System

Abstract

To provide a system.SOLUTION: A system includes: first means for acquiring input words relating to a meal of a user; second means which generates a prompt sentence for outputting a proposal of an appropriate food delivery to the user on the basis of the inputted words; third means for generating the proposal of the appropriate food delivery by inputting the generated prompt sentence to a generative AI model; and fourth means for displaying the generated proposal of the appropriate food delivery on a terminal of the user.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] Traditional chat apps lack guidance on how to start a conversation or what questions to ask when users first start using them, which can make users hesitant to use chat apps. [Means for solving the problem]

[0005] The present invention generates prompt candidates via the web based on user input words, presents the generated prompt candidates to the user, and also incorporates a chatbot that assists in dialogue with the user, making it easier for users to start using the chat app. [Brief explanation of the drawings]

[0006] [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. 2 is a sequence diagram showing a flow of processing in the data processing system according to the first embodiment of the first form example. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Embodiment 1. [Figure 13] FIG. 10 is a sequence diagram showing a processing flow of a data processing system in a second embodiment of the second form example. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 of Embodiment Example 2. [Figure 15] FIG. 10 is a sequence diagram showing the flow of processing in a data processing system according to a third embodiment of the third embodiment. [Figure 16]FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 3 of Embodiment 3. [Figure 17] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the first embodiment of the first form example when an emotion engine is combined. [Figure 18] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Form Example 1 when an emotion engine is combined. [Figure 19] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the second embodiment of the second form example when an emotion engine is combined. [Figure 20] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 of Form Example 2 when an emotion engine is combined. [Figure 21] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the third embodiment of the third form example when an emotion engine is combined. [Figure 22] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 3 of Form Example 3 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0007] 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.

[0008] First, the terms used in the following description will be explained.

[0009] 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, the 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), an APU (Accelerated Processing Unit), or a TPU (TENSOR PROCESSING UNIT (registered trademark)).

[0010] 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.

[0011] 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.

[0012] 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.

[0013] 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."

[0014] [First embodiment]

[0015] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0016] 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.

[0017] 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).

[0018] 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.

[0019] 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.

[0020] 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.

[0021] 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.

[0022] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0023] 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.

[0024] 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.

[0025] 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.

[0026] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0027] "Example 1"

[0028] In one embodiment of the present invention, when a user inputs a specific word into a chat app, the system generates prompt candidates via the web based on the input word. For example, if a user inputs the word "travel," the system generates prompt candidates such as "What are your recommended travel destinations?" and "What do I need to prepare for my trip?"

[0029] "Example 2"

[0030] The generated prompt candidates are presented to the user, allowing the user to receive guidance on how to proceed with the conversation. For example, if a prompt candidate such as "What travel destinations do you recommend?" is presented, the user can respond with a more specific question such as "What overseas places do you recommend?"

[0031] "Example 3"

[0032] Furthermore, the system of the present invention is equipped with a chatbot that supports dialogue with the user. This chatbot converses with the user using prompt candidates generated based on the user's input words. For example, if the user asks, "What overseas places do you recommend?", the chatbot will generate a response such as, "How about Paris, France? You can enjoy visiting museums and gourmet food there."

[0033] The processing flow of each embodiment will be described below.

[0034] "Example 1"

[0035] Step 1: A user enters a specific word in a chat app, for example, the word "travel."

[0036] Step 2: The system generates prompt candidates via the web based on the relevant words. In this case, it generates prompt candidates such as "What travel destinations do you recommend?" and "What do I need to prepare for my trip?"

[0037] Step 3: The generated prompt candidates are presented to the user, providing guidance on how to proceed with the interaction.

[0038] "Example 2"

[0039] Step 1: The user asks a specific question in response to the suggested prompt. For example, in response to the prompt "What travel destinations do you recommend?", the user can ask "What overseas destinations do you recommend?"

[0040] Step 2: The chatbot generates an answer to the user's question, in this case, "How about Paris, France? You can enjoy visiting museums and delicious food."

[0041] "Example 3"

[0042] Step 1: The user asks the chatbot a follow-up question, such as, "What are the best museums to visit in Paris?"

[0043] Step 2: The chatbot generates an answer to the user's question, in this case "The Louvre and the Orsay Museum are must-sees. Also, there are many art studios in Montmartre around the Sacré-Coeur Basilica."

[0044] Example 1

[0045] Next, a description will be given of Example 1 of Form 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."

[0046] Conventional chatbot systems have had difficulty generating appropriate prompt candidates in response to user input, resulting in problems with the dialogue between users and the system not proceeding smoothly. Furthermore, the prompt candidates generated based on user input are limited, making it difficult to respond to diverse user needs.

[0047] 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.

[0048] In this invention, the server includes means for receiving a user's input words and generating prompt candidates using a generative AI model based on the received words, means for presenting the generated prompt candidates to the user, and means for incorporating a dialogue system that supports dialogue with the user. This makes it possible to generate a variety of appropriate prompt candidates in response to the user's input and smoothly progress with dialogue with the user.

[0049] "User" means any person or entity that uses the System to provide input.

[0050] "Input words" are specific words or phrases that users type into a chat app.

[0051] A "generative AI model" is an artificial intelligence model for generating relevant prompt candidates based on user input.

[0052] "Prompt candidates" are questions or phrases that are generated based on the user's input words and serve as a guide for the user's next input.

[0053] A "presenting means" is a method or device for displaying the generated prompt candidates to a user.

[0054] A "dialogue system" is a software or hardware system that assists in dialogue with a user.

[0055] "Server" means a central processing unit for receiving a user's input words, generating prompt candidates using a generative AI model, and presenting the generated prompt candidates to the user.

[0056] MODE FOR CARRYING OUT THE INVENTION

[0057] This invention is a system that, when a user inputs a specific word in a chat app, generates prompt candidates based on the input word using a generative AI model and presents them to the user. A specific embodiment of this system is described below.

[0058] System configuration

[0059] The system includes a terminal that receives input words from a user, a server that analyzes the input words and generates prompt candidates, a means for presenting the generated prompt candidates to the user, and a dialogue system that assists in dialogue with the user.

[0060] Hardware and software used

[0061] Terminal: The device through which the user provides input (e.g., smartphone, tablet, PC).

[0062] Server: Acts as the central processing unit, parses the user's input words, and generates suggested prompts using generative AI models.

[0063] Generative AI models: For example, advanced natural language processing models such as OpenAI's GPT-3.

[0064] Dialogue system: Software that assists in interacting with users.

[0065] Data processing and calculation

[0066] 1. Data reception: The device sends the words the user types into the chat app to the server.

[0067] 2. Data analysis: The server analyzes the received words and extracts relevant keywords and contexts. Natural language processing (NLP) techniques are used for the analysis.

[0068] 3. Prompt generation: The server uses a generative AI model (e.g., GPT-3) to generate prompt candidates based on the extracted keywords and context.

[0069] 4. Data transmission: The server sends the generated prompt candidates to the terminal.

[0070] 5. Data display: The terminal displays the received prompt candidates to the user.

[0071] Specific examples

[0072] For example, if a user types the word "travel" into a chat app, the system will act as follows:

[0073] 1. A user types "travel" into a chat app.

[0074] 2. The device sends the word "travel" to the server.

[0075] 3. The server parses the word "travel" and sends a prompt to the generative AI model.

[0076] 4. The generative AI model generates candidate prompts such as "What are some recommended travel destinations?" and "What do I need to prepare for my trip?"

[0077] 5. The server sends the generated prompt candidates to the device.

[0078] 6. The device displays suggested prompts to the user, such as "What are some recommended travel destinations?" or "What do I need to prepare for my trip?"

[0079] Prompt Sentence Examples

[0080] Examples of prompts to be input to a generative AI model include:

[0081] The user has entered the word "travel", based on which you want to generate relevant prompt suggestions.

[0082] In this way, users can get relevant prompt suggestions through the chat app, which helps to facilitate a smooth dialogue with the user and provide them with the right information according to their needs.

[0083] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0084] Step 1:

[0085] A user types a word into a chat app.

[0086] A user opens a chat app and inputs a specific word (e.g., "travel") based on the user's interests or questions. The input data is saved in text format on the device.

[0087] Step 2:

[0088] The terminal sends the entered word to the server.

[0089] The terminal sends the word entered by the user to the server. At this time, the terminal converts the input data into an appropriate format (for example, JSON format) and sends it to the server via the network. The input data is the user's input word, "travel."

[0090] Step 3:

[0091] The server parses the words and sends prompts to the generative AI model.

[0092] The server analyzes the words received from the device. It uses natural language processing (NLP) technology to understand the meaning and context of the words. Based on the analysis results, it sends a prompt to the generative AI model (e.g., GPT-3). The specific prompt is as follows:

[0093] The user has entered the word "travel", based on which you want to generate relevant prompt suggestions.

[0094] The input data is the user's input word "travel," and the output data is the prompt sentence sent to the generative AI model.

[0095] Step 4:

[0096] A generative AI model generates candidate prompts.

[0097] The generative AI model generates relevant prompt candidates based on the prompt received from the server. For example, candidates such as "What travel destinations do you recommend?" or "What do I need to prepare for my trip?" are generated. The input data is the prompt sentence sent from the server, and the output data is the generated prompt candidates.

[0098] Step 5:

[0099] The server sends the generated prompt candidates to the terminal.

[0100] The server sends the prompt candidates received from the generative AI model to the terminal. At this time, the server converts the data into an appropriate format and sends it to the terminal over the network. The input data is the prompt candidates received from the generative AI model, and the output data is the prompt candidates sent to the terminal.

[0101] Step 6:

[0102] The terminal displays the prompt candidates to the user.

[0103] The terminal displays the prompt candidates received from the server to the user. The user can refer to the displayed prompt candidates to ask more specific questions or enter more specific input. The input data is the prompt candidates received from the server, and the output data is the prompt candidates displayed to the user.

[0104] (Application example 1)

[0105] Next, a description will be given of Application Example 1 of Embodiment 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."

[0106] Conventional chatbot systems have had difficulty generating appropriate prompt candidates for keywords entered by users, resulting in a lack of smooth dialogue with users. They also had issues with not being able to efficiently suggest relevant content that users were looking for, resulting in a lack of improvement in the user experience.

[0107] 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.

[0108] In this invention, the server includes means for receiving words input by a user and generating prompt candidates via the Web based on the words, means for presenting the generated prompt candidates to the user, and means for generating prompt candidates for related content based on keywords input by the user and suggesting them to the user. This makes it possible to generate appropriate prompt candidates for the keywords input by the user and efficiently suggest related content.

[0109] "User" means any individual or entity that uses the System.

[0110] An "input word" is a specific word or phrase that a user enters into the system.

[0111] "Web" refers to information resources on the Internet, including websites and online databases.

[0112] "Prompt suggestions" are suggestions for what to type next or questions that are generated based on the words the user types.

[0113] "Content" is a general term for information, data, media, etc., and includes text, images, video, audio, etc.

[0114] A "chatbot" is a program that automatically interacts with users.

[0115] A "system" is an entire device or program that is composed of a combination of multiple means and functions.

[0116] A "suggestion" is the act of presenting options or information to a user.

[0117] The system for implementing this invention has a function of receiving user input words, generating prompt candidates via the Web based on the received words, and presenting the generated prompt candidates to the user. It also includes a function of generating prompt candidates for related content based on keywords entered by the user and suggesting them to the user.

[0118] The server receives the words entered by the user and retrieves relevant information from the web based on those words. Specifically, it uses a Python library to send web requests and parse the HTML data. It then extracts prompt candidates from the parsed data and presents them to the user.

[0119] When a user enters the keyword "movie," the server retrieves prompt candidates from the web, such as "What movies do you recommend?" or "What are the latest movie reviews?", and suggests them to the user. This allows the user to efficiently obtain related content.

[0120] Examples of specific prompts include the following:

[0121] The user has entered the keyword "movies". Generate relevant prompt suggestions.

[0122] The system uses a smartphone as hardware and Python as software, which allows it to generate appropriate prompt candidates for keywords entered by users and efficiently suggest related content.

[0123] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0124] Step 1:

[0125] The user enters keywords into the terminal. The entered keywords are sent to the system. The input data is the specific words or phrases that the user is interested in.

[0126] Step 2:

[0127] The server sends a web request based on the received keywords. Specifically, it uses a Python library to access a URL to retrieve related information. The input data is the keyword entered by the user, and the output data is HTML data retrieved from the web.

[0128] Step 3:

[0129] The server analyzes the acquired HTML data. Specifically, it analyzes the HTML data and extracts prompt candidates. The input data is the HTML data acquired from the Web, and the output data is the analyzed prompt candidates.

[0130] Step 4:

[0131] The server presents the parsed prompt candidates to the user. Specifically, the server transmits data for displaying the prompt candidates to the user's terminal. The input data is the parsed prompt candidates, and the output data is the prompt candidates presented to the user.

[0132] Step 5:

[0133] The user selects from the presented prompt candidates or enters a new keyword. The selected prompt candidate or new keyword is sent back to the system. The input data is the prompt candidate or new keyword selected by the user, and the output data is the data sent to the system.

[0134] Step 6:

[0135] The server sends a web request again to retrieve relevant information based on the user's selection or new keywords. The input data is the prompt candidates or new keywords selected by the user, and the output data is the retrieved HTML data.

[0136] Step 7:

[0137] The server analyzes the re-obtained HTML data and generates new prompt candidates. The input data is the re-obtained HTML data, and the output data is the newly generated prompt candidates.

[0138] Step 8:

[0139] The server presents the newly generated prompt candidates to the user. The input data is the newly generated prompt candidates, and the output data is the new prompt candidates that are presented to the user.

[0140] In this way, it is possible to generate appropriate prompt candidates for the keywords entered by the user and efficiently suggest related content.

[0141] Example 2

[0142] Next, a description will be given of Example 2 of Form 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."

[0143] Conventional dialogue systems have difficulty generating appropriate prompt candidates for questions or topics entered by users, which can cause the dialogue to not proceed smoothly. In addition, users often have difficulty deciding what question to ask next, which can disrupt the flow of the dialogue.

[0144] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for receiving input data from a user and generating prompt candidates using a generative AI model based on the input data, means for presenting the generated prompt candidates to the user, and means equipped with a dialogue agent for assisting the dialogue with the user. This allows the server to generate appropriate prompt candidates for the question or topic input by the user and present them to the user, thereby enabling the dialogue to proceed smoothly.

[0145] "User" means any person or entity that uses the System to provide input data and interact with the System.

[0146] "Input data" refers to information such as questions and topics that users provide to the system.

[0147] A "generative AI model" is an artificial intelligence algorithm for generating relevant prompt candidates based on user input data.

[0148] "Prompt candidates" are potential next questions or topics generated by a generative AI model and presented to the user.

[0149] A "server" is a computer system that runs a generative AI model, generates prompt candidates, and presents them to the user.

[0150] A "presentation means" is a method or device for displaying the generated prompt candidates to a user.

[0151] A "dialogue agent" is software that supports dialogue with a user and communicates with the user using generated prompt candidates.

[0152] MODE FOR CARRYING OUT THE INVENTION

[0153] This invention is a system that generates prompt candidates using a generative AI model based on data entered by a user and presents them to the user. Specific embodiments of this system are described below.

[0154] Hardware and software used

[0155] The system uses the following hardware and software:

[0156] Server: A computing system for running the generative AI model and generating prompt candidates. For example, a cloud-based server or a dedicated data center can be used.

[0157] Device: The device on which a user provides input data and receives generated prompt suggestions, such as a smartphone, tablet, or computer.

[0158] Generative AI model: An artificial intelligence algorithm for generating prompt candidates based on user input data. For example, a natural language generation model such as OpenAI's GPT-4 (registered trademark) is used.

[0159] Dialogue agent: Software that assists in dialogue with the user and communicates with the user using generated prompt suggestions.

[0160] Data processing and calculation

[0161] The server inputs the input data received from the user into the generative AI model, which then generates relevant prompt candidates based on the input data.The generated prompt candidates are sent from the server to the device and presented to the user.

[0162] Specific examples

[0163] For example, if a user types "What are some recommended travel destinations?", the server uses a generative AI model to generate the following possible prompts:

[0164] What do you recommend overseas?

[0165] What are some hidden gems in the country?

[0166] "What's the best place for a family trip?"

[0167] These prompt candidates are presented to the user via the device. The user can use these prompt candidates to formulate their next question. For example, if the prompt candidate "What are your recommendations for overseas travel?" is selected, the user can enter a more specific question such as "What are some recommended travel destinations in Europe?"

[0168] The system provides users with specific guidance to help them navigate the conversation smoothly.

[0169] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0170] Program processing flow

[0171] Step 1:

[0172] The user enters an initial question or theme.

[0173] Input: A user uses a device to provide input data (e.g., "What are some travel recommendations?").

[0174] Specific operation: The user enters a question into the input field on their smartphone or computer and presses the send button.

[0175] Step 2:

[0176] The server receives the user's input data.

[0177] Input: The input data submitted by the user.

[0178] Data processing: The server analyzes the input data and converts it into a format suitable for the generative AI model.

[0179] Output: Data format for input into a generative AI model.

[0180] Specific operation: The server analyzes the received data and performs preprocessing if necessary.

[0181] Step 3:

[0182] The server generates prompt candidates using a generative AI model.

[0183] Input: Preprocessed user input data.

[0184] Data computation: A generative AI model generates prompt candidates based on input data.

[0185] Output: Generated prompt candidates (e.g., "What are some recommended places abroad?", "What are some hidden gems in Japan?").

[0186] How it works: The server calls the generative AI model, which runs its internal algorithms to generate potential prompts.

[0187] Step 4:

[0188] The server sends the generated prompt candidates to the terminal.

[0189] Input: Generated prompt suggestions.

[0190] Data processing: Converting the prompt candidates into a format for presentation to the user.

[0191] Output: Prompt suggestions sent to the terminal.

[0192] Specific operation: The server converts the prompt candidates into an appropriate format and sends them to the terminal over the Internet.

[0193] Step 5:

[0194] The terminal presents the prompt candidates to the user.

[0195] Input: Prompt suggestions sent by the server.

[0196] Data processing: Display prompt candidates in a format that the user can view.

[0197] Output: The suggested prompts that are presented to the user.

[0198] Specific behavior: The device displays prompt candidates on the screen and allows the user to select one.

[0199] Step 6:

[0200] The user inputs the next question based on the suggested prompts.

[0201] Input: Suggested prompt suggestions.

[0202] Data processing: The user uses the suggested prompts to refine the next question.

[0203] Output: The next question (e.g., "What are the best travel destinations in Europe?").

[0204] Specific operation: The user refers to the presented prompt candidates, enters a new question, and presses the submit button.

[0205] In this way, the user can obtain specific guidance for smoothly progressing the dialogue.

[0206] (Application example 2)

[0207] Next, a description will be given of Application Example 2 of Form 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."

[0208] Conventional conversational agent systems provide insufficient guidance for users to find the content they want to watch, and have difficulty providing appropriate prompt candidates for users to ask specific questions. Furthermore, the generated prompt candidates often do not match the user's needs, resulting in a poor user experience.

[0209] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means for receiving input words from a user and generating prompt candidates via a network based on the input words; means for presenting the generated prompt candidates to the user; a system equipped with a dialogue agent that assists dialogue with the user; means for providing guidance to help the user find content they want to view; means for generating prompt candidates based on the user's input using a generative AI model; and means for presenting the generated prompt candidates to the user. This enables the user to efficiently find content they want to view.

[0210] "User" means an individual or entity that uses the System to provide input to locate content they wish to view.

[0211] "Input words" are keywords or phrases that a user provides to the system and contain information related to the content they wish to view.

[0212] "Network" means an infrastructure for transmitting and receiving data, including the Internet and other communications means.

[0213] "Prompt candidates" are phrases that are generated based on the user's input words and provide guidance for the user to ask specific questions.

[0214] A "conversational agent" is software designed to assist in a dialogue with a user, generating appropriate responses based on the user's input.

[0215] A "system" is a collection of hardware and software for receiving user input, generating prompt candidates, and assisting in user interaction.

[0216] A "Guide" is advice or suggestions provided to help a user find content they want to watch.

[0217] A "generative AI model" is an algorithm or program that uses artificial intelligence technology to generate prompt candidates based on user input.

[0218] A system for implementing this invention receives user input words, generates prompt candidates based on the input words via a network, and presents the generated prompt candidates to the user. The system is equipped with a dialogue agent that assists in dialogue with the user and provides guidance for the user to find content they want to view.

[0219] The server uses a generative AI model to generate prompt candidates based on the user's input. The generative AI model can use OpenAI's API. Specifically, when the user inputs the content they want to watch, the server receives that input and uses the generative AI model to generate appropriate prompt candidates. The generated prompt candidates are then presented to the user.

[0220] Users access the system using a device such as a smartphone and input the content they want to watch. For example, if a user inputs "I want to watch an action movie," the server uses a generative AI model to generate prompt candidates such as "What are some recent recommended action movies?" or "What are some classic action movies?" and presents them to the user.

[0221] The system uses a smartphone (Android®) as its hardware. It uses Python and the OpenAI API as its software. Data processing and calculation involves a series of processes: receiving user input, generating prompt candidates using a generative AI model, and presenting them to the user.

[0222] For example, if a user types "I want to watch an action movie," the generative AI model will generate prompt candidates such as "What are some recent recommended action movies?" or "What are some classic action movies?" By presenting these prompt candidates to users, they can ask more specific questions and efficiently find the content they want to watch.

[0223] An example of a prompt is as follows:

[0224] What action movies do you recommend?

[0225] What are some popular action movies these days?

[0226] What are some classic action movies?

[0227] In this way, users can efficiently find the content they want to watch.

[0228] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0229] Step 1:

[0230] A user accesses the system using a device such as a smartphone and inputs the input words for the content they want to watch. For example, they might input "I want to watch an action movie." This input word is sent to the system.

[0231] Step 2:

[0232] The server receives the input words sent by the user and sends a request to the generative AI model to generate prompt candidates based on the received input words. At this time, the input words are converted into a format that the generative AI model can understand.

[0233] Step 3:

[0234] The generative AI model generates prompt candidates based on the input words received from the server. For example, for the input words "I want to watch an action movie," it generates prompt candidates such as "What are some recent action movies that you recommend?" or "What are some classic action movies?" The generated prompt candidates are returned to the server.

[0235] Step 4:

[0236] The server receives the prompt candidates returned by the generative AI model and formats them for presentation to the user, who then sends the formatted prompt candidates to the user's device.

[0237] Step 5:

[0238] The user's device receives the prompt candidates sent from the server and presents them to the user, who can then ask a specific question based on the presented prompt candidates.

[0239] Step 6:

[0240] When the user enters a specific question based on the suggested prompts, the question is again sent to the server, which again uses the generative AI model to generate an appropriate answer to the user's question.

[0241] Step 7:

[0242] The generative AI model generates an appropriate answer based on the user's specific question and returns the answer to the server, which then sends the generated answer to the user's device.

[0243] Step 8:

[0244] The user's device receives the response sent from the server and presents it to the user, allowing the user to obtain specific information about the content they wish to view.

[0245] Example 3

[0246] Next, a description will be given of a third embodiment of the third embodiment. 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."

[0247] Conventional chatbot systems have difficulty efficiently generating prompt candidates to generate appropriate responses to user input and selecting the most appropriate one. Furthermore, they lack the means to smoothly advance dialogue with users, which has led to issues with the user experience not being improved.

[0248] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0249] In this invention, the server includes a means for receiving user input words and analyzing data based on the words, a means for sending prompts to the generative AI model based on the analysis results, a means for receiving multiple prompt candidates from the generative AI model, a means for selecting the optimal prompt from the received prompt candidates, a means for presenting the selected prompt candidates to the user, and a system equipped with a chatbot that supports dialogue with the user. This makes it possible to efficiently generate appropriate answers to user input and smoothly progress dialogue with the user.

[0250] "User" means an individual or entity that uses the System to ask questions or interact.

[0251] An "input word" is a word or phrase that represents a question or request that a user enters into the system.

[0252] "Means of analyzing data" refers to techniques and methods for analyzing the words entered by the user and understanding their intent and content.

[0253] A "generative AI model" is an artificial intelligence model that generates appropriate prompt candidates and answers based on user input.

[0254] A "prompt" is a form of instruction or question that is input to a generative AI model.

