and peaceful dialogue based on sentiment.

[1255] "Natural language processing means" refers to technologies that analyze input text and enable operations such as translation and sentiment analysis.

[1256] A "machine learning translation model" is a model trained using a large amount of data, and it has the ability to translate text from one language to another.

[1257] "Processing means" refers to an algorithm or program for performing a specific task, and in this invention, it performs text tokenization and decoding.

[1258] "Output means" refers to interfaces or devices for providing processing results to the user, and have functions to display translation results, search results, prediction results, etc.

[1259] A "search engine interface means" is a means of linking with a search engine for obtaining information on the internet.

[1260] A "reliability evaluation method" is a system that executes algorithms and evaluation criteria to select highly reliable information from acquired information sources.

[1261] A "machine learning prediction model" is a prediction model trained using past data, which is used to predict future data.

[1262] "Data processing means" refers to means for dividing input data into training data and test data.

[1263] A "predictive tool" is a means of using a trained model to predict future data.

[1264] "Sentiment analysis techniques" are technologies used to analyze the emotions in text and determine whether they are positive or negative.

[1265] A "recommended action generation method" is a means of generating actions and suggestions to promote peaceful dialogue based on the results of sentiment analysis.

[1266] This invention relates to a system that utilizes natural language processing technology and machine learning models to achieve translation functionality, reliable information acquisition, future prediction, and sentiment analysis and peaceful dialogue on social media. Specific embodiments thereof are described below.

[1267] Translation function

[1268] Specific Embodiments

[1269] 1. The user inputs text through a translation interface and specifies the source and target languages. This interface is implemented as a web browser or mobile application.

[1270] 2. The terminal uses the Python NLTK library to parse and tokenize the input text.

[1271] 3. The device calls the Google Translate API or Microsoft Translator to translate the tokenized text.

[1272] 4. The device decodes the translation result and converts it into a format that the user can understand.

[1273] 5. The server displays the final translation result on the interface and provides it to the user.

[1274] Specific example

[1275] If a user wants to translate the Japanese text "こんにちは" into English, the system receives "こんにちは," uses the Google Translate API to translate it to "Hello," and provides the result to the user.

[1276] Sharing correct knowledge

[1277] Specific Embodiments

[1278] 1. The user enters a question about a specific topic. This question is entered via a web browser or mobile application.

[1279] 2. The server uses the Google Search API and other search engine interfaces to retrieve relevant information from the internet.

[1280] 3. The server runs a reliability evaluation algorithm and filters out reliable information from the retrieved sources.

[1281] 4. The device provides filtered information to the user.

[1282] Specific example

[1283] When searching for information related to "global warming," the system uses the Google Search API to retrieve relevant information, selects the most reliable information, and provides it to the user.

[1284] Future predictions

[1285] Specific Embodiments

[1286] 1. Users upload historical time-series data in CSV files or other digital formats.

[1287] 2. The server uses the Python pandas library to split the input data into training and test data.

[1288] 3. The server uses machine learning frameworks such as TensorFlow and Scikit-learn to train data and create predictive models.

[1289] 4. The server uses the trained model to predict future data.

[1290] 5. The device provides the user with prediction results in graph or table format.

[1291] Specific example

[1292] When a user inputs past sales data, the system uses a TensorFlow LSTM model based on that data to predict future sales and provides the results to the user.

[1293] World peace

[1294] Specific Embodiments

[1295] 1. The user enters the content of their social media post in text format.

[1296] 2. The terminal uses IBM Watson or Microsoft Azure sentiment analysis APIs to analyze the sentiment of the input text.

[1297] 3. Based on the sentiment analysis results, the device uses a recommended action generation algorithm to create suggestions to promote peaceful dialogue.

[1298] 4. The server provides the user with the recommended actions that have been created.

[1299] Specific example

[1300] If a negative post such as "This is so frustrating!" is submitted, IBM Watson will be used to determine that the sentiment is "negative" and generate and provide suggestions to the user to facilitate appropriate dialogue.

[1301] The above describes embodiments of the present invention. This system enables users to achieve highly accurate translation, obtain reliable information, predict the future, and engage in peaceful, emotion-based dialogue.

[1302] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1303] Translation Function Embodiment

[1304] Processing steps

[1305] Step 1:

[1306] The user accesses the system interface and enters the text to be translated, along with the source and target languages. For example, they might enter "こんにちは" (konnichiwa) and specify a translation from Japanese to English. This input is sent to the server in text format.

[1307] Step 2:

[1308] The terminal parses and tokenizes the input text received from the server using Python's NLTK library. The input data is "hello," and through parsing and tokenization, the internal structure of the text is understood. The output is the tokenized text.

[1309] Step 3:

[1310] The device sends tokenized text to the Google Translate API or Microsoft Translator for translation. The input is the tokenized text "こんにちは" (Konnichiwa), and the translation is performed by calling the API. The output is the translated text "Hello".

[1311] Step 4:

[1312] The terminal decodes the translation result returned from the API and converts it into a human-readable format. The input is the translated text "Hello," and the terminal performs the conversion from the encoded format to standard text format. The output is the decoded text "Hello."

[1313] Step 5:

[1314] The server provides the user with the final translation result. The input is the decoded text "Hello," which is returned to the interface for displaying it to the user. The output is the translation result displayed to the user.

[1315] Implementation of sharing correct knowledge

[1316] Processing steps

[1317] Step 1:

[1318] The user enters a question about a specific topic into the system. For example, they might enter, "Tell me about global warming." This input is sent to the server in text format.

[1319] Step 2:

[1320] The server uses the Google Search API to search for relevant information on the internet. The input is the question "Tell me about global warming," and the API is called to retrieve relevant information. The output is the retrieved search results.

[1321] Step 3:

[1322] The server executes a reliability evaluation algorithm to filter the retrieved information for the most reliable information. The input is the retrieved search results, to which the algorithm is applied for reliability evaluation. The output is the result of selecting only the most reliable information.

[1323] Step 4:

[1324] The terminal provides the user with filtered, reliable information. The input is reliable information, which is returned to the interface for displaying it to the user. The output is the reliable information displayed to the user.

[1325] Future prediction implementation

[1326] Processing steps

[1327] Step 1:

[1328] Users upload historical time-series data to the system as a CSV file. For example, they might upload sales data for the past year. This input is sent to the server in file format.

[1329] Step 2:

[1330] The server uses the Python pandas library to split the input data into training and test sets. The input is an uploaded CSV file, which is then divided into 80% training data and 20% test data. The output is the split dataset.

[1331] Step 3:

[1332] The server uses machine learning frameworks such as TensorFlow and Scikit-learn to train data and create predictive models. The input is training data, and the machine learning algorithm generates the predictive model. The output is the trained predictive model.

[1333] Step 4:

[1334] The server evaluates the performance of a model created using test data and predicts future data. The inputs are the test data and the predictive model; performance evaluation and future prediction are performed. The output is the predicted data.

[1335] Step 5:

[1336] The terminal provides prediction results to the user in graph or table format. The input is the predicted data, which is visualized and displayed on the interface. The output is the prediction result displayed to the user.

[1337] A Embodiment of World Peace

[1338] Processing steps

[1339] Step 1:

[1340] The user enters their social media post content into the system in text format. For example, they might enter "This is so frustrating!". This input is then sent to the server in text format.

[1341] Step 2:

[1342] The terminal uses IBM Watson and Microsoft Azure sentiment analysis APIs to analyze the sentiment of the input text. The input is a social media post, "This is so frustrating!", and the sentiment is analyzed by calling the sentiment analysis API. The output is the analyzed sentiment result.

[1343] Step 3:

[1344] The device applies a recommended action generation algorithm to promote peaceful dialogue based on the sentiment analysis results. The input is the sentiment analysis results, and the algorithm is used to generate an appropriate response. The output is the recommended action.

[1345] Step 4:

[1346] The server provides the user with the generated recommended actions. The input is the recommended actions, which are returned to the interface for displaying them to the user. The output is the recommended actions displayed to the user.

[1347] These specific processing steps enable users to achieve highly accurate translations, obtain reliable information, predict the future, and engage in peaceful, emotion-based dialogue.

[1348] (Application Example 1)

[1349] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1350] In many modern situations, communication between people who speak different languages, facilitating emotion-based dialogue, obtaining reliable information, and predicting the future based on diverse data are critical challenges. However, current systems lack an integrated approach to meet these requirements, particularly a lack of dialogue support incorporating real-time sentiment analysis. This makes efficient communication and appropriate decision-making difficult.

[1351] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1352] In this invention, the server includes natural language processing means for translating input text from one language to another, means for tokenizing and decoding the text using a translation model, output means for providing the translation results to the user, sentiment analysis means for analyzing the sentiment of the input text, and dialogue support means for facilitating peaceful dialogue based on the analysis results. This enables real-time sentiment analysis and dialogue support.

[1353] A "natural language processing system" is a means of analyzing input text and translating it from one language to another.

[1354] A "translation model" is a model for tokenizing text and decoding it into another language.

[1355] "Output means" refers to the means of providing translation results or analysis results to the user.

[1356] An "emotion analysis tool" is a means of analyzing the emotions in input text and identifying the type of emotion.

[1357] "Dialogue support methods" are means of creating proposals to promote peaceful dialogue based on the results of emotion analysis.

[1358] A "search engine interface" is a means of obtaining reliable information sources by entering a search query related to a specific topic.

[1359] "Analysis means" are means for filtering reliable information from acquired information sources.

[1360] A "machine learning model" is a model used to predict future trends based on past time-series data.

[1361] "Data processing means" refers to means for dividing input data into training data and test data.

[1362] A "predictive tool" is a means of predicting future data using a trained model.

[1363] This invention can be specifically implemented as follows.

[1364] First, the system receives text input from the user. This input text could be content to be translated, content requiring sentiment analysis, or a search query. Once the user enters the text, the terminal analyzes it using natural language processing tools. At this stage, the main software used includes "Transformers," a widely known natural language processing library.

[1365] Next, if translation is needed, the system uses a translation model to tokenize the text and perform the translation. The translation result is decoded and provided to the user. Open-source translation models such as "Hugging Face" are used.

[1366] Furthermore, the system uses sentiment analysis tools to analyze the sentiment of the input text. Sentiment analysis assigns specific sentiment labels (e.g., positive, negative, neutral). Based on these analysis results, dialogue support tools generate suggestions to facilitate peaceful dialogue.

[1367] When a search query is entered, the server uses a search engine interface to obtain reliable information sources and filters them using analysis tools. The reliable information is then provided to the user.

