System

Abstract

A system is provided.SOLUTION: A system comprising: means for inputting information necessary for creating a recommendation from a terminal of a user; means for receiving information input from the terminal and performing validation; means for transmitting the validated information to a generative artificial intelligence model and generating a recommendation; and means for returning the generated recommendation to the terminal.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Writing recommendation letters and referrals is a difficult, time-consuming and laborious task for many people. It's particularly challenging to adequately express the characteristics and memorable anecdotes of the person recommended. Currently, recommenders must spend a lot of time thinking about the wording, making it difficult to determine the appropriate wording and content. This can lead to a decline in the quality of the recommendation and a lower evaluation of the person recommended. It also places a heavy burden on the recommender, potentially damaging the relationship between the recommender and the person recommended. Given this background, an efficient method for writing recommendation letters is needed. [Means for solving the problem]

[0005] The present invention provides a system that includes a means for inputting information necessary for creating a recommendation from a user's device, receiving that information, and validating it. The system further includes a means for sending the validated information to a generative AI model and generating a recommendation. The system also includes a means for returning the generated recommendation to the user's device. This allows users to easily generate high-quality recommendations, reducing the time and effort required to create them. Specifically, the user's device sends the information necessary for creating a recommendation to a server in JSON format, and the server validates the information and passes it to a generative AI model to generate a recommendation. This allows for the efficient provision of fair and unbiased recommendations.

[0006] A "user terminal" is a device for inputting the information necessary to create a recommendation and transmitting it to a server, and includes devices such as a personal computer and a smartphone.

[0007] A "letter of recommendation" is a document in which the recommender details the characteristics and achievements of the person being recommended and evaluates the person. It is a document used in situations such as going on to higher education, finding employment, or changing jobs.

[0008] "Information" refers to data necessary to create a recommendation letter, including the name, field, characteristics, anecdotes, etc. of the person being recommended.

[0009] "Validation" is the process of ensuring that the information entered is accurate and complete; it is a procedure for checking the consistency of the information.

[0010] A "generative artificial intelligence model" is an artificial intelligence system that generates recommendations using a natural language generation algorithm based on input information.

[0011] The "JSON format" is a lightweight data exchange format for structuring and describing data, and is an abbreviation for JavaScript Object Notation.

[0012] A "server" is a central system that processes information received from user terminals and generates recommendations, and is a device that provides data to clients via a network.

[0013] An "impressive anecdote" is a specific event or experience that demonstrates the nominee's characteristics and skills, and is used in the recommendation letter to highlight the nominee's appeal. [Brief explanation of the drawings]

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

[0015] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

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

[0017] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0018] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0019] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0020] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0021] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0022] [First embodiment]

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

[0024] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0025] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0026] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0027] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0028] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0029] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

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

[0031] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0032] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0033] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0034] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0035] This invention is a system for efficiently creating recommendation and referral text using a generative AI. Below, we will explain the specific system program, its processing content, and specific examples.

[0036] System configuration

[0037] The system consists of three main components:

[0038] 1. User Device

[0039] 2. Server

[0040] 3. Generative AI Model

[0041] User terminal operation

[0042] When a user wants to generate a recommendation using their device, they first access a web or native application. They enter information such as the name, field, characteristics, and memorable episodes of the person they are recommending into the provided input form. After entering all the information, the user presses the "Generate" button. This action converts the entered information into JSON format and sends it to the server as an HTTP POST request.

[0043] Server Operation

[0044] When the server receives a request from a user device, it first validates the data. Specifically, it checks whether all required fields are present and whether the characteristics and episodes are in list format. If the data passes validation, it is passed to the generative AI model, and the recommendation generation process begins.

[0045] The generative AI model takes the received data as input and generates a recommendation that takes into account the characteristics and anecdotes of the person being recommended. This generation process utilizes a natural language generation (NLG) algorithm. The generated recommendation is returned to the server and sent back to the user's device in JSON format.

[0046] How generative artificial intelligence models work

[0047] The generative AI model generates a recommendation based on the input data. This process proceeds as follows: First, the name and field of the person being recommended are inserted at the beginning, and then the recommendation is constructed by listing the characteristics of the person being recommended. Furthermore, memorable anecdotes entered by the user are integrated to enrich the specific content of the recommendation.

[0048] Specific examples

[0049] Let's take a concrete example. For example, suppose a user enters the information of a recommended person named "Yamada Taro" as follows:

[0050] Name: Yamada Taro

[0051] Field: Software Development

[0052] Traits: Hard work, teamwork, leadership

[0053] Episodes: Leadership in Project X, accurate communication in customer relations

[0054] When the user presses the "Generate" button, this information is sent to the server for validation. If validation is successful, the data is passed to a generative AI model, which generates a recommendation like this:

[0055] "Yamada Taro has excellent abilities in the software development field, with particular strengths in diligence, teamwork, and leadership. Specifically, he has demonstrated these qualities through his experience in Project X, such as his leadership and accurate communication in customer relations."

[0056] The generated recommendation is sent back from the server to the user's terminal and displayed on the screen, allowing the user to easily obtain a high-quality recommendation.

[0057] In this way, the present invention is a system that significantly reduces the time and effort required to create a recommendation, and provides users with efficient and high-quality document creation.

[0058] The processing flow will be explained below.

[0059] Step 1:

[0060] A user accesses an input form for writing a recommendation using a terminal, and inputs information about the name, field, characteristics, and memorable episode of the person to be recommended into the form.

[0061] Step 2:

[0062] The user presses the "Generate" button, which converts the entered information into JSON format and sends it to the server as an HTTP POST request.

[0063] Step 3:

[0064] The server parses the JSON data received from the user device and performs validation, checking whether all required fields are present and whether the characteristics and episodes are in list format.

[0065] Step 4:

[0066] The server passes the validated data to the generative AI model, which uses the validated data as input and generates a recommendation using an NLG (natural language generation) algorithm.

[0067] Step 5:

[0068] The generative AI model generates a recommendation based on the name, field, characteristics, and anecdotes of the person being recommended. In the generation process, the recommendation highlights the person's characteristics and integrates anecdotes to enrich the specific content.

[0069] Step 6:

[0070] The server receives the generated recommendation and returns it to the user's device in JSON format.

[0071] Step 7:

[0072] The user terminal displays the recommendation received from the server on the screen, allowing the user to easily obtain high-quality recommendations.

[0073] Example 1

[0074] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0075] Currently, writing testimonials and referrals is a time-consuming and laborious task, and creating high-quality testimonials requires specialized knowledge and experience. There is a need for a system that can solve this problem and automatically and efficiently generate high-quality testimonials.

[0076] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0077] In this invention, the server includes means for inputting information required to create a recommendation from a user's terminal, means for converting the information input from the terminal into JSON format and sending it to the server, means for the server to receive and validate the information, means for sending the validated information to a generative AI model to generate a recommendation, and means for returning the generated recommendation to the terminal. This enables users to easily and quickly create high-quality recommendations without requiring specialized knowledge.

[0078] A "user's terminal" is a digital device used by a user, such as a computer, smartphone, or tablet.

[0079] A "recommendation" is a piece of writing that is written to evaluate or recommend a specific individual or subject.

[0080] The "means for inputting information" refers to an interface (e.g., an input form) through which a user can input elements (such as name, field, characteristics, episodes, etc.) necessary for creating a recommendation.

[0081] The "means for converting to JSON format" is a mechanism for encoding user-entered information into JavaScript Object Notation (JSON) format.

[0082] A "server" is a computer system that receives and processes data sent from a user's terminal.

[0083] "Validation" is the process of checking whether entered data meets certain criteria.

[0084] A "generative artificial intelligence model" is a model that implements an algorithm that generates sentences based on input data using machine learning or deep learning techniques.

[0085] The "means for generating" is a process for generating a recommendation from input data using an artificial intelligence model.

[0086] The "means for sending back" is a mechanism for sending the generated recommendation to the user's terminal and displaying it.

[0087] The "Generate Button" is an interface element that initiates the submission of the information entered by the user and the generation of the recommendation.

[0088] The present invention is a system for efficiently creating recommendations and referrals using a generative AI model. The system includes the following components:

[0089] User terminal operation

[0090] When a user wants to generate a recommendation, they first access a web or native application. They enter information such as the name, field, characteristics, and memorable episodes of the person they are recommending into the provided input form. After entering all the information, the user presses the "Generate" button. This action converts the entered information into JSON format and sends it to the server as an HTTP POST request.

[0091] Server Operation

[0092] When the server receives a request from a user device, it first validates the data. Specifically, it checks whether all required fields are present and whether the characteristics and episodes are in list format. Data that passes validation is passed to a generative AI model, and the recommendation generation process begins. The generative AI model takes the received data as input and generates a recommendation that takes into account the characteristics and episodes of the person being recommended. The generated recommendation is returned to the server and sent back to the user device in JSON format.

[0093] How generative artificial intelligence models work

[0094] The generative AI model generates a recommendation based on the input data. This process works as follows: First, the name and field of the person being recommended are inserted at the beginning, and then the recommendation is constructed by listing the characteristics of the person being recommended. Furthermore, impressive anecdotes entered by the user are integrated to enrich the specific content of the recommendation.

[0095] Specific examples

[0096] Let's take a concrete example. For example, a user enters the information of a recommended person named "Taro Tanaka" as follows:

[0097] Name: Taro Tanaka

[0098] Field: Software Development

[0099] Traits: Hard work, teamwork, leadership

[0100] Episodes: Leadership in Project X, accurate communication in customer relations

[0101] When the user hits the "Generate" button, this information is sent to the server for validation. If validation is successful, the data is passed to a generative AI model, which generates a recommendation like this:

[0102] "Taro Tanaka has excellent abilities in the software development field, and his strengths are his diligence, teamwork, and leadership. Specifically, he has demonstrated these qualities through his experience in Project X, such as his leadership and accurate communication in customer relations."

[0103] The generated recommendation is sent back from the server to the user's terminal and displayed on the screen, allowing the user to easily obtain a high-quality recommendation.

[0104] Prompt Sentence Examples

[0105] An example of an input prompt for the generative AI model in this system might look like this:

[0106] Generate testimonials.

[0107] Name: Taro Tanaka

[0108] Field: Software Development

[0109] Traits: Hard work, teamwork, leadership

[0110] Episodes: Leadership in Project X, accurate communication in customer relations

[0111] In this way, the present invention is a system that significantly reduces the time and effort required to create a recommendation, and provides users with efficient and high-quality document creation.

[0112] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0113] Step 1: Collecting User Input

[0114] To generate a testimonial, a user accesses a web or native application and enters information into a form.

[0115] Input: The user enters the name, field, characteristics, and memorable episode of the person being recommended.

[0116] What happens: A user opens a web browser, such as Google Chrome, visits the system's web application URL, and then enters the following information into the form that appears on the screen:

[0117] Name: Taro Tanaka

[0118] Field: Software Development

[0119] Traits: Hard work, teamwork, leadership

[0120] Episodes: Leadership in Project X, accurate communication in customer relations

[0121] Output: The entered information is saved as form variables.

[0122] Step 2: Submitting input data

[0123] When the user presses the "Generate" button, the entered information is converted to JSON format and sent to the server.

[0124] Input: Information entered in step 1.

[0125] Specific behavior: When the user clicks the "Generate" button, the data is converted to JSON format using JavaScript AJAX (or Fetch API) and sent to the specified API endpoint.

[0126] Output: JSON formatted data is sent to the server as an HTTP POST request.

[0127] Step 3: Validate the data

[0128] The server receives the request from the user device and validates the data.

[0129] Input: The JSON data submitted in step 2.

[0130] What happens: The server (using the Django framework) validates the received JSON data. Specifically, it uses the Python jsonschema package to check the following:

[0131] Does it include names, areas, characteristics, and anecdotes?

[0132] Are the characteristics in list format?

[0133] Output: The validated data or a validation error response.

[0134] Step 4: Request a recommendation

[0135] The server passes the validated data to the generative AI model and makes a request to generate a recommendation.

[0136] Input: Data that passes validation.

[0137] What happens: The server sends the request data to a generative AI model (e.g., OpenAI's GPT-3 API). It uses the Python requests library to make the following request:

[0138] python

[0139] response = requests.post("https: / / api.openai.com / v1 / engines / davinci-codex / completions",

[0140] headers={"Authorization": "Bearer YOUR_API_KEY"},

[0141] json={"prompt": "Generate a testimonial.\nName: Taro Tanaka\nField: Software development\nCharacteristics: Diligence, teamwork, leadership\nAnecdotes: Leadership in Project X, accurate communication in customer relations", "max_tokens": 150})

[0142] Output: A recommendation generation request to the generative AI model.

[0143] Step 5: Generate testimonials

[0144] The generative AI model generates a recommendation based on the data provided.

[0145] Input: The prompt data submitted in step 4.

[0146] How it works: A generative AI model (OpenAI GPT-3) generates a recommendation based on the data provided as a prompt. The model generates text that includes the name, characteristics, and anecdotes of the person being recommended.

[0147] Output: The generated testimonial.

[0148] Step 6: Return and view testimonials

[0149] The server receives the generated recommendation and returns it to the user device in JSON format.

[0150] Input: The testimonial generated in step 5.

[0151] How it works: The server receives the response from the generated AI model and sends it back to the user's device as an HTTP response. The response data is processed by JavaScript on the user's device, and the generated recommendation is displayed at the bottom of the form.

[0152] Output: A testimonial that will be displayed on the user's device.

[0153] (Application example 1)

[0154] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0155] Creating recommendations and product reviews in the past was time-consuming and laborious, and it was difficult to maintain a consistent level of quality. Furthermore, if the information entered by the user was insufficient or inaccurate, the reliability and quality of the generated text declined. Furthermore, systems with no or insufficient validation functions often resulted in incomplete information being used as is. Another issue was the difficulty of accurately reflecting the user's intent when generating recommendations and reviews using natural language generation algorithms.

[0156] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0157] In this invention, the server includes means for inputting information required for creating a recommendation or review from a user's terminal, means for receiving and validating the information input from the terminal, means for sending the validated information to a generative AI model to generate a recommendation or review, and means for returning the generated recommendation or review to the terminal. This ensures the accuracy and completeness of the information input by the user and enables the automatic generation of high-quality recommendations and reviews.

[0158] A "terminal" is an electronic device through which a user enters information and receives generated text.

[0159] "Validation" is the process of ensuring that entered information is accurate and complete.

[0160] A "generative artificial intelligence model" is an artificial intelligence algorithm that automatically generates recommendations and reviews based on input information.

[0161] A "testimonial" is a written description that is written to evaluate and recommend another person.

[0162] A "review" is a document that describes impressions and evaluations of a product.

[0163] A "system" is a structure with a set of functions consisting of terminals, servers, and generative artificial intelligence models.

[0164] A "natural language generation algorithm" is a technology that allows artificial intelligence to generate sentences using human natural language.

[0165] "Information" is data entered by a user to generate a recommendation or review.

[0166] "JSON format" is an abbreviation for JavaScript Object Notation, and is a method of structuring data and representing it in text format.

[0167] An "episode" is a specific event or experience related to the characteristics or usability of the recommended person or product.

[0168] In this invention, the user terminal provides an interface for inputting information for generating a recommendation or product review. The user inputs the product name, reason for purchase, impressions of use, memorable episodes, etc. into an input form on the terminal. For example, the following information is input:

[0169] Product Name: High-Quality Laptop

[0170] Reason for purchase: I needed a powerful laptop for work.

[0171] User experience: Fast operation and long battery life

[0172] Episode: The battery lasted for 8 hours during a business trip, making it useful for presentations.

[0173] This input information is converted to JSON format and sent to the server as an HTTP POST request. The server validates the received data to ensure all required fields are present and the episodes are in a list format. If validation is successful, the server passes the data to a generative AI model to begin the process of generating reviews.

[0174] The server uses OpenAI's API to run a natural language generation (NLG) algorithm to generate a prompt based on the information the user has entered. For example, the following prompt might be generated:

[0175] Product Name: High-Quality Laptop

[0176] Reason for purchase: I needed a powerful laptop for work.

[0177] User experience: Fast operation and long battery life

[0178] Episode: The battery lasted for 8 hours during a business trip, making it useful for presentations.

[0179] Please use the information above to write your product review."

[0180] When this prompt is sent to the OpenAI API, the generated review is sent back to the server and then sent to the user's device in JSON format, allowing users to obtain high-quality reviews in a short amount of time.

[0181] This system uses a terminal, a server, and a generative AI model (e.g., OpenAI's API). The terminal uses a typical smartphone or PC browser, and the server uses a web application using Python and the Flask framework. Information entered by the user is validated on the server and then passed to the generative AI model. The generated text is then sent back to the user's terminal via the server.

[0182] In particular, the automatic generation of product reviews has the advantage of saving users the trouble of writing reviews while providing high-quality reviews. This system makes it possible to consistently improve the quality of reviews and create them quickly.

[0183] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0184] Step 1:

[0185] The user enters information into the device's input form, such as the product name, reason for purchase, impressions on use, memorable episodes, etc. The entered information is converted into JSON format.

[0186] Input: Product name, reason for purchase, impressions of use, memorable episode

[0187] Output: JSON format data

[0188] Step 2:

[0189] The terminal sends the data converted into JSON format to the server as an HTTP POST request.

[0190] Input: JSON format data

[0191] Output: HTTP POST request

[0192] Step 3:

[0193] The server receives the HTTP POST request and performs data validation, ensuring all required fields are present and that the episodes are in a list format.

[0194] Input: JSON data via HTTP POST request

[0195] Output: Validation result (success / failure)

[0196] Step 4:

[0197] If validation is successful, the server sends the data to a generative artificial intelligence model to begin the process of generating the review.

[0198] Input: Validated JSON data

[0199] Output: Input data to a generative AI model

[0200] Step 5:

[0201] The server generates a prompt for the generative AI model (using OpenAI's API) and runs the NLG algorithm to create a prompt based on the information entered by the user.

[0202] Input: Data that has been successfully validated

[0203] Output: Prompt sentence generated by generative AI model

[0204] Step 6:

[0205] A generative AI model receives the prompt and generates a recommendation or review. During this generation process, a review with a specific structure is created based on the user's input.

[0206] Input: prompt statement

[0207] Output: Generated testimonial or review

[0208] Step 7:

[0209] The generated recommendation and review are sent back to the server, which then sends them back to the user's device in JSON format.