[0255] "Prompt candidates" are multiple possible answers or suggestions generated by a generative AI model.

[0256] The "means for selecting the best one" refers to the techniques or methods for selecting the most appropriate one from the generated prompt candidates.

[0257] "User presentation means" refers to the technology or method used to display selected prompt candidates and answers to the user.

[0258] A "chatbot" is an automated response system designed to assist in user interactions.

[0259] A "system" is a collection of devices and software that executes a series of processes to receive user input, analyze it, use a generative AI model to generate an appropriate answer, and present it to the user.

[0260] This invention is a system equipped with a chatbot that supports user interaction. The system executes a series of processes: it receives user input, analyzes data based on the input, generates appropriate prompt candidates using a generative AI model, selects the most suitable one, and presents it to the user.

[0261] Hardware and software used

[0262] This system uses the following hardware and software:

[0263] Server: A server with high-performance computing power is used to analyze user input data and send prompts to the generative AI model.

[0264] Terminal: The device used by the user to provide input (e.g., smartphone, PC, etc.).

[0265] Generative AI models: For example, use generative AI models (such as GPT-3) that excel at natural language processing.

[0266] Natural language processing libraries: Parse user input data.

[0267] Data processing and calculation

[0268] When the server receives the user's input words, it analyzes the data using a natural language processing library. Based on the analysis results, it sends a prompt to the generative AI model. The generative AI model generates multiple prompt candidates based on the received prompt. The server selects the most appropriate prompt from the generated candidates and presents it to the user.

[0269] Specific examples

[0270] As a concrete example, we will explain the processing when a user asks, "What are your recommendations for overseas destinations?"

[0271] 1. The user types "What are your recommendations for overseas destinations?" into their device.

[0272] 2. The terminal sends the input data to the server.

[0273] 3. The server analyzes the received input data and sends the following prompt to the generative AI model:

[0274] "A user asks, 'What are your recommendations abroad?' Generate an appropriate answer."

[0275] 4. The generative AI model generates multiple prompt candidates and returns them to the server. For example, it generates an answer such as, "How about Paris, France? You can enjoy visiting museums and gourmet food."

[0276] 5. The server selects the best prompt from the generated prompt candidates and presents it to the user.

[0277] 6. The device displays the received answer to the user.

[0278] In this way, the system can smoothly advance the dialogue with the user. The flow of the identification process in the third embodiment will be described with reference to FIG.

[0279] Step 1:

[0280] The user enters a question.

[0281] The user types "What do you recommend overseas?" into the chat box on the device. The input data is in text format.

[0282] Step 2:

[0283] The device sends input to the server.

[0284] The terminal converts the user's input data into JSON format and sends it to the server via the Internet. The input is {"question": "What do you recommend overseas?"}, and the output is the JSON data sent to the server.

[0285] Step 3:

[0286] The server parses the input.

[0287] The server parses the received JSON data and uses a natural language processing library to understand the intent of the user's question. The input is the user's question data, and the output is the parsed result.

[0288] Step 4:

[0289] The server sends prompts to the generative AI model.

[0290] The server sends a prompt to the generative AI model based on the analysis results. The prompt text is "The user asked, 'What overseas places do you recommend?' Please generate an appropriate answer." The input is the analysis results, and the output is the prompt text sent to the generative AI model.

[0291] Step 5:

[0292] A generative AI model generates candidate prompts.

[0293] The generative AI model generates multiple candidate answers based on the prompt it receives. For example, it generates an answer such as, "How about Paris, France? You can enjoy visiting museums and gourmet food." The input is the prompt sentence, and the output is multiple candidate prompts.

[0294] Step 6:

[0295] The server selects the best prompt candidate.

[0296] The server selects the most appropriate prompt candidate from the multiple generated prompt candidates. The selection criteria are the relevance to the user's question, the specificity of the information, etc. The input is the prompt candidate, and the output is the optimal prompt candidate.

[0297] Step 7:

[0298] The server sends the answer to the user.

[0299] The server converts the selected answers into JSON format and sends it to the user. The input is the best prompt candidates, and the output is the answer data sent to the user.

[0300] Step 8:

[0301] The terminal displays the answer to the user.

[0302] The device analyzes the received response data and displays it in a chat box. For example, it might say, "How about Paris, France? You can enjoy visiting museums and fine cuisine." The input is the response data, and the output is the response displayed to the user.

[0303] (Application example 3)

[0304] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[0305] Conventional food delivery services require users to search and select menu items themselves, making it difficult to receive optimal suggestions based on user preferences and requests. Furthermore, there is a lack of systems that can provide appropriate suggestions through dialogue with users, and there is a need to improve the user experience.

[0306] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means. In this invention, the server includes means for receiving a user's input words and generating prompt candidates based on the words via a communication network, means for presenting the generated prompt candidates to the user, a system equipped with an interactive agent that supports dialogue with the user, means for making optimal suggestions based on the user's preferences and requests, and means for displaying the generated suggestions to the user. This makes it possible to make optimal food delivery suggestions based on the user's input.

[0307] "User input words" refers to text or keywords that a user enters into the system.

[0308] "Communications Network" means the Internet or other network infrastructure for transmitting and receiving data.

[0309] "Prompt candidates" are response candidates that the system generates based on the user's input words and presents to the user.

[0310] A "conversational agent," also known as a chatbot, is a software program designed to assist users in interacting with the agent.

[0311] "User preferences and requests" refer to the individual preferences and requests that a user expresses to the system.

[0312] "Best Suggestion" refers to the most appropriate response or recommendation generated by the system based on the user's input words and preferences or requests.

[0313] A "generated suggestion" is a specific response or recommendation generated by a conversational agent based on user input.

[0314] A system for implementing this invention includes means for receiving a user's input words and generating prompt candidates based on the words via a communication network, means for presenting the generated prompt candidates to the user, a system equipped with an interactive agent that assists in dialogue with the user, means for making optimal suggestions based on the user's preferences and requests, and means for displaying the generated suggestions to the user.

[0315] Hardware and software used

[0316] Hardware: Smartphone, tablet, or PC

[0317] Software: Python, OpenAI API

[0318] Data processing and calculation

[0319] 1. Obtaining user input: A user uses a device such as a smartphone or PC to enter a question or request into the system. For example, they might enter, "What pizza do you recommend?"

[0320] 2. Prompt generation: The server generates an appropriate prompt based on the user's input. For example, it generates a prompt like, "If the user asks, 'What pizza do you recommend?', please provide an appropriate food delivery suggestion."

[0321] 3. Response generation using an AI model: The server uses OpenAI's GPT-3 model to generate a response based on the generated prompt, for example, "How about a Margherita pizza? It features fresh basil and mozzarella cheese."

[0322] 4. Displaying the response: The generated response is displayed on the user's device, allowing the user to place a food delivery order.

[0323] Specific examples

[0324] Prompt Sentence Examples

[0325] User input: "What pizza do you recommend?"

[0326] Generated prompt: "When a user asks, 'What pizza do you recommend?', provide an appropriate food delivery suggestion."

[0327] Specific response examples

[0328] User input: "What pizza do you recommend?"

[0329] The chatbot responds: "How about a Margherita pizza? Featuring fresh basil and mozzarella cheese."

[0330] In this way, the user experience can be improved by making optimal food delivery suggestions based on user input.

[0331] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[0332] Step 1:

[0333] A user uses a terminal to type a question or request to the system, for example, "What pizza do you recommend?" This input is sent to the system.

[0334] Input: The user's question or request (e.g., "What pizza do you recommend?")

[0335] Output: The user's input word sent to the system

[0336] Step 2:

[0337] The server receives the user's input words and generates a prompt based on the words. For example, it generates a prompt like, "If the user asks, 'What pizza do you recommend?', please suggest an appropriate food delivery option."

[0338] Input: User input (e.g., "What pizza do you recommend?")

[0339] Output: The generated prompt (e.g., "When the user asks 'What pizza do you recommend?', provide an appropriate food delivery suggestion.")

[0340] Step 3:

[0341] The server uses the generated prompt to send a request to OpenAI's GPT-3 model and generate a response, such as "How about a Margherita pizza? Featuring fresh basil and mozzarella cheese."

[0342] Input: Generated prompt text (e.g., "When the user asks 'What pizza do you recommend?', provide appropriate food delivery suggestions.")

[0343] Output: The response generated by the GPT-3 model (e.g., "How about a Margherita pizza? It features fresh basil and mozzarella cheese.")

[0344] Step 4:

[0345] The server generates a response and sends it to the user's terminal, which displays the response.

[0346] Input: The response generated by the GPT-3 model (e.g., "How about a Margherita pizza? It features fresh basil and mozzarella cheese.")

[0347] Output: The response displayed on the user's terminal

[0348] Step 5:

[0349] The user reviews the displayed response and places a food delivery order if desired, for example, ordering a suggested pizza.

[0350] Input: User confirmation and order (e.g. suggested pizza order)

[0351] Output: Food delivery order information

[0352] In this way, a system is realized that processes and calculates data based on user input at each step, ultimately providing the user with optimal suggestions.

[0353] 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.

[0354] "Example 1"

[0355] One embodiment of the present invention is a system that includes a means for receiving user input words and generating prompt candidates via the web based on the received words, a means for presenting the generated prompt candidates to the user, and a chatbot that supports dialogue with the user. This system also includes an emotion engine that recognizes the user's emotions. The emotion engine analyzes the user's input words to determine their emotions, and the chatbot uses the results to engage in dialogue with the user. For example, if a user inputs "Today is the worst day," the emotion engine determines from this input that the user is in a bad mood. Based on the analysis results, the chatbot then engages in appropriate dialogue with the user.

[0356] "Example 2"

[0357] In another embodiment of the present invention, an emotion engine tracks changes in a user's emotions, and the chatbot uses these changes to engage in dialogue with the user. For example, if a user first inputs, "Today is the worst day," and then inputs, "But it was fun meeting my friends," the emotion engine tracks the change in the user's emotions from disappointment to joy. Based on this change in emotion, the chatbot then engages in appropriate dialogue with the user.

[0358] "Example 3"

[0359] In yet another embodiment of the present invention, an emotion engine adjusts prompt candidates based on the user's emotions, and the chatbot uses the adjusted prompt candidates to engage in a dialogue with the user. For example, if a user inputs "Today is the worst day," the emotion engine analyzes this input to indicate that the user is in a bad mood. Based on the analysis result, the emotion engine adjusts the prompt candidates to generate more considerate prompt candidates, such as "That's been tough. Maybe talking about it would help." The chatbot then engages in an appropriate dialogue with the user based on the adjusted prompt candidates.

[0360] The processing flow of each embodiment will be described below.

[0361] "Example 1"

[0362] Step 1: Receive an input word from the user.

[0363] Step 2: Analyze the sentiment using the sentiment engine based on the received words.

[0364] Step 3: Generate prompt candidates based on the analysis results.

[0365] Step 4: Present the generated prompt candidates to the user.

[0366] Step 5: The chatbot interacts with the user.

[0367] "Example 2"

[0368] Step 1: Receive an input word from the user.

[0369] Step 2: Analyze the sentiment using the sentiment engine based on the received words.

[0370] Step 3: Track changes in user sentiment based on the analysis results.

[0371] Step 4: Generate prompt candidates based on changes in emotions.

[0372] Step 5: Present the generated prompt candidates to the user.

[0373] Step 6: The chatbot interacts with the user.

[0374] "Example 3"

[0375] Step 1: Receive an input word from the user.

[0376] Step 2: Analyze the sentiment using the sentiment engine based on the received words.

[0377] Step 3: Adjust the prompt candidates based on the analysis results.

[0378] Step 4: Present the tailored prompt candidates to the user.

[0379] Step 5: The chatbot interacts with the user.

[0380] Example 1

[0381] Next, a description will be given of Example 1 of Form 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."

[0382] Conventional chat systems have had issues with generating appropriate prompt candidates based on user input and not being able to take user emotions into account when conducting dialogue, which leads to a poor user experience and limits the system's usefulness.

[0383] 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.

[0384] In this invention, the server includes means for receiving user input words and generating prompt candidates via an information source based on the words, means for presenting the generated prompt candidates to the user, and means including a dialogue agent that is equipped with an emotion engine that analyzes the user's emotions and supports dialogue with the user based on the analysis results. This makes it possible to generate appropriate prompt candidates based on the user's input and to have a dialogue that takes the user's emotions into consideration.

[0385] "User input words" refers to words or phrases that users type into a chat app.

[0386] An "information source" is an external resource from which information can be obtained, such as an internet database or search engine.

[0387] "Prompt candidates" are questions or phrases that the user might type next, generated based on the words the user types.

[0388] An "emotion engine" is software or algorithm that analyzes emotions from user input words.

[0389] A "conversational agent" is a system equipped with artificial intelligence to assist in dialogue with users.

[0390] This invention is a system that, when a user inputs a specific word into a chat app, generates prompt candidates based on that word via information sources, and also analyzes the user's emotions to support the conversation.

[0391] Hardware and software used

[0392] The system uses the following hardware and software:

[0393] Server: Processes data and calls web search APIs.

[0394] Device: The device (smartphone, tablet, PC, etc.) through which a user operates the chat app.

[0395] Web Search API: An API for retrieving data from sources such as Google® Search API or Bing® Search API.

[0396] Emotion engine: Software for analyzing user emotions, such as IBM Watson® or Microsoft® Azure® sentiment analysis API.

[0397] Conversational Agent: A system equipped with artificial intelligence to assist in dialogue with users.

[0398] Program processing

[0399] When a user enters a specific word into a chat app, the device sends the word to a server. The server uses a web search API to retrieve related information based on the received word. It analyzes the retrieved information and generates prompt candidates such as "What are your recommended travel destinations?" or "What do I need to prepare for my trip?" The generated prompt candidates are presented to the user via the device.

[0400] Furthermore, when a user inputs a word containing an emotion, the device sends this word to the server. The server uses an emotion engine to analyze the user's emotion, and the dialogue agent generates an appropriate response based on the results. For example, if a user inputs "Today is the worst day," the emotion engine will analyze that the user is in a bad mood, and the dialogue agent will generate a response such as "Did something bad happen today?"

[0401] Specific examples

[0402] For example, if a user types "travel" into a chat app, the device sends this word to the server, which uses the Google Search API to retrieve information related to "travel" and generates prompt suggestions such as "What are your recommended travel destinations?" or "What do I need to prepare for my trip?" These suggestions are then displayed to the user via the device.

[0403] For example, if a user types "Today is the worst day," the device sends this phrase to the server. The server uses an emotion engine to analyze the user's emotions and recognizes that the user is in a bad mood. As a result, the conversational agent generates a response such as "Did something bad happen today?" and displays it to the user through the device.

[0404] In this way, the system can generate appropriate prompt candidates based on the user's input and further conduct dialogue that takes the user's emotions into account.

[0405] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0406] Step 1:

[0407] A user types a word into a chat app.

[0408] Input: A user types a word, such as "travel," into an input field in a chat app.

[0409] Output: The entered word is sent to the terminal.

[0410] Specific operation: The user types "travel" into a chat app on their smartphone or PC and presses the send button.

[0411] Step 2:

[0412] The terminal sends the entered word to the server.

[0413] Input: The word entered by the user (e.g., "travel").

[0414] Output: The entered word is sent to the server.

[0415] Specific operation: The device sends the entered word to the server using an HTTP request.

[0416] Step 3:

[0417] The server receives the word and retrieves related information using a web search API.

[0418] Input: A word sent from the device (e.g. "travel").

[0419] Output: Relevant information retrieved from the Web Search API.

[0420] Specific operation: The server calls the Google Search API or Bing Search API to obtain information related to "travel."

[0421] Step 4:

[0422] The server generates prompt candidates based on the information it has obtained.

[0423] Input: Relevant information obtained from the Web Search API.

[0424] Output: Generated prompt suggestions (e.g., "What travel destinations do you recommend?", "What do I need to prepare for my trip?").

[0425] What it does: The server analyzes the information it has retrieved and generates possible prompts for the user to enter next.

[0426] Step 5:

[0427] The server generates prompt candidates and sends them to the terminal.

[0428] Input: Generated prompt suggestions.

[0429] Output: Prompt candidates are sent to the terminal.

[0430] Specific operation: The server sends the generated prompt candidates to the device using an HTTP request.

[0431] Step 6:

[0432] The device displays possible prompts to the user.

[0433] Input: Prompt suggestions sent by the server.

[0434] Output: The suggested prompts are displayed to the user.

[0435] Specific operation: The device displays the prompt candidates received on the chat app screen.

[0436] Step 7:

[0437] The user enters a word that contains a sentiment.

[0438] Input: A user types a sentimental word, such as "Today is the worst day," into an input field in a chat app.

[0439] Output: The entered word is sent to the terminal.

[0440] Specific action: A user types "Today is the worst day" into a chat app on their smartphone or PC and presses the send button.

[0441] Step 8:

[0442] The terminal sends words containing emotions to the server.

[0443] Input: A user-entered sentiment word (e.g., "Today is the worst day").

[0444] Output: The entered word is sent to the server.

[0445] Specific operation: The device sends the entered word to the server using an HTTP request.

[0446] Step 9:

[0447] The server analyzes the emotions using an emotion engine.

[0448] Input: A word containing an emotion sent from the device (e.g., "Today is the worst day").

[0449] Output: Parsed emotion information (e.g., displeased).

[0450] Specific operation: The server calls an emotion engine (e.g., emotion analysis API) and analyzes the user's emotions.

[0451] Step 10:

[0452] The server generates the chatbot's response based on the analysis results.

[0453] Input: Parsed emotion information.

[0454] Output: The generated chatbot response (e.g., "Did something bad happen today?").

[0455] Specific operation: The server generates an appropriate response to the user based on the analysis results.

[0456] Step 11:

[0457] The server generates a response and sends it to the terminal.

[0458] Input: The generated chatbot response.

[0459] Output: The response is sent to the terminal.

[0460] Specific operation: The server uses an HTTP request and sends the generated response to the terminal.

[0461] Step 12:

[0462] The terminal displays the response to the user.

[0463] Input: The chatbot response sent from the server.

[0464] Output: The response is displayed to the user.

[0465] Specific operation: The response received by the device is displayed on the chat app screen.

[0466] (Application example 1)

[0467] Next, a description will be given of Application Example 1 of Embodiment 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."

[0468] Conventional chatbot systems have had difficulty generating appropriate prompt candidates in response to user input, making it difficult to conduct dialogue that takes the user's emotions into account. Furthermore, online shopping sites are required to quickly and accurately recommend products that users want, but conventional systems have not been able to fully achieve this.

[0469] 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.

[0470] In this invention, the server includes a means for receiving user input words and generating prompt candidates via the web based on the words; a means for presenting the generated prompt candidates to the user; a means for incorporating an emotion engine that analyzes the user's emotions and a chatbot that interacts with the user based on the analysis results; and a means for suggesting related products based on the user's input words. This allows for the generation of appropriate prompt candidates in response to the user's input and for a dialogue that takes the user's emotions into consideration. It also allows for the prompting of products desired by the user in a timely and accurate manner.

[0471] "User input words" refers to specific words or phrases that a user inputs into the system.

[0472] "Prompt candidates" are questions or suggestions to present to the user that are generated based on the user's input words.

[0473] An "emotion engine" is software or hardware that analyzes emotions from user input words or text.

[0474] A "chatbot" is a program or system that automatically interacts with users.

[0475] "Product suggestion" refers to the act of recommending related products based on the user's input words.

[0476] The system for implementing this invention has the function of receiving user input words, generating prompt candidates via the web based on the words, and presenting them to the user. It also includes an emotion engine that analyzes the user's emotions and a chatbot that interacts with the user based on the analysis results. It also has the function of suggesting related products based on the user's input words.

[0477] Hardware and software used

[0478] Hardware: Smartphone

[0479] Software: Python

[0480] Data processing and calculation

[0481] Prompt Generation

[0482] The server receives the keywords entered by the user, calls a Web API based on the keywords, and generates relevant prompt candidates, which are then presented to the user.

[0483] Emotion analysis

[0484] The server sends the user's input text to the emotion engine API for emotion analysis. The emotion engine extracts emotions from the user's text and returns the results to the server.

[0485] Dialogue Support

[0486] The server then responds to the user with appropriate dialogue based on the results of the emotion analysis. For example, if a user enters "Today is the worst day," the emotion engine will interpret this as meaning that the user is in a bad mood and respond with appropriate dialogue such as "Did anything special happen today?"

[0487] Product suggestion

[0488] The server suggests related products based on the user's input. For example, if a user inputs "I want a new smartphone," the server generates prompt candidates such as "What brand of smartphone do you like?" and "What is your budget?" and presents them to the user.

[0489] Specific examples

[0490] If a user types "I want a new smartphone," the server generates the following prompt candidates and presents them to the user:

[0491] "Which brand of smartphone do you prefer?"

[0492] "What's your budget?"

[0493] "Which features are important?"

[0494] Also, if a user types, "Today is the worst day," the emotion engine will analyze that the user is in a bad mood, and the server will engage in appropriate dialogue such as, "Did anything special happen today?"

[0495] In this way, the system can generate appropriate prompt candidates based on user input, conduct dialogue that takes into account the user's emotions, and quickly and accurately suggest products that the user wants.

[0496] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0497] Step 1:

[0498] The user inputs a keyword into the terminal.

[0499] Input: The keyword entered by the user (e.g., "I want a new smartphone")

[0500] Output: The device sends the keyword to the server.

[0501] Step 2:

[0502] The server calls a Web API based on the received keywords and generates relevant prompt candidates.

[0503] Input: Keyword sent from the terminal

[0504] Data processing: The server sends keywords to the Web API to obtain related prompt candidates.

[0505] Output: Generated prompt candidates (e.g., "What brand of smartphone do you like?", "What is your budget?")

[0506] Step 3:

[0507] The server transmits the generated prompt candidates to the terminal and presents them to the user.

[0508] Input: Generated prompt suggestions

[0509] Output: The terminal displays the possible prompts to the user.

[0510] Step 4:

[0511] The user inputs text containing emotions into the terminal.

[0512] Input: User-entered text containing emotions (e.g., "Today is the worst day ever").

[0513] Output: The terminal sends text to the server.

[0514] Step 5:

[0515] The server sends the received text to the emotion engine API for emotion analysis.

[0516] Input: Text sent from the terminal

[0517] Data processing: The server sends the text to the emotion engine API and obtains the emotion analysis results.

[0518] Output: Sentiment analysis result (e.g., displeased)

[0519] Step 6:

[0520] The server then engages in appropriate dialogue with the user based on the emotion analysis results.

[0521] Input: Sentiment analysis results

[0522] Data calculation: The server generates appropriate dialogue content based on the results of emotion analysis.

[0523] Output: Generated dialogue (e.g. "Is there anything special happening today?")

[0524] Step 7:

[0525] The server transmits the generated dialogue content to the terminal and presents it to the user.

[0526] Input: Generated dialogue

[0527] Output: The terminal displays the dialogue to the user.

[0528] Step 8:

[0529] The server makes relevant product suggestions based on the user's input words.

[0530] Input: User input word

[0531] Data calculation: The server searches for relevant products based on the input words and generates suggestions.

[0532] Output: Generated product suggestions (e.g., "How about this brand of smartphone?")

[0533] Step 9:

[0534] The server transmits the generated product proposal to the terminal and presents it to the user.

[0535] Input: Generated product suggestions

[0536] Output: The terminal displays the product suggestions to the user.

[0537] Example 2

[0538] Next, a description will be given of Example 2 of Form 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."

[0539] Conventional chatbot systems have difficulty generating appropriate prompt candidates based on user input, which can prevent smooth dialogue with the user. Furthermore, because dialogue is conducted without taking into account changes in the user's emotions, user satisfaction can decline. To solve these problems, a system is needed that can generate appropriate prompt candidates based on the user's input data and track changes in the user's emotions as it engages in dialogue.

[0540] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0541] In this invention, the server includes a means for receiving user input data and generating prompt candidates using a generative AI model based on the data, a means for presenting the generated prompt candidates to the user, a means for analyzing the user's emotions using an emotion engine that tracks changes in the user's emotions, a means for generating a dialogue with the user based on the analyzed changes in emotions, a means for presenting the generated dialogue to the user, and a system equipped with a chatbot that supports the dialogue with the user. This makes it possible to generate appropriate prompt candidates based on the user's input data and to conduct a dialogue that takes into account changes in the user's emotions.