[1368] Furthermore, the system can predict future trends based on historical time-series data. The server divides the input data into training data and test data, and trains the data using a machine learning model. This allows it to predict future data and provide results to the user.

[1369] As a concrete example, suppose a security staff member detected the following statement at the scene:

[1370] “This situation is incredibly frustrating and makes me angry.”

[1371] In this case, the system analyzes the text and detects the emotion "NEGATIVE." It then suggests the following recommended action:

[1372] "Please try to engage in calm dialogue to alleviate the situation."

[1373] Examples of prompt statements include the following:

[1374] "A proposed system for emotional analysis and peaceful dialogue"

[1375] Analyze user sentiment based on their input text and provide appropriate dialogue actions.

[1376] User input: "This situation is incredibly frustrating and makes me angry."

[1377] In this way, the system provides an integrated approach to emotion analysis and dialogue support, enabling it to respond appropriately in real time.

[1378] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1379] Step 1:

[1380] The user provides input text to the terminal. The input can be text to be translated, text requiring sentiment analysis, or a search query. The terminal analyzes this input text using natural language processing. The analyzed text is then sent to the next processing step.

[1381] Step 2:

[1382] The device uses a translation model to tokenize the input text if it requires translation. Tokenization is the process of dividing each word or phrase in the text into individual tokens. The tokens are then sent to the translation model.

[1383] Step 3:

[1384] The translation model translates tokenized text into the specified language. The translation model uses a generative AI model to convert the token array into tokens in the other language. The resulting token array is decoded and converted into human-readable text. The decoded text is returned to the terminal.

[1385] Step 4:

[1386] The terminal provides the translated text to the user through an output device. The user receives the final translation result.

[1387] Step 5:

[1388] The terminal analyzes the sentiment of the input text using sentiment analysis means if the sentiment of the text should be analyzed. The sentiment analysis means analyzes the text and assigns a specific sentiment label such as positive, negative, or neutral. The analysis results and sentiment labels are identified and sent to the next processing step.

[1389] Step 6:

[1390] The device uses dialogue support mechanisms based on the analyzed emotion results to generate suggestions for peaceful dialogue. These dialogue support mechanisms create messages to alleviate negative emotions and messages to maintain a positive atmosphere for positive emotions. These suggestions are then provided to the user.

[1391] Step 7:

[1392] When a user enters a search query, the server uses a search engine interface to retrieve reliable information sources from the internet. These retrieved sources contain data corresponding to the search query.

[1393] Step 8:

[1394] The server uses analysis tools to filter reliable information from the acquired information sources. These filtering tools exclude unreliable information and extract only the reliable information. This filtered information is then provided to the user.

[1395] Step 9:

[1396] When historical time-series data is input, the server uses data processing tools to split the input data into training data and test data. The split data is then used to train a machine learning model.

[1397] Step 10:

[1398] The server uses machine learning models to predict future trends based on historical data. The predicted trend data includes estimates of future sales and demand. The prediction results are provided to the user.

[1399] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1400] This invention relates to a system for translation, filtering out inaccurate information, predicting the future, and processing emotions. Furthermore, the invention implements a form that combines this with an emotion engine that recognizes and utilizes the user's emotional information. Specific embodiments are described below.

[1401] Translation Function Embodiment

[1402] Program processing

[1403] 1. The user enters the text to be translated, the source language, and the target language.

[1404] 2. The terminal receives the input information and analyzes the text using natural language processing tools.

[1405] 3. Use a translation model to tokenize the text and perform the translation.

[1406] 4. The terminal decodes the translation result and converts it into a format that humans can understand.

[1407] 5. The server activates an emotion engine to recognize the user's emotions and performs emotion adjustments appropriate to the translation result.

[1408] 6. The server provides the user with the final translation result.

[1409] Specific example

[1410] For example, if a user wants to translate the Japanese text "こんにちは" (konnichiwa) into English, the system receives "こんにちは," translates it to "Hello" based on its translation model, and then provides it in a form that is adapted to the user's emotional state.

[1411] Implementation of sharing correct knowledge

[1412] Program processing

[1413] 1. The user enters the topic they want to verify as a query.

[1414] 2. The server receives the input query and collects relevant information from the internet through the search engine interface.

[1415] 3. The server filters the collected search results to identify reliable sources and extracts specific URLs.

[1416] 4. The server uses an emotion engine to recognize the user's emotions and adjusts the reliability of the information it provides according to the user's emotional state.

[1417] 5. The device provides users with reliable sources of information and encourages them to obtain additional information as needed.

[1418] Specific example

[1419] For example, when searching for information about "global warming," the system retrieves links from Wikipedia and government-related websites, and then selects and provides appropriate information based on the user's emotional state.

[1420] Future predicted implementation

[1421] Program processing

[1422] 1. The user inputs historical time-series data.

[1423] 2. The server receives the input data, preprocesses it, and splits it into training data and test data.

[1424] 3. Train a machine learning model on the data and create a predictive model.

[1425] 4. The server uses the created predictive model to predict future data based on the test data.

[1426] 5. The server uses an emotion engine to adjust prediction results based on the user's emotional information.

[1427] 6. The device provides the user with prediction results and suggests necessary actions.

[1428] Specific example

[1429] For example, when a user inputs past sales data, the system uses that data to predict future sales and provides the prediction results in a way that is tailored to the user's emotional state.

[1430] A Embodiment of World Peace

[1431] Program processing

[1432] 1. The user enters the content of their social media post.

[1433] 2. The terminal receives the entered text and activates the sentiment analysis model.

[1434] 3. The device uses an emotion analysis model to analyze the emotional state of the text.

[1435] 4. Based on the results of the sentiment analysis, the terminal creates specific suggestions using a recommended action generation mechanism to promote peaceful dialogue.

[1436] 5. The server uses an emotion engine to adjust the optimal recommended action, taking into account the user's emotional state.

[1437] 6. The server provides the user with the generated recommended actions.

[1438] Specific example

[1439] For example, in the case of a positive post like "I love this!" and a negative post like "This is so frustrating!", the system analyzes the emotions in each post and suggests appropriate actions to facilitate dialogue based on the user's emotional state.

[1440] This enables users to achieve highly accurate translations, access to reliable information, predict the future, and engage in peaceful, emotion-based dialogue. Leveraging emotion-driven tools further enhances the user experience and increases the system's practicality.

[1441] The following describes the processing flow.

[1442] Translation Function Embodiment

[1443] Step 1:

[1444] The user enters the text to be translated, the source language, and the target language.

[1445] Step 2:

[1446] The terminal receives the input information and activates the natural language processing system.

[1447] Step 3:

[1448] The terminal tokenizes the source text and converts it into a format suitable for the translation model.

[1449] Step 4:

[1450] The terminal inputs tokenized text into the translation model and performs the translation.

[1451] Step 5:

[1452] The terminal decodes the translation result and converts it into a format that humans can understand.

[1453] Step 6:

[1454] The server activates the emotion engine and recognizes the user's emotional state.

[1455] Step 7:

[1456] The server uses the emotion information recognized by the emotion engine to adjust the translation results and adapt them to the emotions.

[1457] Step 8:

[1458] The server provides the user with the final translation result.

[1459] Implementation of sharing correct knowledge

[1460] Step 1:

[1461] The user enters the topic they want to investigate as a query.

[1462] Step 2:

[1463] The server receives the input query and collects relevant information from the internet through the search engine interface.

[1464] Step 3:

[1465] The server filters the collected search results to identify reliable sources and extracts specific URLs.

[1466] Step 4:

[1467] The server activates the emotion engine and recognizes the user's emotional state.

[1468] Step 5:

[1469] The server uses emotional information to adjust the reliability of the acquired information based on the user's emotional state.

[1470] Step 6:

[1471] The device provides users with reliable sources of information and encourages them to obtain additional information as needed.

[1472] Future predicted implementation

[1473] Step 1:

[1474] The user inputs historical time-series data.

[1475] Step 2:

[1476] The server receives the input data, preprocesses it, and splits it into training and test data.

[1477] Step 3:

[1478] The server uses a machine learning model to create a predictive model based on the training data.

[1479] Step 4:

[1480] The server uses the created predictive model to predict future data based on the test data.

[1481] Step 5:

[1482] The server activates the emotion engine and recognizes the user's emotional state.

[1483] Step 6:

[1484] The server uses the recognized emotion information to adjust the prediction results and adapt them to the user's emotions.

[1485] Step 7:

[1486] The device provides the user with prediction results and suggests necessary actions.

[1487] A Embodiment of World Peace

[1488] Step 1:

[1489] The user enters the content of their social media post.

[1490] Step 2:

[1491] The terminal receives the entered text and activates the sentiment analysis model.

[1492] Step 3:

[1493] The device uses an emotion analysis model to analyze the emotional state of the text.

[1494] Step 4:

[1495] Based on the results of the sentiment analysis, the device creates recommended actions to promote peaceful dialogue.

[1496] Step 5:

[1497] The server activates the emotion engine and recognizes the user's emotional state.

[1498] Step 6:

[1499] The server adjusts recommended actions based on emotional information, adapting to the user's emotions.

[1500] Step 7:

[1501] The server provides the user with a final recommended action.

[1502] This enables users to achieve highly accurate translations, access to reliable information, predict the future, and engage in peaceful, emotion-based dialogue. By leveraging the emotion engine, the user experience is further enhanced, and the system's practicality is increased.

[1503] (Example 2)

[1504] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1505] In today's internet environment, there is a demand for communication that transcends language barriers, the elimination of inaccurate information, future predictions, and responses based on emotions. However, conventional systems struggle to meet these multiple needs simultaneously, and a particular challenge is the lack of consideration for users' emotional information. Specifically, translation results, search results, and future predictions do not adapt to the user's emotional state, resulting in a low quality of user experience. Furthermore, without utilizing emotional information, it becomes difficult to alleviate user anxiety and stress, and to ensure the accuracy and reliability of the information provided.

[1506] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1507] In this invention, the server includes natural language processing means for translating input text from one language to another, means for tokenizing and decoding the text using a translation model, sentiment engine means for adjusting the translation result based on the user's sentiment, and output means for providing the translation result to the user. This makes it possible to provide an appropriate translation result that is in line with the user's sentiment.

[1508] Furthermore, it includes a search engine interface for inputting search queries on a specific topic and obtaining reliable information sources, an analysis means for filtering reliable information from the obtained information sources, an emotion engine means for adjusting the filtered information based on the user's emotions, and an output means for providing that information to the user. This makes it possible to provide reliable information tailored to the user's emotional state.