[0210] Input: Generated testimonial or review

[0211] Output: Generated result data in JSON format

[0212] Step 8:

[0213] The user device receives the generated recommendation and review and displays it to the user, allowing the user to check high-quality reviews.

[0214] Input: Generated result data in JSON format

[0215] Output: Displayed testimonial or review on the user's device

[0216] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0217] The present invention provides a system for recognizing a user's emotions and generating a recommendation sentence by taking the emotion information into consideration. The system includes a user terminal, a server, a generative artificial intelligence model, and an emotion engine.

[0218] System configuration and operation

[0219] User terminal operation

[0220] A user accesses the system using a terminal and enters information into an input form to create a recommendation. The information entered includes the name, field, characteristics, and memorable episodes of the person being recommended. In addition, the user's terminal utilizes an emotion engine to recognize and analyze the user's emotional state (e.g., joy, anger, sadness, etc.) at the time of input.

[0221] Server Operation

[0222] The information obtained from the input form and emotion engine is sent to the server in JSON format. The server has the following functions:

[0223] 1. Data Receipt and Validation:

[0224] The server receives the JSON data sent from the user device and checks whether all required fields are present and whether the characteristics and episodes are in list format.

[0225] 2. Post-validation data processing:

[0226] The data that passes validation and the emotional information obtained from the emotion engine are passed to the generative artificial intelligence model.

[0227] How generative artificial intelligence models work

[0228] The generative AI model generates recommendations based on the received data and sentiment information. This process includes the following steps:

[0229] 1. Integration of basic information:

[0230] The nominee's name, field, characteristics, and anecdote are received as input data.

[0231] 2. Reflecting emotional information:

[0232] The tone and content of the recommendation are adjusted based on the emotional information obtained from the emotion engine. For example, if the user is happy, the recommendation will be generated in a style that reflects that positive emotion.

[0233] 3. Generating Recommendations:

[0234] Finally, a recommendation is finalized using a natural language generation (NLG) algorithm.

[0235] Returning recommendation letters

[0236] The generated recommendation is sent back to the server and sent in JSON format to the user's device, where it is displayed on the screen, allowing the user to obtain a high-quality, emotionally relevant recommendation.

[0237] Specific examples

[0238] Specific examples are shown below.

[0239] 1. If a user wants to write a recommendation, they access the system and enter the information of the person being recommended, "Ichiro Tanaka":

[0240] Name: Ichiro Tanaka

[0241] Area: Marketing

[0242] Traits: Creativity, leadership, analytical skills

[0243] Memorable episodes: Successful new product launch project, winning a marketing strategy competition

[0244] 2. While the user is entering information, the emotion engine detects the user's emotion of joy.

[0245] 3. When the user presses the "Generate" button, the input information and emotion information are sent to the server.

[0246] 4. The server receives the information and validates it.

[0247] 5. The information that passes validation is passed to the generative AI model, which takes into account the emotional information and generates a recommendation like the following:

[0248] "Ichiro Tanaka has outstanding abilities in the field of marketing, with particular strengths in creativity, leadership, and analytical ability. He has a proven track record, including successfully implementing new product launch projects and winning marketing strategy competitions. His positive attitude and energy have a positive impact on the entire team."

[0249] 6. The generated recommendation is sent back from the server to the user's device and displayed on the screen.

[0250] This allows users to easily obtain high-quality recommendation messages that reflect their emotional state, thereby improving the efficiency and accuracy of recommendation message creation.

[0251] The processing flow will be explained below.

[0252] Step 1:

[0253] The user accesses the system's input form using a terminal and inputs the name, field, characteristics, and memorable episode of the person to be recommended.

[0254] Step 2:

[0255] While the user is typing, the device's onboard emotion engine recognizes the user's emotional state, using the camera and microphone to analyze the user's facial expressions and vocal tone.

[0256] Step 3:

[0257] Once the user has entered all the required information, they press the "Generate" button. This action converts the entered information and emotion data into JSON format and sends it to the server as an HTTP POST request.

[0258] Step 4:

[0259] The server parses the JSON data received from the user device and performs data validation, checking whether all required fields are present and whether the characteristics and episodes are in list format.

[0260] Step 5:

[0261] The server passes the validated data and emotion information to the generative AI model. Specifically, the data containing emotion information is input into the model.

[0262] Step 6:

[0263] The generative AI model generates a recommendation based on the received data and emotional information, taking into account the characteristics and anecdotes of the person being recommended and adjusting the writing style and tone based on the emotional information.

[0264] Step 7:

[0265] The generated recommendation is sent back to the server, which then sends it back to the user's device in JSON format.

[0266] Step 8:

[0267] The user's device analyzes the recommendation received from the server and displays it on the screen, allowing the user to obtain high-quality recommendations that reflect their emotional state.

[0268] Example 2

[0269] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0270] Conventional recommendation generation systems update recommendation sentences in a fixed style without considering the user's emotional state, making it difficult to generate persuasive recommendation sentences that fit the user's emotions. Another problem is that processing data without proper validation reduces the quality of the generated recommendation sentences. The purpose of this invention is to solve these problems and generate high-quality recommendation sentences that reflect the user's emotional state.

[0271] The specification process by the specification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for inputting information necessary for creating a recommendation from the user's information processing device, means for analyzing the user's emotional state using an emotion recognition engine, means for receiving and validating the information and emotional information input from the processing device, means for sending the validated information and emotional information to a generative AI model and generating a recommendation, and means for returning the generated recommendation to the processing device. This makes it possible to generate high-quality recommendations that reflect the user's emotional state.

[0272] The "user's information processing device" refers to a device used by the user, such as a computer, smartphone, or tablet, and is a device that has an interface for inputting information for creating a recommendation.

[0273] An "emotion recognition engine" is software or hardware that analyzes a user's emotional state (e.g., joy, anger, sadness, etc.) using the user's typing speed, keystroke strength, facial recognition technology, etc.

[0274] Validation is the process of ensuring that the information entered by the user is accurate, by checking that all required fields are present and that the format is correct.

[0275] A "generative artificial intelligence model" is a machine learning model that generates recommendations using natural language generation algorithms based on input data and emotional information.

[0276] A "recommendation letter" is a document that reflects the characteristics and anecdotes about the recommended person specified by the user, as well as the emotional state of the user at the time, and is used for the purpose of evaluation or introduction.

[0277] A "natural language generation algorithm" is a technology that allows computers to generate human language, and is a method for automatically creating sentences based on input data.

[0278] "JSON format" is an abbreviation for JavaScript Object Notation, a lightweight data exchange format and a text format that is easy to read for both humans and machines.

[0279] The present invention provides a system for recognizing a user's emotions and generating a recommendation sentence by taking the emotion information into consideration. The system includes a user's information processing device, a server, a generative artificial intelligence model, and an emotion recognition engine.

[0280] System configuration and operation

[0281] Operation of user information processing device

[0282] A user inputs necessary information (the name, field, characteristics, and memorable episode of the person being recommended) into an input form for creating a recommendation.

[0283] For example, input may be in the following format:

[0284] Name: Name of the nominee

[0285] Area: Marketing

[0286] Traits: Creativity, leadership, analytical skills

[0287] Memorable episodes: Successful new product launch project, winning a marketing strategy competition

[0288] The user information processing device also uses an emotion recognition engine to analyze the user's emotional state while typing. The emotion recognition engine detects emotions (e.g., joy, anger, sadness) through the user's typing speed, keystroke strength, and facial recognition technology.

[0289] Server Operation

[0290] When the user presses the "Generate" button, the input information and emotion information are sent to the server in JSON format. The server provides the following functions:

[0291] 1. Data reception and validation: The server receives the JSON data sent from the user information processing device and checks whether all required fields are present and whether the characteristics and episodes are in list format.

[0292] 2. Post-validation data processing: The data and sentiment information that have passed validation are passed to the generative AI model.

[0293] How generative artificial intelligence models work

[0294] The generative AI model generates recommendations based on the received data and emotional information. The generative AI model works as follows:

[0295] 1. Integration of basic information: The name, field, characteristics, and anecdotes of the nominee are received as input data.

[0296] 2. Reflection of emotional information: The tone and content of the recommendation are adjusted based on the emotional information obtained from the emotion recognition engine. For example, if the user is happy, the recommendation will be generated in a style that reflects that positive emotion.

[0297] 3. Recommendation Generation: A natural language generation (NLG) algorithm is used to complete the recommendation.

[0298] Returning recommendation letters

[0299] The generated recommendation is returned to the server and sent in JSON format to the user's information processing device, which then displays the received recommendation on its screen, allowing the user to obtain a high-quality recommendation that is tailored to their emotions.

[0300] Specific examples

[0301] When a user wants to write a testimonial, they access the system and enter the following information:

[0302] Name: Ichiro Tanaka

[0303] Area: Marketing

[0304] Traits: Creativity, leadership, analytical skills

[0305] Memorable episodes: Successful new product launch project, winning a marketing strategy competition

[0306] While the user is entering information, the emotion recognition engine detects the user's emotion of "happiness." The input information and emotion information are then sent to the server, where, after validation, they are passed to the generative AI model. The generative AI model generates the following recommendation based on the emotion information:

[0307] "Ichiro Tanaka has outstanding abilities in the field of marketing, with particular strengths in creativity, leadership, and analytical ability. He has a proven track record, including successfully implementing new product launch projects and winning marketing strategy competitions. His positive attitude and energy have a positive impact on the entire team."

[0308] The generated recommendation is sent back from the server to the user's information processing device and displayed on the screen, allowing the user to easily obtain a high-quality recommendation that reflects their emotional state.

[0309] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0310] Step 1:

[0311] Obtaining user input information

[0312] Process description: A user inputs the required information (the name, field, characteristics, and memorable episode of the person being recommended) into an input form to create a recommendation.

[0313] What happens: A user opens a browser and visits a web page on the system. The user enters the following information into the form fields:

[0314] Name: Ichiro Tanaka

[0315] Area: Marketing

[0316] Traits: Creativity, leadership, analytical skills

[0317] Memorable episodes: Successful new product launch project, winning a marketing strategy competition

[0318] Input: The user enters information into an input form on the device.

[0319] Output: The terminal holds the information entered.

[0320] Step 2:

[0321] Emotion recognition using an emotion recognition engine

[0322] Process Description: The emotion recognition engine analyzes the user's emotional state (happiness, anger, sadness, etc.).

[0323] How it works: The emotion recognition engine analyzes the user's emotions through the camera using facial recognition technology, the speed of their typing, and the strength of their keystrokes. The emotion engine detects the emotional state of "joy."

[0324] Input: User input data and facial image data while the user is entering information into an input form.

[0325] Output: Parsed emotion information (e.g., joy).

[0326] Step 3:

[0327] Sending input data and emotion data to the server

[0328] Process description: Send user input information and emotion information to the server in JSON format.

[0329] Specific operation: When the user presses the "Generate" button, the device converts the input information and emotion information into JSON. The device then sends the converted JSON data to the server via HTTPS.

[0330] Input: User input and emotional information.

[0331] Output: The data converted to JSON format is sent to the server.

[0332] Step 4:

[0333] Receiving and validating data by the server

[0334] Process description: The server receives the JSON data and checks whether the entered data is correct.

[0335] What happens: The server parses the received JSON data and verifies that all required fields (name, field, characteristics, episode) are present. Data validation is performed to ensure the format is correct.

[0336] Input: Input and emotion information received in JSON format.

[0337] Output: Data that passes validation.

[0338] Step 5:

[0339] Recommendation generation using generative artificial intelligence models

[0340] Process description: The server passes the validated data and sentiment information to the generative AI model to generate a recommendation.

[0341] Specific operation: The server passes the validated data to the generative AI model. The generative AI model takes into account the user's emotional information (delight) and adjusts the tone and content of the recommendation. The generative AI model generates the recommendation using a natural language generation (NLG) algorithm. For example, a recommendation might read, "Ichiro Tanaka has outstanding abilities in the marketing field, with particular strengths in creativity, leadership, and analytical ability. He has a long list of achievements, including the success of a new product launch project and winning a marketing strategy competition. His proactive attitude and positive energy have a positive influence on the entire team."

[0342] Input: Validated input and sentiment information.

[0343] Output: The generated recommendation.

[0344] Step 6:

[0345] Sending and displaying the recommendation to the user's device

[0346] Process description: The generated recommendation is sent back to the user's device in JSON format from the server and displayed on the screen.

[0347] Specific operation: The server converts the generated recommendation into JSON and sends it to the user's device. The user's device parses the received JSON data and displays the recommendation on the screen.

[0348] Input: JSON data containing the generated testimonials.

[0349] Output: The testimonial displayed on the device.

[0350] (Application example 2)

[0351] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0352] Conventional recommendation generation systems generate recommendation sentences uniformly without considering the user's emotional information, which can result in recommendation sentences that do not match the user's emotions or nuances. This means that users have to spend time and effort writing recommendation sentences, and the generated recommendation sentences do not always fully reflect the user's intentions.

[0353] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0354] In this invention, the server includes means for receiving information necessary for creating a recommendation and user emotional information from the user's terminal and validating the information, means for sending the validated information and emotional information to a generative AI model to generate a recommendation, and means for returning the generated recommendation to the terminal, thereby enabling the generation of high-quality recommendations that reflect the user's emotional information.

[0355] "User terminal" means a device used by a user to access the system and input information, including, for example, a smartphone, tablet, or personal computer.

[0356] "Information" refers to the name, field, characteristics, memorable episodes, etc. of the person being recommended, which are necessary to write the recommendation letter.

[0357] "Emotion information" is data on the user's emotional state (for example, joy, anger, sadness, etc.) extracted based on information input by the user.

[0358] "Validation" is the process of ensuring that the information and sentiment information entered by the user is accurate and complete.

[0359] "Generative AI models" are algorithms or models that generate recommendations based on received information and emotional information. Examples include natural language generation (NLG) algorithms.

[0360] A "recommendation letter" is a piece of writing generated based on information about the person being recommended, and includes content that appealingly conveys the characteristics and anecdotes of the person being recommended.

[0361] A "natural language generation algorithm" is an algorithm for generating understandable and appropriate sentences based on information and emotional information entered by the user.

[0362] "JSON format" is an abbreviation for JavaScript Object Notation, and is a format for expressing structured data in text format.

[0363] "Terminal" refers to any device that accepts user operations and processes information, including smartphones, personal computers, and tablets.

[0364] A "server" is a computer system that receives information sent from a user's terminal, processes it, and returns a final recommendation.

[0365] The present invention relates to a system for generating recommendation sentences taking into account a user's emotional information, which includes a user terminal, a server, a generative artificial intelligence model, and an emotional engine.

[0366] The main components of the system are as follows:

[0367] User terminal

[0368] The user terminal is a device for inputting information and recognizing emotions. Examples include smartphones, tablets, and personal computers. The user uses the terminal to access the system and input information to create a recommendation. The input information includes the name, field, characteristics, and memorable anecdotes of the person being recommended. The system also uses an emotion engine to recognize and analyze the user's emotional state (e.g., joy, anger, sadness, etc.).

[0369] server

[0370] The server is a central device that receives information and emotion information sent from user terminals and generates recommendation statements. Specifically, it has the following functions:

[0371] 1. Information reception and validation: The server receives the information and emotion information entered from the user device in JSON format. It then checks whether all required fields are present and whether the characteristics and episodes are in list format, and performs validation.

[0372] 2. Data processing: Receive the validated information and sentiment information and pass it to the generative AI model.

[0373] Generative AI Model

[0374] The generative AI model generates recommendations based on the received data and emotional information. Specifically, it performs the following processes:

[0375] 1. Basic information integration: Receive the nominee's name, field, characteristics, and anecdotes as input data.

[0376] 2. Reflection of emotional information: The tone and content of the recommendation are adjusted based on the emotional information obtained from the emotion engine. If the user's emotion of joy is detected, the recommendation is generated in a positive tone.

[0377] 3. Recommendation generation: A natural language generation algorithm (e.g., GPT-3) is used to complete the final recommendation.

[0378] Returning recommendation letters

[0379] The generated recommendation is sent back to the server and sent in JSON format to the user's device, where it is displayed on the screen, allowing the user to obtain a high-quality, emotionally relevant recommendation.

[0380] Specific examples

[0381] For example, if a user wants to write a movie review, they enter the following information:

[0382] Username: Taro Nakamura

[0383] Content Type: Movies

[0384] Content Title: Inception

[0385] Details of my viewing experience: I was very moved by the final scene and was moved to tears.

[0386] The emotion engine recognizes the user's emotions and generates a review that looks like this:

[0387] "Taro Nakamura watched the movie 'Inception' and was deeply moved, especially by the final scene. He was moved to tears, and the beauty of the scene and the moving story resonated deeply in his heart. Director Christopher Nolan's outstanding directing skills shine through, and Nakamura has a very high opinion of the film."

[0388] Prompt Sentence Examples

[0389] "Recognize emotions based on the content viewing experience and generate reviews that correspond to the user's emotions."

[0390] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0391] Step 1:

[0392] The user accesses the recommendation writing form using a terminal and inputs information about the person being recommended (name, field, characteristics, memorable episodes), along with details of the user's viewing experience.

[0393] Input: Name, field, characteristics, memorable episodes, details of viewing experience

[0394] Output: Input information and data passed to the emotion recognition engine

[0395] Step 2:

[0396] The device's emotion recognition engine analyzes the details of the user's viewing experience and recognizes the user's emotional state. For example, if the user inputs "I was moved," it will recognize the emotion of being moved.

[0397] Input: Viewing experience details

[0398] Output: User's emotional information (e.g., emotion, joy, anger, etc.)

[0399] Step 3:

[0400] The device converts the input information and recognized emotion information into JSON format and sends it to the server.

[0401] Input: Name, field, characteristics, memorable episodes, user's emotional information

[0402] Output: JSON format data

[0403] Step 4:

[0404] The server parses the data received in JSON format, checks whether all required fields are present, whether the characteristics and episodes are in list format, and performs validation.

[0405] Input: JSON format data

[0406] Output: Successfully validated data or an error message

[0407] Step 5:

[0408] The server passes the validated data and emotion information to the generative artificial intelligence model.

