system

Abstract

To provide a system which reduces load on a person in charge and reduces a time.SOLUTION: A system includes: means for collecting free answers of a questionnaire; means for totalizing and summarizing the collected free answers, using prompt statements instructing generation of a tabulation result and a summary result as a written report and a generative AI model, and generating the written report; means for using a prompt statement instructing generation of a content on the basis of the generated written report and the generative AI model and generating the content; and means for previewing the generated content and receiving editing operation.SELECTED DRAWING: Figure 1

Description

Technical Field

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

Background Art

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

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] [[ID=~35]]There is a problem that since the aggregation and summary of free responses to questionnaires are performed manually, the burden on the person in charge is large and it also takes time.

Means for Solving the Problems

[0005] Provided is a system that collects free responses to questionnaires, automatically aggregates and summarizes them using natural language processing technology, and outputs the results in a form that is easy to visually understand. Thereby, it becomes possible to reduce the burden on the person in charge and shorten the time.

Brief Description of the Drawings

[0006] [Figure 1] It is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11] This is a sequence diagram showing the processing flow of the data processing system in Embodiment 1 of Example 1. [Figure 12] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1 of Form Example 1. [Figure 13] This is a sequence diagram showing the processing flow of the data processing system in Example 2 of Embodiment 2. [Figure 14] This is a sequence diagram showing the processing flow of the data processing system in Application Example 2 of Form Example 2. [Figure 15] This is a sequence diagram showing the processing flow of the data processing system in Embodiment 3 of Example 3. [Figure 16] This is a sequence diagram showing the processing flow of the data processing system in Application Example 3 of Form Example 3. [Figure 17]This is a sequence diagram showing the processing flow of the data processing system in Example 1 of the Form 1 when an emotion engine is combined. [Figure 18] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1 of Form Example 1 when an emotion engine is combined. [Modes for carrying out the invention]

[0007] Hereinafter, an example of an embodiment of the system relating to the technology of this disclosure will be described with reference to the attached drawings.

[0008] First, let's explain the terminology used in the following explanation.

[0009] In the following embodiments, the signed processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Furthermore, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include CPU (Central Processing Unit), GPU (Graphics Processing Unit), GPGPU (General-Purpose computing on Graphics Processing Units), APU (Accelerated Processing Unit), or TPU (TENSOR PROCESSING UNIT®).

[0010] In the following embodiments, signed RAM (Random Access Memory) is a memory that temporarily stores information and is used as work memory by the processor.

[0011] In the following embodiments, the tagged storage is one or more non-volatile storage devices that store various programs, various parameters, and the like. Examples of non-volatile storage devices include flash memories (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0012] In the following embodiments, the tagged communication I / F (Interface) is an interface including a communication processor, an antenna, and the like. The communication I / F controls communication between a plurality of 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), or Bluetooth (registered trademark).

[0013] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B". That is, "A and / or B" means that it may be only A, only B, or a combination of A and B. Also, in this specification, when expressing three or more matters connected by "and / or", the same concept as "A and / or B" is applied.

[0014] [First Embodiment]

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

[0016] As shown in FIG. 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0017] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0018] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0019] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

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

[0021] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

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

[0023] As shown in Figure 2, in the data processing device 12, specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0024] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0025] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

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

[0027] "Example of form 1"

[0028] The system of the present invention includes means for collecting open-ended responses from a questionnaire, means for automatically aggregating and summarizing the collected open-ended responses, and means for outputting the aggregation and summarization results as a report. Specifically, the open-ended responses from the questionnaire are collected using an online questionnaire tool such as Google Forms. The collected open-ended responses are automatically aggregated and summarized using natural language processing technology. This natural language processing technology includes, for example, keyword extraction, topic modeling, and sentiment analysis. The aggregation and summarization results are presented in a visually easy-to-understand format.

[0029] For example, the output can be in the form of graphs or charts. This reduces the workload on the person in charge and saves time.

[0030] The following describes the processing flow for each example of the form.

[0031] "Example of form 1"

[0032] Step 1: Collect open-ended responses from the survey. Specifically, use online survey tools such as Google Forms to collect open-ended responses from users.

[0033] Step 2: Automatically aggregate and summarize the collected open-ended responses. In this step, natural language processing techniques are used to analyze the content of the open-ended responses. Specifically, techniques such as keyword extraction, topic modeling, and sentiment analysis are used to extract the main keywords and phrases from the open-ended responses, and the responses are aggregated and summarized based on these.

[0034] Step 3: Output the aggregated and summarized results as a report. In this step, the aggregated and summarized results obtained in the previous step are output in a visually easy-to-understand format. Specifically, the aggregated and summarized results are displayed in graphs and charts and output as a report. This allows the person in charge to grasp the overview of the open-ended responses simply by looking at the report, thereby reducing the workload on the person in charge and saving time.

[0035] (Example 1)

[0036] Next, we will describe Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0037] Traditional systems had problems such as requiring users to have specialized knowledge when generating programs, and requiring a high level of technical understanding to understand the processing of the generated programs. Furthermore, there was often a lack of concrete examples to understand the specific operation of the generated programs, making it unclear how users should actually use the programs.

[0038] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0039] In this invention, the server includes means for the user to input a prompt, means for generating a program using a generative AI model, means for explaining the processing of the generated program in natural language, and means for adding concrete examples to the explanation. This makes it easier for the user to generate a program and understand its processing content without specialized knowledge, and furthermore, to clearly grasp how to use it in practice through concrete examples.

[0040] A "user" is an entity that uses a system to input prompt messages and receives program generation and explanations.

[0041] A "prompt" is a set of instructions that a user inputs to the system, and it is the text that the generative AI model uses to generate a program.

[0042] A "generative AI model" is an artificial intelligence model that generates program code based on prompt text entered by a user, such as a model that uses natural language processing technology.

[0043] A "program" is the code generated by a generative AI model, consisting of a set of instructions for performing a specific task.

[0044] A "server" is a computer system that receives prompt messages from users, generates programs using a generative AI model, and explains the processing details.

[0045] "Natural language" refers to the language that humans use on a daily basis, and is used to explain the processing content of generated programs in a way that is easy for users to understand.

[0046] A "concrete example" is a specific instance that demonstrates how to actually use the generated program, helping the user understand how the program works.

[0047] This invention relates to a system in which a user inputs a prompt, a generation AI model generates a program, explains the processing content in natural language, and adds concrete examples.

[0048] First, the user accesses the system interface using a terminal and enters a prompt. For example, they might enter a prompt such as, "Generate a program that analyzes the text entered by the user and determines its sentiment."

[0049] Next, the server receives this prompt and generates a program using a generative AI model. This generative AI model, for example, is a model that uses natural language processing technology and generates appropriate program code based on the prompt entered by the user.

[0050] The generated program is written in a programming language such as Python. The server analyzes the processing content of this generated program and explains it in natural language. Specifically, it clearly states what kind of data processing and calculations the program performs. For example, if it uses the NLTK library to analyze the sentiment of text, it will explain the details.

[0051] Furthermore, the server adds concrete examples to the description. For example, if the user enters "Today is a very good day," it specifically shows how the program determines the emotion. This makes it easier for the user to understand the actual behavior of the generated program.

[0052] This system allows users to generate programs without specialized knowledge, easily understand their processing, and clearly grasp their practical usage through concrete examples.

[0053] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0054] Step 1:

[0055] The user enters a prompt message.

[0056] The user accesses the system interface using a terminal and enters prompt messages. For example, they might enter a prompt message such as, "Generate a program that analyzes the text entered by the user and determines its sentiment." The entered prompt message is then sent to the server.

[0057] Step 2:

[0058] The server generates the program using the generated AI model.

[0059] The server inputs the prompt text received from the user into a generative AI model. The generative AI model generates appropriate program code based on the prompt text. For example, a model using natural language processing techniques might generate Python code. The generated program code is stored on the server.

[0060] Step 3:

[0061] The server explains the processing of the generated program in natural language.

[0062] The server analyzes the generated program code and explains its processing in natural language. Specifically, it clearly indicates what kind of data processing and calculations the program performs. For example, if it uses the NLTK library to analyze the sentiment of text, it will explain the details. The explanation is provided to the user.

[0063] Step 4:

[0064] The server adds specific examples to the description.

[0065] The server adds specific examples to the description of the generated program. For example, it specifically shows how the program determines emotion when the user inputs "Today is a very good day." Specific examples are important for demonstrating the actual operation of the program. The description, including the specific examples, is provided to the user.

[0066] The above describes the processing flow of this system's program.

[0067] (Application Example 1)

[0068] Next, we will describe 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 device 14 will be referred to as a "terminal."

[0069] Traditional survey systems often require manual collection, tabulation, and summarization of open-ended responses, resulting in significant time and effort. Furthermore, outputting the results as reports in a visually easy-to-understand format can be challenging. Additionally, creating high-quality ad copy requires specialized knowledge, making it difficult to generate such content quickly. To address these issues, there is a need for a system that efficiently and automatically tabulates and summarizes open-ended responses, generates visually easy-to-understand reports, and further utilizes a generation AI model to generate ad copy based on prompts.

[0070] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0071] In this invention, the server includes means for collecting open-ended responses from a questionnaire, means for automatically aggregating and summarizing the collected open-ended responses, means for outputting the aggregation and summarization results as a report, means for generating ad copy based on prompt sentences using a generation AI model, and means for displaying the generated ad copy. This enables efficient aggregation and summarization of open-ended responses and the generation of visually easy-to-understand reports, as well as the rapid generation of high-quality ad copy.

[0072] "Methods for collecting open-ended responses in surveys" refers to functions for collecting opinions and comments freely written by respondents in surveys as data.

[0073] "Methods for automatically aggregating and summarizing collected open-ended responses" refers to functions that analyze collected open-ended responses, extract key keywords and phrases, and concisely summarize the overall content.

[0074] "Means for outputting aggregated and summarized results as a report" refers to a function for generating and outputting aggregated and summarized data as a report in a visually easy-to-understand format.

[0075] "Means for generating ad copy based on prompt text using a generative AI model" refers to a function that uses a generative AI model to automatically create ad copy based on prompt text entered by the user.

[0076] "Means for displaying generated ad copy" refers to a function for visually displaying ad copy created by a generation AI model to the user.

[0077] The system for implementing this invention includes functions for collecting open-ended responses from questionnaires, automatically aggregating and summarizing the collected responses, and outputting the aggregation and summarization results as a report. It also includes a function for generating advertisement text based on prompt text using a generation AI model and displaying the generated advertisement text.

[0078] Hardware and software configuration

[0079] Hardware:

[0080] server

[0081] User devices (smartphones, smart glasses, etc.)

[0082] software:

[0083] Analysis program using natural language processing technology

[0084] Generative AI models using the OpenAI® API

[0085] Database Management System

[0086] Visualization tools for generating graphs and charts

[0087] Data processing and data calculation

[0088] server:

[0089] 1. The server collects the open-ended responses from the survey submitted by the user's terminal into a database.

[0090] 2. Using natural language processing techniques, the collected open-ended responses are analyzed to extract key keywords and phrases.

[0091] 3. Based on the extracted keywords and phrases, the content of the free-response answers is automatically compiled and summarized.

[0092] 4. Generate a report containing the aggregated and summarized results in a visually easy-to-understand format (e.g., graphs and charts) and send it to the user's terminal.

[0093] User terminal:

[0094] 1. The user enters a prompt and sends it to the generating AI model.

[0095] 2. The server uses the OpenAI API to generate ad text based on the input prompt.

[0096] 3. Display the generated ad copy on the user's device.

[0097] Specific example

[0098] Example of a prompt:

[0099] "Please create ad copy for a new smartphone. Its features include a high-resolution camera and long battery life."