[1509] Furthermore, it includes a machine learning model for predicting future trends based on past time-series data, a data processing means for splitting input data into training data and test data, a prediction means for predicting future data using the trained model, an emotion engine means for adjusting the prediction results based on the user's emotions, and an output means for providing the prediction results to the user. This enables future predictions based on the user's emotional information, making it possible to provide more appropriate action suggestions.

[1510] "Natural language processing means" refers to technologies that analyze input text and convert human language into a format that machines can understand.

[1511] A "translation model" is an algorithm and computational model for converting text from one language to another.

[1512] "Tokenization" is the process of dividing input text into units of words or phrases.

[1513] "Decoding" is the process of reconstructing generated tokens and returning them to a human-understandable format as a result of translation and analysis.

[1514] An "emotional engine" is a technology that recognizes a user's emotional information and adjusts its response based on that information.

[1515] "Output means" refers to interfaces or devices used to provide text or information to users.

[1516] A "search engine interface means" is a technology for searching for information on the internet and obtaining results.

[1517] "Analysis means" refers to techniques for analyzing acquired information and extracting or filtering necessary data.

[1518] A "machine learning model" is a statistical algorithm that learns from past data and uses it to make predictions and analyses of the future.

[1519] "Data processing means" refers to technologies for pre-processing input data and converting it into the required format.

[1520] "Predictive tools" are techniques that use trained models to predict future data and trends.

[1521] This invention relates to a system that provides diverse functions using technologies such as natural language processing, emotion recognition, and machine learning. The specific forms for implementing each function are described below.

[1522] Translation Function Embodiment

[1523] In this system, the user inputs the text to be translated, along with the source and target languages. Specifically, the user inputs the text and selects the language setting through the terminal's interface. The terminal receives the input information and analyzes the text using natural language processing. Next, a translation model (e.g., a common translation API) is used to tokenize the text and perform the translation. The translation result is decoded and converted into a human-readable format. The server activates an emotion engine to recognize the user's emotions and performs emotion adjustments appropriate to the translation result. Finally, the adjusted translation result is provided to the user.

[1524] Specific example

[1525] For example, if a user wants to translate the Japanese text "こんにちは" (konnichiwa) into English, the system receives "こんにちは" as input and translates it to "Hello" based on a translation model (e.g., a general-purpose translation API). Then, using an emotion engine (e.g., a general-purpose emotion recognition API), it adjusts the translation based on the user's emotional state, such as "Hello, how can I assist you today?", and provides it to the user.

[1526] Example of a prompt

[1527] "Translate the following Japanese text 'こんにちは' into English and adjust it to suit the user's mood."

[1528] Implementation of sharing correct knowledge

[1529] In this system, the user inputs the topic they want to verify as a query. The server receives the input query and collects relevant information from the internet through a search engine interface (e.g., a general-purpose search API). It filters the retrieved search results to identify reliable sources and extracts specific URLs. Next, it uses an emotion engine to adjust the reliability of the information to match the user's emotional state. Finally, it provides reliable information to the user via their device.

[1530] Specific example

[1531] For example, when searching for information about "global warming," the system uses a general-purpose search API to collect relevant information and obtain links to Wikipedia and government websites. Then, it uses an emotion engine (e.g., a general-purpose emotion recognition API) to adapt that information to the user's emotional state and present it accordingly.

[1532] Example of a prompt

[1533] "Search for reliable information on 'global warming' and provide the most relevant information tailored to the user's emotions."

[1534] Future predicted implementation

[1535] In this system, the user inputs historical time-series data. The server receives the input data, preprocesses it, and splits it into training and test data. The software used includes, for example, general data processing libraries and machine learning libraries. Next, a machine learning model (for example, a general-purpose machine learning library) is used to train the data and create a predictive model. The prediction results are adjusted based on the user's emotional state using an emotion engine, and finally, the prediction results are provided to the user via a terminal.

[1536] Specific example

[1537] For example, when a user inputs past sales data, the system preprocesses it using a data processing library and then uses a machine learning library to predict future sales. Afterward, an emotion engine (e.g., a general-purpose emotion recognition API) is used to adapt the prediction results to the user's emotional state and provide them to the user.

[1538] Example of a prompt

[1539] "Based on past sales data, we predict future sales and adjust them according to user sentiment."

[1540] A Embodiment of World Peace

[1541] In this system, the user inputs the content of a social media post. The terminal receives the input text and activates a sentiment analysis model (e.g., a general-purpose sentiment analysis API). Using the sentiment analysis model, the emotional state of the text is analyzed, and based on the results, recommended actions to promote peaceful dialogue are generated. Next, the server uses a sentiment engine to adjust the optimal recommended actions, taking into account the user's emotional state. Finally, the generated recommended actions are provided to the user through the terminal.

[1542] Specific example

[1543] For example, in the case of a positive post like "I love this!" and a negative post like "This is so frustrating!", the system uses a sentiment analysis API to analyze the emotions in each post and then uses an emotion engine to suggest an appropriate way to interact with the user.

[1544] Example of a prompt

[1545] "Analyze the sentiment of social media posts and propose ways to promote peaceful dialogue based on user sentiment."

[1546] As a result, the system provided by the present invention can perform multi-functional responses that take into account the user's emotional information, enabling translation, elimination of inaccurate information, future prediction, and promotion of emotion-based dialogue.

[1547] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1548] Translation Function Embodiment

[1549] Program processing

[1550] Step 1:

[1551] The user inputs the text to be translated, along with the source and target languages. The user sets the text "Hello" and the source language to Japanese and the target language to English via the device's interface. Input is done using text boxes and dropdown lists.

[1552] Input: Text "Hello", Source language "Japanese", Target language "English"

[1553] Output: Input information is sent.

[1554] Step 2:

[1555] The terminal receives the input information and analyzes the text using natural language processing techniques. Here, the language of the text is confirmed, and preprocessing is performed, including analysis of words and grammar.

[1556] Input: User input information

[1557] Data processing: Text preprocessing (tokenization, stop word removal, etc.)

[1558] Output: Analysis results (tokenized text)

[1559] Step 3:

[1560] The device sends the analyzed text to a translation model for translation. The translation model uses, for example, a common translation API to convert the tokenized text into the target language.

[1561] Input: Parsed text (tokens)

[1562] Data processing: Call the translation API and perform the translation.

[1563] Output: Translation result (English text "Hello")

[1564] Step 4:

[1565] The terminal receives the translation result, decodes it, and converts it into a human-readable format. Here, it reconstructs another token and formats it into the appropriate language format.

[1566] Input: English token "Hello"

[1567] Data processing: Decode and format tokens.

[1568] Output: Formatted English text "Hello"

[1569] Step 5:

[1570] The server activates the emotion engine, analyzes the translation results, and makes adjustments based on the user's emotional state. Here, sentiment analysis is performed based on the user's previous input and history, and the translation results are adjusted accordingly.

[1571] Input: Translation result "Hello"

[1572] Data processing: Analysis and adjustment using an emotion engine.

[1573] Output: Adjusted translation result: "Hello, how can I assist you today?"

[1574] Step 6:

[1575] The server sends the final translation result to the terminal, which then displays the result to the user. The user's screen displays text that has been adjusted based on their emotions.

[1576] Input: Adjusted translation result

[1577] Output: Translation results displayed to the user

[1578] ---

[1579] Implementation of sharing correct knowledge

[1580] Program processing

[1581] Step 1:

[1582] The user enters a search query related to a specific topic. For example, if they want to find information about "global warming," they would enter "global warming" in the search box.

[1583] Input: Search query "global warming"

[1584] Output: Input query

[1585] Step 2:

[1586] The server receives the input query and collects relevant information from the internet through the search engine interface. For example, it might use a general-purpose search API to retrieve information.

[1587] Input: Search query

[1588] Data processing: Call the search API and collect relevant information.

[1589] Output: Search Results

[1590] Step 3:

[1591] The server filters the collected search results to identify reliable sources. It uses algorithms to evaluate reliability, for example, to extract links to Wikipedia or official government websites.

[1592] Input: Search Results

[1593] Data processing: Reliability assessment and filtering

[1594] Output: Links to reliable sources

[1595] Step 4:

[1596] The server activates the emotion engine and adjusts the reliability of the information according to the user's emotional state. Here, the method and content of information presentation are adapted based on the user's emotional state.

[1597] Input: Links to reliable sources

[1598] Data processing: Analysis and adjustment using an emotion engine.

[1599] Output: Adjusted information

[1600] Step 5:

[1601] The device provides the user with the finalized information and prompts them to obtain additional information as needed.

[1602] Input: Adjusted information

[1603] Output: Information displayed to the user

[1604] ---

[1605] Future predicted implementation

[1606] Program processing

[1607] Step 1:

[1608] Users input historical time-series data. For example, they might upload historical sales data as a CSV file.

[1609] Input: CSV file, historical sales data

[1610] Output: Uploaded data

[1611] Step 2:

[1612] The server receives the input data, preprocesses it, and splits it into training and test data. For example, it performs data cleaning and normalization before splitting the data into training and test sets.

[1613] Input: Uploaded data

[1614] Data processing: Data cleaning, normalization, splitting

[1615] Output: Training data, Test data

[1616] Step 3:

[1617] The server uses training data to train a machine learning model. Here, for example, a general-purpose machine learning library is used to create a model for future prediction.

[1618] Input: Training data

[1619] Data processing: Model training

[1620] Output: Trained predictive model

[1621] Step 4:

[1622] The server uses a trained model to predict future data based on test data. The model is used to predict future sales data, for example.

[1623] Input: Test data

[1624] Data processing: Predicting future data

[1625] Output: Prediction result

[1626] Step 5:

[1627] The server uses an emotion engine to adjust prediction results based on the user's emotional state. For example, it might add encouraging comments to negative prediction results.

[1628] Input: Prediction result

[1629] Data processing: Analysis and adjustment using an emotion engine.

[1630] Output: Adjusted prediction results

[1631] Step 6:

[1632] The device provides the user with the final prediction results and suggests necessary actions.

[1633] Input: Adjusted prediction result

[1634] Output: Prediction results displayed to the user

[1635] ---

[1636] A Embodiment of World Peace

[1637] Program processing

[1638] Step 1:

[1639] The user enters the content of their social media post. For example, they might enter a positive post like "I love this!" in the input field.

[1640] Input: Social media post content

[1641] Output: Input text

[1642] Step 2:

[1643] The terminal receives the entered text and activates the sentiment analysis model. Here, a general-purpose sentiment analysis API is used.

[1644] Input: User's post content

[1645] Data processing: Calling the emotion analysis API

[1646] Output: Sentiment analysis results

[1647] Step 3:

[1648] The device uses an emotion analysis model to analyze the emotional state of the text. For example, it obtains positive, negative, and neutral emotion scores.