[0409] Input: Successfully validated data, emotion information

[0410] Output: The data used to generate the recommendation

[0411] Step 6:

[0412] The generative AI model generates a recommendation based on the received data. Specifically, it integrates basic information (name, field, characteristics, memorable episodes) with emotional information and applies a natural language generation algorithm (NLG) to generate the recommendation.

[0413] Input: Basic information, emotional information

[0414] Output: Generated recommendation

[0415] Step 7:

[0416] The generated recommendation is sent back from the server to the user's device in JSON format.

[0417] Input: Generated testimonial

[0418] Output: Recommendation data in JSON format

[0419] Step 8:

[0420] The user's device analyzes the received recommendation and displays it on the screen, allowing the user to obtain a recommendation that is high quality and emotionally appropriate.

[0421] Input: Recommendation data in JSON format

[0422] Output: Testimonial displayed on screen

[0423] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0424] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0425] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0426] [Second embodiment]

[0427] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0428] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0429] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0430] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0431] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0432] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0433] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0434] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0435] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0436] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0437] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0438] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0439] This invention is a system for efficiently creating recommendation and referral text using a generative AI. Below, we will explain the specific system program, its processing content, and specific examples.

[0440] System configuration

[0441] The system consists of three main components:

[0442] 1. User Device

[0443] 2. Server

[0444] 3. Generative AI Model

[0445] User terminal operation

[0446] When a user wants to generate a recommendation using their device, they first access a web or native application. They enter information such as the name, field, characteristics, and memorable episodes of the person they are recommending into the provided input form. After entering all the information, the user presses the "Generate" button. This action converts the entered information into JSON format and sends it to the server as an HTTP POST request.

[0447] Server Operation

[0448] When the server receives a request from a user device, it first validates the data. Specifically, it checks whether all required fields are present and whether the characteristics and episodes are in list format. If the data passes validation, it is passed to the generative AI model, and the recommendation generation process begins.

[0449] The generative AI model takes the received data as input and generates a recommendation that takes into account the characteristics and anecdotes of the person being recommended. This generation process utilizes a natural language generation (NLG) algorithm. The generated recommendation is returned to the server and sent back to the user's device in JSON format.

[0450] How generative artificial intelligence models work

[0451] The generative AI model generates a recommendation based on the input data. This process proceeds as follows: First, the name and field of the person being recommended are inserted at the beginning, and then the recommendation is constructed by listing the characteristics of the person being recommended. Furthermore, memorable anecdotes entered by the user are integrated to enrich the specific content of the recommendation.

[0452] Specific examples

[0453] Let's take a concrete example. For example, suppose a user enters the information of a recommended person named "Yamada Taro" as follows:

[0454] Name: Yamada Taro

[0455] Field: Software Development

[0456] Traits: Hard work, teamwork, leadership

[0457] Episodes: Leadership in Project X, accurate communication in customer relations

[0458] When the user presses the "Generate" button, this information is sent to the server for validation. If validation is successful, the data is passed to a generative AI model, which generates a recommendation like this:

[0459] "Yamada Taro has excellent abilities in the software development field, with particular strengths in diligence, teamwork, and leadership. Specifically, he has demonstrated these qualities through his experience in Project X, such as his leadership and accurate communication in customer relations."

[0460] The generated recommendation is sent back from the server to the user's terminal and displayed on the screen, allowing the user to easily obtain a high-quality recommendation.

[0461] In this way, the present invention is a system that significantly reduces the time and effort required to create a recommendation, and provides users with efficient and high-quality document creation.

[0462] The processing flow will be explained below.

[0463] Step 1:

[0464] A user accesses an input form for writing a recommendation using a terminal, and inputs information about the name, field, characteristics, and memorable episode of the person to be recommended into the form.

[0465] Step 2:

[0466] The user presses the "Generate" button, which converts the entered information into JSON format and sends it to the server as an HTTP POST request.

[0467] Step 3:

[0468] The server parses the JSON data received from the user device and performs validation, checking whether all required fields are present and whether the characteristics and episodes are in list format.

[0469] Step 4:

[0470] The server passes the validated data to the generative AI model, which uses the validated data as input and generates a recommendation using an NLG (natural language generation) algorithm.

[0471] Step 5:

[0472] The generative AI model generates a recommendation based on the name, field, characteristics, and anecdotes of the person being recommended. In the generation process, the recommendation highlights the person's characteristics and integrates anecdotes to enrich the specific content.

[0473] Step 6:

[0474] The server receives the generated recommendation and returns it to the user's device in JSON format.

[0475] Step 7:

[0476] The user terminal displays the recommendation received from the server on the screen, allowing the user to easily obtain high-quality recommendations.

[0477] Example 1

[0478] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0479] Currently, writing testimonials and referrals is a time-consuming and laborious task, and creating high-quality testimonials requires specialized knowledge and experience. There is a need for a system that can solve this problem and automatically and efficiently generate high-quality testimonials.

[0480] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0481] In this invention, the server includes means for inputting information required to create a recommendation from a user's terminal, means for converting the information input from the terminal into JSON format and sending it to the server, means for the server to receive and validate the information, means for sending the validated information to a generative AI model to generate a recommendation, and means for returning the generated recommendation to the terminal. This enables users to easily and quickly create high-quality recommendations without requiring specialized knowledge.

[0482] A "user's terminal" is a digital device used by a user, such as a computer, smartphone, or tablet.

[0483] A "recommendation" is a piece of writing that is written to evaluate or recommend a specific individual or subject.

[0484] The "means for inputting information" refers to an interface (e.g., an input form) through which a user can input elements (such as name, field, characteristics, episodes, etc.) necessary for creating a recommendation.

[0485] The "means for converting to JSON format" is a mechanism for encoding user-entered information into JavaScript Object Notation (JSON) format.

[0486] A "server" is a computer system that receives and processes data sent from a user's terminal.

[0487] "Validation" is the process of checking whether entered data meets certain criteria.

[0488] A "generative artificial intelligence model" is a model that implements an algorithm that generates sentences based on input data using machine learning or deep learning techniques.

[0489] The "means for generating" is a process for generating a recommendation from input data using an artificial intelligence model.

[0490] The "means for sending back" is a mechanism for sending the generated recommendation to the user's terminal and displaying it.

[0491] The "Generate Button" is an interface element that initiates the submission of the information entered by the user and the generation of the recommendation.

[0492] The present invention is a system for efficiently creating recommendations and referrals using a generative AI model. The system includes the following components:

[0493] User terminal operation

[0494] When a user wants to generate a recommendation, they first access a web or native application. They enter information such as the name, field, characteristics, and memorable episodes of the person they are recommending into the provided input form. After entering all the information, the user presses the "Generate" button. This action converts the entered information into JSON format and sends it to the server as an HTTP POST request.

[0495] Server Operation

[0496] When the server receives a request from a user device, it first validates the data. Specifically, it checks whether all required fields are present and whether the characteristics and episodes are in list format. Data that passes validation is passed to a generative AI model, and the recommendation generation process begins. The generative AI model takes the received data as input and generates a recommendation that takes into account the characteristics and episodes of the person being recommended. The generated recommendation is returned to the server and sent back to the user device in JSON format.

[0497] How generative artificial intelligence models work

[0498] The generative AI model generates a recommendation based on the input data. This process works as follows: First, the name and field of the person being recommended are inserted at the beginning, and then the recommendation is constructed by listing the characteristics of the person being recommended. Furthermore, impressive anecdotes entered by the user are integrated to enrich the specific content of the recommendation.

[0499] Specific examples

[0500] Let's take a concrete example. For example, a user enters the information of a recommended person named "Taro Tanaka" as follows:

[0501] Name: Taro Tanaka

[0502] Field: Software Development

[0503] Traits: Hard work, teamwork, leadership

[0504] Episodes: Leadership in Project X, accurate communication in customer relations

[0505] When the user hits the "Generate" button, this information is sent to the server for validation. If validation is successful, the data is passed to a generative AI model, which generates a recommendation like this:

[0506] "Taro Tanaka has excellent abilities in the software development field, and his strengths are his diligence, teamwork, and leadership. Specifically, he has demonstrated these qualities through his experience in Project X, such as his leadership and accurate communication in customer relations."

[0507] The generated recommendation is sent back from the server to the user's terminal and displayed on the screen, allowing the user to easily obtain a high-quality recommendation.

[0508] Prompt Sentence Examples

[0509] An example of an input prompt for the generative AI model in this system might look like this:

[0510] Generate testimonials.

[0511] Name: Taro Tanaka

[0512] Field: Software Development

[0513] Traits: Hard work, teamwork, leadership

[0514] Episodes: Leadership in Project X, accurate communication in customer relations

[0515] In this way, the present invention is a system that significantly reduces the time and effort required to create a recommendation, and provides users with efficient and high-quality document creation.

[0516] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0517] Step 1: Collecting User Input

[0518] To generate a testimonial, a user accesses a web or native application and enters information into a form.

[0519] Input: The user enters the name, field, characteristics, and memorable episode of the person being recommended.

[0520] What happens: A user opens a web browser, such as Google Chrome, visits the system's web application URL, and then enters the following information into the form that appears on the screen:

[0521] Name: Taro Tanaka

[0522] Field: Software Development

[0523] Traits: Hard work, teamwork, leadership

[0524] Episodes: Leadership in Project X, accurate communication in customer relations

[0525] Output: The entered information is saved as form variables.

[0526] Step 2: Submitting input data

[0527] When the user presses the "Generate" button, the entered information is converted to JSON format and sent to the server.

[0528] Input: Information entered in step 1.

[0529] Specific behavior: When the user clicks the "Generate" button, the data is converted to JSON format using JavaScript AJAX (or Fetch API) and sent to the specified API endpoint.

[0530] Output: JSON formatted data is sent to the server as an HTTP POST request.

[0531] Step 3: Validate the data

[0532] The server receives the request from the user device and validates the data.

[0533] Input: The JSON data submitted in step 2.

[0534] What happens: The server (using the Django framework) validates the received JSON data. Specifically, it uses the Python jsonschema package to check the following:

[0535] Does it include names, areas, characteristics, and anecdotes?

[0536] Are the characteristics in list format?

[0537] Output: The validated data or a validation error response.

[0538] Step 4: Request a recommendation

[0539] The server passes the validated data to the generative AI model and makes a request to generate a recommendation.

[0540] Input: Data that passes validation.

[0541] What happens: The server sends the request data to a generative AI model (e.g., OpenAI's GPT-3 API). It uses the Python requests library to make the following request:

[0542] python

[0543] response = requests.post("https: / / api.openai.com / v1 / engines / davinci-codex / completions",

[0544] headers={"Authorization": "Bearer YOUR_API_KEY"},

[0545] json={"prompt": "Generate a testimonial.\nName: Taro Tanaka\nField: Software development\nCharacteristics: Diligence, teamwork, leadership\nAnecdotes: Leadership in Project X, accurate communication in customer relations", "max_tokens": 150})

[0546] Output: A recommendation generation request to the generative AI model.

[0547] Step 5: Generate testimonials

[0548] The generative AI model generates a recommendation based on the given data.

[0549] Input: The prompt data submitted in step 4.

[0550] How it works: A generative AI model (OpenAI GPT-3) generates a recommendation based on the data provided as a prompt. The model generates text that includes the name, characteristics, and anecdotes of the person being recommended.

[0551] Output: The generated testimonial.

[0552] Step 6: Return and view testimonials

[0553] The server receives the generated recommendation and returns it to the user device in JSON format.

[0554] Input: The testimonial generated in step 5.

[0555] How it works: The server receives the response from the generated AI model and sends it back to the user's device as an HTTP response. The response data is processed by JavaScript on the user's device, and the generated recommendation is displayed at the bottom of the form.

[0556] Output: A testimonial that will be displayed on the user's device.

[0557] (Application example 1)

[0558] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0559] Creating recommendations and product reviews in the past was time-consuming and laborious, and it was difficult to maintain a consistent level of quality. Furthermore, if the information entered by the user was insufficient or inaccurate, the reliability and quality of the generated text declined. Furthermore, systems with no or insufficient validation functions often resulted in incomplete information being used as is. Another issue was the difficulty of accurately reflecting the user's intent when generating recommendations and reviews using natural language generation algorithms.

[0560] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0561] In this invention, the server includes means for inputting information required for creating a recommendation or review from a user's terminal, means for receiving and validating the information input from the terminal, means for sending the validated information to a generative AI model to generate a recommendation or review, and means for returning the generated recommendation or review to the terminal. This ensures the accuracy and completeness of the information input by the user and enables the automatic generation of high-quality recommendations and reviews.

[0562] A "terminal" is an electronic device through which a user enters information and receives generated text.

[0563] "Validation" is the process of ensuring that entered information is accurate and complete.

[0564] A "generative artificial intelligence model" is an artificial intelligence algorithm that automatically generates recommendations and reviews based on input information.

[0565] A "testimonial" is a written description that is written to evaluate and recommend another person.

[0566] A "review" is a document that describes impressions and evaluations of a product.

[0567] A "system" is a structure with a set of functions consisting of terminals, servers, and generative artificial intelligence models.

[0568] A "natural language generation algorithm" is a technology that allows artificial intelligence to generate sentences using human natural language.

[0569] "Information" is data entered by a user to generate a recommendation or review.

[0570] "JSON format" is an abbreviation for JavaScript Object Notation, and is a method of structuring data and representing it in text format.

[0571] An "episode" is a specific event or experience related to the characteristics or usability of the recommended person or product.

[0572] In this invention, the user terminal provides an interface for inputting information for generating a recommendation or product review. The user inputs the product name, reason for purchase, impressions of use, memorable episodes, etc. into an input form on the terminal. For example, the following information is input:

[0573] Product Name: High-Quality Laptop

[0574] Reason for purchase: I needed a powerful laptop for work.

[0575] User experience: Fast operation and long battery life

[0576] Episode: The battery lasted for 8 hours during a business trip, making it useful for presentations.

[0577] This input information is converted to JSON format and sent to the server as an HTTP POST request. The server validates the received data to ensure all required fields are present and the episodes are in a list format. If validation is successful, the server passes the data to a generative AI model to begin the process of generating reviews.

[0578] The server uses OpenAI's API to run a natural language generation (NLG) algorithm to generate a prompt based on the information the user has entered. For example, the following prompt might be generated:

[0579] Product Name: High-Quality Laptop

[0580] Reason for purchase: I needed a powerful laptop for work.

[0581] User experience: Fast operation and long battery life

[0582] Episode: The battery lasted for 8 hours during a business trip, making it useful for presentations.

[0583] Please use the information above to write your product review."

[0584] When this prompt is sent to the OpenAI API, the generated review is sent back to the server and then sent to the user's device in JSON format, allowing users to obtain high-quality reviews in a short amount of time.

[0585] This system uses a terminal, a server, and a generative AI model (e.g., OpenAI's API). The terminal uses a typical smartphone or PC browser, and the server uses a web application using Python and the Flask framework. Information entered by the user is validated on the server and then passed to the generative AI model. The generated text is then sent back to the user's terminal via the server.

[0586] In particular, the automatic generation of product reviews has the advantage of saving users the trouble of writing reviews while providing high-quality reviews. This system makes it possible to consistently improve the quality of reviews and create them quickly.

[0587] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0588] Step 1:

[0589] The user enters information into the device's input form, such as the product name, reason for purchase, impressions on use, memorable episodes, etc. The entered information is converted into JSON format.

[0590] Input: Product name, reason for purchase, impressions of use, memorable episode

[0591] Output: JSON format data

[0592] Step 2:

[0593] The terminal sends the data converted into JSON format to the server as an HTTP POST request.

[0594] Input: JSON format data

[0595] Output: HTTP POST request

[0596] Step 3:

[0597] The server receives the HTTP POST request and performs data validation, ensuring all required fields are present and that the episodes are in a list format.

[0598] Input: JSON data via HTTP POST request

[0599] Output: Validation result (success / failure)

[0600] Step 4:

[0601] If validation is successful, the server sends the data to a generative artificial intelligence model to begin the process of generating the review.

[0602] Input: Validated JSON data

[0603] Output: Input data to a generative AI model

[0604] Step 5:

[0605] The server generates a prompt for the generative AI model (using OpenAI's API) and runs the NLG algorithm to create a prompt based on the information entered by the user.

[0606] Input: Data that has been successfully validated

[0607] Output: Prompt sentence generated by generative AI model

[0608] Step 6:

[0609] A generative AI model receives the prompt and generates a recommendation or review. During this generation process, a review with a specific structure is created based on the user's input.

[0610] Input: prompt statement

[0611] Output: Generated testimonial or review

[0612] Step 7:

[0613] The generated recommendation and review are sent back to the server, which then sends them back to the user's device in JSON format.

[0614] Input: Generated testimonial or review

[0615] Output: Generated result data in JSON format

[0616] Step 8:

[0617] The user device receives the generated recommendation and review and displays it to the user, allowing the user to check high-quality reviews.

[0618] Input: Generated result data in JSON format

[0619] Output: Displayed testimonial or review on the user's device

[0620] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0621] The present invention provides a system for recognizing a user's emotions and generating a recommendation sentence by taking the emotion information into consideration. The system includes a user terminal, a server, a generative artificial intelligence model, and an emotion engine.

[0622] System configuration and operation

[0623] User terminal operation

[0624] A user accesses the system using a terminal and enters information into an input form to create a recommendation. The information entered includes the name, field, characteristics, and memorable episodes of the person being recommended. In addition, the user's terminal utilizes an emotion engine to recognize and analyze the user's emotional state (e.g., joy, anger, sadness, etc.) at the time of input.

[0625] Server Operation

[0626] The information obtained from the input form and emotion engine is sent to the server in JSON format. The server has the following functions:

[0627] 1. Data Receipt and Validation:

[0628] The server receives the JSON data sent from the user device and checks whether all required fields are present and whether the characteristics and episodes are in list format.

[0629] 2. Post-validation data processing:

[0630] The data that passes validation and the emotional information obtained from the emotion engine are passed to the generative artificial intelligence model.

[0631] How generative artificial intelligence models work

[0632] The generative AI model generates recommendations based on the received data and sentiment information. This process includes the following steps:

[0633] 1. Integration of basic information:

[0634] The nominee's name, field, characteristics, and anecdote are received as input data.

[0635] 2. Reflecting emotional information:

[0636] The tone and content of the recommendation are adjusted based on the emotional information obtained from the emotion engine. For example, if the user is happy, the recommendation will be generated in a style that reflects that positive emotion.