[0100] Example of generated ad copy:

[0101] "A new smartphone has arrived! Capture every beautiful moment with its high-resolution camera. Plus, enjoy worry-free use all day long with its long-lasting battery. Get yours now and enjoy the best experience!"

[0102] In this way, users can simply input prompt text, and the AI ​​generation model will quickly generate high-quality ad copy. Furthermore, open-ended survey responses are efficiently compiled and summarized, and output as a visually easy-to-understand report, thus improving operational efficiency.

[0103] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0104] Step 1:

[0105] The user enters their free-response answers to the survey and sends them from their device to the server.

[0106] Input: User-generated free response

[0107] Output: Free-response data sent to the server

[0108] Specific operation: When a user enters free-form answers into a survey form and presses the submit button, the device sends that data to the server.

[0109] Step 2:

[0110] The server saves the received open-ended responses to a database.

[0111] Input: Free-response data sent from the device.

[0112] Output: Open-ended response data stored in the database

[0113] Specific operation: The server stores and saves the received open-ended response data in a database.

[0114] Step 3:

[0115] The server uses natural language processing technology to analyze the open-ended responses and extract key keywords and phrases.

[0116] Input: Open-ended response data stored in the database

[0117] Output: Extracted keywords and phrases

[0118] Specific operation: The server executes a natural language processing algorithm to extract key keywords and phrases from the open-ended text data.

[0119] Step 4:

[0120] The server automatically aggregates and summarizes the free-response content based on the extracted keywords and phrases.

[0121] Input: Extracted keywords or phrases

[0122] Output: Aggregated and summarized data

[0123] Specific operation: The server aggregates the extracted keywords and phrases and uses a summarization algorithm to concisely summarize the content of the open-ended responses.

[0124] Step 5:

[0125] The server generates a report containing the aggregated and summarized results in a visually easy-to-understand format and sends it to the user's terminal.

[0126] Input: Aggregated and summarized data

[0127] Output: A visually easy-to-understand report (in graph and chart format)

[0128] Specific operation: The server uses a visualization tool to generate aggregated and summarized results as a report in graph and chart format, and sends it to the user's terminal.

[0129] Step 6:

[0130] The user enters a prompt and sends it to the generating AI model.

[0131] Input: User-entered prompt message

[0132] Output: Prompt message sent to the server

[0133] Specific operation: When the user enters a prompt message and presses the send button, the terminal sends that data to the server.

[0134] Step 7:

[0135] The server uses the OpenAI API to generate ad copy based on the input prompt text.

[0136] Input: Prompt message sent by the user

[0137] Output: Generated ad copy

[0138] Specific operation: The server calls the OpenAI API, inputs the prompt text into the AI ​​model that generates the text, and retrieves the generated ad text.

[0139] Step 8:

[0140] The server sends the generated ad text to the user's device, and the user's device displays it.

[0141] Input: Generated ad copy

[0142] Output: Ad text displayed on the user's device

[0143] Specific operation: The server sends the generated ad text to the user's terminal, and the user's terminal displays the received ad text on the screen.

[0144] (Example 2)

[0145] Next, we will describe Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0146] Conventional systems have struggled to provide appropriate responses quickly and accurately to user-inputted prompts. Furthermore, the lack of a visually easy-to-understand format for displaying generated responses resulted in low user convenience. This invention aims to solve these problems by providing a system that offers quick and accurate responses to user-inputted prompts and displays those responses in a visually easy-to-understand format.

[0147] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0148] In this invention, the server includes means for the user to input a prompt sentence, means for the terminal to send the prompt sentence to the server, means for the server to pass the prompt sentence to a generation AI model, means for the generation AI model to analyze the prompt sentence and generate a response, means for the server to return the generated response to the terminal, and means for the terminal to display the response to the user. This makes it possible to provide a quick and accurate response to a prompt sentence entered by the user and to display the response in a visually easy-to-understand format.

[0149] A "user" is an individual or group that uses a system to input prompt messages and receives the generated responses.

[0150] A "prompt message" is a sentence containing a question or instruction that a user enters into the system.

[0151] A "terminal" is a device used by a user to input prompt messages and display the generated response. Examples include personal computers and smartphones.

[0152] A "server" is a computer system that receives prompt messages sent from a terminal, passes them to a generation AI model, and returns the generated response to the terminal.

[0153] A "generative AI model" is an artificial intelligence model that analyzes prompt sentences and generates appropriate responses. Examples include models that utilize natural language processing techniques.

[0154] "Natural language processing technology" is a technique that enables computers to understand and analyze human language. This makes it possible to analyze the content of a prompt and generate an appropriate response.

[0155] A "response" refers to the answer or information that a generative AI model generates by analyzing a prompt sentence.

[0156] A "visually easy-to-understand format" is a format that displays the generated response in a way that is easy for the user to understand. Examples include text, graphs, and charts.

[0157] This invention provides a system that offers a rapid and accurate response to a user-inputted prompt and displays that response in a visually easy-to-understand format. The system includes a user, a terminal, a server, and a generative AI model.

[0158] The user enters a prompt message into the terminal's input field. Examples of prompt messages include "What's the weather like today?" or "Please tell me the latest news." The terminal then sends the user's prompt message to the server. At this time, the prompt message is packaged in JSON format.

[0159] The server parses the received prompt and sends a request to the generative AI model's API endpoint. The generative AI model used may be an advanced model employing natural language processing techniques (e.g., GPT-3® or GPT-4®). The server parses the response received from the generative AI model and returns it to the terminal. At this point, the response is again packaged in JSON format.

[0160] The terminal analyzes the response received from the server and displays it to the user. The display format is visually easy to understand, such as text, graphs, and charts. For example, responses such as "Today's weather is sunny" or "The latest news is as follows" may be displayed.

[0161] This system allows users to obtain appropriate responses to their input prompts using a generative AI model. Furthermore, the responses are displayed in a visually easy-to-understand format, improving user convenience.

[0162] As a concrete example, consider a case where a user enters the prompt "How do I write a Hello World program in Python?". In this case, the generative AI model will generate the response "Here's how to write a Hello World program in Python: python print('Hello, World!')" and display it on the terminal.

[0163] In this way, the invention provides a rapid and accurate response to a prompt message entered by the user and displays that response in a visually easy-to-understand format.

[0164] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0165] Step 1:

[0166] The user enters a prompt message.

[0167] The user enters a prompt message into the terminal's input field. For example, they might type, "What's the weather like today?". The entered prompt message is temporarily stored in the terminal's memory.

[0168] Step 2:

[0169] The terminal sends a prompt message to the server.

[0170] The terminal sends the prompt text entered by the user to the server as an HTTP request. The prompt text is packaged in JSON format. The input is the prompt text, and the output is the HTTP request to the server.

[0171] Step 3:

[0172] The server passes the prompt message to the AI ​​model that generates it.

[0173] The server parses the received prompt and sends a request to the generative AI model's API endpoint. For example, it might send the prompt along with the API key as a POST request. The input is the prompt, and the output is the API request to the generative AI model.

[0174] Step 4:

[0175] The generative AI model analyzes the prompt sentence and generates a response.

[0176] The generative AI model analyzes the received prompt and generates an appropriate response. For example, in response to the prompt "What's the weather like today?", it generates the response "The weather is sunny today." The input is the prompt, and the output is the generated response.

[0177] Step 5:

[0178] The server returns the generated response to the terminal.

[0179] The server analyzes the response received from the generated AI model and returns it to the terminal. At this point, the response is again packaged in JSON format. The input is the generated response, and the output is the HTTP response sent to the terminal.

[0180] Step 6:

[0181] The terminal displays a response to the user.

[0182] The terminal analyzes the response received from the server and displays it to the user. For example, "Today's weather is sunny." might be displayed on the screen. The input is the generated response, and the output is what is displayed to the user.

[0183] (Application Example 2)

[0184] Next, we will describe 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 as a "terminal".

[0185] Traditional survey systems often required manual collection, tabulation, and summarization of open-ended responses, resulting in significant time and effort. Furthermore, outputting the results in a visually easy-to-understand report format was challenging. Additionally, automated ad generation lacked a robust mechanism for generating appropriate ads based on user input. This made ad creation cumbersome and hindered the rapid creation of effective advertisements.

[0186] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0187] In this invention, the server includes means for collecting open-ended responses from a questionnaire, means for automatically aggregating and summarizing the collected open-ended responses, means for outputting the aggregation and summarization results as a report, means for automatically generating advertisements based on prompt text using a generation AI model, means for previewing the generated advertisements, means for customizing the generated advertisements, and means for saving and sharing the generated advertisements. This enables efficient collection, aggregation, and summarization of open-ended responses from questionnaires, and allows for the output of reports in a visually easy-to-understand format. Furthermore, it enables the rapid and automatic generation of effective advertisements based on prompt text, and allows users to easily customize and share them.

[0188] "Methods for collecting open-ended responses in surveys" refers to functions for collecting free-form responses from users.

[0189] "Means for automatically aggregating and summarizing collected open-ended responses" refers to a function that automatically analyzes collected open-ended responses, extracts key information, and summarizes it.

[0190] "Means for outputting aggregated and summarized results as a report" refers to a function for outputting aggregated and summarized results in a report format.

[0191] "A means of automatically generating advertisements based on prompt text using a generative AI model" refers to a function that uses a generative AI model to automatically generate advertisements based on prompt text entered by the user.

[0192] "Means for previewing generated advertisements" refers to a function that allows users to preview the generated advertisements.

[0193] "Means for customizing generated ads" refers to features that allow users to edit and customize generated ads.

[0194] "Means for saving and sharing generated ads" refers to features for saving generated ads and sharing them with other users and the platform.

[0195] The system for carrying out this invention has the following configuration: The system includes means for collecting open-ended responses from a questionnaire, means for automatically aggregating and summarizing the collected open-ended responses, means for outputting the aggregation and summarization results as a report, means for automatically generating advertisements based on prompt text using a generation AI model, means for previewing the generated advertisements, means for customizing the generated advertisements, and means for saving and sharing the generated advertisements.

[0196] Hardware and software configuration

[0197] Hardware: Smartphones, servers

[0198] Software: Generative AI models (e.g., OpenAI's GPT-4), mobile app development frameworks (e.g., React Native), cloud storage services (e.g., AWS® S3)

[0199] Data processing and data calculation

[0200] 1. Collection of open-ended survey responses: Users enter their responses via a smartphone app. These responses are sent to a server and stored in a database.

[0201] 2. Automatic aggregation and summarization of open-ended responses: The server analyzes open-ended responses using natural language processing technology and extracts key keywords and phrases. This allows for the aggregation and summarization of the responses.

[0202] 3. Report Output: The aggregated and summarized results are output as a report in a visually easy-to-understand format (e.g., graphs and charts). This report can be viewed by users on their smartphones.

[0203] 4. Automated ad generation: When a user enters a prompt, that prompt is sent to the server. The generation AI model then generates ad text and images based on the prompt.

[0204] 5. Ad preview display: The generated ads are previewed to the user in the smartphone app.

[0205] 6. Ad customization: Users can edit and customize the generated ads within the app.

[0206] 7. Saving and sharing ads: Completed ads are saved to cloud storage, and a link is generated that can be shared via social media or email.

[0207] Specific example

[0208] For example, if a user wants to create an advertisement to promote the opening of a new cafe, they would enter a prompt message like the following:

[0209] Example prompt: "Create an advertisement to promote the opening of a new cafe. The target audience is young people in their 20s and 30s, and the cafe features organic coffee and a relaxing atmosphere."

[0210] When this prompt is input into the AI ​​model, the model generates advertising text and images like the following.