[0542] "User input data" refers to information such as text or voice that a user inputs into the system through a terminal.

[0543] A "generative AI model" is an artificial intelligence model that generates prompt candidates and dialogues based on user input data.

[0544] A "prompt candidate" is a suggestion that indicates what question or interaction the user should have next.

[0545] An "emotion engine" is software or algorithm for analyzing emotional changes from user input data.

[0546] A "chatbot" is an automated response system that assists in conversations with users.

[0547] A "server" is a computer system that receives user input data and generates prompt candidates and dialogue using generative AI models and emotion engines.

[0548] A "terminal" is a device through which a user sends input data to the system and receives generated prompt suggestions and interactions.

[0549] MODE FOR CARRYING OUT THE INVENTION

[0550] The present invention provides a system for generating appropriate prompt candidates based on user input data, and for conducting a dialogue by tracking changes in the user's emotions. Specific embodiments of this system are described below.

[0551] server

[0552] The server receives the user's input data and generates prompt candidates based on the data using a generative AI model. A typical natural language processing model is used as the generative AI model. The server then sends the generated prompt candidates to the device. The server also uses an emotion engine to track changes in the user's emotions and generates a dialogue with the user based on the analyzed changes in emotions. A typical emotion analysis software is used as the emotion engine.

[0553] Terminal

[0554] The terminal receives input data from the user and sends it to the server. The text data and voice data entered by the user are sent to the server via the Internet. The terminal presents the prompt candidates received from the server and the generated dialogue to the user. This allows the user to check the prompt candidates and dialogue displayed on the screen and make the next input.

[0555] user

[0556] Users interact with the chatbot through their devices. For example, if a user inputs, "Today is the worst day," the information is sent to the server via the device. The server uses an emotion engine to analyze this input data and track down that the user's emotion is bad. If the user then inputs, "But it was fun meeting your friends," the server again analyzes it using its emotion engine and tracks that the user's emotion has changed to joy. Based on this information, the server generates an appropriate dialogue such as, "It was great meeting your friends. What fun things are you planning next?" and presents it to the user via the device.

[0557] Specific examples

[0558] Below are some specific examples of prompt sentences to input to the generative AI model.

[0559] "What travel destinations do you recommend?"

[0560] "What recent movies do you recommend?"

[0561] "Teach me a new recipe?"

[0562] These prompts are suggestions that indicate what questions or interactions the user should have next, providing guidance to help the user smoothly proceed with the interaction.

[0563] In this way, the present invention generates appropriate prompt candidates based on the user's input data, enabling a dialogue that takes into account the user's emotional changes, thereby improving user satisfaction and realizing a more natural and effective dialogue.

[0564] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0565] Program processing flow

[0566] Step 1: User types into terminal

[0567] The user uses the terminal to input text data, for example, "Today is the worst day ever." This input data is then sent to the server in the next step.

[0568] Step 2: The device sends the input data to the server

[0569] The terminal transmits the text data entered by the user to the server via the Internet. The data is encrypted before transmission. The input data is the text "Today is the worst day."

[0570] Step 3: The server generates prompt candidates using the generative AI model

[0571] The server uses a generative AI model to generate prompt candidates based on the received user input data. For example, a natural language processing model is used as the generative AI model to generate a prompt candidate such as "What travel destinations do you recommend?" The input data is "Today is the worst day," and the output data is "What travel destinations do you recommend?"

[0572] Step 4: The server sends the generated prompt candidates to the device

[0573] The server sends the generated prompt candidates to the device. At this time, the data is encrypted again before being sent. The input data is "What travel destinations do you recommend?" and the output data is the prompt candidates sent to the device.

[0574] Step 5: The device presents prompt candidates to the user

[0575] The terminal displays the received prompt candidates to the user. The user can check the prompt candidates displayed on the screen. The input data is "What travel destinations do you recommend?" and the output data is the prompt candidates presented to the user.

[0576] Step 6: The user continues the dialogue based on the suggested prompts

[0577] The user can then ask more specific questions or engage in more specific dialogue based on the suggested prompts. For example, the user can type, "What do you recommend for overseas travel?" The input data is "What do you recommend for overseas travel?", and the output data is sent to the server in the next step.

[0578] Step 7: The server uses the emotion engine to track changes in the user's emotions.

[0579] The server uses an emotion engine to analyze changes in emotions from user input data. For example, from inputs such as "Today is the worst day" and "But it was fun meeting my friends," the server tracks changes in the user's emotions from displeasure to joy. The input data is "Today is the worst day" and "But it was fun meeting my friends," and the output data is the analysis result of the change in emotions.

[0580] Step 8: The server generates appropriate dialogue based on the change in emotion.

[0581] The server generates an appropriate dialogue for the user based on the analysis results of the emotion engine. For example, it generates a dialogue such as, "It was nice to meet your friends. What fun things are you planning to do next?" The input data is the analysis results of changes in emotions, and the output data is the generated dialogue.

[0582] Step 9: The server sends the generated dialogue to the device

[0583] The server transmits the generated dialogue to the terminal. At this time, the data is encrypted before transmission. The input data is the generated dialogue, and the output data is the dialogue transmitted to the terminal.

[0584] Step 10: The terminal presents the generated dialogue to the user

[0585] The terminal displays the received dialogue to the user. The user can confirm the dialogue displayed on the screen and continue the dialogue. The input data is the generated dialogue, and the output data is the dialogue presented to the user.

[0586] (Application example 2)

[0587] Next, a description will be given of Application Example 2 of Form 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."

[0588] Conventional chatbot systems have had the problem of being unable to present appropriate prompt candidates in response to user input, preventing users from smoothly progressing through the dialogue. Furthermore, because the dialogue is conducted without taking into account changes in the user's emotions, there is a risk of user satisfaction decreasing. There is a need to provide a system that can solve these problems and allow users to progress through dialogue more comfortably.

[0589] 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.

[0590] In this invention, the server includes means for receiving a user's input words and generating prompt candidates using a generative AI model based on the words, means for presenting the generated prompt candidates to the user, means for analyzing the user's emotional state using an emotion engine that tracks changes in the user's emotions, and means equipped with a chatbot for conducting appropriate dialogue based on the user's emotional state. This makes it possible to present appropriate prompt candidates in response to the user's input and to conduct dialogue according to the user's emotional state.

[0591] "User input words" are text or keywords that a user inputs into the system.

[0592] A "generative AI model" is an algorithm or system that uses artificial intelligence to generate appropriate prompt candidates based on user input.

[0593] "Prompt suggestions" are text that predicts the next question or instruction the user might enter.

[0594] An "emotion engine" is software or algorithms that analyze a user's emotional state from text input and track changes in that state.

[0595] "Emotional state" is information that indicates the user's current emotions and mood.

[0596] A "chatbot" is a program or system that automatically interacts with users.

[0597] "Engaging in dialogue" means generating appropriate responses to user input and maintaining communication with the user.

[0598] "Response to inquiry" refers to information or solutions provided in response to a question or request made by a user to the system.

[0599] A system for implementing this invention includes means for receiving a user's input words and generating prompt candidates using a generative AI model based on the words, means for presenting the generated prompt candidates to the user, means for analyzing the user's emotional state using an emotion engine that tracks changes in the user's emotions, and means equipped with a chatbot that engages in appropriate dialogue based on the user's emotional state.

[0600] System configuration

[0601] 1. A way to receive user input

[0602] A user inputs text into the system using a device such as a smartphone or computer.

[0603] This input word is received as an initial process for the system.

[0604] 2. A method for generating prompt candidates using a generative AI model

[0605] The server sends the received user input words to the generative AI model.

[0606] A generative AI model (e.g., GPT-3) generates possible prompts for the user to type next based on the input words.

[0607] 3. A means of presenting the generated prompt candidates to the user

[0608] The server transmits the generated prompt candidates to the user's terminal and presents them to the user.

[0609] This gives the user guidance on what question or instruction to enter next.

[0610] 4. A means of analyzing the user's emotional state using an emotion engine

[0611] The server sends the user's input words to the emotion engine to analyze the emotional state.

[0612] The emotion engine calculates a user's emotion score from the input text and tracks its changes.

[0613] 5. A chatbot-based approach that provides appropriate dialogue based on the user's emotional state

[0614] The server generates an appropriate dialogue according to the emotional state of the user based on the emotion score obtained from the emotion engine.

[0615] The chatbot interacts with the user based on the generated prompt candidates and emotional state.

[0616] Hardware and software used

[0617] Hardware: smartphones, computers, servers

[0618] Software: OpenAI API (generative AI model), emotion engine

[0619] Specific examples

[0620] For example, if a user inputs "Today is the worst day ever," the emotion engine calculates a low emotion score from this input. The generative AI model generates "What products do you recommend?" as a possible next prompt for the user. The server then displays to the user, "You seem to be feeling unwell. Is there anything I can help you with?" and suggests "What products do you recommend?" as the next possible prompt.

[0621] Prompt Sentence Examples

[0622] "What travel destinations do you recommend?"

[0623] "I'm looking for new fashion items"

[0624] "What are the most popular items these days?"

[0625] In this way, it is possible to present appropriate prompt candidates in response to the user's input and to conduct a dialogue that is appropriate for the user's emotional state.

[0626] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0627] Step 1:

[0628] A user uses a terminal to input text or keywords into the system.

[0629] Input: User input word

[0630] Output: The user's input word is sent to the server

[0631] Specific action: A user enters text into an input field on a smartphone or computer and presses the send button.

[0632] Step 2:

[0633] The server receives the user's input words and sends them to the generative AI model.

[0634] Input: User input word

[0635] Output: The input words sent to the generative AI model

[0636] Specific operation: The server sends the user's input words as an API request to the generative AI model.

[0637] Step 3:

[0638] A generative AI model generates prompt candidates based on the user's input words.

[0639] Input: The input words sent to the generative AI model

[0640] Output: Generated prompt candidates

[0641] How it works: A generative AI model (e.g., GPT-3) analyzes the input words and generates possible prompts the user might type next.

[0642] Step 4:

[0643] The server transmits the generated prompt candidates to the user's terminal and presents them to the user.

[0644] Input: Generated prompt suggestions

[0645] Output: Prompt candidates displayed on the user's terminal

[0646] Specific operation: The server sends the generated prompt candidates to the user's device and displays them on the screen.

[0647] Step 5:

[0648] The server sends the user's input words to the emotion engine to analyze the emotional state.

[0649] Input: User input word

[0650] Output: Sentiment score calculated by the sentiment engine

[0651] Specific operation: The server sends the user's input words to the emotion engine and calculates the emotion score.

[0652] Step 6:

[0653] The server generates an appropriate dialogue according to the emotional state of the user based on the emotion score obtained from the emotion engine.

[0654] Input: Sentiment score calculated by the sentiment engine

[0655] Output: Dialogue content according to the user's emotional state

[0656] Specific operation: The server analyzes the emotion score and generates dialogue content according to the user's emotional state.

[0657] Step 7:

[0658] The chatbot interacts with the user based on the generated prompt candidates and emotional state.

[0659] Input: Generated prompt candidates, dialogue content according to the user's emotional state

[0660] Output: User interaction

[0661] Specific behavior: The chatbot responds appropriately to the user based on the generated prompt candidates and emotional state.

[0662] Example 3

[0663] Next, a description will be given of a third embodiment of the third embodiment. 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."

[0664] Conventional chatbot systems have difficulty generating appropriate prompt candidates in response to user input, and are particularly unable to generate responses that take the user's emotions into account. This results in unnatural interactions with users and a poor user experience.

[0665] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0666] In this invention, the server includes means for receiving user input words and generating prompt candidates using a generative AI model based on the words, means for presenting the generated prompt candidates to the user, means for analyzing the user's emotions and adjusting the prompt candidates based on the analysis results, and means for incorporating a chatbot that interacts with the user using the adjusted prompt candidates. This makes it possible to generate appropriate prompt candidates based on the user's input and provide responses that take the user's emotions into consideration.

[0667] "User input words" refers to the text or message a user sends to a chatbot.

[0668] A "generative AI model" is an artificial intelligence model that uses natural language processing technology to generate appropriate prompt candidates and responses based on user input.

[0669] "Prompt candidates" are potential responses or dialogues that a generative AI model suggests in response to user input.

[0670] "Emotion analysis" refers to the technology of analyzing and identifying a user's emotional state from their input.

[0671] A "chatbot" is a program that automatically interacts with users and has the ability to generate responses using generative AI models and the results of sentiment analysis.

[0672] A "server" is a computer system that receives user input and runs generative AI models and sentiment analysis.

[0673] This invention relates to a system equipped with a chatbot that supports user interactions. The system generates prompt candidates using a generative AI model based on the user's input words, and further analyzes the user's emotions to adjust the prompt candidates to provide more appropriate responses.

[0674] When the server receives the user's input, it first analyzes the input. This analysis uses natural language processing (NLP) technology. Specifically, it uses a common natural language processing model such as GPT-3 as a generative AI model. Based on the analysis results, it generates appropriate prompt candidates.

[0675] For example, if a user types "What do you recommend overseas?", the server will parse this input and input the following prompt sentence into the generative AI model:

[0676] User input: "What are your recommendations for places abroad?"

[0677] Prompt for generative AI model: "A user is asking for international travel recommendations. Please generate an appropriate answer."

[0678] Based on this prompt, the generative AI model generates an answer such as, "How about Paris, France? You can enjoy visiting museums and fine cuisine."

[0679] Furthermore, if an emotion engine is installed, the server analyzes the user's emotions. This analysis is performed using a sentiment analysis algorithm. For example, if a user inputs "Today is the worst day," the emotion engine will determine from this input that the user is in a bad mood. It will then adjust the prompt candidates based on the analysis results. The adjusted prompt candidates might be something like, "That's been tough. Maybe talking about it would help."

[0680] In this way, the server can generate appropriate prompt candidates based on the user's input and provide responses that take the user's emotions into account, making the interaction more natural and effective, and improving the user experience.

[0681] For example, here's what happens when a user asks "What are your recommendations for overseas destinations?":

[0682] 1. The user types "What are your recommendations for overseas?" into their device and presses the send button.

[0683] 2. The server receives this input and analyzes it using NLP techniques.

[0684] 3. Based on the analysis results, the server inputs a prompt statement to the generative AI model: "The user is asking for recommendations for overseas travel. Please generate an appropriate answer."

[0685] 4. The generative AI model generates the answer, "How about Paris, France? You can enjoy visiting museums and fine cuisine."

[0686] 5. The server receives this answer and provides it to the user through the chatbot.

[0687] 6. The user checks this answer on their device.

[0688] This system makes it possible to generate appropriate prompt candidates based on the user's input and provide a response that takes the user's emotions into consideration. The flow of the specification process in the third embodiment will be described with reference to FIG.

[0689] Step 1:

[0690] The user types at the terminal.

[0691] The user uses the device to send a message to the chatbot, for example, by typing "What do you recommend for overseas travel?" and pressing the send button.

[0692] Input: User's message (e.g. "What do you recommend overseas?")

[0693] Output: Message sent from terminal to server

[0694] Step 2:

[0695] The server receives the user's input.

[0696] The server receives input from the user and prepares to parse it.

[0697] Input: User's message sent from the terminal

[0698] Output: Received message

[0699] Step 3:

[0700] The server parses the input.

[0701] The server uses natural language processing (NLP) techniques to analyze the user's input, specifically using a generative AI model (e.g., GPT-3) to tokenize the input sentence and analyze it to understand its context.

[0702] Input: Received message

[0703] Output: Analysis results (e.g., information that the user is asking for recommendations for overseas travel)

[0704] Step 4:

[0705] The server generates the prompt candidates.

[0706] Based on the analysis results, the server inputs a prompt sentence into the generative AI model. For example, the prompt sentence "The user is asking for recommendations for overseas travel. Please generate an appropriate answer." is input into the generative AI model. The generative AI model generates an appropriate answer based on this prompt sentence.

[0707] Input: Analysis results

[0708] Output: Generated prompt candidates (e.g., "How about Paris, France? You can enjoy visiting museums and fine cuisine.")

[0709] Step 5:

[0710] Server performs sentiment analysis (optional).

[0711] If the user's input contains an emotion, the server uses an emotion engine to analyze the emotion. For example, if the user inputs "Today is the worst day," the emotion engine analyzes that the user is in a bad mood.

[0712] Input: Received message

[0713] Output: Sentiment analysis results (e.g., information that the user is in a bad mood)

[0714] Step 6:

[0715] The server optionally adjusts the prompt suggestions.

[0716] Based on the results of the sentiment analysis, the server adjusts the prompt candidates. For example, if the user is in a bad mood, it generates a thoughtful prompt such as, "That must be tough. Maybe talking about it would help."

[0717] Input: Sentiment analysis results

[0718] Output: Tailored prompt suggestions

[0719] Step 7:

[0720] The server generates a response through the chatbot.

[0721] The server receives the answer from the generative AI model and provides it to the user through the chatbot.

[0722] Input: Generated prompt suggestions

[0723] Output: Chatbot response (e.g., "How about Paris, France? You can enjoy visiting museums and delicious food.")

[0724] Step 8:

[0725] The server sends a response to the user.

[0726] The server generates a response and sends it to the user's terminal, where the user can view the response.

[0727] Input: Chatbot response

[0728] Output: The response displayed on the user's terminal

[0729] (Application example 3)

[0730] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[0731] Conventional chatbot systems generate uniform responses without considering the user's emotions, which means they are unable to respond appropriately to the user's feelings. Furthermore, in security services, more considerate responses are required when the user is feeling anxious or scared, but conventional systems have difficulty responding in this way.

[0732] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means. In this invention, the server includes means for receiving words input by a user and generating prompt candidates based on the words via a communication network, means for presenting the generated prompt candidates to the user, a system equipped with a dialogue agent that supports dialogue with the user, emotion analysis means for analyzing the user's emotions, means for adjusting the prompt candidates based on the emotion analysis results, and means for engaging in dialogue with the user using the adjusted prompt candidates. This makes it possible to generate appropriate prompts according to the user's emotions, and to provide security services that take into account the user's anxieties and fears.

[0733] "User input words" are text or keywords that a user inputs into the system.

[0734] A "communications network" is an infrastructure for transmitting and receiving data, including the Internet and other digital networks.

[0735] A "prompt candidate" is a suggested text or question that is generated based on the user's input and presented as the next step in the dialogue.

[0736] A "conversational agent," also known as a chatbot, is a software program designed to assist users in interacting with the agent.

[0737] "Emotion analysis means" refers to technology or algorithms that analyze emotions from the user's input text and determine the type and intensity of those emotions.

[0738] "Emotion analysis results" are data and information about the user's emotions obtained by the emotion analysis means.

[0739] "Adjustment means" refers to technologies and algorithms for appropriately changing or modifying prompt candidates based on the results of sentiment analysis.

[0740] A "means for conducting dialogue" is a technology or algorithm for conducting dialogue with a user using the generated prompt candidates.

[0741] A system for implementing this invention includes means for receiving input words from a user and generating prompt candidates based on the input words via a communication network, means for presenting the generated prompt candidates to the user, a system equipped with a dialogue agent that assists in dialogue with the user, emotion analysis means for analyzing the user's emotions, means for adjusting the prompt candidates based on the emotion analysis results, and means for engaging in dialogue with the user using the adjusted prompt candidates.

[0742] Hardware and software used

[0743] Hardware: smartphones, robots

[0744] Software: Python, OpenAI API

[0745] Data processing and calculation

[0746] Receiving user input words

[0747] A user inputs text into an application installed on a smartphone or robot. For example, the user might input, "I've seen some suspicious people around my house recently."

[0748] Generate prompt candidates

[0749] The server receives the user's input and uses the OpenAI API to generate prompt candidates based on the input. The generated prompt candidates might include, for example, "That's worrying. We recommend you contact the police. Also, check the security cameras in your neighborhood."

[0750] Emotion analysis

[0751] The server analyzes the sentiment of the user's input text, calculating the sentiment polarity (ranging from -1 to 1) and determining the user's emotional state. For example, if the user's input is "I've seen some suspicious people around my house recently," the sentiment analysis result indicates a negative sentiment.

[0752] Adjust prompt suggestions

[0753] Based on the sentiment analysis, the server adjusts the suggested prompts. If the sentiment is negative, it generates a more considerate prompt, such as, "That's worrying. I suggest you contact the police. Also, check the security cameras in your neighborhood."

[0754] User interaction

[0755] Using the tailored prompt candidates, the conversational agent engages in a dialogue with the user, providing appropriate responses to the user's input and taking into consideration the user's anxieties and fears.

[0756] Specific examples

[0757] User input: "I've seen some suspicious people around my house recently."

[0758] Generated prompt: "That's concerning. I suggest you contact the police. Also, check to see if your neighborhood has security cameras."

[0759] The chatbot responds: "That's concerning. I suggest you contact the police. Also, check to see if your neighborhood has security cameras."

[0760] In this way, appropriate prompts can be generated based on the user's emotions, enabling security services to respond in a way that takes into account the user's anxieties and fears.

[0761] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[0762] Step 1:

[0763] The user inputs text into the terminal, for example, "I've seen some suspicious people around my house recently." This input text becomes the input data for the next processing step.

[0764] Step 2:

[0765] The terminal sends the user's input text to the server, which receives the input text and begins processing it as data to generate prompt candidates.

[0766] Step 3:

[0767] The server uses the OpenAI API to generate prompt candidates based on the received input text. The input text is sent to the API, and the generated prompt candidates are received. For example, a prompt candidate might be, "That's worrying. I recommend you contact the police. Also, check the security cameras in your neighborhood."

[0768] Step 4:

[0769] The server analyzes the sentiment of the user's input text and calculates the polarity of the sentiment (ranging from -1 to 1). For example, if the input text is "I've seen some suspicious people around my house recently," the sentiment analysis result will indicate a negative sentiment.

[0770] Step 5:

[0771] The server then adjusts the prompt candidates based on the results of the sentiment analysis. If the sentiment is negative, it generates a more considerate prompt, such as, "That's worrying. I suggest you contact the police. Also, check the security cameras in your neighborhood."

[0772] Step 6:

[0773] The server sends the adjusted prompt candidates to the terminal, which receives the prompt candidates and presents them to the user.

[0774] Step 7:

[0775] The user reviews the suggested prompts and enters further input if necessary. For example, the user might enter, "I've contacted the police, but is there anything else I can do?" This input returns to step 1 and the process repeats.

[0776] In this way, appropriate prompts can be generated according to the user's emotions, enabling security services to respond in a way that takes into account the user's anxiety and fear.

[0777] 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.

[0778] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ) 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.

[0779] Another example of generative AI is Gemini (registered trademark) (Internet search engine). <url: https: gemini.google.com ?hl="ja">) are listed.

[0780] 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.

[0781] [Second embodiment]

[0782] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0783] 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.

[0784] 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).

[0785] 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.

[0786] 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.

[0787] 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).

[0788] 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.

[0789] 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.

[0790] 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.

[0791] 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.

[0792] 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.

[0793] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0794] "Example 1"

[0795] In one embodiment of the present invention, when a user inputs a specific word into a chat app, the system generates prompt candidates via the web based on the input word. For example, if a user inputs the word "travel," the system generates prompt candidates such as "What are your recommended travel destinations?" and "What do I need to prepare for my trip?"

[0796] "Example 2"

[0797] The generated prompt candidates are presented to the user, allowing the user to receive guidance on how to proceed with the conversation. For example, if a prompt candidate such as "What travel destinations do you recommend?" is presented, the user can respond with a more specific question such as "What overseas places do you recommend?"

[0798] "Example 3"

[0799] Furthermore, the system of the present invention is equipped with a chatbot that supports dialogue with the user. This chatbot converses with the user using prompt candidates generated based on the user's input words. For example, if the user asks, "What overseas places do you recommend?", the chatbot will generate a response such as, "How about Paris, France? You can enjoy visiting museums and gourmet food there."

[0800] The processing flow of each embodiment will be described below.

[0801] "Example 1"

[0802] Step 1: A user enters a specific word in a chat app, for example, the word "travel."

[0803] Step 2: The system generates prompt candidates via the web based on the relevant words. In this case, it generates prompt candidates such as "What travel destinations do you recommend?" and "What do I need to prepare for my trip?"

[0804] Step 3: The generated prompt candidates are presented to the user, providing guidance on how to proceed with the interaction.