[0637] 3. Generating Recommendations:

[0638] Finally, a recommendation is finalized using a natural language generation (NLG) algorithm.

[0639] Returning recommendation letters

[0640] The generated recommendation is sent back to the server and sent in JSON format to the user's device, where it is displayed on the screen, allowing the user to obtain a high-quality, emotionally relevant recommendation.

[0641] Specific examples

[0642] Specific examples are shown below.

[0643] 1. If a user wants to write a recommendation, they access the system and enter the information of the person being recommended, "Ichiro Tanaka":

[0644] Name: Ichiro Tanaka

[0645] Area: Marketing

[0646] Traits: Creativity, leadership, analytical skills

[0647] Memorable episodes: Successful new product launch project, winning a marketing strategy competition

[0648] 2. While the user is entering information, the emotion engine detects the user's emotion of joy.

[0649] 3. When the user presses the "Generate" button, the input information and emotion information are sent to the server.

[0650] 4. The server receives the information and validates it.

[0651] 5. The information that passes validation is passed to the generative AI model, which takes into account the emotional information and generates a recommendation like the following:

[0652] "Ichiro Tanaka has outstanding abilities in the field of marketing, with particular strengths in creativity, leadership, and analytical ability. He has a proven track record, including successfully implementing new product launch projects and winning marketing strategy competitions. His positive attitude and energy have a positive impact on the entire team."

[0653] 6. The generated recommendation is sent back from the server to the user's device and displayed on the screen.

[0654] This allows users to easily obtain high-quality recommendation messages that reflect their emotional state, thereby improving the efficiency and accuracy of recommendation message creation.

[0655] The processing flow will be explained below.

[0656] Step 1:

[0657] The user accesses the system's input form using a terminal and inputs the name, field, characteristics, and memorable episode of the person to be recommended.

[0658] Step 2:

[0659] While the user is typing, the device's onboard emotion engine recognizes the user's emotional state, using the camera and microphone to analyze the user's facial expressions and vocal tone.

[0660] Step 3:

[0661] Once the user has entered all the required information, they press the "Generate" button. This action converts the entered information and emotion data into JSON format and sends it to the server as an HTTP POST request.

[0662] Step 4:

[0663] The server parses the JSON data received from the user device and performs data validation, checking whether all required fields are present and whether the characteristics and episodes are in list format.

[0664] Step 5:

[0665] The server passes the validated data and emotion information to the generative AI model. Specifically, the data containing emotion information is input into the model.

[0666] Step 6:

[0667] The generative AI model generates a recommendation based on the received data and emotional information, taking into account the characteristics and anecdotes of the person being recommended and adjusting the writing style and tone based on the emotional information.

[0668] Step 7:

[0669] The generated recommendation is sent back to the server, which then sends it back to the user's device in JSON format.

[0670] Step 8:

[0671] The user's device analyzes the recommendation received from the server and displays it on the screen, allowing the user to obtain high-quality recommendations that reflect their emotional state.

[0672] Example 2

[0673] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0674] Conventional recommendation generation systems update recommendation sentences in a fixed style without considering the user's emotional state, making it difficult to generate persuasive recommendation sentences that fit the user's emotions. Another problem is that processing data without proper validation reduces the quality of the generated recommendation sentences. The purpose of this invention is to solve these problems and generate high-quality recommendation sentences that reflect the user's emotional state.

[0675] The specification process by the specification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for inputting information necessary for creating a recommendation from the user's information processing device, means for analyzing the user's emotional state using an emotion recognition engine, means for receiving and validating the information and emotional information input from the processing device, means for sending the validated information and emotional information to a generative AI model and generating a recommendation, and means for returning the generated recommendation to the processing device. This makes it possible to generate high-quality recommendations that reflect the user's emotional state.

[0676] The "user's information processing device" refers to a device used by the user, such as a computer, smartphone, or tablet, and is a device that has an interface for inputting information for creating a recommendation.

[0677] An "emotion recognition engine" is software or hardware that analyzes a user's emotional state (e.g., joy, anger, sadness, etc.) using the user's typing speed, keystroke strength, facial recognition technology, etc.

[0678] Validation is the process of ensuring that the information entered by the user is accurate, by checking that all required fields are present and that the format is correct.

[0679] A "generative artificial intelligence model" is a machine learning model that generates recommendations using natural language generation algorithms based on input data and emotional information.

[0680] A "recommendation letter" is a document that reflects the characteristics and anecdotes about the recommended person specified by the user, as well as the emotional state of the user at the time, and is used for the purpose of evaluation or introduction.

[0681] A "natural language generation algorithm" is a technology that allows computers to generate human language, and is a method for automatically creating sentences based on input data.

[0682] "JSON format" is an abbreviation for JavaScript Object Notation, a lightweight data exchange format and a text format that is easy to read for both humans and machines.

[0683] The present invention provides a system for recognizing a user's emotions and generating a recommendation sentence by taking the emotion information into consideration. The system includes a user's information processing device, a server, a generative artificial intelligence model, and an emotion recognition engine.

[0684] System configuration and operation

[0685] Operation of user information processing device

[0686] A user inputs necessary information (the name, field, characteristics, and memorable episode of the person being recommended) into an input form for creating a recommendation.

[0687] For example, input may be in the following format:

[0688] Name: Name of the nominee

[0689] Area: Marketing

[0690] Traits: Creativity, leadership, analytical skills

[0691] Memorable episodes: Successful new product launch project, winning a marketing strategy competition

[0692] The user information processing device also uses an emotion recognition engine to analyze the user's emotional state while typing. The emotion recognition engine detects emotions (e.g., joy, anger, sadness) through the user's typing speed, keystroke strength, and facial recognition technology.

[0693] Server Operation

[0694] When the user presses the "Generate" button, the input information and emotion information are sent to the server in JSON format. The server provides the following functions:

[0695] 1. Data reception and validation: The server receives the JSON data sent from the user information processing device and checks whether all required fields are present and whether the characteristics and episodes are in list format.

[0696] 2. Post-validation data processing: The data and sentiment information that have passed validation are passed to the generative AI model.

[0697] How generative artificial intelligence models work

[0698] The generative AI model generates recommendations based on the received data and emotional information. The generative AI model works as follows:

[0699] 1. Integration of basic information: The name, field, characteristics, and anecdotes of the nominee are received as input data.

[0700] 2. Reflection of emotional information: The tone and content of the recommendation are adjusted based on the emotional information obtained from the emotion recognition engine. For example, if the user is happy, the recommendation will be generated in a style that reflects that positive emotion.

[0701] 3. Recommendation Generation: A natural language generation (NLG) algorithm is used to complete the recommendation.

[0702] Returning recommendation letters

[0703] The generated recommendation is returned to the server and sent in JSON format to the user's information processing device, which then displays the received recommendation on its screen, allowing the user to obtain a high-quality recommendation that is tailored to their emotions.

[0704] Specific examples

[0705] When a user wants to write a testimonial, they access the system and enter the following information:

[0706] Name: Ichiro Tanaka

[0707] Area: Marketing

[0708] Traits: Creativity, leadership, analytical skills

[0709] Memorable episodes: Successful new product launch project, winning a marketing strategy competition

[0710] While the user is entering information, the emotion recognition engine detects the user's emotion of "happiness." The input information and emotion information are then sent to the server, where, after validation, they are passed to the generative AI model. The generative AI model generates the following recommendation based on the emotion information:

[0711] "Ichiro Tanaka has outstanding abilities in the field of marketing, with particular strengths in creativity, leadership, and analytical ability. He has a proven track record, including successfully implementing new product launch projects and winning marketing strategy competitions. His positive attitude and energy have a positive impact on the entire team."

[0712] The generated recommendation is sent back from the server to the user's information processing device and displayed on the screen, allowing the user to easily obtain a high-quality recommendation that reflects their emotional state.

[0713] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0714] Step 1:

[0715] Obtaining user input information

[0716] Process description: A user inputs the required information (the name, field, characteristics, and memorable episode of the person being recommended) into an input form to create a recommendation.

[0717] What happens: A user opens a browser and visits a web page on the system. The user enters the following information into the form fields:

[0718] Name: Ichiro Tanaka

[0719] Area: Marketing

[0720] Traits: Creativity, leadership, analytical skills

[0721] Memorable episodes: Successful new product launch project, winning a marketing strategy competition

[0722] Input: The user enters information into an input form on the device.

[0723] Output: The terminal holds the information entered.

[0724] Step 2:

[0725] Emotion recognition using an emotion recognition engine

[0726] Process Description: The emotion recognition engine analyzes the user's emotional state (happiness, anger, sadness, etc.).

[0727] How it works: The emotion recognition engine analyzes the user's emotions through the camera using facial recognition technology, the speed of their typing, and the strength of their keystrokes. The emotion engine detects the emotional state of "joy."

[0728] Input: User input data and facial image data while the user is entering information into an input form.

[0729] Output: Parsed emotion information (e.g., joy).

[0730] Step 3:

[0731] Sending input data and emotion data to the server

[0732] Process description: Send user input information and emotion information to the server in JSON format.

[0733] Specific operation: When the user presses the "Generate" button, the device converts the input information and emotion information into JSON. The device then sends the converted JSON data to the server via HTTPS.

[0734] Input: User input and emotional information.

[0735] Output: The data converted to JSON format is sent to the server.

[0736] Step 4:

[0737] Receiving and validating data by the server

[0738] Process description: The server receives the JSON data and checks whether the entered data is correct.

[0739] What happens: The server parses the received JSON data and verifies that all required fields (name, field, characteristics, episode) are present. Data validation is performed to ensure the format is correct.

[0740] Input: Input and emotion information received in JSON format.

[0741] Output: Data that passes validation.

[0742] Step 5:

[0743] Recommendation generation using generative artificial intelligence models

[0744] Process description: The server passes the validated data and sentiment information to the generative AI model to generate a recommendation.

[0745] Specific operation: The server passes the validated data to the generative AI model. The generative AI model takes into account the user's emotional information (delight) and adjusts the tone and content of the recommendation. The generative AI model generates the recommendation using a natural language generation (NLG) algorithm. For example, a recommendation might read, "Ichiro Tanaka has outstanding abilities in the marketing field, with particular strengths in creativity, leadership, and analytical ability. He has a long list of achievements, including the success of a new product launch project and winning a marketing strategy competition. His proactive attitude and positive energy have a positive influence on the entire team."

[0746] Input: Validated input and sentiment information.

[0747] Output: The generated recommendation.

[0748] Step 6:

[0749] Sending and displaying the recommendation to the user's device

[0750] Process description: The generated recommendation is sent back to the user's device in JSON format from the server and displayed on the screen.

[0751] Specific operation: The server converts the generated recommendation into JSON and sends it to the user's device. The user's device parses the received JSON data and displays the recommendation on the screen.

[0752] Input: JSON data containing the generated testimonials.

[0753] Output: The testimonial displayed on the device.

[0754] (Application example 2)

[0755] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0756] Conventional recommendation generation systems generate recommendation sentences uniformly without considering the user's emotional information, which can result in recommendation sentences that do not match the user's emotions or nuances. This means that users have to spend time and effort writing recommendation sentences, and the generated recommendation sentences do not always fully reflect the user's intentions.

[0757] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0758] In this invention, the server includes means for receiving information necessary for creating a recommendation and user emotional information from the user's terminal and validating the information, means for sending the validated information and emotional information to a generative AI model to generate a recommendation, and means for returning the generated recommendation to the terminal, thereby enabling the generation of high-quality recommendations that reflect the user's emotional information.

[0759] "User terminal" means a device used by a user to access the system and input information, including, for example, a smartphone, tablet, or personal computer.

[0760] "Information" refers to the name, field, characteristics, memorable episodes, etc. of the person being recommended, which are necessary to write the recommendation letter.

[0761] "Emotion information" is data on the user's emotional state (for example, joy, anger, sadness, etc.) extracted based on information input by the user.

[0762] "Validation" is the process of ensuring that the information and sentiment information entered by the user is accurate and complete.

[0763] "Generative AI models" are algorithms or models that generate recommendations based on received information and emotional information. Examples include natural language generation (NLG) algorithms.

[0764] A "recommendation letter" is a piece of writing generated based on information about the person being recommended, and includes content that appealingly conveys the characteristics and anecdotes of the person being recommended.

[0765] A "natural language generation algorithm" is an algorithm for generating understandable and appropriate sentences based on information and emotional information entered by the user.

[0766] "JSON format" is an abbreviation for JavaScript Object Notation, and is a format for expressing structured data in text format.

[0767] "Terminal" refers to any device that accepts user operations and processes information, including smartphones, personal computers, and tablets.

[0768] A "server" is a computer system that receives information sent from a user's terminal, processes it, and returns a final recommendation.

[0769] The present invention relates to a system for generating recommendation sentences taking into account a user's emotional information, which includes a user terminal, a server, a generative artificial intelligence model, and an emotional engine.

[0770] The main components of the system are as follows:

[0771] User terminal

[0772] The user terminal is a device for inputting information and recognizing emotions. Examples include smartphones, tablets, and personal computers. The user uses the terminal to access the system and input information to create a recommendation. The input information includes the name, field, characteristics, and memorable anecdotes of the person being recommended. The system also uses an emotion engine to recognize and analyze the user's emotional state (e.g., joy, anger, sadness, etc.).

[0773] server

[0774] The server is a central device that receives information and emotion information sent from user terminals and generates recommendation statements. Specifically, it has the following functions:

[0775] 1. Information reception and validation: The server receives the information and emotion information entered from the user device in JSON format. It then checks whether all required fields are present and whether the characteristics and episodes are in list format, and performs validation.

[0776] 2. Data processing: Receive the validated information and sentiment information and pass it to the generative AI model.

[0777] Generative AI Model

[0778] The generative AI model generates recommendations based on the received data and emotional information. Specifically, it performs the following processes:

[0779] 1. Basic information integration: Receive the nominee's name, field, characteristics, and anecdotes as input data.

[0780] 2. Reflection of emotional information: The tone and content of the recommendation are adjusted based on the emotional information obtained from the emotion engine. If the user's emotion of joy is detected, the recommendation is generated in a positive tone.

[0781] 3. Recommendation generation: A natural language generation algorithm (e.g., GPT-3) is used to complete the final recommendation.

[0782] Returning recommendation letters

[0783] The generated recommendation is sent back to the server and sent in JSON format to the user's device, where it is displayed on the screen, allowing the user to obtain a high-quality, emotionally relevant recommendation.

[0784] Specific examples

[0785] For example, if a user wants to write a movie review, they enter the following information:

[0786] Username: Taro Nakamura

[0787] Content Type: Movies

[0788] Content Title: Inception

[0789] Details of my viewing experience: I was very moved by the final scene and was moved to tears.

[0790] The emotion engine recognizes the user's emotions and generates a review that looks like this:

[0791] "Taro Nakamura watched the movie 'Inception' and was deeply moved, especially by the final scene. He was moved to tears, and the beauty of the scene and the moving story resonated deeply in his heart. Director Christopher Nolan's outstanding directing skills shine through, and Nakamura has a very high opinion of the film."

[0792] Prompt Sentence Examples

[0793] "Recognize emotions based on the content viewing experience and generate reviews that correspond to the user's emotions."

[0794] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0795] Step 1:

[0796] The user accesses the recommendation writing form using a terminal and inputs information about the person being recommended (name, field, characteristics, memorable episodes), along with details of the user's viewing experience.

[0797] Input: Name, field, characteristics, memorable episodes, details of viewing experience

[0798] Output: Input information and data passed to the emotion recognition engine

[0799] Step 2:

[0800] The device's emotion recognition engine analyzes the details of the user's viewing experience and recognizes the user's emotional state. For example, if the user inputs "I was moved," it will recognize the emotion of being moved.

[0801] Input: Viewing experience details

[0802] Output: User's emotional information (e.g., emotion, joy, anger, etc.)

[0803] Step 3:

[0804] The device converts the input information and recognized emotion information into JSON format and sends it to the server.

[0805] Input: Name, field, characteristics, memorable episodes, user's emotional information

[0806] Output: JSON format data

[0807] Step 4:

[0808] The server parses the data received in JSON format, checks whether all required fields are present, whether the characteristics and episodes are in list format, and performs validation.

[0809] Input: JSON format data

[0810] Output: Successfully validated data or an error message

[0811] Step 5:

[0812] The server passes the validated data and emotion information to the generative artificial intelligence model.

[0813] Input: Successfully validated data, emotion information

[0814] Output: The data used to generate the recommendation

[0815] Step 6:

[0816] The generative AI model generates a recommendation based on the received data. Specifically, it integrates basic information (name, field, characteristics, memorable episodes) with emotional information and applies a natural language generation algorithm (NLG) to generate the recommendation.

[0817] Input: Basic information, emotional information

[0818] Output: Generated recommendation

[0819] Step 7:

[0820] The generated recommendation is sent back from the server to the user's device in JSON format.

[0821] Input: Generated testimonial

[0822] Output: Recommendation data in JSON format

[0823] Step 8:

[0824] The user's device analyzes the received recommendation and displays it on the screen, allowing the user to obtain a recommendation that is high quality and emotionally appropriate.

[0825] Input: Recommendation data in JSON format

[0826] Output: Testimonial displayed on screen

[0827] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0828] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0829] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0830] [Third embodiment]

[0831] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0832] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0833] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0834] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0835] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0836] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0837] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0838] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0839] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0840] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0841] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0842] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0843] This invention is a system for efficiently creating recommendation and referral text using a generative AI. Below, we will explain the specific system program, its processing content, and specific examples.

[0844] System configuration

[0845] The system consists of three main components:

[0846] 1. User Device

[0847] 2. Server

[0848] 3. Generative AI Model

[0849] User terminal operation

[0850] When a user wants to generate a recommendation using their device, they first access a web or native application. They enter information such as the name, field, characteristics, and memorable episodes of the person they are recommending into the provided input form. After entering all the information, the user presses the "Generate" button. This action converts the entered information into JSON format and sends it to the server as an HTTP POST request.

[0851] Server Operation

[0852] When the server receives a request from a user device, it first validates the data. Specifically, it checks whether all required fields are present and whether the characteristics and episodes are in list format. If the data passes validation, it is passed to the generative AI model, and the recommendation generation process begins.