[0211] Example ad text: "A new cafe, 'Relax Cafe,' has opened! With organic coffee and a relaxing atmosphere, it's perfect for young people in their 20s and 30s. Please stop by!"

[0212] Example of advertising images: A generative AI model generates images of cafe interiors and organic coffee, which are then incorporated into advertisements.

[0213] In this way, users can easily create compelling advertisements and effectively reach their target audience.

[0214] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0215] Step 1:

[0216] Users enter free-form answers to a survey via a smartphone app.

[0217] Input: The user enters free-form text.

[0218] Output: The entered free-response answers are sent to the server.

[0219] Specific operation: When a user enters a free-response answer in the app's text input field and presses the "Submit" button, that data is sent to the server.

[0220] Step 2:

[0221] The server collects open-ended responses and saves them to a database.

[0222] Input: Text data of open-ended responses submitted by users.

[0223] Output: Free-response answers stored in the database.

[0224] Specific operation: The server saves the received open-ended responses to the database and generates a confirmation message when saving is complete.

[0225] Step 3:

[0226] The server uses natural language processing technology to analyze the open-ended responses and extract key keywords and phrases.

[0227] Input: Text data of open-ended responses stored in the database.

[0228] Output: Extracted main keywords and phrases.

[0229] Specific operation: The server executes a natural language processing algorithm to extract key keywords and phrases from the open-ended text.

[0230] Step 4:

[0231] The server performs aggregation and summarization based on the extracted keywords and phrases.

[0232] Input: The main keywords or phrases extracted.

[0233] Output: Aggregated and summarized data.

[0234] Specific operation: The server aggregates the extracted keywords and phrases and generates summary data using a summarization algorithm.

[0235] Step 5:

[0236] The server outputs the aggregated and summarized results as a report in a visually easy-to-understand format.

[0237] Input: Aggregated and summarized data.

[0238] Output: Reports in graph and chart format.

[0239] Specific operation: The server converts aggregated and summarized data into graphs and charts and generates them as reports.

[0240] Step 6:

[0241] The user enters a prompt and sends it to the generating AI model.

[0242] Input: The prompt text entered by the user.

[0243] Output: The prompt text sent to the generating AI model.

[0244] Specific operation: When the user enters a prompt in the app's text input field and presses the "Generate" button, that data is sent to the generating AI model.

[0245] Step 7:

[0246] The generative AI model generates ad text and images based on the prompt text.

[0247] Input: The prompt text sent to the generating AI model.

[0248] Output: Generated ad text and images.

[0249] Specific operation: The generative AI model analyzes the prompt text and generates advertising text and images.

[0250] Step 8:

[0251] The server displays a preview of the generated advertisement to the user.

[0252] Input: Generated ad text and images.

[0253] Output: Ad preview displayed on the user's smartphone.

[0254] Specific operation: The server sends the generated advertisement to the user's smartphone and displays a preview in the app.

[0255] Step 9:

[0256] Users can customize the ads that are generated.

[0257] Input: The ad displayed in preview.

[0258] Output: Customized ads.

[0259] Specific operation: Users can edit and customize ad text and images using the app's editing function.

[0260] Step 10:

[0261] The server saves the customized ad to cloud storage and generates a sharing link.

[0262] Input: Customized ad.

[0263] Output: Advertisements and shared links stored in cloud storage.

[0264] Specific operation: The server saves customized advertisements to cloud storage and generates links that can be shared via social media or email.

[0265] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0266] "Example of form 1"

[0267] This invention relates to a system that collects open-ended responses from questionnaires, automatically aggregates and summarizes the collected responses, and outputs the aggregated and summarized results as a report. As a means of automatically aggregating and summarizing the open-ended responses, this system uses natural language processing technology to analyze the content of the responses and extract key keywords and phrases. Furthermore, this system analyzes the content of the open-ended responses using an emotion engine that recognizes user emotions. This emotion engine extracts user emotions from the text data of the open-ended responses and reflects those emotions in the report. Specifically, if a user gives the open-ended response "This product is very good," the emotion engine extracts the emotion "good" from this response and reflects that emotion in the report. The report output means outputs the aggregated and summarized results along with the emotions extracted by the emotion engine in a visually easy-to-understand format, such as a graph or chart. This allows the person in charge to grasp the overview of the open-ended responses and user emotions simply by looking at the report, thereby reducing the workload and saving time.

[0268] The following describes the processing flow for each example of the form.

[0269] "Example of form 1"

[0270] Step 1: Collect open-ended responses from the survey. Specifically, use a survey tool such as Google Forms to collect open-ended responses from users.

[0271] Step 2: Automatically aggregate and summarize the collected open-ended responses. Specifically, natural language processing technology is used to analyze the content of the open-ended responses and extract key keywords and phrases.

[0272] Step 3: Recognize the user's emotions. Specifically, use an emotion engine to analyze the content of open-ended responses and extract the user's emotions.

[0273] Step 4: Reflect the extracted emotions in the report. Specifically, reflect the emotions extracted by the emotion engine in the report and output the emotions in a visually easy-to-understand format, such as in the form of graphs or charts.

[0274] Step 5: Output the aggregation / summary results and the emotion analysis results as a report. Specifically, integrate the aggregation / summary results and the emotion analysis results and output them as one report. As a result, the person in charge

[0275] can grasp the overview of the free-form answers and the user's emotions just by looking at the report.

[0276] (Example 1)

[0277] Next, Example 1 of Form Example 1 will be described. In the following description, the data processing device 12 is referred to as a "server", and the smart device 14 is referred to as a "terminal".

[0278] In the conventional system, the process for generating an appropriate response to the prompt sentence input by the user is complicated and difficult to perform efficiently. In addition, since the quality and display format of the generated results are not consistent, it may be difficult for the user to understand. As a result, there was a problem that the user experience deteriorated and the utility value of the system was limited.

[0279] The specific processing by the specific processing unit 90 of the data processing device 12 in Example 1 is realized by the following means.

[0280] [[ID=D28]]In this invention, the server includes means for receiving a prompt sentence and performing preprocessing, means for inputting the prompt sentence into a generation AI model and performing inference, and means for postprocessing the generated result. As a result, it becomes possible to consistently and efficiently perform the processes from the preprocessing of the prompt sentence to the postprocessing of the generation result.

[0281] A "user" is a subject who operates the system and inputs a prompt sentence.

[0282] The "terminal" is a device for the user to operate, including a PC, smartphone, etc.

[0283] The "prompt text" is the text of instructions or questions input by the user to the generative AI model.

[0284] The "server" is a computer system that receives, preprocesses, inputs to the generative AI model, makes inferences, post-processes, and transmits the results of the prompt text.

[0285] "Preprocessing" is the process of normalizing and tokenizing the text for the prompt text.

[0286] The "generative AI model" is an artificial intelligence model that makes inferences based on the input prompt text and generates results.

[0287] "Inference" is the process by which the generative AI model generates responses or results based on the prompt text.

[0288] "Postprocessing" is the process of adjusting the format and deleting unnecessary information for the generated results.

[0289] The "result" is the response or information generated by the generative AI model based on the prompt text.

[0290] "Display" is the act of visually providing the results received by the terminal from the server to the user.

[0291] This invention is a system in which the user inputs a prompt text, uses a generative AI model to generate a response or result based on the prompt text, and provides it to the user. The following describes specific embodiments of this system.

[0292] First, the user inputs a prompt text using a terminal (such as a PC or smartphone). An example of a prompt text is "Please think of a new recipe using AI."

[0293] The terminal sends the entered prompt text to the server. HTTP requests are used for communication. The server is a computing system with high computing power, specifically a server equipped with a high-performance GPU.

[0294] The server receives prompt messages sent from the terminal and performs preprocessing. Preprocessing includes text normalization (e.g., converting full-width characters to half-width characters) and tokenization (e.g., splitting sentences into words or phrases). Python libraries (e.g., pandas, numpy) are used for preprocessing.

[0295] Next, the server inputs the pre-processed prompt sentences into the generative AI model. For example, GPT-4 is used as the generative AI model. The model performs inference based on the input prompt sentences and generates results. Tensorflow® or PyTorch is used for the model's inference.

[0296] The generated results are post-processed on the server. Post-processing includes formatting the results (e.g., converting to JSON format) and removing unnecessary information.

[0297] The post-processed results are sent from the server to the terminal. HTTP responses are used for communication. The terminal displays the received results to the user. HTML and CSS are used for display.

[0298] As a concrete example, if a user enters the prompt "Please use AI to come up with a new recipe," the server receives this prompt, performs preprocessing, and inputs it into the AI ​​model. The model generates a new recipe, and the server processes the result and sends it to the terminal. The terminal then displays the generated recipe to the user.

[0299] In this way, the user can obtain a new recipe using the generative AI model. This system can consistently and efficiently perform operations from preprocessing of the prompt text to postprocessing of the generated results, and can provide a high-quality response to the user.

[0300] The flow of the specific process in Example 1 will be described using FIG. 15.

[0301] Step 1:

[0302] The user inputs a prompt text.

[0303] The user inputs "Please think of a new recipe using AI." into the input field of the terminal. The input prompt text is saved in the memory of the terminal.

[0304] Step 2:

[0305] The terminal sends the prompt text to the server.

[0306] The terminal sends the input prompt text to the server as an HTTP request. The input is the prompt text, and the output is the request transmission to the server.

[0307] Step 3:

[0308] The server receives the prompt text and performs preprocessing.

[0309] The server receives the prompt text sent from the terminal. For the received prompt text, text normalization (e.g., converting full-width characters to half-width) and tokenization (e.g., splitting the text into words and phrases) are performed. The input is the prompt text, and the output is the preprocessed text data.

[0310] Step 4:

[0311] The server inputs the prompt text into the generative AI model and performs inference.

[0312] The server inputs pre-processed prompt sentences into a generating AI model (e.g., GPT-4). The model performs inference based on the input prompt sentences and generates results. The input is pre-processed text data, and the output is the generated result (e.g., a new recipe).

[0313] Step 5:

[0314] The server then performs post-processing on the generated results.

[0315] The server performs post-processing on the results obtained from the generated AI model. Post-processing includes formatting the results (e.g., converting to JSON format) and removing unnecessary information. The input is the generated result, and the output is the post-processed data.

[0316] Step 6:

[0317] The server sends the results to the terminal.

[0318] The server sends the post-processed results to the terminal as an HTTP response. The input is the post-processed data, and the output is the response sent to the terminal.

[0319] Step 7:

[0320] The device displays the results to the user.

[0321] The terminal displays the results received from the server to the user. HTML and CSS are used for the display. The input is the response data from the server, and the output is the information visually presented to the user.

[0322] (Application Example 1)

[0323] Next, we will describe 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 device 14 will be referred to as a "terminal."

[0324] Traditional survey systems often involve manual collection, tabulation, summarization, and report generation of open-ended responses, resulting in time-consuming and labor-intensive processes. Furthermore, generating advertising content requires specialized knowledge, making it difficult to reach target audiences quickly and effectively. To address these challenges, a system integrating automated tabulation and summarization of open-ended responses with automated advertising content generation is needed.

[0325] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0326] In this invention, the server includes means for collecting open-ended responses to a questionnaire, prompts for aggregating and summarizing the collected open-ended responses and generating a report of the aggregated and summarized results, and means for generating the report using a generation AI model, prompts for generating content based on the generated report and means for generating the content using the generation AI model, means for previewing the generated content and accepting editing operations, means for publishing the edited content, and means for analyzing the performance of the published content. This enables efficient aggregation and summarization of open-ended responses, and automatic generation and publication of advertising content based on prompts.

[0327] "Means for collecting open-ended responses in surveys" refers to interfaces and functions for collecting responses from users in a free-form format.