[0805] "Example 2"

[0806] Step 1: The user asks a specific question in response to the suggested prompt. For example, in response to the prompt "What travel destinations do you recommend?", the user can ask "What overseas destinations do you recommend?"

[0807] Step 2: The chatbot generates an answer to the user's question, in this case, "How about Paris, France? You can enjoy visiting museums and delicious food."

[0808] "Example 3"

[0809] Step 1: The user asks the chatbot a follow-up question, such as, "What are the best museums to visit in Paris?"

[0810] Step 2: The chatbot generates an answer to the user's question, in this case "The Louvre and the Orsay Museum are must-sees. Also, there are many art studios in Montmartre around the Sacré-Coeur Basilica."

[0811] Example 1

[0812] Next, a description will be given of Example 1 of Form 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."

[0813] Conventional chatbot systems have had difficulty generating appropriate prompt candidates in response to user input, resulting in problems with the dialogue between users and the system not proceeding smoothly. Furthermore, the prompt candidates generated based on user input are limited, making it difficult to respond to diverse user needs.

[0814] 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.

[0815] In this invention, the server includes means for receiving a user's input words and generating prompt candidates using a generative AI model based on the received words, means for presenting the generated prompt candidates to the user, and means for incorporating a dialogue system that supports dialogue with the user. This makes it possible to generate a variety of appropriate prompt candidates in response to the user's input and smoothly progress with dialogue with the user.

[0816] "User" means any person or entity that uses the System to provide input.

[0817] "Input words" are specific words or phrases that users type into a chat app.

[0818] A "generative AI model" is an artificial intelligence model for generating relevant prompt candidates based on user input.

[0819] "Prompt candidates" are questions or phrases that are generated based on the user's input words and serve as a guide for the user's next input.

[0820] A "presenting means" is a method or device for displaying the generated prompt candidates to a user.

[0821] A "dialogue system" is a software or hardware system that assists in dialogue with a user.

[0822] "Server" means a central processing unit for receiving a user's input words, generating prompt candidates using a generative AI model, and presenting the generated prompt candidates to the user.

[0823] MODE FOR CARRYING OUT THE INVENTION

[0824] This invention is a system that, when a user inputs a specific word in a chat app, generates prompt candidates based on the input word using a generative AI model and presents them to the user. A specific embodiment of this system is described below.

[0825] System configuration

[0826] The system includes a terminal that receives input words from a user, a server that analyzes the input words and generates prompt candidates, a means for presenting the generated prompt candidates to the user, and a dialogue system that assists in dialogue with the user.

[0827] Hardware and software used

[0828] Terminal: The device through which the user provides input (e.g., smartphone, tablet, PC).

[0829] Server: Acts as the central processing unit, parses the user's input words, and generates suggested prompts using generative AI models.

[0830] Generative AI models: For example, advanced natural language processing models such as OpenAI's GPT-3.

[0831] Dialogue system: Software that assists in interacting with users.

[0832] Data processing and calculation

[0833] 1. Data reception: The device sends the words the user types into the chat app to the server.

[0834] 2. Data analysis: The server analyzes the received words and extracts relevant keywords and contexts. Natural language processing (NLP) techniques are used for the analysis.

[0835] 3. Prompt generation: The server uses a generative AI model (e.g., GPT-3) to generate prompt candidates based on the extracted keywords and context.

[0836] 4. Data transmission: The server sends the generated prompt candidates to the terminal.

[0837] 5. Data display: The terminal displays the received prompt candidates to the user.

[0838] Specific examples

[0839] For example, if a user types the word "travel" into a chat app, the system will act as follows:

[0840] 1. A user types "travel" into a chat app.

[0841] 2. The device sends the word "travel" to the server.

[0842] 3. The server parses the word "travel" and sends a prompt to the generative AI model.

[0843] 4. The generative AI model generates candidate prompts such as "What are some recommended travel destinations?" and "What do I need to prepare for my trip?"

[0844] 5. The server sends the generated prompt candidates to the device.

[0845] 6. The device displays suggested prompts to the user, such as "What are some recommended travel destinations?" or "What do I need to prepare for my trip?"

[0846] Prompt Sentence Examples

[0847] Examples of prompts to be input to a generative AI model include:

[0848] The user has entered the word "travel", based on which you want to generate relevant prompt suggestions.

[0849] In this way, users can get relevant prompt suggestions through the chat app, which helps to facilitate a smooth dialogue with the user and provide them with the right information according to their needs.

[0850] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0851] Step 1:

[0852] A user types a word into a chat app.

[0853] A user opens a chat app and inputs a specific word (e.g., "travel") based on the user's interests or questions. The input data is saved in text format on the device.

[0854] Step 2:

[0855] The terminal sends the entered word to the server.

[0856] The terminal sends the word entered by the user to the server. At this time, the terminal converts the input data into an appropriate format (for example, JSON format) and sends it to the server via the network. The input data is the user's input word, "travel."

[0857] Step 3:

[0858] The server parses the words and sends prompts to the generative AI model.

[0859] The server analyzes the words received from the device. It uses natural language processing (NLP) technology to understand the meaning and context of the words. Based on the analysis results, it sends a prompt to the generative AI model (e.g., GPT-3). The specific prompt is as follows:

[0860] The user has entered the word "travel", based on which you want to generate relevant prompt suggestions.

[0861] The input data is the user's input word "travel," and the output data is the prompt sentence sent to the generative AI model.

[0862] Step 4:

[0863] A generative AI model generates candidate prompts.

[0864] The generative AI model generates relevant prompt candidates based on the prompt received from the server. For example, candidates such as "What travel destinations do you recommend?" or "What do I need to prepare for my trip?" are generated. The input data is the prompt sentence sent from the server, and the output data is the generated prompt candidates.

[0865] Step 5:

[0866] The server sends the generated prompt candidates to the terminal.

[0867] The server sends the prompt candidates received from the generative AI model to the terminal. At this time, the server converts the data into an appropriate format and sends it to the terminal over the network. The input data is the prompt candidates received from the generative AI model, and the output data is the prompt candidates sent to the terminal.

[0868] Step 6:

[0869] The terminal displays the prompt candidates to the user.

[0870] The terminal displays the prompt candidates received from the server to the user. The user can refer to the displayed prompt candidates to ask more specific questions or enter more specific input. The input data is the prompt candidates received from the server, and the output data is the prompt candidates displayed to the user.

[0871] (Application example 1)

[0872] Next, a description will be given of Application Example 1 of Form 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."

[0873] Conventional chatbot systems have had difficulty generating appropriate prompt candidates for keywords entered by users, resulting in a lack of smooth dialogue with users. They also had issues with not being able to efficiently suggest relevant content that users were looking for, resulting in a lack of improvement in the user experience.

[0874] 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.

[0875] In this invention, the server includes means for receiving words input by a user and generating prompt candidates via the Web based on the words, means for presenting the generated prompt candidates to the user, and means for generating prompt candidates for related content based on keywords input by the user and suggesting them to the user. This makes it possible to generate appropriate prompt candidates for the keywords input by the user and efficiently suggest related content.

[0876] "User" means any individual or entity that uses the System.

[0877] An "input word" is a specific word or phrase that a user enters into the system.

[0878] "Web" refers to information resources on the Internet, including websites and online databases.

[0879] "Prompt suggestions" are suggestions for what to type next or questions that are generated based on the words the user types.

[0880] "Content" is a general term for information, data, media, etc., and includes text, images, video, audio, etc.

[0881] A "chatbot" is a program that automatically interacts with users.

[0882] A "system" is an entire device or program that is composed of a combination of multiple means and functions.

[0883] A "suggestion" is the act of presenting options or information to a user.

[0884] The system for implementing this invention has a function of receiving user input words, generating prompt candidates via the Web based on the received words, and presenting the generated prompt candidates to the user. It also includes a function of generating prompt candidates for related content based on keywords entered by the user and suggesting them to the user.

[0885] The server receives the words entered by the user and retrieves relevant information from the web based on those words. Specifically, it uses a Python library to send web requests and parse the HTML data. It then extracts prompt candidates from the parsed data and presents them to the user.

[0886] When a user enters the keyword "movie," the server retrieves prompt candidates from the web, such as "What movies do you recommend?" or "What are the latest movie reviews?", and suggests them to the user. This allows the user to efficiently obtain related content.

[0887] Examples of specific prompts include the following:

[0888] The user has entered the keyword "movies". Generate relevant prompt suggestions.

[0889] The system uses a smartphone as hardware and Python as software, which allows it to generate appropriate prompt candidates for keywords entered by users and efficiently suggest related content.

[0890] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0891] Step 1:

[0892] The user enters keywords into the terminal. The entered keywords are sent to the system. The input data is the specific words or phrases that the user is interested in.

[0893] Step 2:

[0894] The server sends a web request based on the received keywords. Specifically, it uses a Python library to access a URL to retrieve related information. The input data is the keyword entered by the user, and the output data is HTML data retrieved from the web.

[0895] Step 3:

[0896] The server analyzes the acquired HTML data. Specifically, it analyzes the HTML data and extracts prompt candidates. The input data is the HTML data acquired from the Web, and the output data is the analyzed prompt candidates.

[0897] Step 4:

[0898] The server presents the parsed prompt candidates to the user. Specifically, the server transmits data for displaying the prompt candidates to the user's terminal. The input data is the parsed prompt candidates, and the output data is the prompt candidates presented to the user.

[0899] Step 5:

[0900] The user selects from the presented prompt candidates or enters a new keyword. The selected prompt candidate or new keyword is sent back to the system. The input data is the prompt candidate or new keyword selected by the user, and the output data is the data sent to the system.

[0901] Step 6:

[0902] The server sends a web request again to retrieve relevant information based on the user's selection or new keywords. The input data is the prompt candidates or new keywords selected by the user, and the output data is the retrieved HTML data.

[0903] Step 7:

[0904] The server analyzes the re-obtained HTML data and generates new prompt candidates. The input data is the re-obtained HTML data, and the output data is the newly generated prompt candidates.

[0905] Step 8:

[0906] The server presents the newly generated prompt candidates to the user. The input data is the newly generated prompt candidates, and the output data is the new prompt candidates that are presented to the user.

[0907] In this way, it is possible to generate appropriate prompt candidates for the keywords entered by the user and efficiently suggest related content.

[0908] Example 2

[0909] Next, a description will be given of Example 2 of Form 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."

[0910] Conventional dialogue systems have difficulty generating appropriate prompt candidates for questions or topics entered by users, which can cause the dialogue to not proceed smoothly. In addition, users often have difficulty deciding what question to ask next, which can disrupt the flow of the dialogue.

[0911] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for receiving input data from a user and generating prompt candidates using a generative AI model based on the input data, means for presenting the generated prompt candidates to the user, and means equipped with a dialogue agent for assisting the dialogue with the user. This allows the server to generate appropriate prompt candidates for the question or topic input by the user and present them to the user, thereby enabling the dialogue to proceed smoothly.

[0912] "User" means any person or entity that uses the System to provide input data and interact with the System.

[0913] "Input data" refers to information such as questions and topics that users provide to the system.

[0914] A "generative AI model" is an artificial intelligence algorithm for generating relevant prompt candidates based on user input data.

[0915] "Prompt candidates" are potential next questions or topics generated by a generative AI model and presented to the user.

[0916] A "server" is a computer system that runs a generative AI model, generates prompt candidates, and presents them to the user.

[0917] A "presentation means" is a method or device for displaying the generated prompt candidates to a user.

[0918] A "dialogue agent" is software that supports dialogue with a user and communicates with the user using generated prompt candidates.

[0919] MODE FOR CARRYING OUT THE INVENTION

[0920] This invention is a system that generates prompt candidates using a generative AI model based on data entered by a user and presents them to the user. Specific embodiments of this system are described below.

[0921] Hardware and software used

[0922] The system uses the following hardware and software:

[0923] Server: A computing system for running the generative AI model and generating prompt candidates. For example, a cloud-based server or a dedicated data center can be used.

[0924] Device: The device on which a user provides input data and receives generated prompt suggestions, such as a smartphone, tablet, or computer.

[0925] Generative AI model: An artificial intelligence algorithm used to generate prompt candidates based on user input data, such as a natural language generation model like OpenAI's GPT-4.

[0926] Dialogue agent: Software that assists in dialogue with the user and communicates with the user using generated prompt suggestions.

[0927] Data processing and calculation

[0928] The server inputs the input data received from the user into the generative AI model, which then generates relevant prompt candidates based on the input data.The generated prompt candidates are sent from the server to the device and presented to the user.

[0929] Specific examples

[0930] For example, if a user types "What are some recommended travel destinations?", the server uses a generative AI model to generate the following possible prompts:

[0931] What do you recommend overseas?

[0932] What are some hidden gems in the country?

[0933] "What's the best place for a family trip?"

[0934] These prompt candidates are presented to the user via the device. The user can use these prompt candidates to formulate their next question. For example, if the prompt candidate "What are your recommendations for overseas travel?" is selected, the user can enter a more specific question such as "What are some recommended travel destinations in Europe?"

[0935] The system provides users with specific guidance to help them navigate the conversation smoothly.

[0936] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0937] Program processing flow

[0938] Step 1:

[0939] The user enters an initial question or theme.

[0940] Input: A user uses a device to provide input data (e.g., "What are some travel recommendations?").

[0941] Specific operation: The user enters a question into the input field on their smartphone or computer and presses the send button.

[0942] Step 2:

[0943] The server receives the user's input data.

[0944] Input: The input data submitted by the user.

[0945] Data processing: The server analyzes the input data and converts it into a format suitable for the generative AI model.

[0946] Output: Data format for input into a generative AI model.

[0947] Specific operation: The server analyzes the received data and performs preprocessing if necessary.

[0948] Step 3:

[0949] The server generates prompt candidates using a generative AI model.

[0950] Input: Preprocessed user input data.

[0951] Data computation: A generative AI model generates prompt candidates based on input data.

[0952] Output: Generated prompt candidates (e.g., "What are some recommended places abroad?", "What are some hidden gems in Japan?").

[0953] How it works: The server calls the generative AI model, which runs its internal algorithms to generate potential prompts.

[0954] Step 4:

[0955] The server sends the generated prompt candidates to the terminal.

[0956] Input: Generated prompt suggestions.

[0957] Data processing: Converting the prompt candidates into a format for presentation to the user.

[0958] Output: Prompt suggestions sent to the terminal.

[0959] Specific operation: The server converts the prompt candidates into an appropriate format and sends them to the terminal over the Internet.

[0960] Step 5:

[0961] The terminal presents the prompt candidates to the user.

[0962] Input: Prompt suggestions sent by the server.

[0963] Data processing: Display prompt candidates in a format that the user can view.

[0964] Output: The suggested prompts that are presented to the user.

[0965] Specific behavior: The device displays prompt candidates on the screen and allows the user to select one.

[0966] Step 6:

[0967] The user inputs the next question based on the suggested prompts.

[0968] Input: Suggested prompt suggestions.

[0969] Data processing: The user uses the suggested prompts to refine the next question.

[0970] Output: The next question (e.g., "What are the best travel destinations in Europe?").

[0971] Specific operation: The user refers to the presented prompt candidates, enters a new question, and presses the submit button.

[0972] In this way, the user can obtain specific guidance for smoothly progressing the dialogue.

[0973] (Application example 2)

[0974] Next, a description will be given of Application Example 2 of Form 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."

[0975] Conventional conversational agent systems provide insufficient guidance for users to find the content they want to watch, and have difficulty providing appropriate prompt candidates for users to ask specific questions. Furthermore, the generated prompt candidates often do not match the user's needs, resulting in a poor user experience.

[0976] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means for receiving input words from a user and generating prompt candidates via a network based on the input words; means for presenting the generated prompt candidates to the user; a system equipped with a dialogue agent that assists dialogue with the user; means for providing guidance to help the user find content they want to view; means for generating prompt candidates based on the user's input using a generative AI model; and means for presenting the generated prompt candidates to the user. This enables the user to efficiently find content they want to view.

[0977] "User" means an individual or entity that uses the System to provide input to locate content they wish to view.

[0978] "Input words" are keywords or phrases that a user provides to the system and contain information related to the content they wish to view.

[0979] "Network" means an infrastructure for transmitting and receiving data, including the Internet and other communications means.

[0980] "Prompt candidates" are phrases that are generated based on the user's input words and provide guidance for the user to ask specific questions.

[0981] A "conversational agent" is software designed to assist in a dialogue with a user, generating appropriate responses based on the user's input.

[0982] A "system" is a collection of hardware and software for receiving user input, generating prompt candidates, and assisting in user interaction.

[0983] A "Guide" is advice or suggestions provided to help a user find content they want to watch.

[0984] A "generative AI model" is an algorithm or program that uses artificial intelligence technology to generate prompt candidates based on user input.

[0985] A system for implementing this invention receives user input words, generates prompt candidates based on the input words via a network, and presents the generated prompt candidates to the user. The system is equipped with a dialogue agent that assists in dialogue with the user and provides guidance for the user to find content they want to view.

[0986] The server uses a generative AI model to generate prompt candidates based on the user's input. The generative AI model can use OpenAI's API. Specifically, when the user inputs the content they want to watch, the server receives that input and uses the generative AI model to generate appropriate prompt candidates. The generated prompt candidates are then presented to the user.

[0987] Users access the system using a device such as a smartphone and input the content they want to watch. For example, if a user inputs "I want to watch an action movie," the server uses a generative AI model to generate prompt candidates such as "What are some recent recommended action movies?" or "What are some classic action movies?" and presents them to the user.

[0988] The system uses a smartphone (Android) as its hardware, and Python and OpenAI API as its software. Data processing and data calculation involve a series of processes that receive user input, generate prompt candidates using a generative AI model, and present them to the user.

[0989] For example, if a user types "I want to watch an action movie," the generative AI model will generate prompt candidates such as "What are some recent recommended action movies?" or "What are some classic action movies?" By presenting these prompt candidates to users, they can ask more specific questions and efficiently find the content they want to watch.

[0990] An example of a prompt is as follows:

[0991] What action movies do you recommend?

[0992] What are some popular action movies these days?

[0993] What are some classic action movies?

[0994] In this way, users can efficiently find the content they want to watch.

[0995] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0996] Step 1:

[0997] A user accesses the system using a device such as a smartphone and inputs the input words for the content they want to watch. For example, they might input "I want to watch an action movie." This input word is sent to the system.

[0998] Step 2:

[0999] The server receives the input words sent by the user and sends a request to the generative AI model to generate prompt candidates based on the received input words. At this time, the input words are converted into a format that the generative AI model can understand.

[1000] Step 3:

[1001] The generative AI model generates prompt candidates based on the input words received from the server. For example, for the input words "I want to watch an action movie," it generates prompt candidates such as "What are some recent action movies that you recommend?" or "What are some classic action movies?" The generated prompt candidates are returned to the server.

[1002] Step 4:

[1003] The server receives the prompt candidates returned by the generative AI model and formats them for presentation to the user, who then sends the formatted prompt candidates to the user's device.

[1004] Step 5:

[1005] The user's device receives the prompt candidates sent from the server and presents them to the user, who can then ask a specific question based on the presented prompt candidates.

[1006] Step 6:

[1007] When the user enters a specific question based on the suggested prompts, the question is again sent to the server, which again uses the generative AI model to generate an appropriate answer to the user's question.

[1008] Step 7:

[1009] The generative AI model generates an appropriate answer based on the user's specific question and returns the answer to the server, which then sends the generated answer to the user's device.

[1010] Step 8:

[1011] The user's device receives the response sent from the server and presents it to the user, allowing the user to obtain specific information about the content they wish to view.

[1012] Example 3

[1013] Next, a description will be given of Example 3 of Form Example 3. 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."

[1014] Conventional chatbot systems have difficulty efficiently generating prompt candidates to generate appropriate responses to user input and selecting the most appropriate one. Furthermore, they lack the means to smoothly advance dialogue with users, which has led to issues with the user experience not being improved.

[1015] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1016] In this invention, the server includes a means for receiving user input words and analyzing data based on the words, a means for sending prompts to the generative AI model based on the analysis results, a means for receiving multiple prompt candidates from the generative AI model, a means for selecting the optimal prompt from the received prompt candidates, a means for presenting the selected prompt candidates to the user, and a system equipped with a chatbot that supports dialogue with the user. This makes it possible to efficiently generate appropriate answers to user input and smoothly progress dialogue with the user.

[1017] "User" means an individual or entity that uses the System to ask questions or interact.

[1018] An "input word" is a word or phrase that represents a question or request that a user enters into the system.

[1019] "Means of analyzing data" refers to techniques and methods for analyzing the words entered by the user and understanding their intent and content.

[1020] A "generative AI model" is an artificial intelligence model that generates appropriate prompt candidates and answers based on user input.

[1021] A "prompt" is a form of instruction or question that is input to a generative AI model.

[1022] "Prompt candidates" are multiple possible answers or suggestions generated by a generative AI model.

[1023] The "means for selecting the best one" refers to the techniques or methods for selecting the most appropriate one from the generated prompt candidates.

[1024] "User presentation means" refers to the technology or method used to display selected prompt candidates and answers to the user.

[1025] A "chatbot" is an automated response system designed to assist in user interactions.

[1026] A "system" is a collection of devices and software that executes a series of processes to receive user input, analyze it, use a generative AI model to generate an appropriate answer, and present it to the user.

[1027] This invention is a system equipped with a chatbot that supports user interaction. The system executes a series of processes: it receives user input, analyzes data based on the input, generates appropriate prompt candidates using a generative AI model, selects the most suitable one, and presents it to the user.

[1028] Hardware and software used

[1029] This system uses the following hardware and software:

[1030] Server: A server with high-performance computing power is used to analyze user input data and send prompts to the generative AI model.

[1031] Terminal: The device used by the user to provide input (e.g., smartphone, PC, etc.).

[1032] Generative AI models: For example, use generative AI models (such as GPT-3) that excel at natural language processing.

[1033] Natural language processing libraries: Parse user input data.

[1034] Data processing and calculation

[1035] When the server receives the user's input words, it analyzes the data using a natural language processing library. Based on the analysis results, it sends a prompt to the generative AI model. The generative AI model generates multiple prompt candidates based on the received prompt. The server selects the most appropriate prompt from the generated candidates and presents it to the user.

[1036] Specific examples

[1037] As a concrete example, we will explain the processing when a user asks, "What are your recommendations for overseas destinations?"

[1038] 1. The user types "What are your recommendations for overseas destinations?" into their device.

[1039] 2. The terminal sends the input data to the server.

[1040] 3. The server analyzes the received input data and sends the following prompt to the generative AI model:

[1041] "A user asks, 'What are your recommendations abroad?' Generate an appropriate answer."

[1042] 4. The generative AI model generates multiple prompt candidates and returns them to the server. For example, it generates an answer such as, "How about Paris, France? You can enjoy visiting museums and gourmet food."

[1043] 5. The server selects the best prompt from the generated prompt candidates and presents it to the user.

[1044] 6. The device displays the received answer to the user.

[1045] In this way, the system can smoothly advance the dialogue with the user. The flow of the identification process in the third embodiment will be described with reference to FIG.

[1046] Step 1:

[1047] The user enters a question.

[1048] The user types "What do you recommend overseas?" into the chat box on the device. The input data is in text format.

[1049] Step 2:

[1050] The device sends input to the server.

[1051] The terminal converts the user's input data into JSON format and sends it to the server via the Internet. The input is {"question": "What do you recommend overseas?"}, and the output is the JSON data sent to the server.

[1052] Step 3:

[1053] The server parses the input.

[1054] The server parses the received JSON data and uses a natural language processing library to understand the intent of the user's question. The input is the user's question data, and the output is the parsed result.

[1055] Step 4:

[1056] The server sends prompts to the generative AI model.

[1057] The server sends a prompt to the generative AI model based on the analysis results. The prompt text is "The user asked, 'What overseas places do you recommend?' Please generate an appropriate answer." The input is the analysis results, and the output is the prompt text sent to the generative AI model.

[1058] Step 5:

[1059] A generative AI model generates candidate prompts.

[1060] The generative AI model generates multiple candidate answers based on the prompt it receives. For example, it generates an answer such as, "How about Paris, France? You can enjoy visiting museums and gourmet food." The input is the prompt sentence, and the output is multiple candidate prompts.