[0853] The generative AI model takes the received data as input and generates a recommendation that takes into account the characteristics and anecdotes of the person being recommended. This generation process utilizes a natural language generation (NLG) algorithm. The generated recommendation is returned to the server and sent back to the user's device in JSON format.

[0854] How generative artificial intelligence models work

[0855] The generative AI model generates a recommendation based on the input data. This process proceeds as follows: First, the name and field of the person being recommended are inserted at the beginning, and then the recommendation is constructed by listing the characteristics of the person being recommended. Furthermore, memorable anecdotes entered by the user are integrated to enrich the specific content of the recommendation.

[0856] Specific examples

[0857] Let's take a concrete example. For example, suppose a user enters the information of a recommended person named "Yamada Taro" as follows:

[0858] Name: Yamada Taro

[0859] Field: Software Development

[0860] Traits: Hard work, teamwork, leadership

[0861] Episodes: Leadership in Project X, accurate communication in customer relations

[0862] When the user presses the "Generate" button, this information is sent to the server for validation. If validation is successful, the data is passed to a generative AI model, which generates a recommendation like this:

[0863] "Taro Yamada has excellent abilities in the software development field, with particular strengths in diligence, teamwork, and leadership. Specifically, he has demonstrated these qualities through his experience in Project X, such as his leadership and accurate communication in customer relations."

[0864] The generated recommendation is sent back from the server to the user's terminal and displayed on the screen, allowing the user to easily obtain a high-quality recommendation.

[0865] In this way, the present invention is a system that significantly reduces the time and effort required to create a recommendation, and provides users with efficient and high-quality document creation.

[0866] The processing flow will be explained below.

[0867] Step 1:

[0868] A user accesses an input form for writing a recommendation using a terminal, and inputs information about the name, field, characteristics, and memorable episode of the person to be recommended into the form.

[0869] Step 2:

[0870] The user presses the "Generate" button, which converts the entered information into JSON format and sends it to the server as an HTTP POST request.

[0871] Step 3:

[0872] The server parses the JSON data received from the user device and performs validation, checking whether all required fields are present and whether the characteristics and episodes are in list format.

[0873] Step 4:

[0874] The server passes the validated data to the generative AI model, which uses the validated data as input and generates a recommendation using an NLG (natural language generation) algorithm.

[0875] Step 5:

[0876] The generative AI model generates a recommendation based on the name, field, characteristics, and anecdotes of the person being recommended. In the generation process, the recommendation highlights the person's characteristics and integrates anecdotes to enrich the specific content.

[0877] Step 6:

[0878] The server receives the generated recommendation and returns it to the user's device in JSON format.

[0879] Step 7:

[0880] The user terminal displays the recommendation received from the server on the screen, allowing the user to easily obtain high-quality recommendations.

[0881] Example 1

[0882] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0883] Currently, writing testimonials and referrals is a time-consuming and laborious task, and creating high-quality testimonials requires specialized knowledge and experience. There is a need for a system that can solve this problem and automatically and efficiently generate high-quality testimonials.

[0884] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0885] In this invention, the server includes means for inputting information required to create a recommendation from a user's terminal, means for converting the information input from the terminal into JSON format and sending it to the server, means for the server to receive and validate the information, means for sending the validated information to a generative AI model to generate a recommendation, and means for returning the generated recommendation to the terminal. This enables users to easily and quickly create high-quality recommendations without requiring specialized knowledge.

[0886] A "user's terminal" is a digital device used by a user, such as a computer, smartphone, or tablet.

[0887] A "recommendation" is a piece of writing that is written to evaluate or recommend a specific individual or subject.

[0888] The "means for inputting information" refers to an interface (e.g., an input form) through which a user can input elements (such as name, field, characteristics, episodes, etc.) necessary for creating a recommendation.

[0889] The "means for converting to JSON format" is a mechanism for encoding user-entered information into JavaScript Object Notation (JSON) format.

[0890] A "server" is a computer system that receives and processes data sent from a user's terminal.

[0891] "Validation" is the process of checking whether entered data meets certain criteria.

[0892] A "generative artificial intelligence model" is a model that implements an algorithm that generates sentences based on input data using machine learning or deep learning techniques.

[0893] The "means for generating" is a process for generating a recommendation from input data using an artificial intelligence model.

[0894] The "means for sending back" is a mechanism for sending the generated recommendation to the user's terminal and displaying it.

[0895] The "Generate Button" is an interface element that initiates the submission of the information entered by the user and the generation of the recommendation.

[0896] The present invention is a system for efficiently creating recommendations and referrals using a generative AI model. The system includes the following components:

[0897] User terminal operation

[0898] When a user wants to generate a recommendation, they first access a web or native application. They enter information such as the name, field, characteristics, and memorable episodes of the person they are recommending into the provided input form. After entering all the information, the user presses the "Generate" button. This action converts the entered information into JSON format and sends it to the server as an HTTP POST request.

[0899] Server Operation

[0900] When the server receives a request from a user device, it first validates the data. Specifically, it checks whether all required fields are present and whether the characteristics and episodes are in list format. Data that passes validation is passed to a generative AI model, and the recommendation generation process begins. The generative AI model takes the received data as input and generates a recommendation that takes into account the characteristics and episodes of the person being recommended. The generated recommendation is returned to the server and sent back to the user device in JSON format.

[0901] How generative artificial intelligence models work

[0902] The generative AI model generates a recommendation based on the input data. This process works as follows: First, the name and field of the person being recommended are inserted at the beginning, and then the recommendation is constructed by listing the characteristics of the person being recommended. Furthermore, impressive anecdotes entered by the user are integrated to enrich the specific content of the recommendation.

[0903] Specific examples

[0904] Let's take a concrete example. For example, a user enters the information of a recommended person named "Taro Tanaka" as follows:

[0905] Name: Taro Tanaka

[0906] Field: Software Development

[0907] Traits: Hard work, teamwork, leadership

[0908] Episodes: Leadership in Project X, accurate communication in customer relations

[0909] When the user hits the "Generate" button, this information is sent to the server for validation. If validation is successful, the data is passed to a generative AI model, which generates a recommendation like this:

[0910] "Taro Tanaka has excellent abilities in the software development field, and his strengths are his diligence, teamwork, and leadership. Specifically, he has demonstrated these qualities through his experience in Project X, such as his leadership and accurate communication in customer relations."

[0911] The generated recommendation is sent back from the server to the user's terminal and displayed on the screen, allowing the user to easily obtain a high-quality recommendation.

[0912] Prompt Sentence Examples

[0913] An example of an input prompt for the generative AI model in this system might look like this:

[0914] Generate testimonials.

[0915] Name: Taro Tanaka

[0916] Field: Software Development

[0917] Traits: Hard work, teamwork, leadership

[0918] Episodes: Leadership in Project X, accurate communication in customer relations

[0919] In this way, the present invention is a system that significantly reduces the time and effort required to create a recommendation, and provides users with efficient and high-quality document creation.

[0920] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0921] Step 1: Collecting User Input

[0922] To generate a testimonial, a user accesses a web or native application and enters information into a form.

[0923] Input: The user enters the name, field, characteristics, and memorable episode of the person being recommended.

[0924] What happens: A user opens a web browser, such as Google Chrome, visits the system's web application URL, and then enters the following information into the form that appears on the screen:

[0925] Name: Taro Tanaka

[0926] Field: Software Development

[0927] Traits: Hard work, teamwork, leadership

[0928] Episodes: Leadership in Project X, accurate communication in customer relations

[0929] Output: The entered information is saved as form variables.

[0930] Step 2: Submitting input data

[0931] When the user presses the "Generate" button, the entered information is converted to JSON format and sent to the server.

[0932] Input: Information entered in step 1.

[0933] Specific behavior: When the user clicks the "Generate" button, the data is converted to JSON format using JavaScript AJAX (or Fetch API) and sent to the specified API endpoint.

[0934] Output: JSON formatted data is sent to the server as an HTTP POST request.

[0935] Step 3: Validate the data

[0936] The server receives the request from the user device and validates the data.

[0937] Input: The JSON data submitted in step 2.

[0938] What happens: The server (using the Django framework) validates the received JSON data. Specifically, it uses the Python jsonschema package to check the following:

[0939] Does it include names, areas, characteristics, and anecdotes?

[0940] Are the characteristics in list format?

[0941] Output: The validated data or a validation error response.

[0942] Step 4: Request a recommendation

[0943] The server passes the validated data to the generative AI model and makes a request to generate a recommendation.

[0944] Input: Data that passes validation.

[0945] What happens: The server sends the request data to a generative AI model (e.g., OpenAI's GPT-3 API). It uses the Python requests library to make the following request:

[0946] python

[0947] response = requests.post("https: / / api.openai.com / v1 / engines / davinci-codex / completions",

[0948] headers={"Authorization": "Bearer YOUR_API_KEY"},

[0949] json={"prompt": "Generate a testimonial.\nName: Taro Tanaka\nField: Software development\nCharacteristics: Diligence, teamwork, leadership\nAnecdotes: Leadership in Project X, accurate communication in customer relations", "max_tokens": 150})

[0950] Output: A recommendation generation request to the generative AI model.

[0951] Step 5: Generate testimonials

[0952] The generative AI model generates a recommendation based on the data provided.

[0953] Input: The prompt data submitted in step 4.

[0954] How it works: A generative AI model (OpenAI GPT-3) generates a recommendation based on the data provided as a prompt. The model generates text that includes the name, characteristics, and anecdotes of the person being recommended.

[0955] Output: The generated testimonial.

[0956] Step 6: Return and view testimonials

[0957] The server receives the generated recommendation and returns it to the user device in JSON format.

[0958] Input: The testimonial generated in step 5.

[0959] How it works: The server receives the response from the generated AI model and sends it back to the user's device as an HTTP response. The response data is processed by JavaScript on the user's device, and the generated recommendation is displayed at the bottom of the form.

[0960] Output: A testimonial that will be displayed on the user's device.

[0961] (Application example 1)

[0962] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0963] Creating recommendations and product reviews in the past was time-consuming and laborious, and it was difficult to maintain a consistent level of quality. Furthermore, if the information entered by the user was insufficient or inaccurate, the reliability and quality of the generated text declined. Furthermore, systems with no or insufficient validation functions often resulted in incomplete information being used as is. Another issue was the difficulty of accurately reflecting the user's intent when generating recommendations and reviews using natural language generation algorithms.

[0964] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0965] In this invention, the server includes means for inputting information required for creating a recommendation or review from a user's terminal, means for receiving and validating the information input from the terminal, means for sending the validated information to a generative AI model to generate a recommendation or review, and means for returning the generated recommendation or review to the terminal. This ensures the accuracy and completeness of the information input by the user and enables the automatic generation of high-quality recommendations and reviews.

[0966] A "terminal" is an electronic device through which a user enters information and receives generated text.

[0967] "Validation" is the process of ensuring that entered information is accurate and complete.

[0968] A "generative artificial intelligence model" is an artificial intelligence algorithm that automatically generates recommendations and reviews based on input information.

[0969] A "testimonial" is a written description that is written to evaluate and recommend another person.

[0970] A "review" is a document that describes impressions and evaluations of a product.

[0971] A "system" is a structure with a set of functions consisting of terminals, servers, and generative artificial intelligence models.

[0972] A "natural language generation algorithm" is a technology that allows artificial intelligence to generate sentences using human natural language.

[0973] "Information" is data entered by a user to generate a recommendation or review.

[0974] "JSON format" is an abbreviation for JavaScript Object Notation, and is a method of structuring data and representing it in text format.

[0975] An "episode" is a specific event or experience related to the characteristics or usability of the recommended person or product.

[0976] In this invention, the user terminal provides an interface for inputting information for generating a recommendation or product review. The user inputs the product name, reason for purchase, impressions of use, memorable episodes, etc. into an input form on the terminal. For example, the following information is input:

[0977] Product Name: High-Quality Laptop

[0978] Reason for purchase: I needed a powerful laptop for work.

[0979] User experience: Fast operation and long battery life

[0980] Episode: The battery lasted for 8 hours during a business trip, making it useful for presentations.

[0981] This input information is converted to JSON format and sent to the server as an HTTP POST request. The server validates the received data to ensure all required fields are present and the episodes are in a list format. If validation is successful, the server passes the data to a generative AI model to begin the process of generating reviews.

[0982] The server uses OpenAI's API to run a natural language generation (NLG) algorithm to generate a prompt based on the information the user has entered. For example, the following prompt might be generated:

[0983] Product Name: High-Quality Laptop

[0984] Reason for purchase: I needed a powerful laptop for work.

[0985] User experience: Fast operation and long battery life

[0986] Episode: The battery lasted for 8 hours during a business trip, making it useful for presentations.

[0987] Please use the information above to write your product review."

[0988] When this prompt is sent to the OpenAI API, the generated review is sent back to the server and then sent to the user's device in JSON format, allowing users to obtain high-quality reviews in a short amount of time.

[0989] This system uses a terminal, a server, and a generative AI model (e.g., OpenAI's API). The terminal uses a typical smartphone or PC browser, and the server uses a web application using Python and the Flask framework. Information entered by the user is validated on the server and then passed to the generative AI model. The generated text is then sent back to the user's terminal via the server.

[0990] In particular, the automatic generation of product reviews has the advantage of saving users the trouble of writing reviews while providing high-quality reviews. This system makes it possible to consistently improve the quality of reviews and create them quickly.

[0991] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0992] Step 1:

[0993] The user enters information into the device's input form, such as the product name, reason for purchase, impressions on use, memorable episodes, etc. The entered information is converted into JSON format.

[0994] Input: Product name, reason for purchase, impressions of use, memorable episode

[0995] Output: JSON format data

[0996] Step 2:

[0997] The terminal sends the data converted into JSON format to the server as an HTTP POST request.

[0998] Input: JSON format data

[0999] Output: HTTP POST request

[1000] Step 3:

[1001] The server receives the HTTP POST request and performs data validation, ensuring all required fields are present and that the episodes are in a list format.

[1002] Input: JSON data via HTTP POST request

[1003] Output: Validation result (success / failure)

[1004] Step 4:

[1005] If validation is successful, the server sends the data to a generative artificial intelligence model to begin the process of generating the review.

[1006] Input: Validated JSON data

[1007] Output: Input data to a generative AI model

[1008] Step 5:

[1009] The server generates a prompt for the generative AI model (using OpenAI's API) and runs the NLG algorithm to create a prompt based on the information entered by the user.

[1010] Input: Data that has been successfully validated

[1011] Output: Prompt sentence generated by generative AI model

[1012] Step 6:

[1013] A generative AI model receives the prompt and generates a recommendation or review. During this generation process, a review with a specific structure is created based on the user's input.

[1014] Input: prompt statement

[1015] Output: Generated testimonial or review

[1016] Step 7:

[1017] The generated recommendation and review are sent back to the server, which then sends them back to the user's device in JSON format.

[1018] Input: Generated testimonial or review

[1019] Output: Generated result data in JSON format

[1020] Step 8:

[1021] The user device receives the generated recommendation and review and displays it to the user, allowing the user to check high-quality reviews.

[1022] Input: Generated result data in JSON format

[1023] Output: Displayed testimonial or review on the user's device

[1024] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1025] The present invention provides a system for recognizing a user's emotions and generating a recommendation sentence by taking the emotion information into consideration. The system includes a user terminal, a server, a generative artificial intelligence model, and an emotion engine.

[1026] System configuration and operation

[1027] User terminal operation

[1028] A user accesses the system using a terminal and enters information into an input form to create a recommendation. The information entered includes the name, field, characteristics, and memorable episodes of the person being recommended. In addition, the user's terminal utilizes an emotion engine to recognize and analyze the user's emotional state (e.g., joy, anger, sadness, etc.) at the time of input.

[1029] Server Operation

[1030] The information obtained from the input form and emotion engine is sent to the server in JSON format. The server has the following functions:

[1031] 1. Data Receipt and Validation:

[1032] The server receives the JSON data sent from the user device and checks whether all required fields are present and whether the characteristics and episodes are in list format.

[1033] 2. Post-validation data processing:

[1034] The data that passes validation and the emotional information obtained from the emotion engine are passed to the generative artificial intelligence model.

[1035] How generative artificial intelligence models work

[1036] The generative AI model generates recommendations based on the received data and sentiment information. This process includes the following steps:

[1037] 1. Integration of basic information:

[1038] The nominee's name, field, characteristics, and anecdote are received as input data.

[1039] 2. Reflecting emotional information:

[1040] The tone and content of the recommendation are adjusted based on the emotional information obtained from the emotion engine. For example, if the user is happy, the recommendation will be generated in a style that reflects that positive emotion.

[1041] 3. Generating Recommendations:

[1042] Finally, a recommendation is finalized using a natural language generation (NLG) algorithm.

[1043] Returning recommendation letters

[1044] The generated recommendation is sent back to the server and sent in JSON format to the user's device, where it is displayed on the screen, allowing the user to obtain a high-quality, emotionally relevant recommendation.

[1045] Specific examples

[1046] Specific examples are shown below.

[1047] 1. If a user wants to write a recommendation, they access the system and enter the information of the person being recommended, "Ichiro Tanaka":

[1048] Name: Ichiro Tanaka

[1049] Area: Marketing

[1050] Traits: Creativity, leadership, analytical skills

[1051] Memorable episodes: Successful new product launch project, winning a marketing strategy competition

[1052] 2. While the user is entering information, the emotion engine detects the user's emotion of joy.

[1053] 3. When the user presses the "Generate" button, the input information and emotion information are sent to the server.

[1054] 4. The server receives the information and validates it.

[1055] 5. The information that passes validation is passed to the generative AI model, which takes into account the emotional information and generates a recommendation like the following:

[1056] "Ichiro Tanaka has outstanding abilities in the field of marketing, with particular strengths in creativity, leadership, and analytical ability. He has a proven track record, including successfully implementing new product launch projects and winning marketing strategy competitions. His positive attitude and energy have a positive impact on the entire team."

[1057] 6. The generated recommendation is sent back from the server to the user's device and displayed on the screen.

[1058] This allows users to easily obtain high-quality recommendation messages that reflect their emotional state, thereby improving the efficiency and accuracy of recommendation message creation.

[1059] The processing flow will be explained below.

[1060] Step 1:

[1061] The user accesses the system's input form using a terminal and inputs the name, field, characteristics, and memorable episode of the person to be recommended.