[0328] "Means for automatically aggregating and summarizing collected open-ended responses" refers to algorithms or software that analyze collected open-ended response data, extract key information, and summarize it.

[0329] "Means for outputting aggregated and summarized results as a report" refers to a function for generating and outputting aggregated and summarized data as a report in a visually easy-to-understand format.

[0330] "Means for entering prompt text" refers to a text input interface for users to give instructions to the generated AI model.

[0331] "Means of generating content based on prompt text using a generative AI model" refers to algorithms and software that enable a generative AI model to automatically generate content such as text and images based on input prompt text.

[0332] "Means for previewing and editing generated content" refers to interfaces and functions that allow users to review generated content and edit it as needed.

[0333] "Means for publishing generated content" refers to functions for publishing generated and edited content on internet platforms, social networking services, etc.

[0334] "Means of analyzing the performance of published content" refers to tools and software used to collect and analyze performance data such as the number of views and clicks on published content.

[0335] The system for implementing this invention has the following configuration. First, the user inputs free-form answers to a questionnaire about a specific product using a terminal such as a smartphone. The terminal is equipped with means for collecting free-form answers to the questionnaire and transmits the free-form answers entered by the user to a server.

[0336] The server is equipped with a means to automatically aggregate and summarize the collected open-ended responses. This means analyzes the content of the open-ended responses using natural language processing techniques and extracts key keywords and phrases. Specifically, it uses natural language processing libraries (e.g., NLTK, spaCy) to analyze the text data and extract important information. The server is also equipped with a means to output the aggregated and summarized results as a report. This means outputs the aggregated and summarized results in a visually easy-to-understand format, such as graphs and charts. This uses data visualization tools (e.g., Matplotlib, D3.js). In this embodiment, the server generates a report using a prompt message that instructs the server to aggregate and summarize the collected open-ended responses and generate the aggregated and summarized results as a report, and a generation AI model.

[0337] Furthermore, the user provides instructions to the generating AI model using prompts that instruct it to generate content based on the generated report, thereby generating the content. An example of a prompt message is: "Enter the generated report and, using this report as a reference, create an advertisement promoting a new eco-friendly detergent to housewives in their 30s." In this example, the specific product is a detergent.

[0338] The server is equipped with a means for generating content based on prompt text using a generative AI model. This means uses a generative AI model (e.g., OpenAI GPT-4) to generate advertising text and images based on the input prompt text.

[0339] The generated content is provided to the user through means of previewing and editing on their device. Users can review the generated content and edit it as needed. The edited content is then published to internet platforms and social media via the server.

[0340] Finally, the server has a means to analyze the performance of the published content. This means uses tools (e.g., Google Analytics) to collect and analyze performance data such as the number of views and clicks on the published content.

[0341] In this way, efficient aggregation and summarization of open-ended responses, as well as the automatic generation and publication of advertising content based on prompt text, become possible. Alternatively, the content may be modified using a prompt message instructing it to change based on the analyzed performance, along with a generative AI model. This allows the content to be modified in order to improve its performance.

[0342] The flow of a specific process in Application Example 1 will be explained using Figure 16.

[0343] Step 1:

[0344] Users input free-response answers to a survey using a smartphone or other device. The entered free-response answers are sent from the device to the server. The input data is free-response text, and the output is the free-response data sent to the server.

[0345] Step 2:

[0346] The server automatically aggregates and summarizes the collected open-ended responses. It analyzes the content of the open-ended responses using natural language processing techniques and extracts key keywords and phrases. The input data is the open-ended response text, and the output is summarized data containing the key keywords and phrases. Specifically, it analyzes the text data using natural language processing libraries (e.g., NLTK, spaCy). Alternatively, it generates aggregated and summarized results of the collected open-ended responses using a prompt message instructing the server to aggregate and summarize the collected open-ended responses, along with a generative AI model.

[0347] Step 3:

[0348] The server outputs the aggregated and summarized results as a report. The aggregated and summarized results are output in a visually easy-to-understand format, such as graphs or charts. The input data is summarized data, and the output is a visually represented report. Data visualization tools (e.g., Matplotlib, D3.js) are used. Specifically, the report is generated using a prompt message that instructs the server to generate a report based on the generated aggregated and summarized results, and a generating AI model.

[0349] Step 4:

[0350] The user provides instructions to the generating AI model by entering prompts. The input data consists of reports and prompts, and the output consists of reports and prompts sent to the server. An example of a prompt is to input a generated report along with the instruction, "Using this report as a reference, create an advertisement promoting a new eco-friendly detergent to housewives in their 30s."

[0351] Step 5:

[0352] The server generates content based on prompt text using a generative AI model. The input data is the prompt text, and the output is the generated advertisement text and images. A generative AI model (e.g., OpenAI GPT-4) is used to generate content based on the input prompt text.

[0353] Step 6:

[0354] The generated content is provided to the user through means of previewing and editing on the device. The user reviews the generated content and edits it as needed. The input data is the generated content, and the output is the edited content.

[0355] Step 7:

[0356] The edited content is published to internet platforms and social media via a server. The input data is the edited content, and the output is the published content.

[0357] Step 8:

[0358] The server analyzes the performance of published content. It collects and analyzes performance data such as the number of views and clicks on published content. The input data is performance data, and the output is the analysis results. Data analysis tools (e.g., Google Analytics) are used.

[0359] (Example 2)

[0360] Next, we will describe Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0361] Traditional survey systems often involved manual collection, tabulation, and summarization of open-ended responses, resulting in significant time and effort. Furthermore, generating specific programs required programming knowledge, making it difficult for users without specialized expertise. Additionally, the quality and efficiency of the generated program code were often not guaranteed.

[0362] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting free-response answers from a questionnaire, means for automatically aggregating and summarizing the collected free-response answers, means for outputting the aggregation and summarization results as a report, means for the user to input prompt sentences, means for the generating AI model to generate program code based on the prompt sentences, and means for returning the generated program code to the user. This enables efficient aggregation and summarization of free-response answers and allows the user to easily generate high-quality program code.

[0363] A "survey" is a set of questions designed to collect specific information.

[0364] An "open-ended response" is a type of answer in which the respondent writes freely in their own words.

[0365] "Means of collection" refers to the methods and devices used to collect data.

[0366] "Means of automatic aggregation and summarization" refers to methods and devices for mechanically organizing collected data and extracting important information.

[0367] "Means of outputting as a report" refers to methods or devices for providing aggregated and summarized results in document format.

[0368] "User" refers to an individual or group that uses the system.

[0369] A "prompt message" is text used to input specific instructions or questions to a generative AI model.

[0370] A "generative AI model" is an artificial intelligence model that generates program code or other output based on input prompts.

[0371] "Program code" is a set of instructions that a computer executes.

[0372] A "server" is a computer system that provides data and services over a network.

[0373] This invention relates to a system that efficiently collects, compiles, and summarizes open-ended responses from questionnaires, and further enables users to easily generate program code. Specific embodiments of this system are described below.

[0374] First, users enter free-form responses to a survey through the system interface. These responses are collected by the server. The server uses natural language processing technology to automatically aggregate and summarize the collected responses. Specifically, the server extracts key keywords and phrases and then aggregates and summarizes the data based on these.

[0375] Next, the server outputs the aggregated and summarized results as a report. This report is provided in a visually easy-to-understand format, such as graphs and charts. This allows users to intuitively grasp the content of the open-ended responses.

[0376] Furthermore, this system provides a means for the user to input prompts. The user inputs specific instructions or questions to the generating AI model as prompts. For example, the user might input a prompt such as, "Generate a program that counts specific words from a text file."

[0377] The server receives prompt messages from the user and passes them to the generative AI model. The generative AI model generates program code based on the prompt messages. The generated program code is returned to the user via the server. The user can then execute this program code on their own device.

[0378] The following hardware and software will be used to implement this system. The server is a computer system equipped with a high-performance processor and sufficient memory, and features an NVIDIA GPU. For software, deep learning frameworks such as TensorFlow and PyTorch will be used. This will allow the generative AI model to operate efficiently and perform data processing and calculations based on prompts entered by the user.

[0379] As a concrete example, consider the case where a user inputs the following prompt into the AI ​​model:

[0380] Example of a prompt:

[0381] "Generate a program that counts specific words from a text file."

[0382] Based on this prompt, the generative AI model generates Python code and returns it to the user via the server. The user can then run this code on their device to check the frequency of specific words.

[0383] As described above, this system provides an efficient way to process open-ended responses in surveys and a means for users to easily generate program code.

[0384] The flow of the specific processing in Example 2 will be explained using Figure 17.

[0385] Step 1:

[0386] The user enters a prompt message.

[0387] The user inputs prompts for the generating AI model through the system interface. For example, they might input, "Generate a program that counts specific words from a text file." The entered prompts are then sent to the server.

[0388] Step 2:

[0389] The server receives the prompt message and passes it to the generating AI model.

[0390] The server receives prompt messages sent by the user. It then makes appropriate API calls to pass the received prompt messages to the generative AI model. The server may also convert the prompt messages into a format that the generative AI model can understand. The input is a prompt message, and the output is a request to the generative AI model.

[0391] Step 3:

[0392] The generative AI model generates program code based on the prompt statement.

[0393] The generative AI model analyzes the received prompt and generates program code based on it. For example, if a user inputs "Generate a program that counts specific words from a text file," the generative AI model will generate Python code. The input is the prompt, and the output is the generated program code.

[0394] Step 4:

[0395] The server returns the generated program code to the user.

[0396] The server receives the program code returned from the generated AI model and returns it to the user. It provides the code in a user-friendly format (e.g., a text file or text in a code editor). The input is the generated program code, and the output is the transmission of the program code to the user.

[0397] Step 5:

[0398] The user executes the generated program code.

[0399] The user executes program code received from the server on their own terminal. For example, if Python code is received, the user executes the code in their Python environment and checks the result. The input is the generated program code, and the output is the result of the program execution. Specifically, the user opens their Python environment, copies the received code, and executes it. As a result, the frequency of occurrence of a particular word is displayed.

[0400] (Application Example 2)

[0401] Next, we will describe 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 as a "terminal".

[0402] Traditional survey systems suffered from the problem of requiring a great deal of time and effort to collect, compile, and summarize open-ended responses. Furthermore, despite advancements in content generation technology using generative AI models, methods for applying this to survey systems had not been established. Additionally, there was a lack of functionality to automatically generate and display content based on user-inputted prompts.

[0403] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0404] In this invention, the server includes means for collecting open-ended responses from a questionnaire, means for automatically aggregating and summarizing the collected open-ended responses, means for outputting the aggregation and summarization results as a report, means for inputting prompt sentences, means for generating content based on the prompt sentences using a generative AI model, and means for displaying the generated content. This enables efficient collection, aggregation, and summarization of open-ended responses from questionnaires, and further enables content generation and display using a generative AI model.

[0405] "Means for collecting open-ended responses from surveys" refers to devices or software that allow users to input answers in a free-form format and collect that information as data.

[0406] "Means for automatically aggregating and summarizing collected open-ended responses" refers to a device or software that analyzes collected open-ended response data, extracts key keywords and phrases, and concisely summarizes the overall content.

[0407] "Means for outputting aggregated and summarized results as a report" refers to a device or software for generating and outputting aggregated and summarized data as a report in a visually easy-to-understand format.

[0408] "Means for inputting prompt text" refers to a device or software for a user to input text to give instructions to a generated AI model.

[0409] "Means for generating content based on prompt text using a generative AI model" refers to a device or software that uses a generative AI model to automatically generate content such as text and images based on input prompt text.

[0410] "Means for displaying generated content" refers to a device or software for visually displaying content generated by a generative AI model to a user.