[1061] Step 6:

[1062] The server selects the best prompt candidate.

[1063] The server selects the most appropriate prompt candidate from the multiple generated prompt candidates. The selection criteria are the relevance to the user's question, the specificity of the information, etc. The input is the prompt candidate, and the output is the optimal prompt candidate.

[1064] Step 7:

[1065] The server sends the answer to the user.

[1066] The server converts the selected answers into JSON format and sends it to the user. The input is the best prompt candidates, and the output is the answer data sent to the user.

[1067] Step 8:

[1068] The terminal displays the answer to the user.

[1069] The device analyzes the received response data and displays it in a chat box. For example, it might say, "How about Paris, France? You can enjoy visiting museums and fine cuisine." The input is the response data, and the output is the response displayed to the user.

[1070] (Application example 3)

[1071] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[1072] Conventional food delivery services require users to search and select menu items themselves, making it difficult to receive optimal suggestions based on user preferences and requests. Furthermore, there is a lack of systems that can provide appropriate suggestions through dialogue with users, and there is a need to improve the user experience.

[1073] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means. In this invention, the server includes means for receiving a user's input words and generating prompt candidates based on the words via a communication network, means for presenting the generated prompt candidates to the user, a system equipped with an interactive agent that supports dialogue with the user, means for making optimal suggestions based on the user's preferences and requests, and means for displaying the generated suggestions to the user. This makes it possible to make optimal food delivery suggestions based on the user's input.

[1074] "User input words" refers to text or keywords that a user enters into the system.

[1075] "Communications Network" means the Internet or other network infrastructure for transmitting and receiving data.

[1076] "Prompt candidates" are response candidates that the system generates based on the user's input words and presents to the user.

[1077] A "conversational agent," also known as a chatbot, is a software program designed to assist users in interacting with the agent.

[1078] "User preferences and requests" refer to the individual preferences and requests that a user expresses to the system.

[1079] "Best Suggestion" refers to the most appropriate response or recommendation generated by the system based on the user's input words and preferences or requests.

[1080] A "generated suggestion" is a specific response or recommendation generated by a conversational agent based on user input.

[1081] A system for implementing this invention includes means for receiving a user's input words and generating prompt candidates based on the words via a communication network, means for presenting the generated prompt candidates to the user, a system equipped with an interactive agent that assists in dialogue with the user, means for making optimal suggestions based on the user's preferences and requests, and means for displaying the generated suggestions to the user.

[1082] Hardware and software used

[1083] Hardware: Smartphone, tablet, or PC

[1084] Software: Python, OpenAI API

[1085] Data processing and calculation

[1086] 1. Obtaining user input: A user uses a device such as a smartphone or PC to enter a question or request into the system. For example, they might enter, "What pizza do you recommend?"

[1087] 2. Prompt generation: The server generates an appropriate prompt based on the user's input. For example, it generates a prompt like, "If the user asks, 'What pizza do you recommend?', please provide an appropriate food delivery suggestion."

[1088] 3. Response generation using an AI model: The server uses OpenAI's GPT-3 model to generate a response based on the generated prompt, for example, "How about a Margherita pizza? It features fresh basil and mozzarella cheese."

[1089] 4. Displaying the response: The generated response is displayed on the user's device, allowing the user to place a food delivery order.

[1090] Specific examples

[1091] Prompt Sentence Examples

[1092] User input: "What pizza do you recommend?"

[1093] Generated prompt: "When a user asks, 'What pizza do you recommend?', provide an appropriate food delivery suggestion."

[1094] Specific response examples

[1095] User input: "What pizza do you recommend?"

[1096] The chatbot responds: "How about a Margherita pizza? Featuring fresh basil and mozzarella cheese."

[1097] In this way, the user experience can be improved by making optimal food delivery suggestions based on user input.

[1098] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[1099] Step 1:

[1100] A user uses a terminal to type a question or request to the system, for example, "What pizza do you recommend?" This input is sent to the system.

[1101] Input: The user's question or request (e.g., "What pizza do you recommend?")

[1102] Output: The user's input word sent to the system

[1103] Step 2:

[1104] The server receives the user's input words and generates a prompt based on the words. For example, it generates a prompt like, "If the user asks, 'What pizza do you recommend?', please suggest an appropriate food delivery option."

[1105] Input: User input (e.g., "What pizza do you recommend?")

[1106] Output: The generated prompt (e.g., "When the user asks 'What pizza do you recommend?', provide an appropriate food delivery suggestion.")

[1107] Step 3:

[1108] The server uses the generated prompt to send a request to OpenAI's GPT-3 model and generate a response, such as "How about a Margherita pizza? Featuring fresh basil and mozzarella cheese."

[1109] Input: Generated prompt text (e.g., "When the user asks 'What pizza do you recommend?', provide appropriate food delivery suggestions.")

[1110] Output: The response generated by the GPT-3 model (e.g., "How about a Margherita pizza? It features fresh basil and mozzarella cheese.")

[1111] Step 4:

[1112] The server generates a response and sends it to the user's terminal, which displays the response.

[1113] Input: The response generated by the GPT-3 model (e.g., "How about a Margherita pizza? It features fresh basil and mozzarella cheese.")

[1114] Output: The response displayed on the user's terminal

[1115] Step 5:

[1116] The user reviews the displayed response and places a food delivery order if desired, for example, ordering a suggested pizza.

[1117] Input: User confirmation and order (e.g. suggested pizza order)

[1118] Output: Food delivery order information

[1119] In this way, a system is realized that processes and calculates data based on user input at each step, ultimately providing the user with optimal suggestions.

[1120] 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.

[1121] "Example 1"

[1122] One embodiment of the present invention is a system that includes a means for receiving user input words and generating prompt candidates via the web based on the received words, a means for presenting the generated prompt candidates to the user, and a chatbot that supports dialogue with the user. This system also includes an emotion engine that recognizes the user's emotions. The emotion engine analyzes the user's input words to determine their emotions, and the chatbot uses the results to engage in dialogue with the user. For example, if a user inputs "Today is the worst day," the emotion engine determines from this input that the user is in a bad mood. Based on the analysis results, the chatbot then engages in appropriate dialogue with the user.

[1123] "Example 2"

[1124] In another embodiment of the present invention, an emotion engine tracks changes in a user's emotions, and the chatbot uses these changes to engage in dialogue with the user. For example, if a user first inputs, "Today is the worst day," and then inputs, "But it was fun meeting my friends," the emotion engine tracks the change in the user's emotions from disappointment to joy. Based on this change in emotion, the chatbot then engages in appropriate dialogue with the user.

[1125] "Example 3"

[1126] In yet another embodiment of the present invention, an emotion engine adjusts prompt candidates based on the user's emotions, and the chatbot uses the adjusted prompt candidates to engage in a dialogue with the user. For example, if a user inputs "Today is the worst day," the emotion engine analyzes this input to indicate that the user is in a bad mood. Based on the analysis result, the emotion engine adjusts the prompt candidates to generate more considerate prompt candidates, such as "That's been tough. Maybe talking about it would help." The chatbot then engages in an appropriate dialogue with the user based on the adjusted prompt candidates.

[1127] The processing flow of each embodiment will be described below.

[1128] "Example 1"

[1129] Step 1: Receive an input word from the user.

[1130] Step 2: Analyze the sentiment using the sentiment engine based on the received words.

[1131] Step 3: Generate prompt candidates based on the analysis results.

[1132] Step 4: Present the generated prompt candidates to the user.

[1133] Step 5: The chatbot interacts with the user.

[1134] "Example 2"

[1135] Step 1: Receive an input word from the user.

[1136] Step 2: Analyze the sentiment using the sentiment engine based on the received words.

[1137] Step 3: Track changes in user sentiment based on the analysis results.

[1138] Step 4: Generate prompt candidates based on changes in emotions.

[1139] Step 5: Present the generated prompt candidates to the user.

[1140] Step 6: The chatbot interacts with the user.

[1141] "Example 3"

[1142] Step 1: Receive an input word from the user.

[1143] Step 2: Analyze the sentiment using the sentiment engine based on the received words.

[1144] Step 3: Adjust the prompt candidates based on the analysis results.

[1145] Step 4: Present the tailored prompt candidates to the user.

[1146] Step 5: The chatbot interacts with the user.

[1147] Example 1

[1148] Next, a description will be given of Example 1 of Form 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."

[1149] Conventional chat systems have had issues with generating appropriate prompt candidates based on user input and not being able to take user emotions into account when conducting dialogue, which leads to a poor user experience and limits the system's usefulness.

[1150] 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.

[1151] In this invention, the server includes means for receiving user input words and generating prompt candidates via an information source based on the words, means for presenting the generated prompt candidates to the user, and means including a dialogue agent that is equipped with an emotion engine that analyzes the user's emotions and supports dialogue with the user based on the analysis results. This makes it possible to generate appropriate prompt candidates based on the user's input and to have a dialogue that takes the user's emotions into consideration.

[1152] "User input words" refers to words or phrases that users type into a chat app.

[1153] An "information source" is an external resource from which information can be obtained, such as an internet database or search engine.

[1154] "Prompt candidates" are questions or phrases that the user might type next, generated based on the words the user types.

[1155] An "emotion engine" is software or algorithm that analyzes emotions from user input words.

[1156] A "conversational agent" is a system equipped with artificial intelligence to assist in dialogue with users.

[1157] This invention is a system that, when a user inputs a specific word into a chat app, generates prompt candidates based on that word via information sources, and also analyzes the user's emotions to support the conversation.

[1158] Hardware and software used

[1159] The system uses the following hardware and software:

[1160] Server: Processes data and calls web search APIs.

[1161] Device: The device (smartphone, tablet, PC, etc.) through which a user operates the chat app.

[1162] Web Search APIs: APIs for retrieving data from sources such as Google Search API and Bing Search API.

[1163] Sentiment engine: Software for analyzing user emotions, such as IBM Watson or Microsoft Azure's sentiment analysis API.

[1164] Conversational Agent: A system equipped with artificial intelligence to assist in dialogue with users.

[1165] Program processing

[1166] When a user enters a specific word into a chat app, the device sends the word to a server. The server uses a web search API to retrieve related information based on the received word. It analyzes the retrieved information and generates prompt candidates such as "What are your recommended travel destinations?" or "What do I need to prepare for my trip?" The generated prompt candidates are presented to the user via the device.

[1167] Furthermore, when a user inputs a word containing an emotion, the device sends this word to the server. The server uses an emotion engine to analyze the user's emotion, and the dialogue agent generates an appropriate response based on the results. For example, if a user inputs "Today is the worst day," the emotion engine will analyze that the user is in a bad mood, and the dialogue agent will generate a response such as "Did something bad happen today?"

[1168] Specific examples

[1169] For example, if a user types "travel" into a chat app, the device sends this word to the server, which uses the Google Search API to retrieve information related to "travel" and generates prompt suggestions such as "What are your recommended travel destinations?" or "What do I need to prepare for my trip?" These suggestions are then displayed to the user via the device.

[1170] For example, if a user types "Today is the worst day," the device sends this phrase to the server. The server uses an emotion engine to analyze the user's emotions and recognizes that the user is in a bad mood. As a result, the conversational agent generates a response such as "Did something bad happen today?" and displays it to the user through the device.

[1171] In this way, the system can generate appropriate prompt candidates based on the user's input and further conduct dialogue that takes the user's emotions into account.

[1172] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1173] Step 1:

[1174] A user types a word into a chat app.

[1175] Input: A user types a word, such as "travel," into an input field in a chat app.

[1176] Output: The entered word is sent to the terminal.

[1177] Specific operation: The user types "travel" into a chat app on their smartphone or PC and presses the send button.

[1178] Step 2:

[1179] The terminal sends the entered word to the server.

[1180] Input: The word entered by the user (e.g., "travel").

[1181] Output: The entered word is sent to the server.

[1182] Specific operation: The device sends the entered word to the server using an HTTP request.

[1183] Step 3:

[1184] The server receives the word and retrieves related information using a web search API.

[1185] Input: A word sent from the device (e.g. "travel").

[1186] Output: Relevant information retrieved from the Web Search API.

[1187] Specific operation: The server calls the Google Search API or Bing Search API to obtain information related to "travel."

[1188] Step 4:

[1189] The server generates prompt candidates based on the information it has obtained.

[1190] Input: Relevant information obtained from the Web Search API.

[1191] Output: Generated prompt suggestions (e.g., "What travel destinations do you recommend?", "What do I need to prepare for my trip?").

[1192] What it does: The server analyzes the information it has retrieved and generates possible prompts for the user to enter next.

[1193] Step 5:

[1194] The server generates prompt candidates and sends them to the terminal.

[1195] Input: Generated prompt suggestions.

[1196] Output: Prompt candidates are sent to the terminal.

[1197] Specific operation: The server sends the generated prompt candidates to the device using an HTTP request.

[1198] Step 6:

[1199] The device displays possible prompts to the user.

[1200] Input: Prompt suggestions sent by the server.

[1201] Output: The suggested prompts are displayed to the user.

[1202] Specific operation: The device displays the prompt candidates received on the chat app screen.

[1203] Step 7:

[1204] The user enters a word that contains a sentiment.

[1205] Input: A user types a sentimental word, such as "Today is the worst day," into an input field in a chat app.

[1206] Output: The entered word is sent to the terminal.

[1207] Specific action: A user types "Today is the worst day" into a chat app on their smartphone or PC and presses the send button.

[1208] Step 8:

[1209] The terminal sends words containing emotions to the server.

[1210] Input: A user-entered sentiment word (e.g., "Today is the worst day").

[1211] Output: The entered word is sent to the server.

[1212] Specific operation: The device sends the entered word to the server using an HTTP request.

[1213] Step 9:

[1214] The server analyzes the emotions using an emotion engine.

[1215] Input: A word containing an emotion sent from the device (e.g., "Today is the worst day").

[1216] Output: Parsed emotion information (e.g., displeased).

[1217] Specific operation: The server calls an emotion engine (e.g., emotion analysis API) and analyzes the user's emotions.

[1218] Step 10:

[1219] The server generates the chatbot's response based on the analysis results.

[1220] Input: Parsed emotion information.

[1221] Output: The generated chatbot response (e.g., "Did something bad happen today?").

[1222] Specific operation: The server generates an appropriate response to the user based on the analysis results.

[1223] Step 11:

[1224] The server generates a response and sends it to the terminal.

[1225] Input: The generated chatbot response.

[1226] Output: The response is sent to the terminal.

[1227] Specific operation: The server uses an HTTP request and sends the generated response to the terminal.

[1228] Step 12:

[1229] The terminal displays the response to the user.

[1230] Input: The chatbot response sent from the server.

[1231] Output: The response is displayed to the user.

[1232] Specific operation: The response received by the device is displayed on the chat app screen.

[1233] (Application example 1)

[1234] Next, a description will be given of Application Example 1 of Form 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."

[1235] Conventional chatbot systems have had difficulty generating appropriate prompt candidates in response to user input, making it difficult to conduct dialogue that takes the user's emotions into account. Furthermore, online shopping sites are required to quickly and accurately recommend products that users want, but conventional systems have not been able to fully achieve this.

[1236] 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.

[1237] In this invention, the server includes a means for receiving user input words and generating prompt candidates via the web based on the words; a means for presenting the generated prompt candidates to the user; a means for incorporating an emotion engine that analyzes the user's emotions and a chatbot that interacts with the user based on the analysis results; and a means for suggesting related products based on the user's input words. This allows for the generation of appropriate prompt candidates in response to the user's input and for a dialogue that takes the user's emotions into consideration. It also allows for the prompting of products desired by the user in a timely and accurate manner.

[1238] "User input words" refers to specific words or phrases that a user inputs into the system.

[1239] "Prompt candidates" are questions or suggestions to present to the user that are generated based on the user's input words.

[1240] An "emotion engine" is software or hardware that analyzes emotions from user input words or text.

[1241] A "chatbot" is a program or system that automatically interacts with users.

[1242] "Product suggestion" refers to the act of recommending related products based on the user's input words.

[1243] The system for implementing this invention has the function of receiving user input words, generating prompt candidates via the web based on the words, and presenting them to the user. It also includes an emotion engine that analyzes the user's emotions and a chatbot that interacts with the user based on the analysis results. It also has the function of suggesting related products based on the user's input words.

[1244] Hardware and software used

[1245] Hardware: Smartphone

[1246] Software: Python

[1247] Data processing and calculation

[1248] Prompt Generation

[1249] The server receives the keywords entered by the user, calls a Web API based on the keywords, and generates relevant prompt candidates, which are then presented to the user.

[1250] Emotion analysis

[1251] The server sends the user's input text to the emotion engine API for emotion analysis. The emotion engine extracts emotions from the user's text and returns the results to the server.

[1252] Dialogue Support

[1253] The server then responds to the user with appropriate dialogue based on the results of the emotion analysis. For example, if a user enters "Today is the worst day," the emotion engine will interpret this as meaning that the user is in a bad mood and respond with appropriate dialogue such as "Did anything special happen today?"

[1254] Product suggestion

[1255] The server suggests related products based on the user's input. For example, if a user inputs "I want a new smartphone," the server generates prompt candidates such as "What brand of smartphone do you like?" and "What is your budget?" and presents them to the user.

[1256] Specific examples

[1257] If a user types "I want a new smartphone," the server generates the following prompt candidates and presents them to the user:

[1258] "Which brand of smartphone do you prefer?"

[1259] "What's your budget?"

[1260] "Which features are important?"

[1261] Also, if a user types, "Today is the worst day," the emotion engine will analyze that the user is in a bad mood, and the server will engage in appropriate dialogue such as, "Did anything special happen today?"

[1262] In this way, the system can generate appropriate prompt candidates based on user input, conduct dialogue that takes into account the user's emotions, and quickly and accurately suggest products that the user wants.

[1263] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1264] Step 1:

[1265] The user inputs a keyword into the terminal.

[1266] Input: The keyword entered by the user (e.g., "I want a new smartphone")

[1267] Output: The device sends the keyword to the server.

[1268] Step 2:

[1269] The server calls a Web API based on the received keywords and generates relevant prompt candidates.

[1270] Input: Keyword sent from the terminal

[1271] Data processing: The server sends keywords to the Web API to obtain related prompt candidates.

[1272] Output: Generated prompt candidates (e.g., "What brand of smartphone do you like?", "What is your budget?")

[1273] Step 3:

[1274] The server transmits the generated prompt candidates to the terminal and presents them to the user.

[1275] Input: Generated prompt suggestions

[1276] Output: The terminal displays the possible prompts to the user.

[1277] Step 4:

[1278] The user inputs text containing emotions into the terminal.

[1279] Input: User-entered text containing emotions (e.g., "Today is the worst day ever").

[1280] Output: The terminal sends text to the server.

[1281] Step 5:

[1282] The server sends the received text to the emotion engine API for emotion analysis.

[1283] Input: Text sent from the terminal

[1284] Data processing: The server sends the text to the emotion engine API and obtains the emotion analysis results.

[1285] Output: Sentiment analysis result (e.g., displeased)

[1286] Step 6:

[1287] The server then engages in appropriate dialogue with the user based on the emotion analysis results.

[1288] Input: Sentiment analysis results

[1289] Data calculation: The server generates appropriate dialogue content based on the results of emotion analysis.

[1290] Output: Generated dialogue (e.g. "Is there anything special happening today?")

[1291] Step 7:

[1292] The server transmits the generated dialogue content to the terminal and presents it to the user.

[1293] Input: Generated dialogue

[1294] Output: The terminal displays the dialogue to the user.

[1295] Step 8:

[1296] The server makes relevant product suggestions based on the user's input words.

[1297] Input: User input word

[1298] Data calculation: The server searches for relevant products based on the input words and generates suggestions.

[1299] Output: Generated product suggestions (e.g., "How about this brand of smartphone?")

[1300] Step 9:

[1301] The server transmits the generated product proposal to the terminal and presents it to the user.

[1302] Input: Generated product suggestions

[1303] Output: The terminal displays the product suggestions to the user.

[1304] Example 2

[1305] Next, a description will be given of Example 2 of Form 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."

[1306] Conventional chatbot systems have difficulty generating appropriate prompt candidates based on user input, which can prevent smooth dialogue with the user. Furthermore, because dialogue is conducted without taking into account changes in the user's emotions, user satisfaction can decline. To solve these problems, a system is needed that can generate appropriate prompt candidates based on the user's input data and track changes in the user's emotions as it engages in dialogue.

[1307] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1308] In this invention, the server includes a means for receiving user input data and generating prompt candidates using a generative AI model based on the data, a means for presenting the generated prompt candidates to the user, a means for analyzing the user's emotions using an emotion engine that tracks changes in the user's emotions, a means for generating a dialogue with the user based on the analyzed changes in emotions, a means for presenting the generated dialogue to the user, and a system equipped with a chatbot that supports the dialogue with the user. This makes it possible to generate appropriate prompt candidates based on the user's input data and to conduct a dialogue that takes into account changes in the user's emotions.

[1309] "User input data" refers to information such as text or voice that a user inputs into the system through a terminal.

[1310] A "generative AI model" is an artificial intelligence model that generates prompt candidates and dialogues based on user input data.

[1311] A "prompt candidate" is a suggestion that indicates what question or interaction the user should have next.

[1312] An "emotion engine" is software or algorithm for analyzing emotional changes from user input data.

[1313] A "chatbot" is an automated response system that assists in conversations with users.

[1314] A "server" is a computer system that receives user input data and generates prompt candidates and dialogue using generative AI models and emotion engines.

[1315] A "terminal" is a device through which a user sends input data to the system and receives generated prompt suggestions and interactions.

[1316] MODE FOR CARRYING OUT THE INVENTION

[1317] The present invention provides a system for generating appropriate prompt candidates based on user input data, and for conducting a dialogue by tracking changes in the user's emotions. Specific embodiments of this system are described below.

[1318] server

[1319] The server receives the user's input data and generates prompt candidates based on the data using a generative AI model. A typical natural language processing model is used as the generative AI model. The server then sends the generated prompt candidates to the device. The server also uses an emotion engine to track changes in the user's emotions and generates a dialogue with the user based on the analyzed changes in emotions. A typical emotion analysis software is used as the emotion engine.

[1320] Terminal

[1321] The terminal receives input data from the user and sends it to the server. The text data and voice data entered by the user are sent to the server via the Internet. The terminal presents the prompt candidates received from the server and the generated dialogue to the user. This allows the user to check the prompt candidates and dialogue displayed on the screen and make the next input.

[1322] user

[1323] Users interact with the chatbot through their devices. For example, if a user inputs, "Today is the worst day," the information is sent to the server via the device. The server uses an emotion engine to analyze this input data and track down that the user's emotion is bad. If the user then inputs, "But it was fun meeting your friends," the server again analyzes it using its emotion engine and tracks that the user's emotion has changed to joy. Based on this information, the server generates an appropriate dialogue such as, "It was great meeting your friends. What fun things are you planning next?" and presents it to the user via the device.

[1324] Specific examples

[1325] Below are some specific examples of prompt sentences to input to the generative AI model.

[1326] "What travel destinations do you recommend?"

[1327] "What recent movies do you recommend?"

[1328] "Teach me a new recipe?"

[1329] These prompts are suggestions that indicate what questions or interactions the user should have next, providing guidance to help the user smoothly proceed with the interaction.

[1330] In this way, the present invention generates appropriate prompt candidates based on the user's input data, enabling a dialogue that takes into account the user's emotional changes, thereby improving user satisfaction and realizing a more natural and effective dialogue.

[1331] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1332] Program processing flow

[1333] Step 1: User types into terminal

[1334] The user uses the terminal to input text data, for example, "Today is the worst day ever." This input data is then sent to the server in the next step.

[1335] Step 2: The device sends the input data to the server

[1336] The terminal transmits the text data entered by the user to the server via the Internet. The data is encrypted before transmission. The input data is the text "Today is the worst day."

[1337] Step 3: The server generates prompt candidates using the generative AI model

[1338] The server uses a generative AI model to generate prompt candidates based on the received user input data. For example, a natural language processing model is used as the generative AI model to generate a prompt candidate such as "What travel destinations do you recommend?" The input data is "Today is the worst day," and the output data is "What travel destinations do you recommend?"

[1339] Step 4: The server sends the generated prompt candidates to the device

[1340] The server sends the generated prompt candidates to the device. At this time, the data is encrypted again before being sent. The input data is "What travel destinations do you recommend?" and the output data is the prompt candidates sent to the device.