[1062] Step 2:

[1063] While the user is typing, the device's onboard emotion engine recognizes the user's emotional state, using the camera and microphone to analyze the user's facial expressions and vocal tone.

[1064] Step 3:

[1065] Once the user has entered all the required information, they press the "Generate" button. This action converts the entered information and emotion data into JSON format and sends it to the server as an HTTP POST request.

[1066] Step 4:

[1067] The server parses the JSON data received from the user device and performs data validation, checking whether all required fields are present and whether the characteristics and episodes are in list format.

[1068] Step 5:

[1069] The server passes the validated data and emotion information to the generative AI model. Specifically, the data containing emotion information is input into the model.

[1070] Step 6:

[1071] The generative AI model generates a recommendation based on the received data and emotional information, taking into account the characteristics and anecdotes of the person being recommended and adjusting the writing style and tone based on the emotional information.

[1072] Step 7:

[1073] The generated recommendation is sent back to the server, which then sends it back to the user's device in JSON format.

[1074] Step 8:

[1075] The user's device analyzes the recommendation received from the server and displays it on the screen, allowing the user to obtain high-quality recommendations that reflect their emotional state.

[1076] Example 2

[1077] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1078] Conventional recommendation generation systems update recommendation sentences in a fixed style without considering the user's emotional state, making it difficult to generate persuasive recommendation sentences that fit the user's emotions. Another problem is that processing data without proper validation reduces the quality of the generated recommendation sentences. The purpose of this invention is to solve these problems and generate high-quality recommendation sentences that reflect the user's emotional state.

[1079] The specification process by the specification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for inputting information necessary for creating a recommendation from the user's information processing device, means for analyzing the user's emotional state using an emotion recognition engine, means for receiving and validating the information and emotional information input from the processing device, means for sending the validated information and emotional information to a generative AI model and generating a recommendation, and means for returning the generated recommendation to the processing device. This makes it possible to generate high-quality recommendations that reflect the user's emotional state.

[1080] The "user's information processing device" refers to a device used by the user, such as a computer, smartphone, or tablet, and is a device that has an interface for inputting information for creating a recommendation.

[1081] An "emotion recognition engine" is software or hardware that analyzes a user's emotional state (e.g., joy, anger, sadness, etc.) using the user's typing speed, keystroke strength, facial recognition technology, etc.

[1082] Validation is the process of ensuring that the information entered by the user is accurate, by checking that all required fields are present and that the format is correct.

[1083] A "generative artificial intelligence model" is a machine learning model that generates recommendations using natural language generation algorithms based on input data and emotional information.

[1084] A "recommendation letter" is a document that reflects the characteristics and anecdotes about the recommended person specified by the user, as well as the emotional state of the user at the time, and is used for the purpose of evaluation or introduction.

[1085] A "natural language generation algorithm" is a technology that allows computers to generate human language, and is a method for automatically creating sentences based on input data.

[1086] "JSON format" is an abbreviation for JavaScript Object Notation, a lightweight data exchange format and a text format that is easy to read for both humans and machines.

[1087] The present invention provides a system for recognizing a user's emotions and generating a recommendation sentence by taking the emotion information into consideration. The system includes a user's information processing device, a server, a generative artificial intelligence model, and an emotion recognition engine.

[1088] System configuration and operation

[1089] Operation of user information processing device

[1090] A user inputs necessary information (the name, field, characteristics, and memorable episode of the person being recommended) into an input form for creating a recommendation.

[1091] For example, input may be in the following format:

[1092] Name: Name of the nominee

[1093] Area: Marketing

[1094] Traits: Creativity, leadership, analytical skills

[1095] Memorable episodes: Successful new product launch project, winning a marketing strategy competition

[1096] The user information processing device also uses an emotion recognition engine to analyze the user's emotional state while typing. The emotion recognition engine detects emotions (e.g., joy, anger, sadness) through the user's typing speed, keystroke strength, and facial recognition technology.

[1097] Server Operation

[1098] When the user presses the "Generate" button, the input information and emotion information are sent to the server in JSON format. The server provides the following functions:

[1099] 1. Data reception and validation: The server receives the JSON data sent from the user information processing device and checks whether all required fields are present and whether the characteristics and episodes are in list format.

[1100] 2. Post-validation data processing: The data and sentiment information that have passed validation are passed to the generative AI model.

[1101] How generative artificial intelligence models work

[1102] The generative AI model generates recommendations based on the received data and emotional information. The generative AI model works as follows:

[1103] 1. Integration of basic information: The name, field, characteristics, and anecdotes of the nominee are received as input data.

[1104] 2. Reflection of emotional information: The tone and content of the recommendation are adjusted based on the emotional information obtained from the emotion recognition engine. For example, if the user is happy, the recommendation will be generated in a style that reflects that positive emotion.

[1105] 3. Recommendation Generation: A natural language generation (NLG) algorithm is used to complete the recommendation.

[1106] Returning recommendation letters

[1107] The generated recommendation is returned to the server and sent in JSON format to the user's information processing device, which then displays the received recommendation on its screen, allowing the user to obtain a high-quality recommendation that is tailored to their emotions.

[1108] Specific examples

[1109] When a user wants to write a testimonial, they access the system and enter the following information:

[1110] Name: Ichiro Tanaka

[1111] Area: Marketing

[1112] Traits: Creativity, leadership, analytical skills

[1113] Memorable episodes: Successful new product launch project, winning a marketing strategy competition

[1114] While the user is entering information, the emotion recognition engine detects the user's emotion of "happiness." The input information and emotion information are then sent to the server, where, after validation, they are passed to the generative AI model. The generative AI model generates the following recommendation based on the emotion information:

[1115] "Ichiro Tanaka has outstanding abilities in the field of marketing, with particular strengths in creativity, leadership, and analytical ability. He has a proven track record, including successfully implementing new product launch projects and winning marketing strategy competitions. His positive attitude and energy have a positive impact on the entire team."

[1116] The generated recommendation is sent back from the server to the user's information processing device and displayed on the screen, allowing the user to easily obtain a high-quality recommendation that reflects their emotional state.

[1117] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1118] Step 1:

[1119] Obtaining user input information

[1120] Process description: A user inputs the required information (the name, field, characteristics, and memorable episode of the person being recommended) into an input form to create a recommendation.

[1121] What happens: A user opens a browser and visits a web page on the system. The user enters the following information into the form fields:

[1122] Name: Ichiro Tanaka

[1123] Area: Marketing

[1124] Traits: Creativity, leadership, analytical skills

[1125] Memorable episodes: Successful new product launch project, winning a marketing strategy competition

[1126] Input: The user enters information into an input form on the device.

[1127] Output: The terminal holds the information entered.

[1128] Step 2:

[1129] Emotion recognition using an emotion recognition engine

[1130] Process Description: The emotion recognition engine analyzes the user's emotional state (happiness, anger, sadness, etc.).

[1131] How it works: The emotion recognition engine analyzes the user's emotions through the camera using facial recognition technology, the speed of their typing, and the strength of their keystrokes. The emotion engine detects the emotional state of "joy."

[1132] Input: User input data and facial image data while the user is entering information into an input form.

[1133] Output: Parsed emotion information (e.g., joy).

[1134] Step 3:

[1135] Sending input data and emotion data to the server

[1136] Process description: Send user input information and emotion information to the server in JSON format.

[1137] Specific operation: When the user presses the "Generate" button, the device converts the input information and emotion information into JSON. The device then sends the converted JSON data to the server via HTTPS.

[1138] Input: User input and emotional information.

[1139] Output: The data converted to JSON format is sent to the server.

[1140] Step 4:

[1141] Receiving and validating data by the server

[1142] Process description: The server receives the JSON data and checks whether the entered data is correct.

[1143] What happens: The server parses the received JSON data and verifies that all required fields (name, field, characteristics, episode) are present. Data validation is performed to ensure the format is correct.

[1144] Input: Input and emotion information received in JSON format.

[1145] Output: Data that passes validation.

[1146] Step 5:

[1147] Recommendation generation using generative artificial intelligence models

[1148] Process description: The server passes the validated data and sentiment information to the generative AI model to generate a recommendation.

[1149] Specific operation: The server passes the validated data to the generative AI model. The generative AI model takes into account the user's emotional information (delight) and adjusts the tone and content of the recommendation. The generative AI model generates the recommendation using a natural language generation (NLG) algorithm. For example, a recommendation might read, "Ichiro Tanaka has outstanding abilities in the marketing field, with particular strengths in creativity, leadership, and analytical ability. He has a long list of achievements, including the success of a new product launch project and winning a marketing strategy competition. His proactive attitude and positive energy have a positive influence on the entire team."

[1150] Input: Validated input and sentiment information.

[1151] Output: The generated recommendation.

[1152] Step 6:

[1153] Sending and displaying the recommendation to the user's device

[1154] Process description: The generated recommendation is sent back to the user's device in JSON format from the server and displayed on the screen.

[1155] Specific operation: The server converts the generated recommendation into JSON and sends it to the user's device. The user's device parses the received JSON data and displays the recommendation on the screen.

[1156] Input: JSON data containing the generated testimonials.

[1157] Output: The testimonial displayed on the device.

[1158] (Application example 2)

[1159] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1160] Conventional recommendation generation systems generate recommendation sentences uniformly without considering the user's emotional information, which can result in recommendation sentences that do not match the user's emotions or nuances. This means that users have to spend time and effort writing recommendation sentences, and the generated recommendation sentences do not always fully reflect the user's intentions.

[1161] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1162] In this invention, the server includes means for receiving information necessary for creating a recommendation and user emotional information from the user's terminal and validating the information, means for sending the validated information and emotional information to a generative AI model to generate a recommendation, and means for returning the generated recommendation to the terminal, thereby enabling the generation of high-quality recommendations that reflect the user's emotional information.

[1163] "User terminal" means a device used by a user to access the system and input information, including, for example, a smartphone, tablet, or personal computer.

[1164] "Information" refers to the name, field, characteristics, memorable episodes, etc. of the person being recommended, which are necessary to write the recommendation letter.

[1165] "Emotion information" is data on the user's emotional state (for example, joy, anger, sadness, etc.) extracted based on information input by the user.

[1166] "Validation" is the process of ensuring that the information and sentiment information entered by the user is accurate and complete.

[1167] "Generative AI models" are algorithms or models that generate recommendations based on received information and emotional information. Examples include natural language generation (NLG) algorithms.

[1168] A "recommendation letter" is a piece of writing generated based on information about the person being recommended, and includes content that appealingly conveys the characteristics and anecdotes of the person being recommended.

[1169] A "natural language generation algorithm" is an algorithm for generating understandable and appropriate sentences based on information and emotional information entered by the user.

[1170] "JSON format" is an abbreviation for JavaScript Object Notation, and is a format for expressing structured data in text format.

[1171] "Terminal" refers to any device that accepts user operations and processes information, including smartphones, personal computers, and tablets.

[1172] A "server" is a computer system that receives information sent from a user's terminal, processes it, and returns a final recommendation.

[1173] The present invention relates to a system for generating recommendation sentences taking into account a user's emotional information, which includes a user terminal, a server, a generative artificial intelligence model, and an emotional engine.

[1174] The main components of the system are as follows:

[1175] User terminal

[1176] The user terminal is a device for inputting information and recognizing emotions. Examples include smartphones, tablets, and personal computers. The user uses the terminal to access the system and input information to create a recommendation. The input information includes the name, field, characteristics, and memorable anecdotes of the person being recommended. The system also uses an emotion engine to recognize and analyze the user's emotional state (e.g., joy, anger, sadness, etc.).

[1177] server

[1178] The server is a central device that receives information and emotion information sent from user terminals and generates recommendation statements. Specifically, it has the following functions:

[1179] 1. Information reception and validation: The server receives the information and emotion information entered from the user device in JSON format. It then checks whether all required fields are present and whether the characteristics and episodes are in list format, and performs validation.

[1180] 2. Data processing: Receive the validated information and sentiment information and pass it to the generative AI model.

[1181] Generative AI Model

[1182] The generative AI model generates recommendations based on the received data and emotional information. Specifically, it performs the following processes:

[1183] 1. Basic information integration: Receive the nominee's name, field, characteristics, and anecdotes as input data.

[1184] 2. Reflection of emotional information: The tone and content of the recommendation are adjusted based on the emotional information obtained from the emotion engine. If the user's emotion of joy is detected, the recommendation is generated in a positive tone.

[1185] 3. Recommendation generation: A natural language generation algorithm (e.g., GPT-3) is used to complete the final recommendation.

[1186] Returning recommendation letters

[1187] The generated recommendation is sent back to the server and sent in JSON format to the user's device, where it is displayed on the screen, allowing the user to obtain a high-quality, emotionally relevant recommendation.

[1188] Specific examples

[1189] For example, if a user wants to write a movie review, they enter the following information:

[1190] Username: Taro Nakamura

[1191] Content Type: Movies

[1192] Content Title: Inception

[1193] Details of my viewing experience: I was very moved by the final scene and was moved to tears.

[1194] The emotion engine recognizes the user's emotions and generates a review that looks like this:

[1195] "Taro Nakamura watched the movie 'Inception' and was deeply moved, especially by the final scene. He was moved to tears, and the beauty of the scene and the moving story resonated deeply in his heart. Director Christopher Nolan's outstanding directing skills shine through, and Nakamura has a very high opinion of the film."

[1196] Prompt Sentence Examples

[1197] "Recognize emotions based on the content viewing experience and generate reviews that correspond to the user's emotions."

[1198] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1199] Step 1:

[1200] The user accesses the recommendation writing form using a terminal and inputs information about the person being recommended (name, field, characteristics, memorable episodes), along with details of the user's viewing experience.

[1201] Input: Name, field, characteristics, memorable episodes, details of viewing experience

[1202] Output: Input information and data passed to the emotion recognition engine

[1203] Step 2:

[1204] The device's emotion recognition engine analyzes the details of the user's viewing experience and recognizes the user's emotional state. For example, if the user inputs "I was moved," it will recognize the emotion of being moved.

[1205] Input: Viewing experience details

[1206] Output: User's emotional information (e.g., emotion, joy, anger, etc.)

[1207] Step 3:

[1208] The device converts the input information and recognized emotion information into JSON format and sends it to the server.

[1209] Input: Name, field, characteristics, memorable episodes, user's emotional information

[1210] Output: JSON format data

[1211] Step 4:

[1212] The server parses the data received in JSON format, checks whether all required fields are present, whether the characteristics and episodes are in list format, and performs validation.

[1213] Input: JSON format data

[1214] Output: Successfully validated data or an error message

[1215] Step 5:

[1216] The server passes the validated data and emotion information to the generative artificial intelligence model.

[1217] Input: Successfully validated data, emotion information

[1218] Output: The data used to generate the recommendation

[1219] Step 6:

[1220] The generative AI model generates a recommendation based on the received data. Specifically, it integrates basic information (name, field, characteristics, memorable episodes) with emotional information and applies a natural language generation algorithm (NLG) to generate the recommendation.

[1221] Input: Basic information, emotional information

[1222] Output: Generated recommendation

[1223] Step 7:

[1224] The generated recommendation is sent back from the server to the user's device in JSON format.

[1225] Input: Generated testimonial

[1226] Output: Recommendation data in JSON format

[1227] Step 8:

[1228] The user's device analyzes the received recommendation and displays it on the screen, allowing the user to obtain a recommendation that is high quality and emotionally appropriate.

[1229] Input: Recommendation data in JSON format

[1230] Output: Testimonial displayed on screen

[1231] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1232] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1233] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1234] [Fourth embodiment]

[1235] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1236] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1237] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1238] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1239] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1240] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1241] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1242] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1243] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1244] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1245] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1246] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1247] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1248] This invention is a system for efficiently creating recommendation and referral text using a generative AI. Below, we will explain the specific system program, its processing content, and specific examples.

[1249] System configuration

[1250] The system consists of three main components:

[1251] 1. User Device

[1252] 2. Server

[1253] 3. Generative AI Model

[1254] User terminal operation

[1255] When a user wants to generate a recommendation using their device, they first access a web or native application. They enter information such as the name, field, characteristics, and memorable episodes of the person they are recommending into the provided input form. After entering all the information, the user presses the "Generate" button. This action converts the entered information into JSON format and sends it to the server as an HTTP POST request.

[1256] Server Operation

[1257] When the server receives a request from a user device, it first validates the data. Specifically, it checks whether all required fields are present and whether the characteristics and episodes are in list format. If the data passes validation, it is passed to the generative AI model, and the recommendation generation process begins.

[1258] The generative AI model takes the received data as input and generates a recommendation that takes into account the characteristics and anecdotes of the person being recommended. This generation process utilizes a natural language generation (NLG) algorithm. The generated recommendation is returned to the server and sent back to the user's device in JSON format.

[1259] How generative artificial intelligence models work

[1260] The generative AI model generates a recommendation based on the input data. This process proceeds as follows: First, the name and field of the person being recommended are inserted at the beginning, and then the recommendation is constructed by listing the characteristics of the person being recommended. Furthermore, memorable anecdotes entered by the user are integrated to enrich the specific content of the recommendation.

[1261] Specific examples

[1262] Let's take a concrete example. For example, suppose a user enters the information of a recommended person named "Yamada Taro" as follows:

[1263] Name: Yamada Taro

[1264] Field: Software Development

[1265] Traits: Hard work, teamwork, leadership

[1266] Episodes: Leadership in Project X, accurate communication in customer relations

[1267] When the user presses the "Generate" button, this information is sent to the server for validation. If validation is successful, the data is passed to a generative AI model, which generates a recommendation like this:

[1268] "Taro Yamada has excellent abilities in the software development field, with particular strengths in diligence, teamwork, and leadership. Specifically, he has demonstrated these qualities through his experience in Project X, such as his leadership and accurate communication in customer relations."

[1269] The generated recommendation is sent back from the server to the user's terminal and displayed on the screen, allowing the user to easily obtain a high-quality recommendation.

[1270] In this way, the present invention is a system that significantly reduces the time and effort required to create a recommendation, and provides users with efficient and high-quality document creation.

[1271] The processing flow will be explained below.

[1272] Step 1:

[1273] A user accesses an input form for writing a recommendation using a terminal, and inputs information about the name, field, characteristics, and memorable episode of the person to be recommended into the form.

[1274] Step 2:

[1275] The user presses the "Generate" button, which converts the entered information into JSON format and sends it to the server as an HTTP POST request.