[0411] The system for implementing this invention has the following configuration. First, the user inputs free-response answers to a questionnaire using a terminal such as a smartphone. The terminal sends the inputted free-response answers to the server. The server uses natural language processing technology to automatically aggregate and summarize the collected free-response answers. Specifically, it extracts key keywords and phrases and concisely summarizes the overall content.

[0412] Next, the server generates a report containing the aggregated and summarized results, outputting it in a visually easy-to-understand format, such as graphs and charts. This allows users to intuitively grasp the survey results.

[0413] Furthermore, the user provides instructions to the generative AI model by entering prompt text. An example of a prompt text is "The story of a brave knight battling a dragon." The server uses the generative AI model to generate content based on this prompt text. The generated content is in the form of text, images, etc., and is displayed visually to the user.

[0414] The following hardware and software are required to implement this system. Hardware includes smartphones and servers. Software includes Python and the OpenAI API. Python is a programming language for implementing natural language processing techniques, and the OpenAI API is an interface for using generative AI models.

[0415] As a concrete example, if a user enters the prompt "A story about a brave knight fighting a dragon," the server will use a generative AI model to generate a story like the following.

[0416] Once upon a time, there was a brave knight named Arthur. Arthur decided to fight a fearsome dragon to protect his kingdom. Reaching the dragon's lair, Arthur drew his sword and confronted the dragon. After a fierce battle, Arthur defeated the dragon and brought peace to the kingdom.

[0417] In this way, users can enjoy generating and displaying content using generative AI models.

[0418] The flow of a specific process in Application Example 2 will be explained using Figure 18.

[0419] Step 1:

[0420] Users enter free-form responses to a survey using a device such as a smartphone. The entered responses are sent from the device to the server. The input data is in text format and is sent to the server.

[0421] Step 2:

[0422] The server analyzes the received open-ended response data using natural language processing techniques. Specifically, it extracts key keywords and phrases and summarizes the overall content concisely. The input data is the open-ended response text, and the output data is a summarized text containing the key keywords and phrases.

[0423] Step 3:

[0424] The server generates a report containing the aggregated and summarized results. The report is output in a visually easy-to-understand format, such as graphs and charts. The input data is a summarized text, while the output data is a visual report.

[0425] Step 4:

[0426] The user provides instructions to the generated AI model by entering prompts. These prompts are entered in text format and sent from the terminal to the server. The input data is the text of the prompts.

[0427] Step 5:

[0428] The server generates content based on prompt text using a generative AI model. The generative AI model automatically generates content such as text and images based on the input prompt text. The input data is the text of the prompt, and the output data is the generated content.

[0429] Step 6:

[0430] The server visually displays the generated content to the user. The generated content is sent to the terminal and displayed on the user's screen. The input data is the generated content, and the output data is the content displayed on the user's screen.

[0431] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0432] Data generation model 58 is a form of so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0433] Other examples of generative AI include Gemini® (registered trademark) (Internet search). <url: https: gemini.google.com ?hl="ja">) are some examples.

[0434] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0435] [Second Embodiment]

[0436] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0437] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0438] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0439] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0440] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0441] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0442] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0443] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0444] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0445] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0446] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0447] Next, the identification process performed by the identification processing unit 290 of the data processing device 12 will be described.

[0448] "Example of form 1"

[0449] The system of the present invention includes means for collecting open-ended responses from a questionnaire, means for automatically aggregating and summarizing the collected open-ended responses, and means for outputting the aggregation and summarization results as a report. Specifically, the open-ended responses from the questionnaire are collected using an online questionnaire tool such as Google Forms. The collected open-ended responses are automatically aggregated and summarized using natural language processing technology. This natural language processing technology includes, for example, keyword extraction, topic modeling, and sentiment analysis. The aggregation and summarization results are output in a visually easy-to-understand format, such as graphs or charts. This reduces the workload on the person in charge and saves time.

[0450] The following describes the processing flow for each example of the form.

[0451] "Example of form 1"

[0452] Step 1: Collect open-ended responses from the survey. Specifically, use online survey tools such as Google Forms to collect open-ended responses from users.

[0453] Step 2: Automatically aggregate and summarize the collected open-ended responses. In this step, natural language processing techniques are used to analyze the content of the open-ended responses. Specifically, techniques such as keyword extraction, topic modeling, and sentiment analysis are used to extract the main keywords and phrases from the open-ended responses, and the responses are aggregated and summarized based on these.

[0454] Step 3: Output the aggregated and summarized results as a report. In this step, the aggregated and summarized results obtained in the previous step are output in a visually easy-to-understand format. Specifically, the aggregated and summarized results are displayed in graphs and charts and output as a report. This allows the person in charge to grasp the overview of the open-ended responses simply by looking at the report, thereby reducing the workload on the person in charge and saving time.

[0455] (Example 1)

[0456] Next, we will describe Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0457] Traditional systems had problems such as requiring users to have specialized knowledge when generating programs, and requiring a high level of technical understanding to understand the processing of the generated programs. Furthermore, there was often a lack of concrete examples to understand the specific operation of the generated programs, making it unclear how users should actually use the programs.

[0458] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0459] In this invention, the server includes means for the user to input a prompt, means for generating a program using a generative AI model, means for explaining the processing of the generated program in natural language, and means for adding concrete examples to the explanation. This makes it easier for the user to generate a program and understand its processing content without specialized knowledge, and furthermore, to clearly grasp how to use it in practice through concrete examples.

[0460] A "user" is an entity that uses a system to input prompt messages and receives program generation and explanations.

[0461] A "prompt" is a set of instructions that a user inputs to the system, and it is the text that the generative AI model uses to generate a program.

[0462] A "generative AI model" is an artificial intelligence model that generates program code based on prompt text entered by a user, such as a model that uses natural language processing technology.

[0463] A "program" is the code generated by a generative AI model, consisting of a set of instructions for performing a specific task.

[0464] A "server" is a computer system that receives prompt messages from users, generates programs using a generative AI model, and explains the processing details.

[0465] "Natural language" refers to the language that humans use on a daily basis, and is used to explain the processing content of generated programs in a way that is easy for users to understand.

[0466] A "concrete example" is a specific instance that demonstrates how to actually use the generated program, helping the user understand how the program works.

[0467] This invention relates to a system in which a user inputs a prompt, a generation AI model generates a program, explains the processing content in natural language, and adds concrete examples.

[0468] First, the user accesses the system interface using a terminal and enters a prompt. For example, they might enter a prompt such as, "Generate a program that analyzes the text entered by the user and determines its sentiment."

[0469] Next, the server receives this prompt and generates a program using a generative AI model. This generative AI model, for example, is a model that uses natural language processing technology and generates appropriate program code based on the prompt entered by the user.

[0470] The generated program is written in a programming language such as Python. The server analyzes the processing content of this generated program and explains it in natural language. Specifically, it clearly states what kind of data processing and calculations the program performs. For example, if it uses the NLTK library to analyze the sentiment of text, it will explain the details.

[0471] Furthermore, the server adds concrete examples to the description. For example, if the user enters "Today is a very good day," it specifically shows how the program determines the emotion. This makes it easier for the user to understand the actual behavior of the generated program.

[0472] This system allows users to generate programs without specialized knowledge, easily understand their processing, and clearly grasp their practical usage through concrete examples.

[0473] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0474] Step 1:

[0475] The user enters a prompt message.

[0476] The user accesses the system interface using a terminal and enters prompt messages. For example, they might enter a prompt message such as, "Generate a program that analyzes the text entered by the user and determines its sentiment." The entered prompt message is then sent to the server.

[0477] Step 2:

[0478] The server generates the program using the generated AI model.

[0479] The server inputs the prompt text received from the user into a generative AI model. The generative AI model generates appropriate program code based on the prompt text. For example, a model using natural language processing techniques might generate Python code. The generated program code is stored on the server.

[0480] Step 3:

[0481] The server explains the processing of the generated program in natural language.

[0482] The server analyzes the generated program code and explains its processing in natural language. Specifically, it clearly indicates what kind of data processing and calculations the program performs. For example, if it uses the NLTK library to analyze the sentiment of text, it will explain the details. The explanation is provided to the user.

[0483] Step 4:

[0484] The server adds specific examples to the description.

[0485] The server adds specific examples to the description of the generated program. For example, it specifically shows how the program determines emotion when the user inputs "Today is a very good day." Specific examples are important for demonstrating the actual operation of the program. The description, including the specific examples, is provided to the user.

[0486] The above describes the processing flow of this system's program.

[0487] (Application Example 1)

[0488] Next, we will describe Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0489] Traditional survey systems often require manual collection, tabulation, and summarization of open-ended responses, resulting in significant time and effort. Furthermore, outputting the results as reports in a visually easy-to-understand format can be challenging. Additionally, creating high-quality ad copy requires specialized knowledge, making it difficult to generate such content quickly. To address these issues, there is a need for a system that efficiently and automatically tabulates and summarizes open-ended responses, generates visually easy-to-understand reports, and further utilizes a generation AI model to generate ad copy based on prompts.

[0490] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0491] In this invention, the server includes means for collecting open-ended responses from a questionnaire, means for automatically aggregating and summarizing the collected open-ended responses, means for outputting the aggregation and summarization results as a report, means for generating ad copy based on prompt sentences using a generation AI model, and means for displaying the generated ad copy. This enables efficient aggregation and summarization of open-ended responses and the generation of visually easy-to-understand reports, as well as the rapid generation of high-quality ad copy.

[0492] "Methods for collecting open-ended responses in surveys" refers to functions for collecting opinions and comments freely written by respondents in surveys as data.

[0493] "Methods for automatically aggregating and summarizing collected open-ended responses" refers to functions that analyze collected open-ended responses, extract key keywords and phrases, and concisely summarize the overall content.

[0494] "Means for outputting aggregated and summarized results as a report" refers to a function for generating and outputting aggregated and summarized data as a report in a visually easy-to-understand format.

[0495] "Means for generating ad copy based on prompt text using a generative AI model" refers to a function that uses a generative AI model to automatically create ad copy based on prompt text entered by the user.

[0496] "Means for displaying generated ad copy" refers to a function for visually displaying ad copy created by a generation AI model to the user.

[0497] The system for implementing this invention includes functions for collecting open-ended responses from questionnaires, automatically aggregating and summarizing the collected responses, and outputting the aggregation and summarization results as a report. It also includes a function for generating advertisement text based on prompt text using a generation AI model and displaying the generated advertisement text.

[0498] Hardware and software configuration

[0499] Hardware:

[0500] server

[0501] User devices (smartphones, smart glasses, etc.)

[0502] software:

[0503] Analysis program using natural language processing technology

[0504] Generative AI model using the OpenAI API

[0505] Database Management System

[0506] Visualization tools for generating graphs and charts

[0507] Data processing and data calculation

[0508] server:

[0509] 1. The server collects the open-ended responses from the survey submitted by the user's terminal into a database.

[0510] 2. Using natural language processing techniques, the collected open-ended responses are analyzed to extract key keywords and phrases.

[0511] 3. Based on the extracted keywords and phrases, the content of the free-response answers is automatically compiled and summarized.

[0512] 4. Generate a report containing the aggregated and summarized results in a visually easy-to-understand format (e.g., graphs and charts) and send it to the user's terminal.

[0513] User terminal:

[0514] 1. The user enters a prompt and sends it to the generating AI model.

[0515] 2. The server uses the OpenAI API to generate ad text based on the input prompt.

[0516] 3. Display the generated ad copy on the user's device.

[0517] Specific example

[0518] Example of a prompt:

[0519] "Please create ad copy for a new smartphone. Its features include a high-resolution camera and long battery life."

[0520] Example of generated ad copy:

[0521] "A new smartphone has arrived! Capture every beautiful moment with its high-resolution camera. Plus, enjoy worry-free use all day long with its long-lasting battery. Get yours now and enjoy the best experience!"