[1341] Step 5: The device presents prompt candidates to the user

[1342] The terminal displays the received prompt candidates to the user. The user can check the prompt candidates displayed on the screen. The input data is "What travel destinations do you recommend?" and the output data is the prompt candidates presented to the user.

[1343] Step 6: The user continues the dialogue based on the suggested prompts

[1344] The user can then ask more specific questions or engage in more specific dialogue based on the suggested prompts. For example, the user can type, "What do you recommend for overseas travel?" The input data is "What do you recommend for overseas travel?", and the output data is sent to the server in the next step.

[1345] Step 7: The server uses the emotion engine to track changes in the user's emotions.

[1346] The server uses an emotion engine to analyze changes in emotions from user input data. For example, from inputs such as "Today is the worst day" and "But it was fun meeting my friends," the server tracks changes in the user's emotions from displeasure to joy. The input data is "Today is the worst day" and "But it was fun meeting my friends," and the output data is the analysis result of the change in emotions.

[1347] Step 8: The server generates appropriate dialogue based on the change in emotion.

[1348] The server generates an appropriate dialogue for the user based on the analysis results of the emotion engine. For example, it generates a dialogue such as, "It was nice to meet your friends. What fun things are you planning to do next?" The input data is the analysis results of changes in emotions, and the output data is the generated dialogue.

[1349] Step 9: The server sends the generated dialogue to the device

[1350] The server transmits the generated dialogue to the terminal. At this time, the data is encrypted before transmission. The input data is the generated dialogue, and the output data is the dialogue transmitted to the terminal.

[1351] Step 10: The terminal presents the generated dialogue to the user

[1352] The terminal displays the received dialogue to the user. The user can confirm the dialogue displayed on the screen and continue the dialogue. The input data is the generated dialogue, and the output data is the dialogue presented to the user.

[1353] (Application example 2)

[1354] Next, a description will be given of Application Example 2 of Form 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."

[1355] Conventional chatbot systems have had the problem of being unable to present appropriate prompt candidates in response to user input, preventing users from smoothly progressing through the dialogue. Furthermore, because the dialogue is conducted without taking into account changes in the user's emotions, there is a risk of user satisfaction decreasing. There is a need to provide a system that can solve these problems and allow users to progress through dialogue more comfortably.

[1356] 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.

[1357] In this invention, the server includes means for receiving a user's input words and generating prompt candidates using a generative AI model based on the words, means for presenting the generated prompt candidates to the user, means for analyzing the user's emotional state using an emotion engine that tracks changes in the user's emotions, and means equipped with a chatbot for conducting appropriate dialogue based on the user's emotional state. This makes it possible to present appropriate prompt candidates in response to the user's input and to conduct dialogue according to the user's emotional state.

[1358] "User input words" are text or keywords that a user inputs into the system.

[1359] A "generative AI model" is an algorithm or system that uses artificial intelligence to generate appropriate prompt candidates based on user input.

[1360] "Prompt suggestions" are text that predicts the next question or instruction the user might enter.

[1361] An "emotion engine" is software or algorithms that analyze a user's emotional state from text input and track changes in that state.

[1362] "Emotional state" is information that indicates the user's current emotions and mood.

[1363] A "chatbot" is a program or system that automatically interacts with users.

[1364] "Engaging in dialogue" means generating appropriate responses to user input and maintaining communication with the user.

[1365] "Response to inquiry" refers to information or solutions provided in response to a question or request made by a user to the system.

[1366] A system for implementing this invention includes means for receiving a user's input words and generating prompt candidates using a generative AI model based on the words, means for presenting the generated prompt candidates to the user, means for analyzing the user's emotional state using an emotion engine that tracks changes in the user's emotions, and means equipped with a chatbot that engages in appropriate dialogue based on the user's emotional state.

[1367] System configuration

[1368] 1. A way to receive user input

[1369] A user inputs text into the system using a device such as a smartphone or computer.

[1370] This input word is received as an initial process for the system.

[1371] 2. A method for generating prompt candidates using a generative AI model

[1372] The server sends the received user input words to the generative AI model.

[1373] A generative AI model (e.g., GPT-3) generates possible prompts for the user to type next based on the input words.

[1374] 3. A means of presenting the generated prompt candidates to the user

[1375] The server transmits the generated prompt candidates to the user's terminal and presents them to the user.

[1376] This gives the user guidance on what question or instruction to enter next.

[1377] 4. A means of analyzing the user's emotional state using an emotion engine

[1378] The server sends the user's input words to the emotion engine to analyze the emotional state.

[1379] The emotion engine calculates a user's emotion score from the input text and tracks its changes.

[1380] 5. A chatbot-based approach that provides appropriate dialogue based on the user's emotional state

[1381] The server generates an appropriate dialogue according to the emotional state of the user based on the emotion score obtained from the emotion engine.

[1382] The chatbot interacts with the user based on the generated prompt candidates and emotional state.

[1383] Hardware and software used

[1384] Hardware: smartphones, computers, servers

[1385] Software: OpenAI API (generative AI model), emotion engine

[1386] Specific examples

[1387] For example, if a user inputs "Today is the worst day ever," the emotion engine calculates a low emotion score from this input. The generative AI model generates "What products do you recommend?" as a possible next prompt for the user. The server then displays to the user, "You seem to be feeling unwell. Is there anything I can help you with?" and suggests "What products do you recommend?" as the next possible prompt.

[1388] Prompt Sentence Examples

[1389] "What travel destinations do you recommend?"

[1390] "I'm looking for new fashion items"

[1391] "What are the most popular items these days?"

[1392] In this way, it is possible to present appropriate prompt candidates in response to the user's input and to conduct a dialogue that is appropriate for the user's emotional state.

[1393] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1394] Step 1:

[1395] A user uses a terminal to input text or keywords into the system.

[1396] Input: User input word

[1397] Output: The user's input word is sent to the server

[1398] Specific action: A user enters text into an input field on a smartphone or computer and presses the send button.

[1399] Step 2:

[1400] The server receives the user's input words and sends them to the generative AI model.

[1401] Input: User input word

[1402] Output: The input words sent to the generative AI model

[1403] Specific operation: The server sends the user's input words as an API request to the generative AI model.

[1404] Step 3:

[1405] A generative AI model generates prompt candidates based on the user's input words.

[1406] Input: The input words sent to the generative AI model

[1407] Output: Generated prompt candidates

[1408] How it works: A generative AI model (e.g., GPT-3) analyzes the input words and generates possible prompts the user might type next.

[1409] Step 4:

[1410] The server transmits the generated prompt candidates to the user's terminal and presents them to the user.

[1411] Input: Generated prompt suggestions

[1412] Output: Prompt candidates displayed on the user's terminal

[1413] Specific operation: The server sends the generated prompt candidates to the user's device and displays them on the screen.

[1414] Step 5:

[1415] The server sends the user's input words to the emotion engine to analyze the emotional state.

[1416] Input: User input word

[1417] Output: Sentiment score calculated by the sentiment engine

[1418] Specific operation: The server sends the user's input words to the emotion engine and calculates the emotion score.

[1419] Step 6:

[1420] The server generates an appropriate dialogue according to the emotional state of the user based on the emotion score obtained from the emotion engine.

[1421] Input: Sentiment score calculated by the sentiment engine

[1422] Output: Dialogue content according to the user's emotional state

[1423] Specific operation: The server analyzes the emotion score and generates dialogue content according to the user's emotional state.

[1424] Step 7:

[1425] The chatbot interacts with the user based on the generated prompt candidates and emotional state.

[1426] Input: Generated prompt candidates, dialogue content according to the user's emotional state

[1427] Output: User interaction

[1428] Specific behavior: The chatbot responds appropriately to the user based on the generated prompt candidates and emotional state.

[1429] Example 3

[1430] Next, a description will be given of Example 3 of Form Example 3. 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."

[1431] Conventional chatbot systems have difficulty generating appropriate prompt candidates in response to user input, and are particularly unable to generate responses that take the user's emotions into account. This results in unnatural interactions with users and a poor user experience.

[1432] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1433] In this invention, the server includes means for receiving user input words and generating prompt candidates using a generative AI model based on the words, means for presenting the generated prompt candidates to the user, means for analyzing the user's emotions and adjusting the prompt candidates based on the analysis results, and means for incorporating a chatbot that interacts with the user using the adjusted prompt candidates. This makes it possible to generate appropriate prompt candidates based on the user's input and provide responses that take the user's emotions into consideration.

[1434] "User input words" refers to the text or message a user sends to a chatbot.

[1435] A "generative AI model" is an artificial intelligence model that uses natural language processing technology to generate appropriate prompt candidates and responses based on user input.

[1436] "Prompt candidates" are potential responses or dialogues that a generative AI model suggests in response to user input.

[1437] "Emotion analysis" refers to the technology of analyzing and identifying a user's emotional state from their input.

[1438] A "chatbot" is a program that automatically interacts with users and has the ability to generate responses using generative AI models and the results of sentiment analysis.

[1439] A "server" is a computer system that receives user input and runs generative AI models and sentiment analysis.

[1440] This invention relates to a system equipped with a chatbot that supports user interactions. The system generates prompt candidates using a generative AI model based on the user's input words, and further analyzes the user's emotions to adjust the prompt candidates to provide more appropriate responses.

[1441] When the server receives the user's input, it first analyzes the input. This analysis uses natural language processing (NLP) technology. Specifically, it uses a common natural language processing model such as GPT-3 as a generative AI model. Based on the analysis results, it generates appropriate prompt candidates.

[1442] For example, if a user types "What do you recommend overseas?", the server will parse this input and input the following prompt sentence into the generative AI model:

[1443] User input: "What are your recommendations for places abroad?"

[1444] Prompt for generative AI model: "A user is asking for international travel recommendations. Please generate an appropriate answer."

[1445] Based on this prompt, the generative AI model generates an answer such as, "How about Paris, France? You can enjoy visiting museums and fine cuisine."

[1446] Furthermore, if an emotion engine is installed, the server analyzes the user's emotions. This analysis is performed using a sentiment analysis algorithm. For example, if a user inputs "Today is the worst day," the emotion engine will determine from this input that the user is in a bad mood. It will then adjust the prompt candidates based on the analysis results. The adjusted prompt candidates might be something like, "That's been tough. Maybe talking about it would help."

[1447] In this way, the server can generate appropriate prompt candidates based on the user's input and provide responses that take the user's emotions into account, making the interaction more natural and effective, and improving the user experience.

[1448] For example, here's what happens when a user asks "What are your recommendations for overseas destinations?":

[1449] 1. The user types "What are your recommendations for overseas?" into their device and presses the send button.

[1450] 2. The server receives this input and analyzes it using NLP techniques.

[1451] 3. Based on the analysis results, the server inputs a prompt statement to the generative AI model: "The user is asking for recommendations for overseas travel. Please generate an appropriate answer."

[1452] 4. The generative AI model generates the answer, "How about Paris, France? You can enjoy visiting museums and fine cuisine."

[1453] 5. The server receives this answer and provides it to the user through the chatbot.

[1454] 6. The user checks this answer on their device.

[1455] This system makes it possible to generate appropriate prompt candidates based on the user's input and provide a response that takes the user's emotions into consideration. The flow of the specification process in the third embodiment will be described with reference to FIG.

[1456] Step 1:

[1457] The user types at the terminal.

[1458] The user uses the device to send a message to the chatbot, for example, by typing "What do you recommend for overseas travel?" and pressing the send button.

[1459] Input: User's message (e.g. "What do you recommend overseas?")

[1460] Output: Message sent from terminal to server

[1461] Step 2:

[1462] The server receives the user's input.

[1463] The server receives input from the user and prepares to parse it.

[1464] Input: User's message sent from the terminal

[1465] Output: Received message

[1466] Step 3:

[1467] The server parses the input.

[1468] The server uses natural language processing (NLP) techniques to analyze the user's input, specifically using a generative AI model (e.g., GPT-3) to tokenize the input sentence and analyze it to understand its context.

[1469] Input: Received message

[1470] Output: Analysis results (e.g., information that the user is asking for recommendations for overseas travel)

[1471] Step 4:

[1472] The server generates the prompt candidates.

[1473] Based on the analysis results, the server inputs a prompt sentence into the generative AI model. For example, the prompt sentence "The user is asking for recommendations for overseas travel. Please generate an appropriate answer." is input into the generative AI model. The generative AI model generates an appropriate answer based on this prompt sentence.

[1474] Input: Analysis results

[1475] Output: Generated prompt candidates (e.g., "How about Paris, France? You can enjoy visiting museums and fine cuisine.")

[1476] Step 5:

[1477] Server performs sentiment analysis (optional).

[1478] If the user's input contains an emotion, the server uses an emotion engine to analyze the emotion. For example, if the user inputs "Today is the worst day," the emotion engine analyzes that the user is in a bad mood.

[1479] Input: Received message

[1480] Output: Sentiment analysis results (e.g., information that the user is in a bad mood)

[1481] Step 6:

[1482] The server optionally adjusts the prompt suggestions.

[1483] Based on the results of the sentiment analysis, the server adjusts the prompt candidates. For example, if the user is in a bad mood, it generates a thoughtful prompt such as, "That must be tough. Maybe talking about it would help."

[1484] Input: Sentiment analysis results

[1485] Output: Tailored prompt suggestions

[1486] Step 7:

[1487] The server generates a response through the chatbot.

[1488] The server receives the answer from the generative AI model and provides it to the user through the chatbot.

[1489] Input: Generated prompt suggestions

[1490] Output: Chatbot response (e.g., "How about Paris, France? You can enjoy visiting museums and delicious food.")

[1491] Step 8:

[1492] The server sends a response to the user.

[1493] The server generates a response and sends it to the user's terminal, where the user can view the response.

[1494] Input: Chatbot response

[1495] Output: The response displayed on the user's terminal

[1496] (Application example 3)

[1497] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[1498] Conventional chatbot systems generate uniform responses without considering the user's emotions, which means they are unable to respond appropriately to the user's feelings. Furthermore, in security services, more considerate responses are required when the user is feeling anxious or scared, but conventional systems have difficulty responding in this way.

[1499] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means. In this invention, the server includes means for receiving words input by a user and generating prompt candidates based on the words via a communication network, means for presenting the generated prompt candidates to the user, a system equipped with a dialogue agent that supports dialogue with the user, emotion analysis means for analyzing the user's emotions, means for adjusting the prompt candidates based on the emotion analysis results, and means for engaging in dialogue with the user using the adjusted prompt candidates. This makes it possible to generate appropriate prompts according to the user's emotions, and to provide security services that take into account the user's anxieties and fears.

[1500] "User input words" are text or keywords that a user inputs into the system.

[1501] A "communications network" is an infrastructure for transmitting and receiving data, including the Internet and other digital networks.

[1502] A "prompt candidate" is a suggested text or question that is generated based on the user's input and presented as the next step in the dialogue.

[1503] A "conversational agent," also known as a chatbot, is a software program designed to assist users in interacting with the agent.

[1504] "Emotion analysis means" refers to technology or algorithms that analyze emotions from the user's input text and determine the type and intensity of those emotions.

[1505] "Emotion analysis results" are data and information about the user's emotions obtained by the emotion analysis means.

[1506] "Adjustment means" refers to technologies and algorithms for appropriately changing or modifying prompt candidates based on the results of sentiment analysis.

[1507] A "means for conducting dialogue" is a technology or algorithm for conducting dialogue with a user using the generated prompt candidates.

[1508] A system for implementing this invention includes means for receiving input words from a user and generating prompt candidates based on the input words via a communication network, means for presenting the generated prompt candidates to the user, a system equipped with a dialogue agent that assists in dialogue with the user, emotion analysis means for analyzing the user's emotions, means for adjusting the prompt candidates based on the emotion analysis results, and means for engaging in dialogue with the user using the adjusted prompt candidates.

[1509] Hardware and software used

[1510] Hardware: smartphones, robots

[1511] Software: Python, OpenAI API

[1512] Data processing and calculation

[1513] Receiving user input words

[1514] A user inputs text into an application installed on a smartphone or robot. For example, the user might input, "I've seen some suspicious people around my house recently."

[1515] Generate prompt candidates

[1516] The server receives the user's input and uses the OpenAI API to generate prompt candidates based on the input. The generated prompt candidates might include, for example, "That's worrying. We recommend you contact the police. Also, check the security cameras in your neighborhood."

[1517] Emotion analysis

[1518] The server analyzes the sentiment of the user's input text, calculating the sentiment polarity (ranging from -1 to 1) and determining the user's emotional state. For example, if the user's input is "I've seen some suspicious people around my house recently," the sentiment analysis result indicates a negative sentiment.

[1519] Adjust prompt suggestions

[1520] Based on the sentiment analysis, the server adjusts the suggested prompts. If the sentiment is negative, it generates a more considerate prompt, such as, "That's worrying. I suggest you contact the police. Also, check the security cameras in your neighborhood."

[1521] User interaction

[1522] Using the tailored prompt candidates, the conversational agent engages in a dialogue with the user, providing appropriate responses to the user's input and taking into consideration the user's anxieties and fears.

[1523] Specific examples

[1524] User input: "I've seen some suspicious people around my house recently."

[1525] Generated prompt: "That's concerning. I suggest you contact the police. Also, check to see if your neighborhood has security cameras."

[1526] The chatbot responds: "That's concerning. I suggest you contact the police. Also, check to see if your neighborhood has security cameras."

[1527] In this way, appropriate prompts can be generated based on the user's emotions, enabling security services to respond in a way that takes into account the user's anxieties and fears.

[1528] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[1529] Step 1:

[1530] The user inputs text into the terminal, for example, "I've seen some suspicious people around my house recently." This input text becomes the input data for the next processing step.

[1531] Step 2:

[1532] The terminal sends the user's input text to the server, which receives the input text and begins processing it as data to generate prompt candidates.

[1533] Step 3:

[1534] The server uses the OpenAI API to generate prompt candidates based on the received input text. The input text is sent to the API, and the generated prompt candidates are received. For example, a prompt candidate might be, "That's worrying. I recommend you contact the police. Also, check the security cameras in your neighborhood."

[1535] Step 4:

[1536] The server analyzes the sentiment of the user's input text and calculates the polarity of the sentiment (ranging from -1 to 1). For example, if the input text is "I've seen some suspicious people around my house recently," the sentiment analysis result will indicate a negative sentiment.

[1537] Step 5:

[1538] The server then adjusts the prompt candidates based on the results of the sentiment analysis. If the sentiment is negative, it generates a more considerate prompt, such as, "That's worrying. I suggest you contact the police. Also, check the security cameras in your neighborhood."

[1539] Step 6:

[1540] The server sends the adjusted prompt candidates to the terminal, which receives the prompt candidates and presents them to the user.

[1541] Step 7:

[1542] The user reviews the suggested prompts and enters further input if necessary. For example, the user might enter, "I've contacted the police, but is there anything else I can do?" This input returns to step 1 and the process repeats.

[1543] In this way, appropriate prompts can be generated according to the user's emotions, enabling security services to respond in a way that takes into account the user's anxiety and fear.

[1544] 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.

[1545] 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> ) 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.

[1546] Another example of generative AI is Gemini (internet search engine). <url: https: gemini.google.com ?hl="ja">) are listed.

[1547] 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.

[1548] [Third embodiment]

[1549] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1550] 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.

[1551] 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).

[1552] 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.

[1553] 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.

[1554] 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).

[1555] 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.

[1556] 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.

[1557] 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.

[1558] 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.

[1559] 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.

[1560] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[1561] "Example 1"

[1562] In one embodiment of the present invention, when a user inputs a specific word into a chat app, the system generates prompt candidates via the web based on the input word. For example, if a user inputs the word "travel," the system generates prompt candidates such as "What are your recommended travel destinations?" and "What do I need to prepare for my trip?"

[1563] "Example 2"

[1564] The generated prompt candidates are presented to the user, allowing the user to receive guidance on how to proceed with the conversation. For example, if a prompt candidate such as "What travel destinations do you recommend?" is presented, the user can respond with a more specific question such as "What overseas places do you recommend?"

[1565] "Example 3"

[1566] Furthermore, the system of the present invention is equipped with a chatbot that supports dialogue with the user. This chatbot converses with the user using prompt candidates generated based on the user's input words. For example, if the user asks, "What overseas places do you recommend?", the chatbot will generate a response such as, "How about Paris, France? You can enjoy visiting museums and gourmet food there."

[1567] The processing flow of each embodiment will be described below.

[1568] "Example 1"

[1569] Step 1: A user enters a specific word in a chat app, for example, the word "travel."

[1570] Step 2: The system generates prompt candidates via the web based on the relevant words. In this case, it generates prompt candidates such as "What travel destinations do you recommend?" and "What do I need to prepare for my trip?"

[1571] Step 3: The generated prompt candidates are presented to the user, providing guidance on how to proceed with the interaction.

[1572] "Example 2"

[1573] Step 1: The user asks a specific question in response to the suggested prompt. For example, in response to the prompt "What travel destinations do you recommend?", the user can ask "What overseas destinations do you recommend?"

[1574] Step 2: The chatbot generates an answer to the user's question, in this case, "How about Paris, France? You can enjoy visiting museums and delicious food."

[1575] "Example 3"

[1576] Step 1: The user asks the chatbot a follow-up question, such as, "What are the best museums to visit in Paris?"

[1577] Step 2: The chatbot generates an answer to the user's question, in this case "The Louvre and the Orsay Museum are must-sees. Also, there are many art studios in Montmartre around the Sacré-Coeur Basilica."

[1578] Example 1

[1579] Next, a description will be given of Example 1 of Form 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."

[1580] Conventional chatbot systems have had difficulty generating appropriate prompt candidates in response to user input, resulting in problems with the dialogue between users and the system not proceeding smoothly. Furthermore, the prompt candidates generated based on user input are limited, making it difficult to respond to diverse user needs.

[1581] 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.

[1582] In this invention, the server includes means for receiving a user's input words and generating prompt candidates using a generative AI model based on the received words, means for presenting the generated prompt candidates to the user, and means for incorporating a dialogue system that supports dialogue with the user. This makes it possible to generate a variety of appropriate prompt candidates in response to the user's input and smoothly progress with dialogue with the user.

[1583] "User" means any person or entity that uses the System to provide input.

[1584] "Input words" are specific words or phrases that users type into a chat app.

[1585] A "generative AI model" is an artificial intelligence model for generating relevant prompt candidates based on user input.

[1586] "Prompt candidates" are questions or phrases that are generated based on the user's input words and serve as a guide for the user's next input.

[1587] A "presenting means" is a method or device for displaying the generated prompt candidates to a user.

[1588] A "dialogue system" is a software or hardware system that assists in dialogue with a user.

[1589] "Server" means a central processing unit for receiving a user's input words, generating prompt candidates using a generative AI model, and presenting the generated prompt candidates to the user.

[1590] MODE FOR CARRYING OUT THE INVENTION

[1591] This invention is a system that, when a user inputs a specific word in a chat app, generates prompt candidates based on the input word using a generative AI model and presents them to the user. A specific embodiment of this system is described below.

[1592] System configuration

[1593] The system includes a terminal that receives input words from a user, a server that analyzes the input words and generates prompt candidates, a means for presenting the generated prompt candidates to the user, and a dialogue system that assists in dialogue with the user.

[1594] Hardware and software used

[1595] Terminal: The device through which the user provides input (e.g., smartphone, tablet, PC).

[1596] Server: Acts as the central processing unit, parses the user's input words, and generates suggested prompts using generative AI models.

[1597] Generative AI models: For example, advanced natural language processing models such as OpenAI's GPT-3.

[1598] Dialogue system: Software that assists in interacting with users.

[1599] Data processing and calculation

[1600] 1. Data reception: The device sends the words the user types into the chat app to the server.

[1601] 2. Data analysis: The server analyzes the received words and extracts relevant keywords and contexts. Natural language processing (NLP) techniques are used for the analysis.

[1602] 3. Prompt generation: The server uses a generative AI model (e.g., GPT-3) to generate prompt candidates based on the extracted keywords and context.

[1603] 4. Data transmission: The server sends the generated prompt candidates to the terminal.

[1604] 5. Data display: The terminal displays the received prompt candidates to the user.

[1605] Specific examples

[1606] For example, if a user types the word "travel" into a chat app, the system will act as follows:

[1607] 1. A user types "travel" into a chat app.