[1276] Step 3:

[1277] The server parses the JSON data received from the user device and performs validation, checking whether all required fields are present and whether the characteristics and episodes are in list format.

[1278] Step 4:

[1279] The server passes the validated data to the generative AI model, which uses the validated data as input and generates a recommendation using an NLG (natural language generation) algorithm.

[1280] Step 5:

[1281] The generative AI model generates a recommendation based on the name, field, characteristics, and anecdotes of the person being recommended. In the generation process, the recommendation highlights the person's characteristics and integrates anecdotes to enrich the specific content.

[1282] Step 6:

[1283] The server receives the generated recommendation and returns it to the user's device in JSON format.

[1284] Step 7:

[1285] The user terminal displays the recommendation received from the server on the screen, allowing the user to easily obtain high-quality recommendations.

[1286] Example 1

[1287] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1288] Currently, writing testimonials and referrals is a time-consuming and laborious task, and creating high-quality testimonials requires specialized knowledge and experience. There is a need for a system that can solve this problem and automatically and efficiently generate high-quality testimonials.

[1289] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1290] In this invention, the server includes means for inputting information required to create a recommendation from a user's terminal, means for converting the information input from the terminal into JSON format and sending it to the server, means for the server to receive and validate the information, means for sending the validated information to a generative AI model to generate a recommendation, and means for returning the generated recommendation to the terminal. This enables users to easily and quickly create high-quality recommendations without requiring specialized knowledge.

[1291] A "user's terminal" is a digital device used by a user, such as a computer, smartphone, or tablet.

[1292] A "recommendation" is a piece of writing that is written to evaluate or recommend a specific individual or subject.

[1293] The "means for inputting information" refers to an interface (e.g., an input form) through which a user can input elements (such as name, field, characteristics, episodes, etc.) necessary for creating a recommendation.

[1294] The "means for converting to JSON format" is a mechanism for encoding user-entered information into JavaScript Object Notation (JSON) format.

[1295] A "server" is a computer system that receives and processes data sent from a user's terminal.

[1296] "Validation" is the process of checking whether entered data meets certain criteria.

[1297] A "generative artificial intelligence model" is a model that implements an algorithm that generates sentences based on input data using machine learning or deep learning techniques.

[1298] The "means for generating" is a process for generating a recommendation from input data using an artificial intelligence model.

[1299] The "means for sending back" is a mechanism for sending the generated recommendation to the user's terminal and displaying it.

[1300] The "Generate Button" is an interface element that initiates the submission of the information entered by the user and the generation of the recommendation.

[1301] The present invention is a system for efficiently creating recommendations and referrals using a generative AI model. The system includes the following components:

[1302] User terminal operation

[1303] When a user wants to generate a recommendation, they first access a web or native application. They enter information such as the name, field, characteristics, and memorable episodes of the person they are recommending into the provided input form. After entering all the information, the user presses the "Generate" button. This action converts the entered information into JSON format and sends it to the server as an HTTP POST request.

[1304] Server Operation

[1305] When the server receives a request from a user device, it first validates the data. Specifically, it checks whether all required fields are present and whether the characteristics and episodes are in list format. Data that passes validation is passed to a generative AI model, and the recommendation generation process begins. The generative AI model takes the received data as input and generates a recommendation that takes into account the characteristics and episodes of the person being recommended. The generated recommendation is returned to the server and sent back to the user device in JSON format.

[1306] How generative artificial intelligence models work

[1307] The generative AI model generates a recommendation based on the input data. This process works as follows: First, the name and field of the person being recommended are inserted at the beginning, and then the recommendation is constructed by listing the characteristics of the person being recommended. Furthermore, impressive anecdotes entered by the user are integrated to enrich the specific content of the recommendation.

[1308] Specific examples

[1309] Let's take a concrete example. For example, a user enters the information of a recommended person named "Taro Tanaka" as follows:

[1310] Name: Taro Tanaka

[1311] Field: Software Development

[1312] Traits: Hard work, teamwork, leadership

[1313] Episodes: Leadership in Project X, accurate communication in customer relations

[1314] When the user hits the "Generate" button, this information is sent to the server for validation. If validation is successful, the data is passed to a generative AI model, which generates a recommendation like this:

[1315] "Taro Tanaka has excellent abilities in the software development field, and his strengths are his diligence, teamwork, and leadership. Specifically, he has demonstrated these qualities through his experience in Project X, such as his leadership and accurate communication in customer relations."

[1316] The generated recommendation is sent back from the server to the user's terminal and displayed on the screen, allowing the user to easily obtain a high-quality recommendation.

[1317] Prompt Sentence Examples

[1318] An example of an input prompt for the generative AI model in this system might look like this:

[1319] Generate testimonials.

[1320] Name: Taro Tanaka

[1321] Field: Software Development

[1322] Traits: Hard work, teamwork, leadership

[1323] Episodes: Leadership in Project X, accurate communication in customer relations

[1324] In this way, the present invention is a system that significantly reduces the time and effort required to create a recommendation, and provides users with efficient and high-quality document creation.

[1325] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1326] Step 1: Collecting User Input

[1327] To generate a testimonial, a user accesses a web or native application and enters information into a form.

[1328] Input: The user enters the name, field, characteristics, and memorable episode of the person being recommended.

[1329] What happens: A user opens a web browser, such as Google Chrome, visits the system's web application URL, and then enters the following information into the form that appears on the screen:

[1330] Name: Taro Tanaka

[1331] Field: Software Development

[1332] Traits: Hard work, teamwork, leadership

[1333] Episodes: Leadership in Project X, accurate communication in customer relations

[1334] Output: The entered information is saved as form variables.

[1335] Step 2: Submitting input data

[1336] When the user presses the "Generate" button, the entered information is converted to JSON format and sent to the server.

[1337] Input: Information entered in step 1.

[1338] Specific behavior: When the user clicks the "Generate" button, the data is converted to JSON format using JavaScript AJAX (or Fetch API) and sent to the specified API endpoint.

[1339] Output: JSON formatted data is sent to the server as an HTTP POST request.

[1340] Step 3: Validate the data

[1341] The server receives the request from the user device and validates the data.

[1342] Input: The JSON data submitted in step 2.

[1343] What happens: The server (using the Django framework) validates the received JSON data. Specifically, it uses the Python jsonschema package to check the following:

[1344] Does it include names, areas, characteristics, and anecdotes?

[1345] Are the characteristics in list format?

[1346] Output: The validated data or a validation error response.

[1347] Step 4: Request a recommendation

[1348] The server passes the validated data to the generative AI model and makes a request to generate a recommendation.

[1349] Input: Data that passes validation.

[1350] What happens: The server sends the request data to a generative AI model (e.g., OpenAI's GPT-3 API). It uses the Python requests library to make the following request:

[1351] python

[1352] response = requests.post("https: / / api.openai.com / v1 / engines / davinci-codex / completions",

[1353] headers={"Authorization": "Bearer YOUR_API_KEY"},

[1354] json={"prompt": "Generate a testimonial.\nName: Taro Tanaka\nField: Software development\nCharacteristics: Diligence, teamwork, leadership\nAnecdotes: Leadership in Project X, accurate communication in customer relations", "max_tokens": 150})

[1355] Output: A recommendation generation request to the generative AI model.

[1356] Step 5: Generate testimonials

[1357] The generative AI model generates a recommendation based on the data provided.

[1358] Input: The prompt data submitted in step 4.

[1359] How it works: A generative AI model (OpenAI GPT-3) generates a recommendation based on the data provided as a prompt. The model generates text that includes the name, characteristics, and anecdotes of the person being recommended.

[1360] Output: The generated testimonial.

[1361] Step 6: Return and view testimonials

[1362] The server receives the generated recommendation and returns it to the user device in JSON format.

[1363] Input: The testimonial generated in step 5.

[1364] How it works: The server receives the response from the generated AI model and sends it back to the user's device as an HTTP response. The response data is processed by JavaScript on the user's device, and the generated recommendation is displayed at the bottom of the form.

[1365] Output: A testimonial that will be displayed on the user's device.

[1366] (Application example 1)

[1367] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1368] Creating recommendations and product reviews in the past was time-consuming and laborious, and it was difficult to maintain a consistent level of quality. Furthermore, if the information entered by the user was insufficient or inaccurate, the reliability and quality of the generated text declined. Furthermore, systems with no or insufficient validation functions often resulted in incomplete information being used as is. Another issue was the difficulty of accurately reflecting the user's intent when generating recommendations and reviews using natural language generation algorithms.

[1369] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1370] In this invention, the server includes means for inputting information required for creating a recommendation or review from a user's terminal, means for receiving and validating the information input from the terminal, means for sending the validated information to a generative AI model to generate a recommendation or review, and means for returning the generated recommendation or review to the terminal. This ensures the accuracy and completeness of the information input by the user and enables the automatic generation of high-quality recommendations and reviews.

[1371] A "terminal" is an electronic device through which a user enters information and receives generated text.

[1372] "Validation" is the process of ensuring that entered information is accurate and complete.

[1373] A "generative artificial intelligence model" is an artificial intelligence algorithm that automatically generates recommendations and reviews based on input information.

[1374] A "testimonial" is a written description that is written to evaluate and recommend another person.

[1375] A "review" is a document that describes impressions and evaluations of a product.

[1376] A "system" is a structure with a set of functions consisting of terminals, servers, and generative artificial intelligence models.

[1377] A "natural language generation algorithm" is a technology that uses artificial intelligence to generate sentences using human natural language.

[1378] "Information" is data entered by a user to generate a recommendation or review.

[1379] "JSON format" is an abbreviation for JavaScript Object Notation, and is a method of structuring data and representing it in text format.

[1380] An "episode" is a specific event or experience related to the characteristics or usability of the recommended person or product.

[1381] In this invention, the user terminal provides an interface for inputting information for generating a recommendation or product review. The user inputs the product name, reason for purchase, impressions of use, memorable episodes, etc. into an input form on the terminal. For example, the following information is input:

[1382] Product Name: High-Quality Laptop

[1383] Reason for purchase: I needed a powerful laptop for work.

[1384] User experience: Fast operation and long battery life

[1385] Episode: The battery lasted for 8 hours during a business trip, making it useful for presentations.

[1386] This input information is converted to JSON format and sent to the server as an HTTP POST request. The server validates the received data to ensure all required fields are present and the episodes are in a list format. If validation is successful, the server passes the data to a generative AI model to begin the process of generating reviews.

[1387] The server uses OpenAI's API to run a natural language generation (NLG) algorithm to generate a prompt based on the information the user has entered. For example, the following prompt might be generated:

[1388] Product name: High-quality laptop

[1389] Reason for purchase: I needed a powerful laptop for work.

[1390] User experience: Fast operation and long battery life

[1391] Episode: The battery lasted for 8 hours during a business trip, making it useful for presentations.

[1392] Please use the information above to write your product review."

[1393] When this prompt is sent to the OpenAI API, the generated review is sent back to the server and then sent to the user's device in JSON format, allowing users to obtain high-quality reviews in a short amount of time.

[1394] This system uses a terminal, a server, and a generative AI model (e.g., OpenAI's API). The terminal uses a typical smartphone or PC browser, and the server uses a web application using Python and the Flask framework. Information entered by the user is validated on the server and then passed to the generative AI model. The generated text is then sent back to the user's terminal via the server.

[1395] In particular, the automatic generation of product reviews has the advantage of saving users the trouble of writing reviews while providing high-quality reviews. This system makes it possible to consistently improve the quality of reviews and create them quickly.

[1396] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1397] Step 1:

[1398] The user enters information into the device's input form, such as the product name, reason for purchase, impressions on use, memorable episodes, etc. The entered information is converted into JSON format.

[1399] Input: Product name, reason for purchase, impressions of use, memorable episode

[1400] Output: JSON format data

[1401] Step 2:

[1402] The terminal sends the data converted into JSON format to the server as an HTTP POST request.

[1403] Input: JSON format data

[1404] Output: HTTP POST request

[1405] Step 3:

[1406] The server receives the HTTP POST request and performs data validation, ensuring all required fields are present and that the episodes are in a list format.

[1407] Input: JSON data via HTTP POST request

[1408] Output: Validation result (success / failure)

[1409] Step 4:

[1410] If validation is successful, the server sends the data to a generative artificial intelligence model to begin the process of generating the review.

[1411] Input: Validated JSON data

[1412] Output: Input data to a generative AI model

[1413] Step 5:

[1414] The server generates a prompt for the generative AI model (using OpenAI's API) and runs the NLG algorithm to create a prompt based on the information entered by the user.

[1415] Input: Data that has been successfully validated

[1416] Output: Prompt sentence generated by generative AI model

[1417] Step 6:

[1418] A generative AI model receives the prompt and generates a recommendation or review. During this generation process, a review with a specific structure is created based on the user's input.

[1419] Input: prompt statement

[1420] Output: Generated testimonial or review

[1421] Step 7:

[1422] The generated recommendation and review are sent back to the server, which then sends them back to the user's device in JSON format.

[1423] Input: Generated testimonial or review

[1424] Output: Generated result data in JSON format

[1425] Step 8:

[1426] The user device receives the generated recommendation and review and displays it to the user, allowing the user to check high-quality reviews.

[1427] Input: JSON format generated result data

[1428] Output: Displayed testimonial or review on the user's device

[1429] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1430] The present invention provides a system for recognizing a user's emotions and generating a recommendation sentence by taking the emotion information into consideration. The system includes a user terminal, a server, a generative artificial intelligence model, and an emotion engine.

[1431] System configuration and operation

[1432] User terminal operation

[1433] A user accesses the system using a terminal and enters information into an input form to create a recommendation. The information entered includes the name, field, characteristics, and memorable episodes of the person being recommended. In addition, the user's terminal utilizes an emotion engine to recognize and analyze the user's emotional state (e.g., joy, anger, sadness, etc.) at the time of input.

[1434] Server Operation

[1435] The information obtained from the input form and emotion engine is sent to the server in JSON format. The server has the following functions:

[1436] 1. Data Receipt and Validation:

[1437] The server receives the JSON data sent from the user device and checks whether all required fields are present and whether the characteristics and episodes are in list format.

[1438] 2. Post-validation data processing:

[1439] The data that passes validation and the emotional information obtained from the emotion engine are passed to the generative artificial intelligence model.

[1440] How generative artificial intelligence models work

[1441] The generative AI model generates recommendations based on the received data and sentiment information. This process includes the following steps:

[1442] 1. Integration of basic information:

[1443] The nominee's name, field, characteristics, and anecdote are received as input data.

[1444] 2. Reflecting emotional information:

[1445] The tone and content of the recommendation are adjusted based on the emotional information obtained from the emotion engine. For example, if the user is happy, the recommendation will be generated in a style that reflects that positive emotion.

[1446] 3. Generating Recommendations:

[1447] Finally, a recommendation is finalized using a natural language generation (NLG) algorithm.

[1448] Returning recommendation letters

[1449] The generated recommendation is sent back to the server and sent in JSON format to the user's device, where it is displayed on the screen, allowing the user to obtain a high-quality, emotionally relevant recommendation.

[1450] Specific examples

[1451] Specific examples are shown below.

[1452] 1. If a user wants to write a recommendation, they access the system and enter the information of the person being recommended, "Ichiro Tanaka":

[1453] Name: Ichiro Tanaka

[1454] Area: Marketing

[1455] Traits: Creativity, leadership, analytical skills

[1456] Memorable episodes: Successful new product launch project, winning a marketing strategy competition

[1457] 2. While the user is entering information, the emotion engine detects the user's emotion of joy.

[1458] 3. When the user presses the "Generate" button, the input information and emotion information are sent to the server.

[1459] 4. The server receives the information and validates it.

[1460] 5. The information that passes validation is passed to the generative AI model, which takes into account the emotional information and generates a recommendation like the following:

[1461] "Ichiro Tanaka has outstanding abilities in the field of marketing, with particular strengths in creativity, leadership, and analytical ability. He has a proven track record, including successfully implementing new product launch projects and winning marketing strategy competitions. His positive attitude and energy have a positive impact on the entire team."

[1462] 6. The generated recommendation is sent back from the server to the user's device and displayed on the screen.

[1463] This allows users to easily obtain high-quality recommendation messages that reflect their emotional state, thereby improving the efficiency and accuracy of recommendation message creation.

[1464] The processing flow will be explained below.

[1465] Step 1:

[1466] The user accesses the system's input form using a terminal and inputs the name, field, characteristics, and memorable episode of the person to be recommended.

[1467] Step 2:

[1468] While the user is typing, the device's onboard emotion engine recognizes the user's emotional state, using the camera and microphone to analyze the user's facial expressions and vocal tone.

[1469] Step 3:

[1470] Once the user has entered all the required information, they press the "Generate" button. This action converts the entered information and emotion data into JSON format and sends it to the server as an HTTP POST request.

[1471] Step 4:

[1472] The server parses the JSON data received from the user device and performs data validation, checking whether all required fields are present and whether the characteristics and episodes are in list format.

[1473] Step 5:

[1474] The server passes the validated data and emotion information to the generative AI model. Specifically, the data containing emotion information is input into the model.

[1475] Step 6:

[1476] The generative AI model generates a recommendation based on the received data and emotional information, taking into account the characteristics and anecdotes of the person being recommended and adjusting the writing style and tone based on the emotional information.

[1477] Step 7:

[1478] The generated recommendation is sent back to the server, which then sends it back to the user's device in JSON format.

[1479] Step 8:

[1480] The user's device analyzes the recommendation received from the server and displays it on the screen, allowing the user to obtain high-quality recommendations that reflect their emotional state.

[1481] Example 2

[1482] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1483] Conventional recommendation generation systems update recommendation sentences in a fixed style without considering the user's emotional state, making it difficult to generate persuasive recommendation sentences that fit the user's emotions. Another problem is that processing data without proper validation reduces the quality of the generated recommendation sentences. The purpose of this invention is to solve these problems and generate high-quality recommendation sentences that reflect the user's emotional state.

[1484] The specification process by the specification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for inputting information necessary for creating a recommendation from the user's information processing device, means for analyzing the user's emotional state using an emotion recognition engine, means for receiving and validating the information and emotional information input from the processing device, means for sending the validated information and emotional information to a generative AI model and generating a recommendation, and means for returning the generated recommendation to the processing device. This makes it possible to generate high-quality recommendations that reflect the user's emotional state.