[1649] Input: User's post content

[1650] Data processing: Calculation of sentiment score

[1651] Output: Emotional state score

[1652] Step 4:

[1653] The device generates recommended actions to promote peaceful dialogue based on sentiment analysis results. For example, it might suggest "Share more!" for positive posts.

[1654] Input: Emotional state score

[1655] Data processing: Generating recommended actions

[1656] Output: Recommended Action

[1657] Step 5:

[1658] The server uses an emotion engine to adjust the optimal recommended action based on the user's emotional state. For example, it might take into account past posting history.

[1659] Input: Recommended Action

[1660] Data processing: Analysis and adjustment using an emotion engine.

[1661] Output: Optimized Recommended Actions

[1662] Step 6:

[1663] The server sends the final recommended action to the terminal, which then provides it to the user.

[1664] Input: Optimized Recommended Actions

[1665] Output: Recommended actions displayed to the user

[1666] (Application Example 2)

[1667] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1668] Conventional systems lack the ability to translate, provide information, or predict future trends that take into account user emotions, resulting in insufficient individual optimization of the user experience. Furthermore, e-commerce sites are currently unable to provide appropriate product recommendations based on user emotional states, limiting their sales promotion effectiveness.

[1669] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes natural language processing means for translating input text from one language to another, means for tokenizing and decoding the text using a translation model, sentiment analysis means for analyzing the user's sentiment information, sentiment engine means for adjusting the translation result based on the analyzed sentiment information, and output means for providing the translation result to the user. This makes it possible to provide individually optimized translation results based on the user's sentiment information.

[1670] "Natural language processing means" refers to technologies that analyze input text and perform specific processing.

[1671] A "translation model" is an algorithm that converts text from one language to another.

[1672] "Tokenization" is the process of analyzing text and dividing it into the smallest units of words or phrases.

[1673] "Decoding methods" refer to techniques for reconstructing tokenized text into natural language.

[1674] "Emotional analysis methods" refer to technologies that analyze a user's emotional information from input text and data.

[1675] An "emotional engine" is a technology that adjusts the system's response and output based on analyzed emotional information.

[1676] "Output means" refers to the technologies and devices that a system uses to display results to the user.

[1677] A "search engine interface" is a system for collecting information from the internet.

[1678] "Reliable information" refers to verified data obtained from trustworthy sources.

[1679] A "machine learning model" is an algorithm that learns patterns from past data and uses them to predict the future.

[1680] "Data processing means" refers to techniques that preprocess input data and optimize it for analysis and prediction.

[1681] A "predictive tool" is a technique that uses trained machine learning models to estimate future data.

[1682] System Overview

[1683] This invention is a system that analyzes user emotional information and optimizes translation results, information provision, future predictions, etc., based on that analysis. This system includes the following main components.

[1684] Hardware and software

[1685] 1. Hardware: Smartphones are primarily used. Smartphones provide an interface for users to input text and have a display that shows analysis results and recommendations.

[1686] 2. Software: The Python language and TextBlob library are used for sentiment analysis and natural language processing. The server implements a sentiment engine, natural language processing, data processing, and machine learning models.

[1687] Processing Overview

[1688] 1. Sentiment Analysis: The text entered by the user is analyzed using the TextBlob library to calculate an emotion score. Based on the emotion score, the user's emotional state is understood.

[1689] 2. Natural Language Processing: Natural language processing tools are used to translate text entered by the user from one language to another. The translation model performs tokenization and decoding between languages.

[1690] 3. Emotion Engine: Based on the analyzed emotional information, it adjusts the translation results and information provided. For example, if the emotional state is relaxed, it recommends relaxation-related products.

[1691] Specific example

[1692] (Specific example 1)

[1693] Suppose a user enters "I've been feeling really stressed out lately." This input text is then analyzed by the TextBlob library to calculate a negative emotion score, which is then categorized as "stress relief" by the emotion analysis tool. Next, the emotion engine is activated, and products such as stress balls or therapy sessions are recommended from the stress relief category.

[1694] (Specific example 2)

[1695] If a user enters "I'm feeling great today!", based on this positive emotion score, an aroma diffuser or relaxing tea will be recommended from the relaxation category.

[1696] Example of a prompt

[1697] This is an example where a user enters "I'm feeling overwhelmed" as their current emotional state, and relaxation or stress-relief products are recommended based on their emotional score. The product list is as follows:

[1698] Relaxation products: Aroma diffuser, Relaxing tea, Yoga mat

[1699] Stress relief products: Stress Ball, Meditation App Subscription, Therapy Sessions

[1700] If the emotional analysis score is positive, relaxation products will be recommended; if it is negative, stress relief products will be recommended.

[1701] Effects of implementation

[1702] This invention enables the provision of appropriate information and product recommendations based on the user's emotional state, thereby providing a more personalized user experience. Furthermore, it is expected to have a positive impact on sales promotion on e-commerce sites.

[1703] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1704] Step 1:

[1705] The terminal receives text input from the user. This input text reflects the user's emotional state. This text serves as the starting point for system processing. For example, the input data might be "I've been feeling really stressed out lately."

[1706] Step 2:

[1707] The terminal sends the received text to the server. The server receives the text and analyzes the sentiment information of the input text using sentiment analysis tools. It calculates the sentiment score of the text using the TextBlob library. The sentiment score is output as the calculation result. For example, a negative sentiment score may be obtained.

[1708] Step 3:

[1709] The server determines the emotional category of the text based on its emotional score. The emotional analysis tool classifies the text into the appropriate emotional category, such as "relaxation" or "stress relief." The classification result is then passed on to the next processing step. For example, if the emotional score is negative, it is classified into the "stress relief" category.

[1710] Step 4:

[1711] The server activates an emotion engine and selects the most suitable product for the user based on the analyzed emotion information. It extracts products matching the emotion category from the product list and generates a recommendation list. Random sampling or predefined rules are used to generate the recommendation list. For example, stress balls and therapy sessions might be extracted from the "stress relief" category.

[1712] Step 5:

[1713] The server sends the generated recommendation list to the terminal. The terminal displays the received recommendation list to the user. Specifically, information about the recommended products is visualized on the display device. For example, detailed information about "stress balls" and "therapy sessions" is displayed.

[1714] Step 6:

[1715] Users can make selection and purchase actions based on the displayed recommended products. The system records user selections and uses them to improve future recommendation algorithms.

[1716] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1717] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1718] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[1719] [Fourth Embodiment]

[1720] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1721] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1722] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1723] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[1724] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1725] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1726] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1727] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[1728] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1729] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1730] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1731] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1732] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1733] This invention relates to a system for performing translation, filtering out inaccurate information, predicting the future, and processing emotions. Specific embodiments thereof are described below.

[1734] Translation Function Embodiment

[1735] Program processing

[1736] 1. The user enters the text they want to translate, along with the source language and the target language.

[1737] 2. The terminal receives the input information and analyzes the text using natural language processing tools.

[1738] 3. Use a translation model to tokenize the text and perform the translation.

[1739] 4. The terminal decodes the translation result and converts it into a format that humans can understand.

[1740] 5. The server provides the user with the final translation result.

[1741] Specific example

[1742] For example, if a user wants to translate the Japanese text "こんにちは" into English, the system receives "こんにちは," translates it to "Hello" based on its translation model, and provides it to the user.

[1743] Implementation of sharing correct knowledge

[1744] Program processing

[1745] 1. The user enters a question about a specific topic.

[1746] 2. The server uses a search engine interface to obtain relevant information from the internet.

[1747] 3. The server performs analysis to filter reliable information from the acquired information sources.

[1748] 4. The device provides users with filtered, reliable information.

[1749] Specific example

[1750] For example, when searching for information about "global warming," the system retrieves links to Wikipedia and government-related websites and provides them to the user.

[1751] Future predicted implementation

[1752] Program processing

[1753] 1. The user inputs historical time-series data.

[1754] 2. The server uses data processing equipment to split the input data into training data and test data.

[1755] 3. Train a machine learning model on the data and create a predictive model.

[1756] 4. The server uses the created model to predict future data.

[1757] 5. The device provides the user with the prediction results.

[1758] Specific example

[1759] For example, when a user inputs past sales data, the system uses that data to predict future sales and provides the results to the user.

[1760] A Embodiment of World Peace

[1761] Program processing

[1762] 1. The user enters the content of their social media post.

[1763] 2. The terminal uses an emotion analysis model to analyze the emotion of the input text.

[1764] 3. The terminal uses a recommended action generation mechanism based on the sentiment analysis results to create suggestions to promote peaceful dialogue.

[1765] 4. The server provides the user with the recommended actions that have been created.

[1766] Specific example

[1767] For example, in the case of a positive post like "I love this!" and a negative post like "This is so frustrating!", the system analyzes the emotions in each post, determines whether they are "positive" or "negative," and suggests actions to the user to facilitate appropriate dialogue.

[1768] These embodiments enable users to achieve highly accurate translation, access to reliable information, predict the future, and engage in peaceful, emotion-based dialogue.

[1769] The following describes the processing flow.

[1770] Translation Function Embodiment

[1771] Step 1:

[1772] The user enters the text to be translated, the source language, and the target language.

[1773] Step 2:

[1774] The terminal receives input from the user and activates natural language processing for translation.

[1775] Step 3:

[1776] The terminal tokenizes the source text and converts it into a format compatible with the language model.

[1777] Step 4:

[1778] The terminal inputs tokenized text into the translation model and performs the translation.

[1779] Step 5:

[1780] The device decodes the translation result and converts it into a natural language format that the user can understand.

[1781] Step 6:

[1782] The server provides the user with the final translation result.

[1783] Implementation of sharing correct knowledge

[1784] Step 1:

[1785] The user enters the topic they want to investigate as a query.

[1786] Step 2:

[1787] The server receives the input query and collects relevant information from the internet through the search engine interface.

[1788] Step 3:

[1789] The server filters the collected search results to identify reliable sources and extracts specific URLs.

[1790] Step 4:

[1791] The device provides users with reliable sources of information and encourages them to obtain additional information as needed.

[1792] Future predicted implementation

[1793] Step 1:

[1794] The user inputs historical time-series data.

[1795] Step 2:

[1796] The server receives the input data, preprocesses it, and splits it into training data and test data.

[1797] Step 3:

[1798] The server uses a machine learning model to create a predictive model based on the training data.

[1799] Step 4:

[1800] The server uses the created predictive model to predict future data based on the test data.