[1608] 2. The device sends the word "travel" to the server.

[1609] 3. The server parses the word "travel" and sends a prompt to the generative AI model.

[1610] 4. The generative AI model generates candidate prompts such as "What are some recommended travel destinations?" and "What do I need to prepare for my trip?"

[1611] 5. The server sends the generated prompt candidates to the device.

[1612] 6. The device displays suggested prompts to the user, such as "What are some recommended travel destinations?" or "What do I need to prepare for my trip?"

[1613] Prompt Sentence Examples

[1614] Examples of prompts to be input to a generative AI model include:

[1615] The user has entered the word "travel", based on which you want to generate relevant prompt suggestions.

[1616] In this way, users can get relevant prompt suggestions through the chat app, which helps to facilitate a smooth dialogue with the user and provide them with the right information according to their needs.

[1617] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1618] Step 1:

[1619] A user types a word into a chat app.

[1620] A user opens a chat app and inputs a specific word (e.g., "travel") based on the user's interests or questions. The input data is saved in text format on the device.

[1621] Step 2:

[1622] The terminal sends the entered word to the server.

[1623] The terminal sends the word entered by the user to the server. At this time, the terminal converts the input data into an appropriate format (for example, JSON format) and sends it to the server via the network. The input data is the user's input word, "travel."

[1624] Step 3:

[1625] The server parses the words and sends prompts to the generative AI model.

[1626] The server analyzes the words received from the device. It uses natural language processing (NLP) technology to understand the meaning and context of the words. Based on the analysis results, it sends a prompt to the generative AI model (e.g., GPT-3). The specific prompt is as follows:

[1627] The user has entered the word "travel", based on which you want to generate relevant prompt suggestions.

[1628] The input data is the user's input word "travel," and the output data is the prompt sentence sent to the generative AI model.

[1629] Step 4:

[1630] A generative AI model generates candidate prompts.

[1631] The generative AI model generates relevant prompt candidates based on the prompt received from the server. For example, candidates such as "What travel destinations do you recommend?" or "What do I need to prepare for my trip?" are generated. The input data is the prompt sentence sent from the server, and the output data is the generated prompt candidates.

[1632] Step 5:

[1633] The server sends the generated prompt candidates to the terminal.

[1634] The server sends the prompt candidates received from the generative AI model to the terminal. At this time, the server converts the data into an appropriate format and sends it to the terminal over the network. The input data is the prompt candidates received from the generative AI model, and the output data is the prompt candidates sent to the terminal.

[1635] Step 6:

[1636] The terminal displays the prompt candidates to the user.

[1637] The terminal displays the prompt candidates received from the server to the user. The user can refer to the displayed prompt candidates to ask more specific questions or enter more specific input. The input data is the prompt candidates received from the server, and the output data is the prompt candidates displayed to the user.

[1638] (Application example 1)

[1639] Next, a description will be given of Application Example 1 of Form 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."

[1640] Conventional chatbot systems have had difficulty generating appropriate prompt candidates for keywords entered by users, resulting in a lack of smooth dialogue with users. They also had issues with not being able to efficiently suggest relevant content that users were looking for, resulting in a lack of improvement in the user experience.

[1641] 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.

[1642] In this invention, the server includes means for receiving words input by a user and generating prompt candidates via the Web based on the words, means for presenting the generated prompt candidates to the user, and means for generating prompt candidates for related content based on keywords input by the user and suggesting them to the user. This makes it possible to generate appropriate prompt candidates for the keywords input by the user and efficiently suggest related content.

[1643] "User" means any individual or entity that uses the System.

[1644] An "input word" is a specific word or phrase that a user enters into the system.

[1645] "Web" refers to information resources on the Internet, including websites and online databases.

[1646] "Prompt suggestions" are suggestions for what to type next or questions that are generated based on the words the user types.

[1647] "Content" is a general term for information, data, media, etc., and includes text, images, video, audio, etc.

[1648] A "chatbot" is a program that automatically interacts with users.

[1649] A "system" is an entire device or program that is composed of a combination of multiple means and functions.

[1650] A "suggestion" is the act of presenting options or information to a user.

[1651] The system for implementing this invention has a function of receiving user input words, generating prompt candidates via the Web based on the received words, and presenting the generated prompt candidates to the user. It also includes a function of generating prompt candidates for related content based on keywords entered by the user and suggesting them to the user.

[1652] The server receives the words entered by the user and retrieves relevant information from the web based on those words. Specifically, it uses a Python library to send web requests and parse the HTML data. It then extracts prompt candidates from the parsed data and presents them to the user.

[1653] When a user enters the keyword "movie," the server retrieves prompt candidates from the web, such as "What movies do you recommend?" or "What are the latest movie reviews?", and suggests them to the user. This allows the user to efficiently obtain related content.

[1654] Examples of specific prompts include the following:

[1655] The user has entered the keyword "movies". Generate relevant prompt suggestions.

[1656] The system uses a smartphone as hardware and Python as software, which allows it to generate appropriate prompt candidates for keywords entered by users and efficiently suggest related content.

[1657] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1658] Step 1:

[1659] The user enters keywords into the terminal. The entered keywords are sent to the system. The input data is the specific words or phrases that the user is interested in.

[1660] Step 2:

[1661] The server sends a web request based on the received keywords. Specifically, it uses a Python library to access a URL to retrieve related information. The input data is the keyword entered by the user, and the output data is HTML data retrieved from the web.

[1662] Step 3:

[1663] The server analyzes the acquired HTML data. Specifically, it analyzes the HTML data and extracts prompt candidates. The input data is the HTML data acquired from the Web, and the output data is the analyzed prompt candidates.

[1664] Step 4:

[1665] The server presents the parsed prompt candidates to the user. Specifically, the server transmits data for displaying the prompt candidates to the user's terminal. The input data is the parsed prompt candidates, and the output data is the prompt candidates presented to the user.

[1666] Step 5:

[1667] The user selects from the presented prompt candidates or enters a new keyword. The selected prompt candidate or new keyword is sent back to the system. The input data is the prompt candidate or new keyword selected by the user, and the output data is the data sent to the system.

[1668] Step 6:

[1669] The server sends a web request again to retrieve relevant information based on the user's selection or new keywords. The input data is the prompt candidates or new keywords selected by the user, and the output data is the retrieved HTML data.

[1670] Step 7:

[1671] The server analyzes the re-obtained HTML data and generates new prompt candidates. The input data is the re-obtained HTML data, and the output data is the newly generated prompt candidates.

[1672] Step 8:

[1673] The server presents the newly generated prompt candidates to the user. The input data is the newly generated prompt candidates, and the output data is the new prompt candidates that are presented to the user.

[1674] In this way, it is possible to generate appropriate prompt candidates for the keywords entered by the user and efficiently suggest related content.

[1675] Example 2

[1676] Next, a description will be given of Example 2 of Form 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."

[1677] Conventional dialogue systems have difficulty generating appropriate prompt candidates for questions or topics entered by users, which can cause the dialogue to not proceed smoothly. In addition, users often have difficulty deciding what question to ask next, which can disrupt the flow of the dialogue.

[1678] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for receiving input data from a user and generating prompt candidates using a generative AI model based on the input data, means for presenting the generated prompt candidates to the user, and means equipped with a dialogue agent for assisting the dialogue with the user. This allows the server to generate appropriate prompt candidates for the question or topic input by the user and present them to the user, thereby enabling the dialogue to proceed smoothly.

[1679] "User" means any person or entity that uses the System to provide input data and interact with the System.

[1680] "Input data" refers to information such as questions and topics that users provide to the system.

[1681] A "generative AI model" is an artificial intelligence algorithm for generating relevant prompt candidates based on user input data.

[1682] "Prompt candidates" are potential next questions or topics generated by a generative AI model and presented to the user.

[1683] A "server" is a computer system that runs a generative AI model, generates prompt candidates, and presents them to the user.

[1684] A "presentation means" is a method or device for displaying the generated prompt candidates to a user.

[1685] A "dialogue agent" is software that supports dialogue with a user and communicates with the user using generated prompt candidates.

[1686] MODE FOR CARRYING OUT THE INVENTION

[1687] This invention is a system that generates prompt candidates using a generative AI model based on data entered by a user and presents them to the user. Specific embodiments of this system are described below.

[1688] Hardware and software used

[1689] The system uses the following hardware and software:

[1690] Server: A computing system for running the generative AI model and generating prompt candidates. For example, a cloud-based server or a dedicated data center can be used.

[1691] Device: The device on which a user provides input data and receives generated prompt suggestions, such as a smartphone, tablet, or computer.

[1692] Generative AI model: An artificial intelligence algorithm used to generate prompt candidates based on user input data, such as a natural language generation model like OpenAI's GPT-4.

[1693] Dialogue agent: Software that assists in dialogue with the user and communicates with the user using generated prompt suggestions.

[1694] Data processing and calculation

[1695] The server inputs the input data received from the user into the generative AI model, which then generates relevant prompt candidates based on the input data.The generated prompt candidates are sent from the server to the device and presented to the user.

[1696] Specific examples

[1697] For example, if a user types "What are some recommended travel destinations?", the server uses a generative AI model to generate the following possible prompts:

[1698] What do you recommend overseas?

[1699] What are some hidden gems in the country?

[1700] "What's the best place for a family trip?"

[1701] These prompt candidates are presented to the user via the device. The user can use these prompt candidates to formulate their next question. For example, if the prompt candidate "What are your recommendations for overseas travel?" is selected, the user can enter a more specific question such as "What are some recommended travel destinations in Europe?"

[1702] The system provides users with specific guidance to help them navigate the conversation smoothly.

[1703] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1704] Program processing flow

[1705] Step 1:

[1706] The user enters an initial question or theme.

[1707] Input: A user uses a device to provide input data (e.g., "What are some travel recommendations?").

[1708] Specific operation: The user enters a question into the input field on their smartphone or computer and presses the send button.

[1709] Step 2:

[1710] The server receives the user's input data.

[1711] Input: The input data submitted by the user.

[1712] Data processing: The server analyzes the input data and converts it into a format suitable for the generative AI model.

[1713] Output: Data format for input into a generative AI model.

[1714] Specific operation: The server analyzes the received data and performs preprocessing if necessary.

[1715] Step 3:

[1716] The server generates prompt candidates using a generative AI model.

[1717] Input: Preprocessed user input data.

[1718] Data computation: A generative AI model generates prompt candidates based on input data.

[1719] Output: Generated prompt candidates (e.g., "What are some recommended places abroad?", "What are some hidden gems in Japan?").

[1720] How it works: The server calls the generative AI model, which runs its internal algorithms to generate potential prompts.

[1721] Step 4:

[1722] The server sends the generated prompt candidates to the terminal.

[1723] Input: Generated prompt suggestions.

[1724] Data processing: Converting the prompt candidates into a format for presentation to the user.

[1725] Output: Prompt suggestions sent to the terminal.

[1726] Specific operation: The server converts the prompt candidates into an appropriate format and sends them to the terminal over the Internet.

[1727] Step 5:

[1728] The terminal presents the prompt candidates to the user.

[1729] Input: Prompt suggestions sent by the server.

[1730] Data processing: Display prompt candidates in a format that the user can view.

[1731] Output: The suggested prompts that are presented to the user.

[1732] Specific behavior: The device displays prompt candidates on the screen and allows the user to select one.

[1733] Step 6:

[1734] The user inputs the next question based on the suggested prompts.

[1735] Input: Suggested prompt suggestions.

[1736] Data processing: The user uses the suggested prompts to refine the next question.

[1737] Output: The next question (e.g., "What are the best travel destinations in Europe?").

[1738] Specific operation: The user refers to the presented prompt candidates, enters a new question, and presses the submit button.

[1739] In this way, the user can obtain specific guidance for smoothly progressing the dialogue.

[1740] (Application example 2)

[1741] Next, a description will be given of Application Example 2 of Form 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."

[1742] Conventional conversational agent systems provide insufficient guidance for users to find the content they want to watch, and have difficulty providing appropriate prompt candidates for users to ask specific questions. Furthermore, the generated prompt candidates often do not match the user's needs, resulting in a poor user experience.

[1743] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes: means for receiving input words from a user and generating prompt candidates via a network based on the input words; means for presenting the generated prompt candidates to the user; a system equipped with a dialogue agent that assists dialogue with the user; means for providing guidance to help the user find content they want to view; means for generating prompt candidates based on the user's input using a generative AI model; and means for presenting the generated prompt candidates to the user. This enables the user to efficiently find content they want to view.

[1744] "User" means an individual or entity that uses the System to provide input to locate content they wish to view.

[1745] "Input words" are keywords or phrases that a user provides to the system and contain information related to the content they wish to view.

[1746] "Network" means an infrastructure for transmitting and receiving data, including the Internet and other communications means.

[1747] "Prompt candidates" are phrases that are generated based on the user's input words and provide guidance for the user to ask specific questions.

[1748] A "conversational agent" is software designed to assist in a dialogue with a user, generating appropriate responses based on the user's input.

[1749] A "system" is a collection of hardware and software for receiving user input, generating prompt candidates, and assisting in user interaction.

[1750] A "Guide" is advice or suggestions provided to help a user find content they want to watch.

[1751] A "generative AI model" is an algorithm or program that uses artificial intelligence technology to generate prompt candidates based on user input.

[1752] A system for implementing this invention receives user input words, generates prompt candidates based on the input words via a network, and presents the generated prompt candidates to the user. The system is equipped with a dialogue agent that assists in dialogue with the user and provides guidance for the user to find content they want to view.

[1753] The server uses a generative AI model to generate prompt candidates based on the user's input. The generative AI model can use OpenAI's API. Specifically, when the user inputs the content they want to watch, the server receives that input and uses the generative AI model to generate appropriate prompt candidates. The generated prompt candidates are then presented to the user.

[1754] Users access the system using a device such as a smartphone and input the content they want to watch. For example, if a user inputs "I want to watch an action movie," the server uses a generative AI model to generate prompt candidates such as "What are some recent recommended action movies?" or "What are some classic action movies?" and presents them to the user.

[1755] The system uses a smartphone (Android) as its hardware, and Python and OpenAI API as its software. Data processing and data calculation involve a series of processes that receive user input, generate prompt candidates using a generative AI model, and present them to the user.

[1756] For example, if a user types "I want to watch an action movie," the generative AI model will generate prompt candidates such as "What are some recent recommended action movies?" or "What are some classic action movies?" By presenting these prompt candidates to users, they can ask more specific questions and efficiently find the content they want to watch.

[1757] An example of a prompt is as follows:

[1758] What action movies do you recommend?

[1759] What are some popular action movies these days?

[1760] What are some classic action movies?

[1761] In this way, users can efficiently find the content they want to watch.

[1762] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1763] Step 1:

[1764] A user accesses the system using a device such as a smartphone and inputs the input words for the content they want to watch. For example, they might input "I want to watch an action movie." This input word is sent to the system.

[1765] Step 2:

[1766] The server receives the input words sent by the user and sends a request to the generative AI model to generate prompt candidates based on the received input words. At this time, the input words are converted into a format that the generative AI model can understand.

[1767] Step 3:

[1768] The generative AI model generates prompt candidates based on the input words received from the server. For example, for the input words "I want to watch an action movie," it generates prompt candidates such as "What are some recent action movies that you recommend?" or "What are some classic action movies?" The generated prompt candidates are returned to the server.

[1769] Step 4:

[1770] The server receives the prompt candidates returned by the generative AI model and formats them for presentation to the user, who then sends the formatted prompt candidates to the user's device.

[1771] Step 5:

[1772] The user's device receives the prompt candidates sent from the server and presents them to the user, who can then ask a specific question based on the presented prompt candidates.

[1773] Step 6:

[1774] When the user enters a specific question based on the suggested prompts, the question is again sent to the server, which again uses the generative AI model to generate an appropriate answer to the user's question.

[1775] Step 7:

[1776] The generative AI model generates an appropriate answer based on the user's specific question and returns the answer to the server, which then sends the generated answer to the user's device.

[1777] Step 8:

[1778] The user's device receives the response sent from the server and presents it to the user, allowing the user to obtain specific information about the content they wish to view.

[1779] Example 3

[1780] Next, a third embodiment of the third embodiment will be described. 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."

[1781] Conventional chatbot systems have difficulty efficiently generating prompt candidates to generate appropriate responses to user input and selecting the most appropriate one. Furthermore, they lack the means to smoothly advance dialogue with users, which has led to issues with the user experience not being improved.

[1782] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1783] In this invention, the server includes a means for receiving user input words and analyzing data based on the words, a means for sending prompts to the generative AI model based on the analysis results, a means for receiving multiple prompt candidates from the generative AI model, a means for selecting the optimal prompt from the received prompt candidates, a means for presenting the selected prompt candidates to the user, and a system equipped with a chatbot that supports dialogue with the user. This makes it possible to efficiently generate appropriate answers to user input and smoothly progress dialogue with the user.

[1784] "User" means an individual or entity that uses the System to ask questions or interact.

[1785] An "input word" is a word or phrase that represents a question or request that a user enters into the system.

[1786] "Means of analyzing data" refers to techniques and methods for analyzing the words entered by the user and understanding their intent and content.

[1787] A "generative AI model" is an artificial intelligence model that generates appropriate prompt candidates and answers based on user input.

[1788] A "prompt" is a form of instruction or question that is input to a generative AI model.

[1789] "Prompt candidates" are multiple possible answers or suggestions generated by a generative AI model.

[1790] The "means for selecting the best one" refers to the techniques or methods for selecting the most appropriate one from the generated prompt candidates.

[1791] "User presentation means" refers to the technology or method used to display selected prompt candidates and answers to the user.

[1792] A "chatbot" is an automated response system designed to assist in user interactions.

[1793] A "system" is a collection of devices and software that executes a series of processes to receive user input, analyze it, use a generative AI model to generate an appropriate answer, and present it to the user.

[1794] This invention is a system equipped with a chatbot that supports user interaction. The system executes a series of processes: it receives user input, analyzes data based on the input, generates appropriate prompt candidates using a generative AI model, selects the most suitable one, and presents it to the user.

[1795] Hardware and software used

[1796] This system uses the following hardware and software:

[1797] Server: A server with high-performance computing power is used to analyze user input data and send prompts to the generative AI model.

[1798] Terminal: The device used by the user to provide input (e.g., smartphone, PC, etc.).

[1799] Generative AI models: For example, use generative AI models (such as GPT-3) that excel at natural language processing.

[1800] Natural language processing libraries: Parse user input data.

[1801] Data processing and calculation

[1802] When the server receives the user's input words, it analyzes the data using a natural language processing library. Based on the analysis results, it sends a prompt to the generative AI model. The generative AI model generates multiple prompt candidates based on the received prompt. The server selects the most appropriate prompt from the generated candidates and presents it to the user.

[1803] Specific examples

[1804] As a concrete example, we will explain the processing when a user asks, "What are your recommendations for overseas destinations?"

[1805] 1. The user types "What are your recommendations for overseas destinations?" into their device.

[1806] 2. The terminal sends the input data to the server.

[1807] 3. The server analyzes the received input data and sends the following prompt to the generative AI model:

[1808] "A user asks, 'What are your recommendations abroad?' Generate an appropriate answer."

[1809] 4. The generative AI model generates multiple prompt candidates and returns them to the server. For example, it generates an answer such as, "How about Paris, France? You can enjoy visiting museums and gourmet food."

[1810] 5. The server selects the best prompt from the generated prompt candidates and presents it to the user.

[1811] 6. The device displays the received answer to the user.

[1812] In this way, the system can smoothly advance the dialogue with the user. The flow of the identification process in the third embodiment will be described with reference to FIG.

[1813] Step 1:

[1814] The user enters a question.

[1815] The user types "What do you recommend overseas?" into the chat box on the device. The input data is in text format.

[1816] Step 2:

[1817] The device sends input to the server.

[1818] The terminal converts the user's input data into JSON format and sends it to the server via the Internet. The input is {"question": "What do you recommend overseas?"}, and the output is the JSON data sent to the server.

[1819] Step 3:

[1820] The server parses the input.

[1821] The server parses the received JSON data and uses a natural language processing library to understand the intent of the user's question. The input is the user's question data, and the output is the parsed result.

[1822] Step 4:

[1823] The server sends prompts to the generative AI model.

[1824] The server sends a prompt to the generative AI model based on the analysis results. The prompt text is "The user asked, 'What overseas places do you recommend?' Please generate an appropriate answer." The input is the analysis results, and the output is the prompt text sent to the generative AI model.

[1825] Step 5:

[1826] A generative AI model generates candidate prompts.

[1827] The generative AI model generates multiple candidate answers based on the prompt it receives. For example, it generates an answer such as, "How about Paris, France? You can enjoy visiting museums and gourmet food." The input is the prompt sentence, and the output is multiple candidate prompts.

[1828] Step 6:

[1829] The server selects the best prompt candidate.

[1830] The server selects the most appropriate prompt candidate from the multiple generated prompt candidates. The selection criteria are the relevance to the user's question, the specificity of the information, etc. The input is the prompt candidate, and the output is the optimal prompt candidate.

[1831] Step 7:

[1832] The server sends the answer to the user.

[1833] The server converts the selected answers into JSON format and sends it to the user. The input is the best prompt candidates, and the output is the answer data sent to the user.

[1834] Step 8:

[1835] The terminal displays the answer to the user.

[1836] The device analyzes the received response data and displays it in a chat box. For example, it might say, "How about Paris, France? You can enjoy visiting museums and fine cuisine." The input is the response data, and the output is the response displayed to the user.

[1837] (Application example 3)

[1838] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[1839] Conventional food delivery services require users to search and select menu items themselves, making it difficult to receive optimal suggestions based on user preferences and requests. Furthermore, there is a lack of systems that can provide appropriate suggestions through dialogue with users, and there is a need to improve the user experience.

[1840] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means. In this invention, the server includes means for receiving a user's input words and generating prompt candidates based on the words via a communication network, means for presenting the generated prompt candidates to the user, a system equipped with an interactive agent that supports dialogue with the user, means for making optimal suggestions based on the user's preferences and requests, and means for displaying the generated suggestions to the user. This makes it possible to make optimal food delivery suggestions based on the user's input.

[1841] "User input words" refers to text or keywords that a user enters into the system.

[1842] "Communications Network" means the Internet or other network infrastructure for transmitting and receiving data.

[1843] "Prompt candidates" are response candidates that the system generates based on the user's input words and presents to the user.

[1844] A "conversational agent," also known as a chatbot, is a software program designed to assist users in interacting with the agent.

[1845] "User preferences and requests" refer to the individual preferences and requests that a user expresses to the system.

[1846] "Best Suggestion" refers to the most appropriate response or recommendation generated by the system based on the user's input words and preferences or requests.

[1847] A "generated suggestion" is a specific response or recommendation generated by a conversational agent based on user input.

[1848] A system for implementing this invention includes means for receiving a user's input words and generating prompt candidates based on the words via a communication network, means for presenting the generated prompt candidates to the user, a system equipped with an interactive agent that assists in dialogue with the user, means for making optimal suggestions based on the user's preferences and requests, and means for displaying the generated suggestions to the user.

[1849] Hardware and software used

[1850] Hardware: Smartphone, tablet, or PC

[1851] Software: Python, OpenAI API

[1852] Data processing and calculation

[1853] 1. Obtaining user input: A user uses a device such as a smartphone or PC to enter a question or request into the system. For example, they might enter, "What pizza do you recommend?"

[1854] 2. Prompt generation: The server generates an appropriate prompt based on the user's input. For example, it generates a prompt like, "If the user asks, 'What pizza do you recommend?', please provide an appropriate food delivery suggestion."

[1855] 3. Response generation using an AI model: The server uses OpenAI's GPT-3 model to generate a response based on the generated prompt, for example, "How about a Margherita pizza? It features fresh basil and mozzarella cheese."

[1856] 4. Displaying the response: The generated response is displayed on the user's device, allowing the user to place a food delivery order.

[1857] Specific examples

[1858] Prompt Sentence Examples

[1859] User input: "What pizza do you recommend?"

[1860] Generated prompt: "When a user asks, 'What pizza do you recommend?', provide an appropriate food delivery suggestion."

[1861] Specific response examples

[1862] User input: "What pizza do you recommend?"

[1863] The chatbot responds: "How about a Margherita pizza? Featuring fresh basil and mozzarella cheese."