[0522] In this way, users can simply input prompt text, and the AI ​​generation model will quickly generate high-quality ad copy. Furthermore, open-ended survey responses are efficiently compiled and summarized, and output as a visually easy-to-understand report, thus improving operational efficiency.

[0523] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0524] Step 1:

[0525] The user enters their free-response answers to the survey and sends them from their device to the server.

[0526] Input: User-generated free response

[0527] Output: Free-response data sent to the server

[0528] Specific operation: When a user enters free-form answers into a survey form and presses the submit button, the device sends that data to the server.

[0529] Step 2:

[0530] The server saves the received open-ended responses to a database.

[0531] Input: Free-response data sent from the device.

[0532] Output: Open-ended response data stored in the database

[0533] Specific operation: The server stores and saves the received open-ended response data in a database.

[0534] Step 3:

[0535] The server uses natural language processing technology to analyze the open-ended responses and extract key keywords and phrases.

[0536] Input: Open-ended response data stored in the database

[0537] Output: Extracted keywords and phrases

[0538] Specific operation: The server executes a natural language processing algorithm to extract key keywords and phrases from the open-ended text data.

[0539] Step 4:

[0540] The server automatically aggregates and summarizes the free-response content based on the extracted keywords and phrases.

[0541] Input: Extracted keywords or phrases

[0542] Output: Aggregated and summarized data

[0543] Specific operation: The server aggregates the extracted keywords and phrases and uses a summarization algorithm to concisely summarize the content of the open-ended responses.

[0544] Step 5:

[0545] The server generates a report containing the aggregated and summarized results in a visually easy-to-understand format and sends it to the user's terminal.

[0546] Input: Aggregated and summarized data

[0547] Output: A visually easy-to-understand report (in graph and chart format)

[0548] Specific operation: The server uses a visualization tool to generate aggregated and summarized results as a report in graph and chart format, and sends it to the user's terminal.

[0549] Step 6:

[0550] The user enters a prompt and sends it to the generating AI model.

[0551] Input: User-entered prompt message

[0552] Output: Prompt message sent to the server

[0553] Specific operation: When the user enters a prompt message and presses the send button, the terminal sends that data to the server.

[0554] Step 7:

[0555] The server uses the OpenAI API to generate ad copy based on the input prompt text.

[0556] Input: Prompt message sent by the user

[0557] Output: Generated ad copy

[0558] Specific operation: The server calls the OpenAI API, inputs the prompt text into the AI ​​model that generates the text, and retrieves the generated ad text.

[0559] Step 8:

[0560] The server sends the generated ad text to the user's device, and the user's device displays it.

[0561] Input: Generated ad copy

[0562] Output: Ad text displayed on the user's device

[0563] Specific operation: The server sends the generated ad text to the user's terminal, and the user's terminal displays the received ad text on the screen.

[0564] (Example 2)

[0565] Next, we will describe Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0566] Conventional systems have struggled to provide appropriate responses quickly and accurately to user-inputted prompts. Furthermore, the lack of a visually easy-to-understand format for displaying generated responses resulted in low user convenience. This invention aims to solve these problems by providing a system that offers quick and accurate responses to user-inputted prompts and displays those responses in a visually easy-to-understand format.

[0567] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0568] In this invention, the server includes means for the user to input a prompt sentence, means for the terminal to send the prompt sentence to the server, means for the server to pass the prompt sentence to a generation AI model, means for the generation AI model to analyze the prompt sentence and generate a response, means for the server to return the generated response to the terminal, and means for the terminal to display the response to the user. This makes it possible to provide a quick and accurate response to a prompt sentence entered by the user and to display the response in a visually easy-to-understand format.

[0569] A "user" is an individual or group that uses a system to input prompt messages and receives the generated responses.

[0570] A "prompt message" is a sentence containing a question or instruction that a user enters into the system.

[0571] A "terminal" is a device used by a user to input prompt messages and display the generated response. Examples include personal computers and smartphones.

[0572] A "server" is a computer system that receives prompt messages sent from a terminal, passes them to a generation AI model, and returns the generated response to the terminal.

[0573] A "generative AI model" is an artificial intelligence model that analyzes prompt sentences and generates appropriate responses. Examples include models that utilize natural language processing techniques.

[0574] "Natural language processing technology" is a technique that enables computers to understand and analyze human language. This makes it possible to analyze the content of a prompt and generate an appropriate response.

[0575] A "response" refers to the answer or information that a generative AI model generates by analyzing a prompt sentence.

[0576] A "visually easy-to-understand format" is a format that displays the generated response in a way that is easy for the user to understand. Examples include text, graphs, and charts.

[0577] This invention provides a system that offers a rapid and accurate response to a user-inputted prompt and displays that response in a visually easy-to-understand format. The system includes a user, a terminal, a server, and a generative AI model.

[0578] The user enters a prompt message into the terminal's input field. Examples of prompt messages include "What's the weather like today?" or "Please tell me the latest news." The terminal then sends the user's prompt message to the server. At this time, the prompt message is packaged in JSON format.

[0579] The server parses the received prompt and sends a request to the generative AI model's API endpoint. The generative AI model used is, for example, an advanced model employing natural language processing techniques (e.g., GPT-3 or GPT-4). The server parses the response received from the generative AI model and returns it to the terminal. At this point, the response is again packaged in JSON format.

[0580] The terminal analyzes the response received from the server and displays it to the user. The display format is visually easy to understand, such as text, graphs, and charts. For example, responses such as "Today's weather is sunny" or "The latest news is as follows" may be displayed.

[0581] This system allows users to obtain appropriate responses to their input prompts using a generative AI model. Furthermore, the responses are displayed in a visually easy-to-understand format, improving user convenience.

[0582] As a concrete example, consider a case where a user enters the prompt "How do I write a Hello World program in Python?". In this case, the generative AI model will generate the response "Here's how to write a Hello World program in Python: python print('Hello, World!')" and display it on the terminal.

[0583] In this way, the invention provides a rapid and accurate response to a prompt message entered by the user and displays that response in a visually easy-to-understand format.

[0584] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0585] Step 1:

[0586] The user enters a prompt message.

[0587] The user enters a prompt message into the terminal's input field. For example, they might type, "What's the weather like today?". The entered prompt message is temporarily stored in the terminal's memory.

[0588] Step 2:

[0589] The terminal sends a prompt message to the server.

[0590] The terminal sends the prompt text entered by the user to the server as an HTTP request. The prompt text is packaged in JSON format. The input is the prompt text, and the output is the HTTP request to the server.

[0591] Step 3:

[0592] The server passes the prompt message to the AI ​​model that generates it.

[0593] The server parses the received prompt and sends a request to the generative AI model's API endpoint. For example, it might send the prompt along with the API key as a POST request. The input is the prompt, and the output is the API request to the generative AI model.

[0594] Step 4:

[0595] The generative AI model analyzes the prompt sentence and generates a response.

[0596] The generative AI model analyzes the received prompt and generates an appropriate response. For example, in response to the prompt "What's the weather like today?", it generates the response "The weather is sunny today." The input is the prompt, and the output is the generated response.

[0597] Step 5:

[0598] The server returns the generated response to the terminal.

[0599] The server analyzes the response received from the generated AI model and returns it to the terminal. At this point, the response is again packaged in JSON format. The input is the generated response, and the output is the HTTP response sent to the terminal.

[0600] Step 6:

[0601] The terminal displays a response to the user.

[0602] The terminal analyzes the response received from the server and displays it to the user. For example, "Today's weather is sunny." might be displayed on the screen. The input is the generated response, and the output is what is displayed to the user.

[0603] (Application Example 2)

[0604] Next, we will describe application example 2 of form example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 as the "terminal".

[0605] Traditional survey systems often required manual collection, tabulation, and summarization of open-ended responses, resulting in significant time and effort. Furthermore, outputting the results in a visually easy-to-understand report format was challenging. Additionally, automated ad generation lacked a robust mechanism for generating appropriate ads based on user input. This made ad creation cumbersome and hindered the rapid creation of effective advertisements.

[0606] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0607] In this invention, the server includes means for collecting open-ended responses from a questionnaire, means for automatically aggregating and summarizing the collected open-ended responses, means for outputting the aggregation and summarization results as a report, means for automatically generating advertisements based on prompt text using a generation AI model, means for previewing the generated advertisements, means for customizing the generated advertisements, and means for saving and sharing the generated advertisements. This enables efficient collection, aggregation, and summarization of open-ended responses from questionnaires, and allows for the output of reports in a visually easy-to-understand format. Furthermore, it enables the rapid and automatic generation of effective advertisements based on prompt text, and allows users to easily customize and share them.

[0608] "Methods for collecting open-ended responses in surveys" refers to functions for collecting free-form responses from users.

[0609] "Means for automatically aggregating and summarizing collected open-ended responses" refers to a function that automatically analyzes collected open-ended responses, extracts key information, and summarizes it.

[0610] "Means for outputting aggregated and summarized results as a report" refers to a function for outputting aggregated and summarized results in a report format.

[0611] "A means of automatically generating advertisements based on prompt text using a generative AI model" refers to a function that uses a generative AI model to automatically generate advertisements based on prompt text entered by the user.

[0612] "Means for previewing generated advertisements" refers to a function that allows users to preview the generated advertisements.

[0613] "Means for customizing generated ads" refers to features that allow users to edit and customize generated ads.

[0614] "Means for saving and sharing generated ads" refers to features for saving generated ads and sharing them with other users and the platform.

[0615] The system for carrying out this invention has the following configuration: The system includes means for collecting open-ended responses from a questionnaire, means for automatically aggregating and summarizing the collected open-ended responses, means for outputting the aggregation and summarization results as a report, means for automatically generating advertisements based on prompt text using a generation AI model, means for previewing the generated advertisements, means for customizing the generated advertisements, and means for saving and sharing the generated advertisements.

[0616] Hardware and software configuration

[0617] Hardware: Smartphones, servers

[0618] Software: Generative AI models (e.g., OpenAI's GPT-4), mobile app development frameworks (e.g., React Native), cloud storage services (e.g., AWS S3)

[0619] Data processing and data calculation

[0620] 1. Collection of open-ended survey responses: Users enter their responses via a smartphone app. These responses are sent to a server and stored in a database.

[0621] 2. Automatic aggregation and summarization of open-ended responses: The server analyzes open-ended responses using natural language processing technology and extracts key keywords and phrases. This allows for the aggregation and summarization of the responses.

[0622] 3. Report Output: The aggregated and summarized results are output as a report in a visually easy-to-understand format (e.g., graphs and charts). This report can be viewed by users on their smartphones.

[0623] 4. Automated ad generation: When a user enters a prompt, that prompt is sent to the server. The generation AI model then generates ad text and images based on the prompt.

[0624] 5. Ad preview display: The generated ads are previewed to the user in the smartphone app.

[0625] 6. Ad customization: Users can edit and customize the generated ads within the app.

[0626] 7. Saving and sharing ads: Completed ads are saved to cloud storage, and a link is generated that can be shared via social media or email.

[0627] Specific example

[0628] For example, if a user wants to create an advertisement to promote the opening of a new cafe, they would enter a prompt message like the following:

[0629] Example prompt: "Create an advertisement to promote the opening of a new cafe. The target audience is young people in their 20s and 30s, and the cafe features organic coffee and a relaxing atmosphere."

[0630] When this prompt is input into the AI ​​model, the model generates advertising text and images like the following.

[0631] Example ad text: "A new cafe, 'Relax Cafe,' has opened! With organic coffee and a relaxing atmosphere, it's perfect for young people in their 20s and 30s. Please stop by!"