[1801] Step 5:

[1802] The device provides the user with prediction results and suggests necessary actions.

[1803] A Embodiment of World Peace

[1804] Step 1:

[1805] The user enters the content of their social media post.

[1806] Step 2:

[1807] The terminal receives the entered text and activates the sentiment analysis model.

[1808] Step 3:

[1809] The device uses an emotion analysis model to analyze the emotional state of the text.

[1810] Step 4:

[1811] Based on the results of the sentiment analysis, the device creates recommended actions to promote peaceful dialogue.

[1812] Step 5:

[1813] The server provides the user with the generated recommended actions.

[1814] This allows users to effectively translate text, obtain reliable information, predict the future, and engage in emotion-based conversations.

[1815] (Example 1)

[1816] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1817] In recent years, the importance of global communication has been steadily increasing, but handling multiple different languages ​​requires advanced translation technology. Furthermore, there is a need for systems that can quickly obtain reliable information, predict future trends, and facilitate emotion-based dialogue. To meet these needs, systems that efficiently combine highly accurate natural language processing technologies and machine learning models are essential.

[1818] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1819] In this invention, the server includes natural language processing means for translating input text from one language to another; processing means for tokenizing and decoding the text using a machine learning translation model; output means for providing the translation results to the user; search engine interface means for inputting search queries on a specific topic and obtaining relevant information; reliability evaluation means for filtering reliable information from the obtained information sources; output means for providing the filtered information to the user; machine learning prediction model means for predicting future data based on past time series data; data processing means for splitting input data into training data and test data; prediction means for predicting future data using the trained model; output means for providing the prediction results to the user; sentiment analysis means for inputting social media posts and analyzing the sentiment of the text; recommendation action generation means for promoting peaceful dialogue based on the sentiment analysis results; and output means for providing the generated recommendation actions to the user. This enables highly accurate translation, acquisition of reliable information, future prediction, and peaceful dialogue based on sentiment.

[1820] "Natural language processing means" refers to technologies that analyze input text and enable operations such as translation and sentiment analysis.

[1821] A "machine learning translation model" is a model trained using a large amount of data, and it has the ability to translate text from one language to another.

[1822] "Processing means" refers to an algorithm or program for performing a specific task, and in this invention, it performs text tokenization and decoding.

[1823] "Output means" refers to interfaces or devices for providing processing results to the user, and have functions to display translation results, search results, prediction results, etc.

[1824] A "search engine interface means" is a means of linking with a search engine for obtaining information on the internet.

[1825] A "reliability evaluation method" is a system that executes algorithms and evaluation criteria to select highly reliable information from acquired information sources.

[1826] A "machine learning prediction model" is a prediction model trained using past data, which is used to predict future data.

[1827] "Data processing means" refers to means for dividing input data into training data and test data.

[1828] A "predictive tool" is a means of using a trained model to predict future data.

[1829] "Sentiment analysis techniques" are technologies used to analyze the emotions in text and determine whether they are positive or negative.

[1830] A "recommended action generation method" is a means of generating actions and suggestions to promote peaceful dialogue based on the results of sentiment analysis.

[1831] This invention relates to a system that utilizes natural language processing technology and machine learning models to achieve translation functionality, reliable information acquisition, future prediction, and sentiment analysis and peaceful dialogue on social media. Specific embodiments thereof are described below.

[1832] Translation function

[1833] Specific Embodiments

[1834] 1. The user inputs text through a translation interface and specifies the source and target languages. This interface is implemented as a web browser or mobile application.

[1835] 2. The terminal uses the Python NLTK library to parse and tokenize the input text.

[1836] 3. The device calls the Google Translate API or Microsoft Translator to translate the tokenized text.

[1837] 4. The device decodes the translation result and converts it into a format that the user can understand.

[1838] 5. The server displays the final translation result on the interface and provides it to the user.

[1839] Specific example

[1840] If a user wants to translate the Japanese text "こんにちは" into English, the system receives "こんにちは," uses the Google Translate API to translate it to "Hello," and provides the result to the user.

[1841] Sharing correct knowledge

[1842] Specific Embodiments

[1843] 1. The user enters a question about a specific topic. This question is entered via a web browser or mobile application.

[1844] 2. The server uses the Google Search API and other search engine interfaces to retrieve relevant information from the internet.

[1845] 3. The server runs a reliability evaluation algorithm and filters out reliable information from the retrieved sources.

[1846] 4. The device provides filtered information to the user.

[1847] Specific example

[1848] When searching for information related to "global warming," the system uses the Google Search API to retrieve relevant information, selects the most reliable information, and provides it to the user.

[1849] Future predictions

[1850] Specific Embodiments

[1851] 1. Users upload historical time-series data in CSV files or other digital formats.

[1852] 2. The server uses the Python pandas library to split the input data into training and test data.

[1853] 3. The server uses machine learning frameworks such as TensorFlow and Scikit-learn to train data and create predictive models.

[1854] 4. The server uses the trained model to predict future data.

[1855] 5. The device provides the user with prediction results in graph or table format.

[1856] Specific example

[1857] When a user inputs past sales data, the system uses a TensorFlow LSTM model based on that data to predict future sales and provides the results to the user.

[1858] World peace

[1859] Specific Embodiments

[1860] 1. The user enters the content of their social media post in text format.

[1861] 2. The terminal uses IBM Watson or Microsoft Azure sentiment analysis APIs to analyze the sentiment of the input text.

[1862] 3. Based on the sentiment analysis results, the device uses a recommended action generation algorithm to create suggestions to promote peaceful dialogue.

[1863] 4. The server provides the user with the recommended actions that have been created.

[1864] Specific example

[1865] If a negative post such as "This is so frustrating!" is submitted, IBM Watson will be used to determine that the sentiment is "negative" and generate and provide suggestions to the user to facilitate appropriate dialogue.

[1866] The above describes embodiments of the present invention. This system enables users to achieve highly accurate translation, obtain reliable information, predict the future, and engage in peaceful, emotion-based dialogue.

[1867] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1868] Translation Function Embodiment

[1869] Processing steps

[1870] Step 1:

[1871] The user accesses the system interface and enters the text to be translated, along with the source and target languages. For example, they might enter "こんにちは" (konnichiwa) and specify a translation from Japanese to English. This input is sent to the server in text format.

[1872] Step 2:

[1873] The terminal parses and tokenizes the input text received from the server using Python's NLTK library. The input data is "hello," and through parsing and tokenization, the internal structure of the text is understood. The output is the tokenized text.

[1874] Step 3:

[1875] The device sends tokenized text to the Google Translate API or Microsoft Translator for translation. The input is the tokenized text "こんにちは" (Konnichiwa), and the translation is performed by calling the API. The output is the translated text "Hello".

[1876] Step 4:

[1877] The terminal decodes the translation result returned from the API and converts it into a human-readable format. The input is the translated text "Hello," and the terminal performs the conversion from the encoded format to standard text format. The output is the decoded text "Hello."

[1878] Step 5:

[1879] The server provides the user with the final translation result. The input is the decoded text "Hello," which is returned to the interface for displaying it to the user. The output is the translation result displayed to the user.

[1880] Implementation of sharing correct knowledge

[1881] Processing steps

[1882] Step 1:

[1883] The user enters a question about a specific topic into the system. For example, they might enter, "Tell me about global warming." This input is sent to the server in text format.

[1884] Step 2:

[1885] The server uses the Google Search API to search for relevant information on the internet. The input is the question "Tell me about global warming," and the API is called to retrieve relevant information. The output is the retrieved search results.

[1886] Step 3:

[1887] The server executes a reliability evaluation algorithm to filter the retrieved information for the most reliable information. The input is the retrieved search results, to which the algorithm is applied for reliability evaluation. The output is the result of selecting only the most reliable information.

[1888] Step 4:

[1889] The terminal provides the user with filtered, reliable information. The input is reliable information, which is returned to the interface for displaying it to the user. The output is the reliable information displayed to the user.

[1890] Future prediction implementation

[1891] Processing steps

[1892] Step 1:

[1893] Users upload historical time-series data to the system as a CSV file. For example, they might upload sales data for the past year. This input is sent to the server in file format.

[1894] Step 2:

[1895] The server uses the Python pandas library to split the input data into training and test sets. The input is an uploaded CSV file, which is then divided into 80% training data and 20% test data. The output is the split dataset.

[1896] Step 3:

[1897] The server uses machine learning frameworks such as TensorFlow and Scikit-learn to train data and create predictive models. The input is training data, and the machine learning algorithm generates the predictive model. The output is the trained predictive model.

[1898] Step 4:

[1899] The server evaluates the performance of a model created using test data and predicts future data. The inputs are the test data and the predictive model; performance evaluation and future prediction are performed. The output is the predicted data.

[1900] Step 5:

[1901] The terminal provides prediction results to the user in graph or table format. The input is the predicted data, which is visualized and displayed on the interface. The output is the prediction result displayed to the user.

[1902] A Embodiment of World Peace

[1903] Processing steps

[1904] Step 1:

[1905] The user enters their social media post content into the system in text format. For example, they might enter "This is so frustrating!". This input is then sent to the server in text format.

[1906] Step 2:

[1907] The terminal uses IBM Watson and Microsoft Azure sentiment analysis APIs to analyze the sentiment of the input text. The input is a social media post, "This is so frustrating!", and the sentiment is analyzed by calling the sentiment analysis API. The output is the analyzed sentiment result.

[1908] Step 3:

[1909] The device applies a recommended action generation algorithm to promote peaceful dialogue based on the sentiment analysis results. The input is the sentiment analysis results, and the algorithm is used to generate an appropriate response. The output is the recommended action.

[1910] Step 4:

[1911] The server provides the user with the generated recommended actions. The input is the recommended actions, which are returned to the interface for displaying them to the user. The output is the recommended actions displayed to the user.

[1912] These specific processing steps enable users to achieve highly accurate translations, obtain reliable information, predict the future, and engage in peaceful, emotion-based dialogue.

[1913] (Application Example 1)

[1914] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1915] In many modern situations, communication between people who speak different languages, facilitating emotion-based dialogue, obtaining reliable information, and predicting the future based on diverse data are critical challenges. However, current systems lack an integrated approach to meet these requirements, particularly a lack of dialogue support incorporating real-time sentiment analysis. This makes efficient communication and appropriate decision-making difficult.

[1916] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1917] In this invention, the server includes natural language processing means for translating input text from one language to another, means for tokenizing and decoding the text using a translation model, output means for providing the translation results to the user, sentiment analysis means for analyzing the sentiment of the input text, and dialogue support means for facilitating peaceful dialogue based on the analysis results. This enables real-time sentiment analysis and dialogue support.