[1864] In this way, the user experience can be improved by making optimal food delivery suggestions based on user input.

[1865] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[1866] Step 1:

[1867] A user uses a terminal to type a question or request to the system, for example, "What pizza do you recommend?" This input is sent to the system.

[1868] Input: The user's question or request (e.g., "What pizza do you recommend?")

[1869] Output: The user's input word sent to the system

[1870] Step 2:

[1871] The server receives the user's input words and generates a prompt based on the words. For example, it generates a prompt like, "If the user asks, 'What pizza do you recommend?', please suggest an appropriate food delivery option."

[1872] Input: User input (e.g., "What pizza do you recommend?")

[1873] Output: The generated prompt (e.g., "When the user asks 'What pizza do you recommend?', provide an appropriate food delivery suggestion.")

[1874] Step 3:

[1875] The server uses the generated prompt to send a request to OpenAI's GPT-3 model and generate a response, such as "How about a Margherita pizza? Featuring fresh basil and mozzarella cheese."

[1876] Input: Generated prompt text (e.g., "When the user asks 'What pizza do you recommend?', provide appropriate food delivery suggestions.")

[1877] Output: The response generated by the GPT-3 model (e.g., "How about a Margherita pizza? It features fresh basil and mozzarella cheese.")

[1878] Step 4:

[1879] The server generates a response and sends it to the user's terminal, which displays the response.

[1880] Input: The response generated by the GPT-3 model (e.g., "How about a Margherita pizza? It features fresh basil and mozzarella cheese.")

[1881] Output: The response displayed on the user's terminal

[1882] Step 5:

[1883] The user reviews the displayed response and places a food delivery order if desired, for example, ordering a suggested pizza.

[1884] Input: User confirmation and order (e.g. suggested pizza order)

[1885] Output: Food delivery order information

[1886] In this way, a system is realized that processes and calculates data based on user input at each step, ultimately providing the user with optimal suggestions.

[1887] 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.

[1888] "Example 1"

[1889] One embodiment of the present invention is a system that includes a means for receiving user input words and generating prompt candidates via the web based on the received words, a means for presenting the generated prompt candidates to the user, and a chatbot that supports dialogue with the user. This system also includes an emotion engine that recognizes the user's emotions. The emotion engine analyzes the user's input words to determine their emotions, and the chatbot uses the results to engage in dialogue with the user. For example, if a user inputs "Today is the worst day," the emotion engine determines from this input that the user is in a bad mood. Based on the analysis results, the chatbot then engages in appropriate dialogue with the user.

[1890] "Example 2"

[1891] In another embodiment of the present invention, an emotion engine tracks changes in a user's emotions, and the chatbot uses these changes to engage in dialogue with the user. For example, if a user first inputs, "Today is the worst day," and then inputs, "But it was fun meeting my friends," the emotion engine tracks the change in the user's emotions from disappointment to joy. Based on this change in emotion, the chatbot then engages in appropriate dialogue with the user.

[1892] "Example 3"

[1893] In yet another embodiment of the present invention, an emotion engine adjusts prompt candidates based on the user's emotions, and the chatbot uses the adjusted prompt candidates to engage in a dialogue with the user. For example, if a user inputs "Today is the worst day," the emotion engine analyzes this input to indicate that the user is in a bad mood. Based on the analysis result, the emotion engine adjusts the prompt candidates to generate more considerate prompt candidates, such as "That's been tough. Maybe talking about it would help." The chatbot then engages in an appropriate dialogue with the user based on the adjusted prompt candidates.

[1894] The processing flow of each embodiment will be described below.

[1895] "Example 1"

[1896] Step 1: Receive an input word from the user.

[1897] Step 2: Analyze the sentiment using the sentiment engine based on the received words.

[1898] Step 3: Generate prompt candidates based on the analysis results.

[1899] Step 4: Present the generated prompt candidates to the user.

[1900] Step 5: The chatbot interacts with the user.

[1901] "Example 2"

[1902] Step 1: Receive an input word from the user.

[1903] Step 2: Analyze the sentiment using the sentiment engine based on the received words.

[1904] Step 3: Track changes in user sentiment based on the analysis results.

[1905] Step 4: Generate prompt candidates based on changes in emotions.

[1906] Step 5: Present the generated prompt candidates to the user.

[1907] Step 6: The chatbot interacts with the user.

[1908] "Example 3"

[1909] Step 1: Receive an input word from the user.

[1910] Step 2: Analyze the sentiment using the sentiment engine based on the received words.

[1911] Step 3: Adjust the prompt candidates based on the analysis results.

[1912] Step 4: Present the tailored prompt candidates to the user.

[1913] Step 5: The chatbot interacts with the user.

[1914] Example 1

[1915] Next, a description will be given of Example 1 of Form 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."

[1916] Conventional chat systems have had issues with generating appropriate prompt candidates based on user input and not being able to take user emotions into account when conducting dialogue, which leads to a poor user experience and limits the system's usefulness.

[1917] 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.

[1918] In this invention, the server includes means for receiving user input words and generating prompt candidates via an information source based on the words, means for presenting the generated prompt candidates to the user, and means including a dialogue agent that is equipped with an emotion engine that analyzes the user's emotions and supports dialogue with the user based on the analysis results. This makes it possible to generate appropriate prompt candidates based on the user's input and to have a dialogue that takes the user's emotions into consideration.

[1919] "User input words" refers to words or phrases that users type into a chat app.

[1920] An "information source" is an external resource from which information can be obtained, such as an internet database or search engine.

[1921] "Prompt candidates" are questions or phrases that the user might type next, generated based on the words the user types.

[1922] An "emotion engine" is software or algorithm that analyzes emotions from user input words.

[1923] A "conversational agent" is a system equipped with artificial intelligence to assist in dialogue with users.

[1924] This invention is a system that, when a user inputs a specific word into a chat app, generates prompt candidates based on that word via information sources, and also analyzes the user's emotions to support the conversation.

[1925] Hardware and software used

[1926] The system uses the following hardware and software:

[1927] Server: Processes data and calls web search APIs.

[1928] Device: The device (smartphone, tablet, PC, etc.) through which a user operates the chat app.

[1929] Web Search APIs: APIs for retrieving data from sources such as Google Search API and Bing Search API.

[1930] Sentiment engine: Software for analyzing user emotions, such as IBM Watson or Microsoft Azure's sentiment analysis API.

[1931] Conversational Agent: A system equipped with artificial intelligence to assist in dialogue with users.

[1932] Program processing

[1933] When a user enters a specific word into a chat app, the device sends the word to a server. The server uses a web search API to retrieve related information based on the received word. It analyzes the retrieved information and generates prompt candidates such as "What are your recommended travel destinations?" or "What do I need to prepare for my trip?" The generated prompt candidates are presented to the user via the device.

[1934] Furthermore, when a user inputs a word containing an emotion, the device sends this word to the server. The server uses an emotion engine to analyze the user's emotion, and the dialogue agent generates an appropriate response based on the results. For example, if a user inputs "Today is the worst day," the emotion engine will analyze that the user is in a bad mood, and the dialogue agent will generate a response such as "Did something bad happen today?"

[1935] Specific examples

[1936] For example, if a user types "travel" into a chat app, the device sends this word to the server, which uses the Google Search API to retrieve information related to "travel" and generates prompt suggestions such as "What are your recommended travel destinations?" or "What do I need to prepare for my trip?" These suggestions are then displayed to the user via the device.

[1937] For example, if a user types "Today is the worst day," the device sends this phrase to the server. The server uses an emotion engine to analyze the user's emotions and recognizes that the user is in a bad mood. As a result, the conversational agent generates a response such as "Did something bad happen today?" and displays it to the user through the device.

[1938] In this way, the system can generate appropriate prompt candidates based on the user's input and further conduct dialogue that takes the user's emotions into account.

[1939] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1940] Step 1:

[1941] A user types a word into a chat app.

[1942] Input: A user types a word, such as "travel," into an input field in a chat app.

[1943] Output: The entered word is sent to the terminal.

[1944] Specific operation: The user types "travel" into a chat app on their smartphone or PC and presses the send button.

[1945] Step 2:

[1946] The terminal sends the entered word to the server.

[1947] Input: The word entered by the user (e.g., "travel").

[1948] Output: The entered word is sent to the server.

[1949] Specific operation: The device sends the entered word to the server using an HTTP request.

[1950] Step 3:

[1951] The server receives the word and retrieves related information using a web search API.

[1952] Input: A word sent from the device (e.g. "travel").

[1953] Output: Relevant information retrieved from the Web Search API.

[1954] Specific operation: The server calls the Google Search API or Bing Search API to obtain information related to "travel."

[1955] Step 4:

[1956] The server generates prompt candidates based on the information it has obtained.

[1957] Input: Relevant information obtained from the Web Search API.

[1958] Output: Generated prompt suggestions (e.g., "What travel destinations do you recommend?", "What do I need to prepare for my trip?").

[1959] What it does: The server analyzes the information it has retrieved and generates possible prompts for the user to enter next.

[1960] Step 5:

[1961] The server generates prompt candidates and sends them to the terminal.

[1962] Input: Generated prompt suggestions.

[1963] Output: Prompt candidates are sent to the terminal.

[1964] Specific operation: The server sends the generated prompt candidates to the device using an HTTP request.

[1965] Step 6:

[1966] The device displays possible prompts to the user.

[1967] Input: Prompt suggestions sent by the server.

[1968] Output: The suggested prompts are displayed to the user.

[1969] Specific operation: The device displays the prompt candidates received on the chat app screen.

[1970] Step 7:

[1971] The user enters a word that contains a sentiment.

[1972] Input: A user types a sentimental word, such as "Today is the worst day," into an input field in a chat app.

[1973] Output: The entered word is sent to the terminal.

[1974] Specific action: A user types "Today is the worst day" into a chat app on their smartphone or PC and presses the send button.

[1975] Step 8:

[1976] The terminal sends words containing emotions to the server.

[1977] Input: A user-entered sentiment word (e.g., "Today is the worst day").

[1978] Output: The entered word is sent to the server.

[1979] Specific operation: The device sends the entered word to the server using an HTTP request.

[1980] Step 9:

[1981] The server analyzes the emotions using an emotion engine.

[1982] Input: A word containing an emotion sent from the device (e.g., "Today is the worst day").

[1983] Output: Parsed emotion information (e.g., displeased).

[1984] Specific operation: The server calls an emotion engine (e.g., emotion analysis API) and analyzes the user's emotions.

[1985] Step 10:

[1986] The server generates the chatbot's response based on the analysis results.

[1987] Input: Parsed emotion information.

[1988] Output: The generated chatbot response (e.g., "Did something bad happen today?").

[1989] Specific operation: The server generates an appropriate response to the user based on the analysis results.

[1990] Step 11:

[1991] The server generates a response and sends it to the terminal.

[1992] Input: The generated chatbot response.

[1993] Output: The response is sent to the terminal.

[1994] Specific operation: The server uses an HTTP request and sends the generated response to the terminal.

[1995] Step 12:

[1996] The terminal displays the response to the user.

[1997] Input: The chatbot response sent from the server.

[1998] Output: The response is displayed to the user.

[1999] Specific operation: The response received by the device is displayed on the chat app screen.

[2000] (Application example 1)

[2001] Next, a description will be given of Application Example 1 of Form 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."

[2002] Conventional chatbot systems have had difficulty generating appropriate prompt candidates in response to user input, making it difficult to conduct dialogue that takes the user's emotions into account. Furthermore, online shopping sites are required to quickly and accurately recommend products that users want, but conventional systems have not been able to fully achieve this.

[2003] 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.

[2004] In this invention, the server includes a means for receiving user input words and generating prompt candidates via the web based on the words; a means for presenting the generated prompt candidates to the user; a means for incorporating an emotion engine that analyzes the user's emotions and a chatbot that interacts with the user based on the analysis results; and a means for suggesting related products based on the user's input words. This allows for the generation of appropriate prompt candidates in response to the user's input and for a dialogue that takes the user's emotions into consideration. It also allows for the prompting of products desired by the user in a timely and accurate manner.

[2005] "User input words" refers to specific words or phrases that a user inputs into the system.

[2006] "Prompt candidates" are questions or suggestions to present to the user that are generated based on the user's input words.

[2007] An "emotion engine" is software or hardware that analyzes emotions from user input words or text.

[2008] A "chatbot" is a program or system that automatically interacts with users.

[2009] "Product suggestion" refers to the act of recommending related products based on the user's input words.

[2010] The system for implementing this invention has the function of receiving user input words, generating prompt candidates via the web based on the words, and presenting them to the user. It also includes an emotion engine that analyzes the user's emotions and a chatbot that interacts with the user based on the analysis results. It also has the function of suggesting related products based on the user's input words.

[2011] Hardware and software used

[2012] Hardware: Smartphone

[2013] Software: Python

[2014] Data processing and calculation

[2015] Prompt Generation

[2016] The server receives the keywords entered by the user, calls a Web API based on the keywords, and generates relevant prompt candidates, which are then presented to the user.

[2017] Emotion analysis

[2018] The server sends the user's input text to the emotion engine API for emotion analysis. The emotion engine extracts emotions from the user's text and returns the results to the server.

[2019] Dialogue Support

[2020] The server then responds to the user with appropriate dialogue based on the results of the emotion analysis. For example, if a user enters "Today is the worst day," the emotion engine will interpret this as meaning that the user is in a bad mood and respond with appropriate dialogue such as "Did anything special happen today?"

[2021] Product suggestion

[2022] The server suggests related products based on the user's input. For example, if a user inputs "I want a new smartphone," the server generates prompt candidates such as "What brand of smartphone do you like?" and "What is your budget?" and presents them to the user.

[2023] Specific examples

[2024] If a user types "I want a new smartphone," the server generates the following prompt candidates and presents them to the user:

[2025] "Which brand of smartphone do you prefer?"

[2026] "What's your budget?"

[2027] "Which features are important?"

[2028] Also, if a user types, "Today is the worst day," the emotion engine will analyze that the user is in a bad mood, and the server will engage in appropriate dialogue such as, "Did anything special happen today?"

[2029] In this way, the system can generate appropriate prompt candidates based on user input, conduct dialogue that takes into account the user's emotions, and quickly and accurately suggest products that the user wants.

[2030] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[2031] Step 1:

[2032] The user inputs a keyword into the terminal.

[2033] Input: The keyword entered by the user (e.g., "I want a new smartphone")

[2034] Output: The device sends the keyword to the server.

[2035] Step 2:

[2036] The server calls a Web API based on the received keywords and generates relevant prompt candidates.

[2037] Input: Keyword sent from the terminal

[2038] Data processing: The server sends keywords to the Web API to obtain related prompt candidates.

[2039] Output: Generated prompt candidates (e.g., "What brand of smartphone do you like?", "What is your budget?")

[2040] Step 3:

[2041] The server transmits the generated prompt candidates to the terminal and presents them to the user.

[2042] Input: Generated prompt suggestions

[2043] Output: The terminal displays the possible prompts to the user.

[2044] Step 4:

[2045] The user inputs text containing emotions into the terminal.

[2046] Input: User-entered text containing emotions (e.g., "Today is the worst day ever").

[2047] Output: The terminal sends text to the server.

[2048] Step 5:

[2049] The server sends the received text to the emotion engine API for emotion analysis.

[2050] Input: Text sent from the terminal

[2051] Data processing: The server sends the text to the emotion engine API and obtains the emotion analysis results.

[2052] Output: Sentiment analysis result (e.g., displeased)

[2053] Step 6:

[2054] The server then engages in appropriate dialogue with the user based on the emotion analysis results.

[2055] Input: Sentiment analysis results

[2056] Data calculation: The server generates appropriate dialogue content based on the results of emotion analysis.

[2057] Output: Generated dialogue (e.g. "Is there anything special happening today?")

[2058] Step 7:

[2059] The server transmits the generated dialogue content to the terminal and presents it to the user.

[2060] Input: Generated dialogue

[2061] Output: The terminal displays the dialogue to the user.

[2062] Step 8:

[2063] The server makes relevant product suggestions based on the user's input words.

[2064] Input: User input word

[2065] Data calculation: The server searches for relevant products based on the input words and generates suggestions.

[2066] Output: Generated product suggestions (e.g., "How about this brand of smartphone?")

[2067] Step 9:

[2068] The server transmits the generated product proposal to the terminal and presents it to the user.

[2069] Input: Generated product suggestions

[2070] Output: The terminal displays the product suggestions to the user.

[2071] Example 2

[2072] Next, a description will be given of Example 2 of Form 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."

[2073] Conventional chatbot systems have difficulty generating appropriate prompt candidates based on user input, which can prevent smooth dialogue with the user. Furthermore, because dialogue is conducted without taking into account changes in the user's emotions, user satisfaction can decline. To solve these problems, a system is needed that can generate appropriate prompt candidates based on the user's input data and track changes in the user's emotions as it engages in dialogue.

[2074] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[2075] In this invention, the server includes a means for receiving user input data and generating prompt candidates using a generative AI model based on the data, a means for presenting the generated prompt candidates to the user, a means for analyzing the user's emotions using an emotion engine that tracks changes in the user's emotions, a means for generating a dialogue with the user based on the analyzed changes in emotions, a means for presenting the generated dialogue to the user, and a system equipped with a chatbot that supports the dialogue with the user. This makes it possible to generate appropriate prompt candidates based on the user's input data and to conduct a dialogue that takes into account changes in the user's emotions.

[2076] "User input data" refers to information such as text or voice that a user inputs into the system through a terminal.

[2077] A "generative AI model" is an artificial intelligence model that generates prompt candidates and dialogues based on user input data.

[2078] A "prompt candidate" is a suggestion that indicates what question or interaction the user should have next.

[2079] An "emotion engine" is software or algorithm for analyzing emotional changes from user input data.

[2080] A "chatbot" is an automated response system that assists in conversations with users.

[2081] A "server" is a computer system that receives user input data and generates prompt candidates and dialogue using generative AI models and emotion engines.

[2082] A "terminal" is a device through which a user sends input data to the system and receives generated prompt suggestions and interactions.

[2083] MODE FOR CARRYING OUT THE INVENTION

[2084] The present invention provides a system for generating appropriate prompt candidates based on user input data, and for conducting a dialogue by tracking changes in the user's emotions. Specific embodiments of this system are described below.

[2085] server

[2086] The server receives the user's input data and generates prompt candidates based on the data using a generative AI model. A typical natural language processing model is used as the generative AI model. The server then sends the generated prompt candidates to the device. The server also uses an emotion engine to track changes in the user's emotions and generates a dialogue with the user based on the analyzed changes in emotions. A typical emotion analysis software is used as the emotion engine.

[2087] Terminal

[2088] The terminal receives input data from the user and sends it to the server. The text data and voice data entered by the user are sent to the server via the Internet. The terminal presents the prompt candidates received from the server and the generated dialogue to the user. This allows the user to check the prompt candidates and dialogue displayed on the screen and make the next input.

[2089] user

[2090] Users interact with the chatbot through their devices. For example, if a user inputs, "Today is the worst day," the information is sent to the server via the device. The server uses an emotion engine to analyze this input data and track down that the user's emotion is bad. If the user then inputs, "But it was fun meeting your friends," the server again analyzes it using its emotion engine and tracks that the user's emotion has changed to joy. Based on this information, the server generates an appropriate dialogue such as, "It was great meeting your friends. What fun things are you planning next?" and presents it to the user via the device.

[2091] Specific examples

[2092] Below are some specific examples of prompt sentences to input to the generative AI model.

[2093] "What travel destinations do you recommend?"

[2094] "What recent movies do you recommend?"

[2095] "Teach me a new recipe?"

[2096] These prompts are suggestions that indicate what questions or interactions the user should have next, providing guidance to help the user smoothly proceed with the interaction.

[2097] In this way, the present invention generates appropriate prompt candidates based on the user's input data, enabling a dialogue that takes into account the user's emotional changes, thereby improving user satisfaction and realizing a more natural and effective dialogue.

[2098] The flow of the identification process in the second embodiment will be described with reference to FIG.

[2099] Program processing flow

[2100] Step 1: User types into terminal

[2101] The user uses the terminal to input text data, for example, "Today is the worst day ever." This input data is then sent to the server in the next step.

[2102] Step 2: The device sends the input data to the server

[2103] The terminal transmits the text data entered by the user to the server via the Internet. The data is encrypted before transmission. The input data is the text "Today is the worst day."

[2104] Step 3: The server generates prompt candidates using the generative AI model

[2105] The server uses a generative AI model to generate prompt candidates based on the received user input data. For example, a natural language processing model is used as the generative AI model to generate a prompt candidate such as "What travel destinations do you recommend?" The input data is "Today is the worst day," and the output data is "What travel destinations do you recommend?"

[2106] Step 4: The server sends the generated prompt candidates to the device

[2107] The server sends the generated prompt candidates to the device. At this time, the data is encrypted again before being sent. The input data is "What travel destinations do you recommend?" and the output data is the prompt candidates sent to the device.

[2108] Step 5: The device presents prompt candidates to the user

[2109] The terminal displays the received prompt candidates to the user. The user can check the prompt candidates displayed on the screen. The input data is "What travel destinations do you recommend?" and the output data is the prompt candidates presented to the user.

[2110] Step 6: The user continues the dialogue based on the suggested prompts

[2111] The user can then ask more specific questions or engage in more specific dialogue based on the suggested prompts. For example, the user can type, "What do you recommend for overseas travel?" The input data is "What do you recommend for overseas travel?", and the output data is sent to the server in the next step.

[2112] Step 7: The server uses the emotion engine to track changes in the user's emotions.

[2113] The server uses an emotion engine to analyze changes in emotions from user input data. For example, from inputs such as "Today is the worst day" and "But it was fun meeting my friends," the server tracks changes in the user's emotions from displeasure to joy. The input data is "Today is the worst day" and "But it was fun meeting my friends," and the output data is the analysis result of the change in emotions.

[2114] Step 8: The server generates appropriate dialogue based on the change in emotion.

[2115] The server generates an appropriate dialogue for the user based on the analysis results of the emotion engine. For example, it generates a dialogue such as, "It was nice to meet your friends. What fun things are you planning to do next?" The input data is the analysis results of changes in emotions, and the output data is the generated dialogue.

[2116] Step 9: The server sends the generated dialogue to the device

[2117] The server transmits the generated dialogue to the terminal. At this time, the data is encrypted before transmission. The input data is the generated dialogue, and the output data is the dialogue transmitted to the terminal.

[2118] Step 10: The terminal presents the generated dialogue to the user

[2119] The terminal displays the received dialogue to the user. The user can confirm the dialogue displayed on the screen and continue the dialogue. The input data is the generated dialogue, and the output data is the dialogue presented to the user.

[2120] (Application example 2)

[2121] Next, a description will be given of Application Example 2 of Form 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."

[2122] Conventional chatbot systems have had the problem of being unable to present appropriate prompt candidates in response to user input, preventing users from smoothly progressing through the dialogue. Furthermore, because the dialogue is conducted without taking into account changes in the user's emotions, there is a risk of user satisfaction decreasing. There is a need to provide a system that can solve these problems and allow users to progress through dialogue more comfortably.

[2123] 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.

[2124] In this invention, the server includes means for receiving a user's input words and generating prompt candidates using a generative AI model based on the words, means for presenting the generated prompt candidates to the user, means for analyzing the user's emotional state using an emotion engine that tracks changes in the user's emotions, and means equipped with a chatbot for conducting appropriate dialogue based on the user's emotional state. This makes it possible to present appropriate prompt candidates in response to the user's input and to conduct dialogue according to the user's emotional state.

[2125] "User input words" are text or keywords that a user inputs into the syste...

Claims

1. A first means for acquiring input words related to meals of a user; a second means for generating a prompt sentence for outputting an appropriate food delivery suggestion to the user based on the input word; A third means for inputting the generated prompt sentence into a generative AI model to generate the appropriate food delivery suggestion. A fourth means for displaying the generated suitable food delivery suggestions on a terminal of the user; A system including:

2. A fifth means for analyzing the user's emotion from the input words using an emotion engine; A sixth means for generating an appropriate dialogue for the user based on the emotion of the user; The system of claim 1 , comprising:

3. a seventh means for analyzing a change in the user's emotion from the input words using the emotion engine; The sixth means generates an appropriate dialogue for the user based on a change in the user's emotion. The system of claim 2.

Images