[0632] Example of advertising images: A generative AI model generates images of cafe interiors and organic coffee, which are then incorporated into advertisements.

[0633] In this way, users can easily create compelling advertisements and effectively reach their target audience.

[0634] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0635] Step 1:

[0636] Users enter free-form answers to a survey via a smartphone app.

[0637] Input: The user enters free-form text.

[0638] Output: The entered free-response answers are sent to the server.

[0639] Specific operation: When a user enters a free-response answer in the app's text input field and presses the "Submit" button, that data is sent to the server.

[0640] Step 2:

[0641] The server collects open-ended responses and saves them to a database.

[0642] Input: Text data of open-ended responses submitted by users.

[0643] Output: Free-response answers stored in the database.

[0644] Specific operation: The server saves the received open-ended responses to the database and generates a confirmation message when saving is complete.

[0645] Step 3:

[0646] The server uses natural language processing technology to analyze the open-ended responses and extract key keywords and phrases.

[0647] Input: Text data of open-ended responses stored in the database.

[0648] Output: Extracted main keywords and phrases.

[0649] Specific operation: The server executes a natural language processing algorithm to extract key keywords and phrases from the open-ended text.

[0650] Step 4:

[0651] The server performs aggregation and summarization based on the extracted keywords and phrases.

[0652] Input: The main keywords or phrases extracted.

[0653] Output: Aggregated and summarized data.

[0654] Specific operation: The server aggregates the extracted keywords and phrases and generates summary data using a summarization algorithm.

[0655] Step 5:

[0656] The server outputs the aggregated and summarized results as a report in a visually easy-to-understand format.

[0657] Input: Aggregated and summarized data.

[0658] Output: Reports in graph and chart format.

[0659] Specific operation: The server converts aggregated and summarized data into graphs and charts and generates them as reports.

[0660] Step 6:

[0661] The user enters a prompt and sends it to the generating AI model.

[0662] Input: The prompt text entered by the user.

[0663] Output: The prompt text sent to the generating AI model.

[0664] Specific operation: When the user enters a prompt in the app's text input field and presses the "Generate" button, that data is sent to the generating AI model.

[0665] Step 7:

[0666] The generative AI model generates ad text and images based on the prompt text.

[0667] Input: The prompt text sent to the generating AI model.

[0668] Output: Generated ad text and images.

[0669] Specific operation: The generative AI model analyzes the prompt text and generates advertising text and images.

[0670] Step 8:

[0671] The server displays a preview of the generated advertisement to the user.

[0672] Input: Generated ad text and images.

[0673] Output: Ad preview displayed on the user's smartphone.

[0674] Specific operation: The server sends the generated advertisement to the user's smartphone and displays a preview in the app.

[0675] Step 9:

[0676] Users can customize the ads that are generated.

[0677] Input: The ad displayed in preview.

[0678] Output: Customized ads.

[0679] Specific operation: Users can edit and customize ad text and images using the app's editing function.

[0680] Step 10:

[0681] The server saves the customized ad to cloud storage and generates a sharing link.

[0682] Input: Customized ad.

[0683] Output: Advertisements and shared links stored in cloud storage.

[0684] Specific operation: The server saves customized advertisements to cloud storage and generates links that can be shared via social media or email.

[0685] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0686] "Example of form 1"

[0687] This invention relates to a system that collects open-ended responses from questionnaires, automatically aggregates and summarizes the collected responses, and outputs the aggregated and summarized results as a report. As a means of automatically aggregating and summarizing the open-ended responses, this system uses natural language processing technology to analyze the content of the responses and extract key keywords and phrases. Furthermore, this system analyzes the content of the open-ended responses using an emotion engine that recognizes user emotions. This emotion engine extracts user emotions from the text data of the open-ended responses and reflects those emotions in the report. Specifically, if a user gives the open-ended response "This product is very good," the emotion engine extracts the emotion "good" from this response and reflects that emotion in the report. The report output means outputs the aggregated and summarized results along with the emotions extracted by the emotion engine in a visually easy-to-understand format, such as a graph or chart. This allows the person in charge to grasp the overview of the open-ended responses and user emotions simply by looking at the report, thereby reducing the workload and saving time.

[0688] The following describes the processing flow for each example of the form.

[0689] "Example of form 1"

[0690] Step 1: Collect open-ended responses from the survey. Specifically, use a survey tool such as Google Forms to collect open-ended responses from users.

[0691] Step 2: Automatically aggregate and summarize the collected open-ended responses. Specifically, natural language processing technology is used to analyze the content of the open-ended responses and extract key keywords and phrases.

[0692] Step 3: Recognize the user's emotions. Specifically, use an emotion engine to analyze the content of open-ended responses and extract the user's emotions.

[0693] Step 4: Reflect the extracted emotions in the report. Specifically, the emotions extracted by the emotion engine are reflected in the report and output in a visually easy-to-understand format, such as a graph or chart.

[0694] Step 5: Output the aggregated / summarized results and sentiment analysis results as a report. Specifically, the aggregated / summarized results and sentiment analysis results are integrated and output as a single report. This allows the person in charge to understand the overview of the open-ended responses and the users' sentiments simply by looking at the report.

[0695] (Example 1)

[0696] Next, we will describe Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0697] In conventional systems, the process of generating appropriate responses to user-inputted prompts was complex and difficult to perform efficiently. Furthermore, the quality and display format of the generated results were inconsistent, sometimes making them difficult for users to understand. This resulted in a degraded user experience and limited the system's usefulness.

[0698] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0699] In this invention, the server includes means for receiving prompt statements and performing preprocessing, means for inputting prompt statements into a generation AI model and performing inference, and means for post-processing the generated results. This makes it possible to consistently and efficiently perform everything from preprocessing prompt statements to post-processing of the generated results.

[0700] A "user" is the entity that operates the system and enters prompt messages.

[0701] A "terminal" is a device used by a user to operate something, and includes PCs, smartphones, and other similar devices.

[0702] A "prompt" is the text of instructions or questions that a user inputs to the generated AI model.

[0703] A "server" is a computing system that receives prompt messages, performs preprocessing, inputs data into a generative AI model, performs inference, performs postprocessing, and sends the results.

[0704] "Preprocessing" refers to the process of normalizing and tokenizing the text of a prompt message.

[0705] A "generative AI model" is an artificial intelligence model that performs inference based on input prompt sentences and generates results.

[0706] "Inference" is the process by which a generative AI model generates responses or results based on a prompt.

[0707] "Post-processing" refers to the process of formatting the generated results and removing unnecessary information.

[0708] "Results" refer to the responses and information generated by the generative AI model based on the prompt text.

[0709] "Display" refers to the act of providing the user with a visual representation of the results received by the terminal from the server.

[0710] This invention is a system in which a user inputs a prompt, and a generation AI model generates a response or result based on that prompt, which is then provided to the user. Specific embodiments of this system are described below.

[0711] First, the user enters a prompt using a device (such as a PC or smartphone). An example of a prompt is, "Please use AI to come up with a new recipe."

[0712] The terminal sends the entered prompt text to the server. HTTP requests are used for communication. The server is a computing system with high computing power, specifically a server equipped with a high-performance GPU.

[0713] The server receives prompt messages sent from the terminal and performs preprocessing. Preprocessing includes text normalization (e.g., converting full-width characters to half-width characters) and tokenization (e.g., splitting sentences into words or phrases). Python libraries (e.g., pandas, numpy) are used for preprocessing.

[0714] Next, the server inputs the pre-processed prompt sentences into the generative AI model. For example, GPT-4 is used as the generative AI model. The model performs inference based on the input prompt sentences and generates results. TensorFlow or PyTorch is used for the model's inference.

[0715] The generated results are post-processed on the server. Post-processing includes formatting the results (e.g., converting to JSON format) and removing unnecessary information.

[0716] The post-processed results are sent from the server to the terminal. HTTP responses are used for communication. The terminal displays the received results to the user. HTML and CSS are used for display.

[0717] As a concrete example, if a user enters the prompt "Please use AI to come up with a new recipe," the server receives this prompt, performs preprocessing, and inputs it into the AI ​​model. The model generates a new recipe, and the server processes the result and sends it to the terminal. The terminal then displays the generated recipe to the user.

[0718] In this way, users can obtain new recipes using the generative AI model. This system can efficiently handle everything from pre-processing of prompts to post-processing of the generated results, providing users with high-quality responses.

[0719] The flow of the specific processing in Example 1 will be explained using Figure 15.

[0720] Step 1:

[0721] The user enters a prompt message.

[0722] The user enters "Use AI to come up with a new recipe" into the terminal's input field. The entered prompt is saved to the terminal's memory.

[0723] Step 2:

[0724] The terminal sends a prompt message to the server.

[0725] The terminal sends the entered prompt text to the server as an HTTP request. The input is the prompt text, and the output is the request sent to the server.

[0726] Step 3:

[0727] The server receives the prompt message and performs preprocessing.

[0728] The server receives prompt messages sent from the terminal. It then performs text normalization (e.g., converting full-width characters to half-width characters) and tokenization (e.g., splitting sentences into words and phrases) on the received prompt messages. The input is the prompt message, and the output is pre-processed text data.

[0729] Step 4:

[0730] The server inputs prompt messages into the generated AI model and performs inference.

[0731] The server inputs pre-processed prompt sentences into a generating AI model (e.g., GPT-4). The model performs inference based on the input prompt sentences and generates results. The input is pre-processed text data, and the output is the generated result (e.g., a new recipe).

[0732] Step 5:

[0733] The server then performs post-processing on the generated results.

[0734] The server performs post-processing on the results obtained from the generated AI model. Post-processing includes formatting the results (e.g., converting to JSON format) and removing unnecessary information. The input is the generated result, and the output is the post-processed data.

[0735] Step 6:

[0736] The server sends the results to the terminal.

[0737] The server sends the post-processed results to the terminal as an HTTP response. The input is the post-processed data, and the output is the response sent to the terminal.

[0738] Step 7:

[0739] The device displays the results to the user.

[0740] The terminal displays the results received from the server to the user. HTML and CSS are used for the display. The input is the response data from the server, and the output is the information visually presented to the user.

[0741] (Application Example 1)

[0742] Next, we will describe Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0743] Traditional survey systems often involve manual collection, tabulation, summarization, and report generation of open-ended responses, resulting in time-consuming and labor-intensive processes. Furthermore, generating advertising content requires specialized knowledge, making it difficult to reach target audiences quickly and effectively. To address these challenges, a system integrating automated tabulation and summarization of open-ended responses with automated advertising content generation is needed.

[0744] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0745] In this invention, the server includes means for collecting open-ended responses to a questionnaire, prompts for aggregating and summarizing the collected open-ended responses and generating a report of the aggregated and summarized results, and means for generating the report using a generation AI model, prompts for generating content based on the generated report and means for generating the content using the generation AI model, means for previewing the generated content and accepting editing operations, means for publishing the edited content, and means for analyzing the performance of the published content. This enables efficient aggregation and summarization of open-ended responses, and automatic generation and publication of advertising content based on prompts.

[0746] "Means for collecting open-ended responses in surveys" refers to interfaces and functions for collecting responses from users in a free-form format.

[0747] "Means for automatically aggregating and summarizing collected open-ended responses" refers to algorithms or software that analyze collected open-ended response data, extract key information, and summarize it.

[0748] "Means for outputting aggregated and summarized results as a report" refers to a function for generating and outputting aggregated and summarized data as a report in a visually easy-to-understand format.

[0749] "Means for entering prompt text" refers to a text input interface for users to give instructions to the generated AI model.

[0750] "Means of generating content based on prompt text using a generative AI model" refers to algorithms and software that enable a generative AI model to automatically generate content such as text and images based on input prompt text.