[1918] A "natural language processing system" is a means of analyzing input text and translating it from one language to another.

[1919] A "translation model" is a model for tokenizing text and decoding it into another language.

[1920] "Output means" refers to the means of providing translation results or analysis results to the user.

[1921] An "emotion analysis tool" is a means of analyzing the emotions in input text and identifying the type of emotion.

[1922] "Dialogue support methods" are means of creating proposals to promote peaceful dialogue based on the results of emotion analysis.

[1923] A "search engine interface" is a means of obtaining reliable information sources by entering a search query related to a specific topic.

[1924] "Analysis means" are means for filtering reliable information from acquired information sources.

[1925] A "machine learning model" is a model used to predict future trends based on past time-series data.

[1926] "Data processing means" refers to means for dividing input data into training data and test data.

[1927] A "predictive tool" is a means of predicting future data using a trained model.

[1928] This invention can be specifically implemented as follows.

[1929] First, the system receives text input from the user. This input text could be content to be translated, content requiring sentiment analysis, or a search query. Once the user enters the text, the terminal analyzes it using natural language processing tools. At this stage, the main software used includes "Transformers," a widely known natural language processing library.

[1930] Next, if translation is needed, the system uses a translation model to tokenize the text and perform the translation. The translation result is decoded and provided to the user. Open-source translation models such as "Hugging Face" are used.

[1931] Furthermore, the system uses sentiment analysis tools to analyze the sentiment of the input text. Sentiment analysis assigns specific sentiment labels (e.g., positive, negative, neutral). Based on these analysis results, dialogue support tools generate suggestions to facilitate peaceful dialogue.

[1932] When a search query is entered, the server uses a search engine interface to obtain reliable information sources and filters them using analysis tools. The reliable information is then provided to the user.

[1933] Furthermore, the system can predict future trends based on historical time-series data. The server divides the input data into training data and test data, and trains the data using a machine learning model. This allows it to predict future data and provide results to the user.

[1934] As a concrete example, suppose a security staff member detected the following statement at the scene:

[1935] “This situation is incredibly frustrating and makes me angry.”

[1936] In this case, the system analyzes the text and detects the emotion "NEGATIVE." It then suggests the following recommended action:

[1937] "Please try to engage in calm dialogue to alleviate the situation."

[1938] Examples of prompt statements include the following:

[1939] "A proposed system for emotional analysis and peaceful dialogue"

[1940] Analyze user sentiment based on their input text and provide appropriate dialogue actions.

[1941] User input: "This situation is incredibly frustrating and makes me angry."

[1942] In this way, the system provides an integrated approach to emotion analysis and dialogue support, enabling it to respond appropriately in real time.

[1943] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1944] Step 1:

[1945] The user provides input text to the terminal. The input can be text to be translated, text requiring sentiment analysis, or a search query. The terminal analyzes this input text using natural language processing. The analyzed text is then sent to the next processing step.

[1946] Step 2:

[1947] The device uses a translation model to tokenize the input text if it requires translation. Tokenization is the process of dividing each word or phrase in the text into individual tokens. The tokens are then sent to the translation model.

[1948] Step 3:

[1949] The translation model translates tokenized text into the specified language. The translation model uses a generative AI model to convert the token array into tokens in the other language. The resulting token array is decoded and converted into human-readable text. The decoded text is returned to the terminal.

[1950] Step 4:

[1951] The terminal provides the translated text to the user through an output device. The user receives the final translation result.

[1952] Step 5:

[1953] The terminal analyzes the sentiment of the input text using sentiment analysis means if the sentiment of the text should be analyzed. The sentiment analysis means analyzes the text and assigns a specific sentiment label such as positive, negative, or neutral. The analysis results and sentiment labels are identified and sent to the next processing step.

[1954] Step 6:

[1955] The device uses dialogue support mechanisms based on the analyzed emotion results to generate suggestions for peaceful dialogue. These dialogue support mechanisms create messages to alleviate negative emotions and messages to maintain a positive atmosphere for positive emotions. These suggestions are then provided to the user.

[1956] Step 7:

[1957] When a user enters a search query, the server uses a search engine interface to retrieve reliable information sources from the internet. These retrieved sources contain data corresponding to the search query.

[1958] Step 8:

[1959] The server uses analysis tools to filter reliable information from the acquired information sources. These filtering tools exclude unreliable information and extract only the reliable information. This filtered information is then provided to the user.

[1960] Step 9:

[1961] When historical time-series data is input, the server uses data processing tools to split the input data into training data and test data. The split data is then used to train a machine learning model.

[1962] Step 10:

[1963] The server uses machine learning models to predict future trends based on historical data. The predicted trend data includes estimates of future sales and demand. The prediction results are provided to the user.

[1964] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1965] This invention relates to a system for translation, filtering out inaccurate information, predicting the future, and processing emotions. Furthermore, the invention implements a form that combines this with an emotion engine that recognizes and utilizes the user's emotional information. Specific embodiments are described below.

[1966] Translation Function Embodiment

[1967] Program processing

[1968] 1. The user enters the text to be translated, the source language, and the target language.

[1969] 2. The terminal receives the input information and analyzes the text using natural language processing tools.

[1970] 3. Use a translation model to tokenize the text and perform the translation.

[1971] 4. The terminal decodes the translation result and converts it into a format that humans can understand.

[1972] 5. The server activates an emotion engine to recognize the user's emotions and performs emotion adjustments appropriate to the translation result.

[1973] 6. The server provides the user with the final translation result.

[1974] Specific example

[1975] For example, if a user wants to translate the Japanese text "こんにちは" (konnichiwa) into English, the system receives "こんにちは," translates it to "Hello" based on its translation model, and then provides it in a form that is adapted to the user's emotional state.

[1976] Implementation of sharing correct knowledge

[1977] Program processing

[1978] 1. The user enters the topic they want to verify as a query.

[1979] 2. The server receives the input query and collects relevant information from the internet through the search engine interface.

[1980] 3. The server filters the collected search results to identify reliable sources and extracts specific URLs.

[1981] 4. The server uses an emotion engine to recognize the user's emotions and adjusts the reliability of the information it provides according to the user's emotional state.

[1982] 5. The device provides users with reliable sources of information and encourages them to obtain additional information as needed.

[1983] Specific example

[1984] For example, when searching for information about "global warming," the system retrieves links from Wikipedia and government-related websites, and then selects and provides appropriate information based on the user's emotional state.

[1985] Future predicted implementation

[1986] Program processing

[1987] 1. The user inputs historical time-series data.

[1988] 2. The server receives the input data, preprocesses it, and splits it into training data and test data.

[1989] 3. Train a machine learning model on the data and create a predictive model.

[1990] 4. The server uses the created predictive model to predict future data based on the test data.

[1991] 5. The server uses an emotion engine to adjust prediction results based on the user's emotional information.

[1992] 6. The device provides the user with prediction results and suggests necessary actions.

[1993] Specific example

[1994] For example, when a user inputs past sales data, the system uses that data to predict future sales and provides the prediction results in a way that is tailored to the user's emotional state.

[1995] A Embodiment of World Peace

[1996] Program processing

[1997] 1. The user enters the content of their social media post.

[1998] 2. The terminal receives the entered text and activates the sentiment analysis model.

[1999] 3. The device uses an emotion analysis model to analyze the emotional state of the text.

[2000] 4. Based on the results of the sentiment analysis, the terminal creates specific suggestions using a recommended action generation mechanism to promote peaceful dialogue.

[2001] 5. The server uses an emotion engine to adjust the optimal recommended action, taking into account the user's emotional state.

[2002] 6. The server provides the user with the generated recommended actions.

[2003] Specific example

[2004] For example, in the case of a positive post like "I love this!" and a negative post like "This is so frustrating!", the system analyzes the emotions in each post and suggests appropriate actions to facilitate dialogue based on the user's emotional state.

[2005] This enables users to achieve highly accurate translations, access to reliable information, predict the future, and engage in peaceful, emotion-based dialogue. Leveraging emotion-driven tools further enhances the user experience and increases the system's practicality.

[2006] The following describes the processing flow.

[2007] Translation Function Embodiment

[2008] Step 1:

[2009] The user enters the text to be translated, the source language, and the target language.

[2010] Step 2:

[2011] The terminal receives the input information and activates the natural language processing system.

[2012] Step 3:

[2013] The terminal tokenizes the source text and converts it into a format suitable for the translation model.

[2014] Step 4:

[2015] The terminal inputs tokenized text into the translation model and performs the translation.

[2016] Step 5:

[2017] The terminal decodes the translation result and converts it into a format that humans can understand.

[2018] Step 6:

[2019] The server activates the emotion engine and recognizes the user's emotional state.

[2020] Step 7:

[2021] The server uses the emotion information recognized by the emotion engine to adjust the translation results and adapt them to the emotions.

[2022] Step 8:

[2023] The server provides the user with the final translation result.

[2024] Implementation of sharing correct knowledge

[2025] Step 1:

[2026] The user enters the topic they want to investigate as a query.

[2027] Step 2:

[2028] The server receives the input query and collects relevant information from the internet through the search engine interface.

[2029] Step 3:

[2030] The server filters the collected search results to identify reliable sources and extracts specific URLs.

[2031] Step 4:

[2032] The server activates the emotion engine and recognizes the user's emotional state.

[2033] Step 5:

[2034] The server uses emotional information to adjust the reliability of the acquired information based on the user's emotional state.

[2035] Step 6:

[2036] The device provides users with reliable sources of information and encourages them to obtain additional information as needed.

[2037] Future predicted implementation

[2038] Step 1:

[2039] The user inputs historical time-series data.

[2040] Step 2:

[2041] The server receives the input data, preprocesses it, and splits it into training and test data.

[2042] Step 3:

[2043] The server uses a machine learning model to create a predictive model based on the training data.

[2044] Step 4:

[2045] The server uses the created predictive model to predict future data based on the test data.

[2046] Step 5:

[2047] The server activates the emotion engine and recognizes the user's emotional state.

[2048] Step 6:

[2049] The server uses the recognized emotion information to adjust the prediction results and adapt them to the user's emotions.

[2050] Step 7:

[2051] The device provides the user with prediction results and suggests necessary actions.

[2052] A Embodiment of World Peace

[2053] Step 1:

[2054] The user enters the content of their social media post.

[2055] Step 2:

[2056] The terminal receives the entered text and activates the sentiment analysis model.

[2057] Step 3:

[2058] The device uses an emotion analysis model to analyze the emotional state of the text.