[1485] The "user's information processing device" refers to a device used by the user, such as a computer, smartphone, or tablet, and is a device that has an interface for inputting information for creating a recommendation.

[1486] An "emotion recognition engine" is software or hardware that analyzes a user's emotional state (e.g., joy, anger, sadness, etc.) using the user's typing speed, keystroke strength, facial recognition technology, etc.

[1487] Validation is the process of ensuring that the information entered by the user is accurate, by checking that all required fields are present and that the format is correct.

[1488] A "generative artificial intelligence model" is a machine learning model that generates recommendations using natural language generation algorithms based on input data and emotional information.

[1489] A "recommendation letter" is a document that reflects the characteristics and anecdotes about the recommended person specified by the user, as well as the emotional state of the user at the time, and is used for the purpose of evaluation or introduction.

[1490] A "natural language generation algorithm" is a technology that allows computers to generate human language, and is a method for automatically creating sentences based on input data.

[1491] "JSON format" is an abbreviation for JavaScript Object Notation, a lightweight data exchange format and a text format that is easy to read for both humans and machines.

[1492] The present invention provides a system for recognizing a user's emotions and generating a recommendation sentence by taking the emotion information into consideration. The system includes a user's information processing device, a server, a generative artificial intelligence model, and an emotion recognition engine.

[1493] System configuration and operation

[1494] Operation of user information processing device

[1495] A user inputs necessary information (the name, field, characteristics, and memorable episode of the person being recommended) into an input form for creating a recommendation.

[1496] For example, input may be in the following format:

[1497] Name: Name of the nominee

[1498] Area: Marketing

[1499] Traits: Creativity, leadership, analytical skills

[1500] Memorable episodes: Successful new product launch project, winning a marketing strategy competition

[1501] The user information processing device also uses an emotion recognition engine to analyze the user's emotional state while typing. The emotion recognition engine detects emotions (e.g., joy, anger, sadness) through the user's typing speed, keystroke strength, and facial recognition technology.

[1502] Server Operation

[1503] When the user presses the "Generate" button, the input information and emotion information are sent to the server in JSON format. The server provides the following functions:

[1504] 1. Data reception and validation: The server receives the JSON data sent from the user information processing device and checks whether all required fields are present and whether the characteristics and episodes are in list format.

[1505] 2. Post-validation data processing: The data and sentiment information that have passed validation are passed to the generative AI model.

[1506] How generative artificial intelligence models work

[1507] The generative AI model generates recommendations based on the received data and emotional information. The generative AI model works as follows:

[1508] 1. Integration of basic information: The name, field, characteristics, and anecdotes of the nominee are received as input data.

[1509] 2. Reflection of emotional information: The tone and content of the recommendation are adjusted based on the emotional information obtained from the emotion recognition engine. For example, if the user is happy, the recommendation will be generated in a style that reflects that positive emotion.

[1510] 3. Recommendation Generation: A natural language generation (NLG) algorithm is used to complete the recommendation.

[1511] Returning recommendation letters

[1512] The generated recommendation is returned to the server and sent in JSON format to the user's information processing device, which then displays the received recommendation on its screen, allowing the user to obtain a high-quality recommendation that is tailored to their emotions.

[1513] Specific examples

[1514] When a user wants to write a testimonial, they access the system and enter the following information:

[1515] Name: Ichiro Tanaka

[1516] Area: Marketing

[1517] Traits: Creativity, leadership, analytical skills

[1518] Memorable episodes: Successful new product launch project, winning a marketing strategy competition

[1519] While the user is entering information, the emotion recognition engine detects the user's emotion of "happiness." The input information and emotion information are then sent to the server, where, after validation, they are passed to the generative AI model. The generative AI model generates the following recommendation based on the emotion information:

[1520] "Ichiro Tanaka has outstanding abilities in the field of marketing, with particular strengths in creativity, leadership, and analytical ability. He has a proven track record, including successfully implementing new product launch projects and winning marketing strategy competitions. His positive attitude and energy have a positive impact on the entire team."

[1521] The generated recommendation is sent back from the server to the user's information processing device and displayed on the screen, allowing the user to easily obtain a high-quality recommendation that reflects their emotional state.

[1522] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1523] Step 1:

[1524] Obtaining user input information

[1525] Process description: A user inputs the required information (the name, field, characteristics, and memorable episode of the person being recommended) into an input form to create a recommendation.

[1526] What happens: A user opens a browser and visits a web page on the system. The user enters the following information into the form fields:

[1527] Name: Ichiro Tanaka

[1528] Area: Marketing

[1529] Traits: Creativity, leadership, analytical skills

[1530] Memorable episodes: Successful new product launch project, winning a marketing strategy competition

[1531] Input: The user enters information into an input form on the device.

[1532] Output: The terminal holds the information entered.

[1533] Step 2:

[1534] Emotion recognition using an emotion recognition engine

[1535] Process Description: The emotion recognition engine analyzes the user's emotional state (happiness, anger, sadness, etc.).

[1536] How it works: The emotion recognition engine analyzes the user's emotions through the camera using facial recognition technology, the speed of their typing, and the strength of their keystrokes. The emotion engine detects the emotional state of "joy."

[1537] Input: User input data and facial image data while the user is entering information into an input form.

[1538] Output: Parsed emotion information (e.g., joy).

[1539] Step 3:

[1540] Sending input data and emotion data to the server

[1541] Process description: Send user input information and emotion information to the server in JSON format.

[1542] Specific operation: When the user presses the "Generate" button, the device converts the input information and emotion information into JSON. The device then sends the converted JSON data to the server via HTTPS.

[1543] Input: User input and emotional information.

[1544] Output: The data converted to JSON format is sent to the server.

[1545] Step 4:

[1546] Receiving and validating data by the server

[1547] Process description: The server receives the JSON data and checks whether the entered data is correct.

[1548] What happens: The server parses the received JSON data and verifies that all required fields (name, field, characteristics, episode) are present. Data validation is performed to ensure the format is correct.

[1549] Input: Input and emotion information received in JSON format.

[1550] Output: Data that passes validation.

[1551] Step 5:

[1552] Recommendation generation using generative artificial intelligence models

[1553] Process description: The server passes the validated data and sentiment information to the generative AI model to generate a recommendation.

[1554] Specific operation: The server passes the validated data to the generative AI model. The generative AI model takes into account the user's emotional information (delight) and adjusts the tone and content of the recommendation. The generative AI model generates the recommendation using a natural language generation (NLG) algorithm. For example, a recommendation might read, "Ichiro Tanaka has outstanding abilities in the marketing field, with particular strengths in creativity, leadership, and analytical ability. He has a long list of achievements, including the success of a new product launch project and winning a marketing strategy competition. His proactive attitude and positive energy have a positive influence on the entire team."

[1555] Input: Validated input and sentiment information.

[1556] Output: The generated recommendation.

[1557] Step 6:

[1558] Sending and displaying the recommendation to the user's device

[1559] Process description: The generated recommendation is sent back to the user's device in JSON format from the server and displayed on the screen.

[1560] Specific operation: The server converts the generated recommendation into JSON and sends it to the user's device. The user's device parses the received JSON data and displays the recommendation on the screen.

[1561] Input: JSON data containing the generated testimonials.

[1562] Output: The testimonial displayed on the device.

[1563] (Application example 2)

[1564] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1565] Conventional recommendation generation systems generate recommendation sentences uniformly without considering the user's emotional information, which can result in recommendation sentences that do not match the user's emotions or nuances. This means that users have to spend time and effort writing recommendation sentences, and the generated recommendation sentences do not always fully reflect the user's intentions.

[1566] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1567] In this invention, the server includes means for receiving information necessary for creating a recommendation and user emotional information from the user's terminal and validating the information, means for sending the validated information and emotional information to a generative AI model to generate a recommendation, and means for returning the generated recommendation to the terminal, thereby enabling the generation of high-quality recommendations that reflect the user's emotional information.

[1568] "User terminal" means a device used by a user to access the system and input information, including, for example, a smartphone, tablet, or personal computer.

[1569] "Information" refers to the name, field, characteristics, memorable episodes, etc. of the person being recommended, which are necessary to write the recommendation letter.

[1570] "Emotion information" is data on the user's emotional state (for example, joy, anger, sadness, etc.) extracted based on information input by the user.

[1571] "Validation" is the process of ensuring that the information and sentiment information entered by the user is accurate and complete.

[1572] "Generative AI models" are algorithms or models that generate recommendations based on received information and emotional information. Examples include natural language generation (NLG) algorithms.

[1573] A "recommendation letter" is a piece of writing generated based on information about the person being recommended, and includes content that appealingly conveys the characteristics and anecdotes of the person being recommended.

[1574] A "natural language generation algorithm" is an algorithm for generating understandable and appropriate sentences based on information and emotional information entered by the user.

[1575] "JSON format" is an abbreviation for JavaScript Object Notation, and is a format for expressing structured data in text format.

[1576] "Terminal" refers to any device that accepts user operations and processes information, including smartphones, personal computers, and tablets.

[1577] A "server" is a computer system that receives information sent from a user's terminal, processes it, and returns a final recommendation.

[1578] The present invention relates to a system for generating recommendation sentences taking into account a user's emotional information, which includes a user terminal, a server, a generative artificial intelligence model, and an emotional engine.

[1579] The main components of the system are as follows:

[1580] User terminal

[1581] The user terminal is a device for inputting information and recognizing emotions. Examples include smartphones, tablets, and personal computers. The user uses the terminal to access the system and input information to create a recommendation. The input information includes the name, field, characteristics, and memorable anecdotes of the person being recommended. The system also uses an emotion engine to recognize and analyze the user's emotional state (e.g., joy, anger, sadness, etc.).

[1582] server

[1583] The server is a central device that receives information and emotion information sent from user terminals and generates recommendation statements. Specifically, it has the following functions:

[1584] 1. Information reception and validation: The server receives the information and emotion information entered from the user device in JSON format. It then checks whether all required fields are present and whether the characteristics and episodes are in list format, and performs validation.

[1585] 2. Data processing: Receive the validated information and sentiment information and pass it to the generative AI model.

[1586] Generative AI Model

[1587] The generative AI model generates recommendations based on the received data and emotional information. Specifically, it performs the following processes:

[1588] 1. Basic information integration: Receive the nominee's name, field, characteristics, and anecdotes as input data.

[1589] 2. Reflection of emotional information: The tone and content of the recommendation are adjusted based on the emotional information obtained from the emotion engine. If the user's emotion of joy is detected, the recommendation is generated in a positive tone.

[1590] 3. Recommendation generation: A natural language generation algorithm (e.g., GPT-3) is used to complete the final recommendation.

[1591] Returning recommendation letters

[1592] The generated recommendation is sent back to the server and sent in JSON format to the user's device, where it is displayed on the screen, allowing the user to obtain a high-quality, emotionally relevant recommendation.

[1593] Specific examples

[1594] For example, if a user wants to write a movie review, they enter the following information:

[1595] Username: Taro Nakamura

[1596] Content Type: Movies

[1597] Content Title: Inception

[1598] Details of my viewing experience: I was very moved by the final scene and was moved to tears.

[1599] The emotion engine recognizes the user's emotions and generates a review that looks like this:

[1600] "Taro Nakamura watched the movie 'Inception' and was deeply moved, especially by the final scene. He was moved to tears, and the beauty of the scene and the moving story resonated deeply in his heart. Director Christopher Nolan's outstanding directing skills shine through, and Nakamura has a very high opinion of the film."

[1601] Prompt Sentence Examples

[1602] "Recognize emotions based on the content viewing experience and generate reviews that correspond to the user's emotions."

[1603] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1604] Step 1:

[1605] The user accesses the recommendation writing form using a terminal and inputs information about the person being recommended (name, field, characteristics, memorable episodes), along with details of the user's viewing experience.

[1606] Input: Name, field, characteristics, memorable episodes, details of viewing experience

[1607] Output: Input information and data passed to the emotion recognition engine

[1608] Step 2:

[1609] The device's emotion recognition engine analyzes the details of the user's viewing experience and recognizes the user's emotional state. For example, if the user inputs "I was moved," it will recognize the emotion of being moved.

[1610] Input: Viewing experience details

[1611] Output: User's emotional information (e.g., emotion, joy, anger, etc.)

[1612] Step 3:

[1613] The device converts the input information and recognized emotion information into JSON format and sends it to the server.

[1614] Input: Name, field, characteristics, memorable episodes, user's emotional information

[1615] Output: JSON format data

[1616] Step 4:

[1617] The server parses the data received in JSON format, checks whether all required fields are present, whether the characteristics and episodes are in list format, and performs validation.

[1618] Input: JSON format data

[1619] Output: Successfully validated data or an error message

[1620] Step 5:

[1621] The server passes the validated data and emotion information to the generative artificial intelligence model.

[1622] Input: Successfully validated data, emotion information

[1623] Output: The data used to generate the recommendation

[1624] Step 6:

[1625] The generative AI model generates a recommendation based on the received data. Specifically, it integrates basic information (name, field, characteristics, memorable episodes) with emotional information and applies a natural language generation algorithm (NLG) to generate the recommendation.

[1626] Input: Basic information, emotional information

[1627] Output: Generated recommendation

[1628] Step 7:

[1629] The generated recommendation is sent back from the server to the user's device in JSON format.

[1630] Input: Generated testimonial

[1631] Output: Recommendation data in JSON format

[1632] Step 8:

[1633] The user's device analyzes the received recommendation and displays it on the screen, allowing the user to obtain a recommendation that is high quality and emotionally appropriate.

[1634] Input: Recommendation data in JSON format

[1635] Output: Testimonial displayed on screen

[1636] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1637] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1638] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1639] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1640] FIG. 9 illustrates an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and behaviors arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1641] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1642] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1643] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1644] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1645] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1646] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1647] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1648] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1649] 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.

[1650] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1651] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1652] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1653] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1654] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1655] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1656] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1657] The following is further disclosed regarding the above embodiment.

[1658] (Claim 1)

[1659] A means for inputting information necessary for creating a recommendation from a user's terminal;

[1660] means for receiving and validating information input from the terminal;

[1661] A means for transmitting the validated information to a generative artificial intelligence model to generate a recommendation;

[1662] means for returning the generated recommendation to the terminal;

[1663] A system including:

[1664] (Claim 2)

[1665] The system of claim 1, wherein the generative artificial intelligence model generates a recommendation by applying a natural language generation algorithm based on the characteristics and anecdotes of the person being recommended.

[1666] (Claim 3)

[1667] 2. The system according to claim 1, wherein the terminal has means for transmitting information necessary for creating the recommendation to a server in JSON format.

[1668] "Example 1"

[1669] Claims

[1670] (Claim 1)

[1671] A means for inputting information necessary for creating a recommendation from a user's terminal;

[1672] means for converting information input from the terminal into JSON format and transmitting the converted information to a server;

[1673] means for the server to receive and validate the information;

[1674] A means for transmitting the validated information to a generative artificial intelligence model to generate a recommendation;

[1675] means for returning the generated recommendation to the terminal;

[1676] A system including:

[1677] (Claim 2)

[1678] The system of claim 1, wherein the generative artificial intelligence model generates a recommendation by applying a natural language generation algorithm based on the characteristics and anecdotes of the person being recommended.

[1679] (Claim 3)

[1680] 2. The system according to claim 1, wherein the terminal has means for transmitting information required for creating the recommendation to a server by pressing a create button after inputting the information required for creating the recommendation.

[1681] "Application Example 1"

[1682] (Claim 1)

[1683] A means for inputting information necessary for creating a recommendation or review from a user's device;

[1684] means for receiving and validating information input from the terminal;

[1685] A means for transmitting the validated information to a generative artificial intelligence model to generate a recommendation or review;

[1686] means for returning the generated recommendation or review to the terminal;

[1687] A system including:

[1688] (Claim 2)

[1689] The system of claim 1, wherein the generative artificial intelligence model generates a recommendation or review by applying a natural language generation algorithm based on the characteristics and anecdotes of the recommended person or product.

[1690] (Claim 3)

[1691] 2. The system according to claim 1, wherein the terminal has means for transmitting information required for creating the recommendation or review to a server in JSON format.

[1692] "Example 2: Combining Emotion Engines"

[1693] (Claim 1)

[1694] A means for inputting information required for creating a recommendation from a user's information processing device;

[1695] means for analyzing the emotional state of the user using an emotion recognition engine;

[1696] means for receiving and validating information and emotion information input from the processing device;

[1697] A means for transmitting the validated information and emotion information to a generative artificial intelligence model to generate a recommendation;

[1698] means for returning the generated recommendation to the processing device;

[1699] A system including:

[1700] (Claim 2)

[1701] The system of claim 1, wherein the generative artificial intelligence model generates a recommendation by applying a natural language generation algorithm based on the characteristics and anecdotes of the recommended person as well as the user's emotional state analyzed using the emotion recognition engine.

[1702] (Claim 3)

[1703] 2. The system according to claim 1, wherein the processing device has means for transmitting the information necessary for creating the recommendation statement and the emotion information to the data processing device in JSON format.

[1704] "Application example 2 when combining emotion engines"

[1705] (Claim 1)

[1706] A means for inputting information necessary for creating a recommendation from a user's terminal;

[1707] means for receiving and validating information input from the terminal and user emotion information;

[1708] A means for transmitting the validated information and emotion information to a generative artificial intelligence model to generate a recommendation;

[1709] means for returning the generated recommendation to the terminal;

[1710] A system including:

[1711] (Claim 2)

[1712] The system of claim 1, wherein the generative artificial intelligence model generates a recommendation by applying a natural language generation algorithm based on the characteristics and anecdotes of the recommended person and the user's emotional information.

[1713] (Claim 3)

[1714] 2. The system according to claim 1, wherein the terminal has means for transmitting information necessary for creating the recommendation to a server in JSON format. [Explanation of symbols]

[1715] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means for inputting information necessary for creating a recommendation from a user's terminal; means for receiving and validating information input from the terminal; A means for transmitting the validated information to a generative artificial intelligence model to generate a recommendation; means for returning the generated recommendation to the terminal; A system including:

2. The system according to claim 1 , wherein the generative artificial intelligence model generates a recommendation by applying a natural language generation algorithm based on the characteristics and anecdotes of the person being recommended.

3. The system according to claim 1 , wherein the terminal has means for transmitting information required for creating the recommendation to a server in JSON format.

Images