[2059] Step 4:

[2060] Based on the results of the sentiment analysis, the device creates recommended actions to promote peaceful dialogue.

[2061] Step 5:

[2062] The server activates the emotion engine and recognizes the user's emotional state.

[2063] Step 6:

[2064] The server adjusts recommended actions based on emotional information, adapting to the user's emotions.

[2065] Step 7:

[2066] The server provides the user with a final recommended action.

[2067] This enables users to achieve highly accurate translations, access to reliable information, predict the future, and engage in peaceful, emotion-based dialogue. By leveraging the emotion engine, the user experience is further enhanced, and the system's practicality is increased.

[2068] (Example 2)

[2069] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[2070] In today's internet environment, there is a demand for communication that transcends language barriers, the elimination of inaccurate information, future predictions, and responses based on emotions. However, conventional systems struggle to meet these multiple needs simultaneously, and a particular challenge is the lack of consideration for users' emotional information. Specifically, translation results, search results, and future predictions do not adapt to the user's emotional state, resulting in a low quality of user experience. Furthermore, without utilizing emotional information, it becomes difficult to alleviate user anxiety and stress, and to ensure the accuracy and reliability of the information provided.

[2071] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[2072] In this invention, the server includes natural language processing means for translating input text from one language to another, means for tokenizing and decoding the text using a translation model, sentiment engine means for adjusting the translation result based on the user's sentiment, and output means for providing the translation result to the user. This makes it possible to provide an appropriate translation result that is in line with the user's sentiment.

[2073] Furthermore, it includes a search engine interface for inputting search queries on a specific topic and obtaining reliable information sources, an analysis means for filtering reliable information from the obtained information sources, an emotion engine means for adjusting the filtered information based on the user's emotions, and an output means for providing that information to the user. This makes it possible to provide reliable information tailored to the user's emotional state.

[2074] Furthermore, it includes a machine learning model for predicting future trends based on past time-series data, a data processing means for splitting input data into training data and test data, a prediction means for predicting future data using the trained model, an emotion engine means for adjusting the prediction results based on the user's emotions, and an output means for providing the prediction results to the user. This enables future predictions based on the user's emotional information, making it possible to provide more appropriate action suggestions.

[2075] "Natural language processing means" refers to technologies that analyze input text and convert human language into a format that machines can understand.

[2076] A "translation model" is an algorithm and computational model for converting text from one language to another.

[2077] "Tokenization" is the process of dividing input text into units of words or phrases.

[2078] "Decoding" is the process of reconstructing generated tokens and returning them to a human-understandable format as a result of translation and analysis.

[2079] An "emotional engine" is a technology that recognizes a user's emotional information and adjusts its response based on that information.

[2080] "Output means" refers to interfaces or devices used to provide text or information to users.

[2081] A "search engine interface means" is a technology for searching for information on the internet and obtaining results.

[2082] "Analysis means" refers to techniques for analyzing acquired information and extracting or filtering necessary data.

[2083] A "machine learning model" is a statistical algorithm that learns from past data and uses it to make predictions and analyses of the future.

[2084] "Data processing means" refers to technologies for pre-processing input data and converting it into the required format.

[2085] "Predictive tools" are techniques that use trained models to predict future data and trends.

[2086] This invention relates to a system that provides diverse functions using technologies such as natural language processing, emotion recognition, and machine learning. The specific forms for implementing each function are described below.

[2087] Translation Function Embodiment

[2088] In this system, the user inputs the text to be translated, along with the source and target languages. Specifically, the user inputs the text and selects the language setting through the terminal's interface. The terminal receives the input information and analyzes the text using natural language processing. Next, a translation model (e.g., a common translation API) is used to tokenize the text and perform the translation. The translation result is decoded and converted into a human-readable format. The server activates an emotion engine to recognize the user's emotions and performs emotion adjustments appropriate to the translation result. Finally, the adjusted translation result is provided to the user.

[2089] Specific example

[2090] For example, if a user wants to translate the Japanese text "こんにちは" (konnichiwa) into English, the system receives "こんにちは" as input and translates it to "Hello" based on a translation model (e.g., a general-purpose translation API). Then, using an emotion engine (e.g., a general-purpose emotion recognition API), it adjusts the translation based on the user's emotional state, such as "Hello, how can I assist you today?", and provides it to the user.

[2091] Example of a prompt

[2092] "Translate the following Japanese text 'こんにちは' into English and adjust it to suit the user's mood."

[2093] Implementation of sharing correct knowledge

[2094] In this system, the user inputs the topic they want to verify as a query. The server receives the input query and collects relevant information from the internet through a search engine interface (e.g., a general-purpose search API). It filters the retrieved search results to identify reliable sources and extracts specific URLs. Next, it uses an emotion engine to adjust the reliability of the information to match the user's emotional state. Finally, it provides reliable information to the user via their device.

[2095] Specific example

[2096] For example, when searching for information about "global warming," the system uses a general-purpose search API to collect relevant information and obtain links to Wikipedia and government websites. Then, it uses an emotion engine (e.g., a general-purpose emotion recognition API) to adapt that information to the user's emotional state and present it accordingly.

[2097] Example of a prompt

[2098] "Search for reliable information on 'global warming' and provide the most relevant information tailored to the user's emotions."

[2099] Future predicted implementation

[2100] In this system, the user inputs historical time-series data. The server receives the input data, preprocesses it, and splits it into training and test data. The software used includes, for example, general data processing libraries and machine learning libraries. Next, a machine learning model (for example, a general-purpose machine learning library) is used to train the data and create a predictive model. The prediction results are adjusted based on the user's emotional state using an emotion engine, and finally, the prediction results are provided to the user via a terminal.

[2101] Specific example

[2102] For example, when a user inputs past sales data, the system preprocesses it using a data processing library and then uses a machine learning library to predict future sales. Afterward, an emotion engine (e.g., a general-purpose emotion recognition API) is used to adapt the prediction results to the user's emotional state and provide them to the user.

[2103] Example of a prompt

[2104] "Based on past sales data, we predict future sales and adjust them according to user sentiment."

[2105] A Embodiment of World Peace

[2106] In this system, the user inputs the content of a social media post. The terminal receives the input text and activates a sentiment analysis model (e.g., a general-purpose sentiment analysis API). Using the sentiment analysis model, the emotional state of the text is analyzed, and based on the results, recommended actions to promote peaceful dialogue are generated. Next, the server uses a sentiment engine to adjust the optimal recommended actions, taking into account the user's emotional state. Finally, the generated recommended actions are provided to the user through the terminal.

[2107] Specific example

[2108] For example, in the case of a positive post like "I love this!" and a negative post like "This is so frustrating!", the system uses a sentiment analysis API to analyze the emotions in each post and then uses an emotion engine to suggest an appropriate way to interact with the user.

[2109] Example of a prompt

[2110] "Analyze the sentiment of social media posts and propose ways to promote peaceful dialogue based on user sentiment."

[2111] As a result, the system provided by the present invention can perform multi-functional responses that take into account the user's emotional information, enabling translation, elimination of inaccurate information, future prediction, and promotion of emotion-based dialogue.

[2112] The flow of the specific processing in Example 2 will be explained using Figure 13.

[2113] Translation Function Embodiment

[2114] Program processing

[2115] Step 1:

[2116] The user inputs the text to be translated, along with the source and target languages. The user sets the text "Hello" and the source language to Japanese and the target language to English via the device's interface. Input is done using text boxes and dropdown lists.

[2117] Input: Text "Hello", Source language "Japanese", Target language "English"

[2118] Output: Input information is sent.

[2119] Step 2:

[2120] The terminal receives the input information and analyzes the text using natural language processing techniques. Here, the language of the text is confirmed, and preprocessing is performed, including analysis of words and grammar.

[2121] Input: User input information

[2122] Data processing: Text preprocessing (tokenization, stop word removal, etc.)

[2123] Output: Analysis results (tokenized text)

[2124] Step 3:

[2125] The device sends the analyzed text to a translation model for translation. The translation model uses, for example, a common translation API to convert the tokenized text into the target language.

[2126] Input: Parsed text (tokens)

[2127] Data processing: Call the translation API and perform the translation.

[2128] Output: Translation result (English text "Hello")

[2129] Step 4:

[2130] The terminal receives the translation result, decodes it, and converts it into a human-readable format. Here, it reconstructs another token and formats it into the appropriate language format.

[2131] Input: English token "Hello"

[2132] Data processing: Decode and format tokens.

[2133] Output: Formatted English text "Hello"

[2134] Step 5:

[2135] The server activates the emotion engine, analyzes the translation results, and makes adjustments based on the user's emotional state. Here, sentiment analysis is performed based on the user's previous input and history, and the translation results are adjusted accordingly.

[2136] Input: Translation result "Hello"

[2137] Data processing: Analysis and adjustment using an emotion engine.

[2138] Output: Adjusted translation result: "Hello, how can I assist you today?"

[2139] Step 6:

[2140] The server sends the final translation result to the terminal, which then displays the result to the user. The user's screen displays text that has been adjusted based on their emotions.

[2141] Input: Adjusted translation result

[2142] Output: Translation results displayed to the user

[2143] ---

[2144] Implementation of sharing correct knowledge

[2145] Program processing

[2146] Step 1:

[2147] The user enters a search query related to a specific topic. For example, if they want to find information about "global warming," they would enter "global warming" in the search box.

[2148] Input: Search query "global warming"

[2149] Output: Input query

[2150] Step 2:

[2151] The server receives the input query and collects relevant information from the internet through the search engine interface. For example, it might use a general-purpose search API to retrieve information.

[2152] Input: Search query

[2153] Data processing: Call the search API and collect relevant information.

[2154] Output: Search Results

[2155] Step 3:

[2156] The server filters the collected search results to identify reliable sources. It uses algorithms to evaluate reliability, for example, to extract links to Wikipedia or official government websites.

[2157] Input: Search Results

[2158] Data processing: Reliability assessment and filtering

[2159] Output: Links to reliable sources

[2160] Step 4:

[2161] The server activates the emotion engine and adjusts the reliability of the information according to the user's emotional state. Here, the method and content of information presentation are adapted based on the user's emotional state.

[2162] Input: Links to reliable sources

[2163] Data processing: Analysis and adjustment using an emotion engine.

[2164] Output: Adjusted information

[2165] Step 5:

[2166] The device provides the user with the finalized information and prompts them to obtain additional information as needed.

[2167] Input: Adjusted information

[2168] Output: Information displayed to the user

[2169] ---

[2170] Future predicted implementation

[2171] Program processing

[2172] Step 1:

[2173] Users input historical time-series data. For example, they might upload historical sales data as a CSV file.

[2174] Input: CSV file, historical sales data

[2175] Output: Uploaded data

[2176] Step 2:

[2177] The server receives the input data, preprocesses it, and splits it into training and test data. For example, it performs data cleaning and normalization before splitting the data into training and test sets.

[2178] Input: Uploaded data

[2179] Data processing: Data cleaning, normalization, splitting

[2180] Output: Training data, Test data

[2181] Step 3:

[2182] The server uses training data to train a machine learning model. Here, for example, a general-purpose machine learning library is used to create a model for future prediction.

[2183] Input: Training data

[2184] Data processing: Model training

[2185] Output: Trained predictive model

[2186] Step 4:

[2187] The server uses a trained model to predict future data based on test data. The model is used to predict future sales data, for example.

[2188] Input: Test data

[2189] Data processing: Predicting future data

[2190] Output: Prediction result

[2191] Step 5:

[2192] The server uses an emotion engine to adjust prediction results based on the user's emotional state. For example, it might add encouraging comments to negative prediction results.

[2193] Input: Prediction result

[2194] Data processing: Analysis and adjustment using an emotion engine.

[2195] Output: Adjusted prediction results

[2196] Step 6:

[2197] The device provides the user with the final prediction results and suggests necessary actions.

[2198] Input: Adjusted prediction result

[2199] Output: Prediction results displayed to the user

[2200] ---

[2201] A Embodiment of World Peace

[2202] Program processing

[2203] Step 1:

[2204] The user enters the content of their social media post. For example, they might enter a positive post like "I love this!" in the input field.

[2205] Input: Social media post content

[2206] Output: Input text

[2207] Step 2:

[2208] The terminal receives the entered text and activates the sentiment analysis model. Here, a general-purpose sentiment analysis API is used.

[2209] Input: User's post content

[2210] Data processing: Calling the emotion analysis API

[2211] Output: Sentiment analysis results

[2212] Step 3:

[2213] The device uses an emotion analysis model to analyze the emotional state of the text. For example, it obtains positive, negative, and neutral emotion scores.

[2214] Input: User's post content

[2215] Data processing: Calculation of sentiment score

[2216] Output: Emotional state score

[2217] Step 4:

[2218] The device generates recommended actions to promote peaceful dialogue based on sentiment analysis results. For example, it might suggest "Share more!" for positive posts.

[2219] Input: Emotional state score

[2220] Data processing: Generating recommended actions

[2221] Output: Recommended Action

[2222] Step 5:

[2223] The server uses an emotion engine to adjust the optimal recommended action based on the user's emotional state. For example, it might take into account past posting history.

[2224] Input: Recommended Action

[2225] Data processing: Analysis and adjustment using an emotion engine.

[2226] Output: Optimized Recommended Actions

[2227] Step 6:

[2228] The server sends the final recommended action to the terminal, which then provides it to the user.

[2229] Input: Optimized Recommended Actions

[2230] Output: Recommended actions displayed to the user

[2231] (Application Example 2)

[2232] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[2233] Conventional systems lack the ability to translate, provide information, or predict future trends that take into account user emotions, resulting in insufficient individual optimization of the user experience. Furthermore, e-commerce sites are currently unable to provide appropriate product recommendations based on user emotional states, limiting their sales promotion effectiveness.

[2234] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes natural language processing means for translating input text from one language to another, means for tokenizing and decoding the text using a translation model, sentiment analysis means for analyzing the user's sentiment information, sentiment engine means for adjusting the translation result based on the analyzed sentiment information, and output means for providing the translation result to the user. This makes it possible to provide individually optimized translation results based on the user's sentiment information.

[2235] "Natural language processing means" refers to technologies that analyze input text and perform specific processing.

[2236] A "translation model" is an algorithm that converts text from one language to another.

[2237] "Tokenization" is the process of analyzing text and dividing it into the smallest units of words or phrases.

[2238] "Decoding methods" refer to techniques for reconstructing tokenized text into natural language.

[2239] "Emotional analysis methods" refer to technologies that analyze a user's emotional information from input text and data.

[2240] An "emotional engine" is a technology that adjusts the system's response and output based on analyzed emotional information.

[2241] "Output means" refers to the technologies and devices that a system uses to display results to the user.

[2242] A "search engine interface" is a system for collecting information from the internet.

[2243] "Reliable information" refers to verified data obtained from trustworthy sources.

[2244] A "machine learning model" is an algorithm that learns patterns from past data and uses them to predict the future.

[2245] "Data processing means" refers to techniques that preprocess input data and optimize it for analysis and prediction.

[2246] A "predictive tool" is a technique that uses trained machine learning models to estimate future data.

[2247] System Overview

[2248] This invention is a system that analyzes user emotional information and optimizes translation results, information provision, future predictions, etc., based on that analysis. This system includes the following main components.

[2249] Hardware and software

[2250] 1. Hardware: Smartphones are primarily used. Smartphones provide an interface for users to input text and have a display that shows analysis results and recommendations.

[2251] 2. Software: The Python language and TextBlob library are used for sentiment analysis and natural language processing. The server implements a sentiment engine, natural language processing, data processing, and machine learning models.

[2252] Processing Overview

[2253] 1. Sentiment Analysis: The text entered by the user is analyzed using the TextBlob library to calculate an emotion score. Based on the emotion score, the user's emotional state is understood.

[2254] 2. Natural Language Processing: Natural language processing tools are used to translate text entered by the user from one language to another. The translation model performs tokenization and decoding between languages.

[2255] 3. Emotion Engine: Based on the analyzed emotional information, it adjusts the translation results and information provided. For example, if the emotional state is relaxed, it recommends relaxation-related products.

[2256] Specific example

[2257] (Specific example 1)

[2258] Suppose a user enters "I've been feeling really stressed out lately." This input text is then analyzed by the TextBlob library to calculate a negative emotion score, which is then categorized as "stress relief" by the emotion analysis tool. Next, the emotion engine is activated, and products such as stress balls or therapy sessions are recommended from the stress relief category.

[2259] (Specific example 2)

[2260] If a user enters "I'm feeling great today!", based on this positive emotion score, an aroma diffuser or relaxing tea will be recommended from the relaxation category.

[2261] Example of a prompt

[2262] This is an example where a user enters "I'm feeling overwhelmed" as their current emotional state, and relaxation or stress-relief products are recommended based on their emotional score. The product list is as follows:

[2263] Relaxation products: Aroma diffuser, Relaxing tea, Yoga mat

[2264] Stress relief products: Stress Ball, Meditation App Subscription, Therapy Sessions

[2265] If the emotional analysis score is positive, relaxation products will be recommended; if it is negative, stress relief products will be recommended.

[2266] Effects of implementation

[2267] This invention enables the provision of appropriate information and product recommendations based on the user's emotional state, thereby providing a more personalized user experience. Furthermore, it is expected to have a positive impact on sales promotion on e-commerce sites.

[2268] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[2269] Step 1:

[2270] The terminal receives text input from the user. This input text reflects the user's emotional state. This text serves as the starting point for system processing. For example, the input data might be "I've been feeling really stressed out lately."

[2271] Step 2:

[2272] The terminal sends the received text to the server. The server receives the text and analyzes the sentiment information of the input text using sentiment analysis tools. It calculates the sentiment score of the text using the TextBlob library. The sentiment score is output as the calculation result. For example, a negative sentiment score may be obtained.

[2273] Step 3:

[2274] The server determines the emotional category of the text based on its emotional score. The emotional analysis tool classifies the text into the appropriate emotional category, such as "relaxation" or "stress relief." The classification result is then passed on to the next processing step. For example, if the emotional score is negative, it is classified into the "stress relief" category.

[2275] Step 4:

[2276] The server activates an emotion engine and selects the most suitable product for the user based on the analyzed emotion information. It extracts products matching the emotion category from the product list and generates a recommendation list. Random sampling or predefined rules are used to generate the recommendation list. For example, stress balls and therapy sessions might be extracted from the "stress relief" category.

[2277] Step 5:

[2278] The server sends the generated recommendation list to the terminal. The terminal displays the received recommendation list to the user. Specifically, information about the recommended products is visualized on the display device. For example, detailed information about "stress balls" and "therapy sessions" is displayed.

[2279] Step 6:

[2280] Users can make selection and purchase actions based on the displayed recommended products. The system records user selections and uses them to improve future recommendation algorithms.

[2281] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[2282] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[2283] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[2284] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[2285] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[2286] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[2287] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[2288] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[2289] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[2290] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[2291] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[2292] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[2293] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[2294] 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.

[2295] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[2296] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[2297] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[2298] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[2299] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[2300] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[2301] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[2302] The following is further disclosed regarding the embodiments described above.

[2303] (Claim 1)

[2304] A natural language processing system for translating input text from one language to another,

[2305] A means of tokenizing and decoding text using a translation model,

[2306] An output means for providing the translation result to the user,

[2307] A system that includes this.

[2308] (Claim 2)

[2309] A search engine interface means for entering search queries on a specific topic and obtaining reliable information sources,

[2310] An analytical means for filtering reliable information from acquired information sources,

[2311] An output means for providing filtered information to the user,

[2312] The system according to claim 1, including the following:

[2313] (Claim 3)

[2314] A machine learning model for predicting future trends based on past time-series data,

[2315] A data processing means for splitting input data into training data and test data,

[2316] Predictive means for predicting future data using a trained model,

[2317] Output means for providing prediction results to the user,

[2318] The system according to claim 1, including the following:

[2319] (Claim 4)

[2320] A sentiment analysis model for performing sentiment analysis on social media posts,

[2321] A text processing means for converting input text into a format suitable for an analysis model,

[2322] A means for generating recommended actions to promote peaceful dialogue based on the results of sentiment analysis,

[2323] ...

Claims

1. A natural language processing system for translating input text from one language to another, A means of tokenizing and decoding text using a translation model, An output means for providing the translation result to the user, A system that includes this.

2. A search engine interface means for entering search queries on a specific topic and obtaining reliable information sources, An analytical means for filtering reliable information from acquired information sources, An output means for providing filtered information to the user, The system according to claim 1, including the following:

3. A machine learning model for predicting future trends based on past time-series data, A data processing means for splitting input data into training data and test data, Predictive means for predicting future data using a trained model, Output means for providing prediction results to the user, The system according to claim 1, including the following:

4. A sentiment analysis model for performing sentiment analysis on social media posts, A text processing means for converting input text into a format suitable for an analysis model, A means for generating recommended actions to promote peaceful dialogue based on the results of sentiment analysis, Output means for providing recommended actions to the user, The system according to claim 1, including the following:

Images