[0751] "Means for previewing and editing generated content" refers to interfaces and functions that allow users to review generated content and edit it as needed.

[0752] "Means for publishing generated content" refers to functions for publishing generated and edited content on internet platforms, social networking services, etc.

[0753] "Means of analyzing the performance of published content" refers to tools and software used to collect and analyze performance data such as the number of views and clicks on published content.

[0754] The system for implementing this invention has the following configuration. First, the user inputs free-form answers to a questionnaire about a specific product using a terminal such as a smartphone. The terminal is equipped with means for collecting free-form answers to the questionnaire and transmits the free-form answers entered by the user to a server.

[0755] The server is equipped with a means to automatically aggregate and summarize the collected open-ended responses. This means analyzes the content of the open-ended responses using natural language processing techniques and extracts key keywords and phrases. Specifically, it uses natural language processing libraries (e.g., NLTK, spaCy) to analyze the text data and extract important information. The server is also equipped with a means to output the aggregated and summarized results as a report. This means outputs the aggregated and summarized results in a visually easy-to-understand format, such as graphs and charts. This uses data visualization tools (e.g., Matplotlib, D3.js). In this embodiment, the server generates a report using a prompt message that instructs the server to aggregate and summarize the collected open-ended responses and generate the aggregated and summarized results as a report, and a generation AI model.

[0756] Furthermore, the user provides instructions to the generating AI model using prompts that instruct it to generate content based on the generated report, thereby generating the content. An example of a prompt message is: "Enter the generated report and, using this report as a reference, create an advertisement promoting a new eco-friendly detergent to housewives in their 30s." In this example, the specific product is a detergent.

[0757] The server is equipped with a means for generating content based on prompt text using a generative AI model. This means uses a generative AI model (e.g., OpenAI GPT-4) to generate advertising text and images based on the input prompt text.

[0758] The generated content is provided to the user through means of previewing and editing on their device. Users can review the generated content and edit it as needed. The edited content is then published to internet platforms and social media via the server.

[0759] Finally, the server has a means to analyze the performance of the published content. This means uses tools (e.g., Google Analytics) to collect and analyze performance data such as the number of views and clicks on the published content.

[0760] In this way, efficient aggregation and summarization of open-ended responses, as well as the automatic generation and publication of advertising content based on prompt text, become possible. Alternatively, the content may be modified using a prompt message instructing it to change based on the analyzed performance, along with a generative AI model. This allows the content to be modified in order to improve its performance.

[0761] The flow of a specific process in Application Example 1 will be explained using Figure 16.

[0762] Step 1:

[0763] Users input free-response answers to a survey using a smartphone or other device. The entered free-response answers are sent from the device to the server. The input data is free-response text, and the output is the free-response data sent to the server.

[0764] Step 2:

[0765] The server automatically aggregates and summarizes the collected open-ended responses. It analyzes the content of the open-ended responses using natural language processing techniques and extracts key keywords and phrases. The input data is the open-ended response text, and the output is summarized data containing the key keywords and phrases. Specifically, it analyzes the text data using natural language processing libraries (e.g., NLTK, spaCy). Alternatively, it generates aggregated and summarized results of the collected open-ended responses using a prompt message instructing the server to aggregate and summarize the collected open-ended responses, along with a generative AI model.

[0766] Step 3:

[0767] The server outputs the aggregated and summarized results as a report. The aggregated and summarized results are output in a visually easy-to-understand format, such as graphs or charts. The input data is summarized data, and the output is a visually represented report. Data visualization tools (e.g., Matplotlib, D3.js) are used. Specifically, the report is generated using a prompt message that instructs the server to generate a report based on the generated aggregated and summarized results, and a generating AI model.

[0768] Step 4:

[0769] The user provides instructions to the generating AI model by entering prompts. The input data consists of reports and prompts, and the output consists of reports and prompts sent to the server. An example of a prompt is to input a generated report along with the instruction, "Using this report as a reference, create an advertisement promoting a new eco-friendly detergent to housewives in their 30s."

[0770] Step 5:

[0771] The server generates content based on prompt text using a generative AI model. The input data is the prompt text, and the output is the generated advertisement text and images. A generative AI model (e.g., OpenAI GPT-4) is used to generate content based on the input prompt text.

[0772] Step 6:

[0773] The generated content is provided to the user through means of previewing and editing on the device. The user reviews the generated content and edits it as needed. The input data is the generated content, and the output is the edited content.

[0774] Step 7:

[0775] The edited content is published to internet platforms and social media via a server. The input data is the edited content, and the output is the published content.

[0776] Step 8:

[0777] The server analyzes the performance of published content. It collects and analyzes performance data such as the number of views and clicks on published content. The input data is performance data, and the output is the analysis results. Data analysis tools (e.g., Google Analytics) are used.

[0778] (Example 2)

[0779] Next, we will describe Example 2 of Form Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0780] Traditional survey systems often involved manual collection, tabulation, and summarization of open-ended responses, resulting in significant time and effort. Furthermore, generating specific programs required programming knowledge, making it difficult for users without specialized expertise. Additionally, the quality and efficiency of the generated program code were often not guaranteed.

[0781] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting free-response answers from a questionnaire, means for automatically aggregating and summarizing the collected free-response answers, means for outputting the aggregation and summarization results as a report, means for the user to input prompt sentences, means for the generating AI model to generate program code based on the prompt sentences, and means for returning the generated program code to the user. This enables efficient aggregation and summarization of free-response answers and allows the user to easily generate high-quality program code.

[0782] A "survey" is a set of questions designed to collect specific information.

[0783] An "open-ended response" is a type of answer in which the respondent writes freely in their own words.

[0784] "Means of collection" refers to the methods and devices used to collect data.

[0785] "Means of automatic aggregation and summarization" refers to methods and devices for mechanically organizing collected data and extracting important information.

[0786] "Means of outputting as a report" refers to methods or devices for providing aggregated and summarized results in document format.

[0787] "User" refers to an individual or group that uses the system.

[0788] A "prompt message" is text used to input specific instructions or questions to a generative AI model.

[0789] A "generative AI model" is an artificial intelligence model that generates program code or other output based on input prompts.

[0790] "Program code" is a set of instructions that a computer executes.

[0791] A "server" is a computer system that provides data and services over a network.

[0792] This invention relates to a system that efficiently collects, compiles, and summarizes open-ended responses from questionnaires, and further enables users to easily generate program code. Specific embodiments of this system are described below.

[0793] First, users enter free-form responses to a survey through the system interface. These responses are collected by the server. The server uses natural language processing technology to automatically aggregate and summarize the collected responses. Specifically, the server extracts key keywords and phrases and then aggregates and summarizes the data based on these.

[0794] Next, the server outputs the aggregated and summarized results as a report. This report is provided in a visually easy-to-understand format, such as graphs and charts. This allows users to intuitively grasp the content of the open-ended responses.

[0795] Furthermore, this system provides a means for the user to input prompts. The user inputs specific instructions or questions to the generating AI model as prompts. For example, the user might input a prompt such as, "Generate a program that counts specific words from a text file."

[0796] The server receives prompt messages from the user and passes them to the generative AI model. The generative AI model generates program code based on the prompt messages. The generated program code is returned to the user via the server. The user can then execute this program code on their own device.

[0797] The following hardware and software will be used to implement this system. The server is a computer system equipped with a high-performance processor and sufficient memory, and features an NVIDIA GPU. For software, deep learning frameworks such as TensorFlow and PyTorch will be used. This will allow the generative AI model to operate efficiently and perform data processing and calculations based on prompts entered by the user.

[0798] As a concrete example, consider the case where a user inputs the following prompt into the AI ​​model:

[0799] Example of a prompt:

[0800] "Generate a program that counts specific words from a text file."

[0801] Based on this prompt, the generative AI model generates Python code and returns it to the user via the server. The user can then run this code on their device to check the frequency of specific words.

[0802] As described above, this system provides an efficient way to process open-ended responses in surveys and a means for users to easily generate program code.

[0803] The flow of the specific processing in Example 2 will be explained using Figure 17.

[0804] Step 1:

[0805] The user enters a prompt message.

[0806] The user inputs prompts for the generating AI model through the system interface. For example, they might input, "Generate a program that counts specific words from a text file." The entered prompts are then sent to the server.

[0807] Step 2:

[0808] The server receives the prompt message and passes it to the generating AI model.

[0809] The server receives prompt messages sent by the user. It then makes appropriate API calls to pass the received prompt messages to the generative AI model. The server may also convert the prompt messages into a format that the generative AI model can understand. The input is a prompt message, and the output is a request to the generative AI model.

[0810] Step 3:

[0811] The generative AI model generates program code based on the prompt statement.

[0812] The generative AI model analyzes the received prompt and generates program code based on it. For example, if a user inputs "Generate a program that counts specific words from a text file," the generative AI model will generate Python code. The input is the prompt, and the output is the generated program code.

[0813] Step 4:

[0814] The server returns the generated program code to the user.

[0815] The server receives the program code returned from the generated AI model and returns it to the user. It provides the code in a user-friendly format (e.g., a text file or text in a code editor). The input is the generated program code, and the output is the transmission of the program code to the user.

[0816] Step 5:

[0817] The user executes the generated program code.

[0818] The user executes program code received from the server on their own terminal. For example, if Python code is received, the user executes the code in their Python environment and checks the result. The input is the generated program code, and the output is the result of the program execution. Specifically, the user opens their Python environment, copies the received code, and executes it. As a result, the frequency of occurrence of a particular word is displayed.

[0819] (Application Example 2)

[0820] Next, we will describe application example 2 of form example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 as the "terminal".

[0821] Traditional survey systems suffered from the problem of requiring a great deal of time and effort to collect, compile, and summarize open-ended responses. Furthermore, despite advancements in content generation technology using generative AI models, methods for applying this to survey systems had not been established. Additionally, there was a lack of functionality to automatically generate and display content based on user-inputted prompts.

[0822] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0823] In this invention, the server includes means for collecting open-ended responses from a questionnaire, means for automatically aggregating and summarizing the collected open-ended responses, means for outputting the aggregation and summarization results as a report, means for inputting prompt sentences, means for generating content based on the prompt sentences using a generative AI model, and means for displaying the generated content. This enables efficient collection, aggregation, and summarization of open-ended responses from questionnaires, and further enables content generation and display using a generative AI model.

[0824] "Means for collecting open-ended responses from surveys" refers to devices or software that allow users to input answers in a free-form format and collect that information as data.

[0825] "Means for automatically aggregating and summarizing collected open-ended responses" refers to a device or software that analyzes collected open-ended response data, extracts key keywords and phrases, and concisely summarizes the overall content.

[0826] "Means for outputting aggregated and summarized results as a report" refers to a device or software for generating and outputting aggregated and summarized data as a report in a visually easy-to-understand format.

[0827] "Means for inputting prompt text" refers to a device or software for a user to input text to give instructions to a generated AI model.

[0828] "Means for generating content based on prompt text using a generative AI model" refers to a device or software that uses a generative AI model to automatically generate content such as text and images based on input prompt text.

[0829] "Means for displaying generated content" refers to a device or software for visually displaying content generated by a generative AI model to a user.

[0830] The system for implementing this invention has the following configuration. First, the user inputs free-response answers to a questionnaire using a terminal such as a smartphone. The terminal sends the inputted free-response answers to the server. The server uses natural language processing technology to automatically aggregate and summarize the collected free-response answers. Specifically, it extracts key keywords and phrases and concisely summarizes the overall content.

[0831] Next, the server generates a report containing the aggregated and summarized results, outputting it in a visually easy-to-understand format, such as graphs and charts. This allows users to intuitively grasp the survey results.

[0832] Furthermore, the user provides instructions to the generative AI model by entering prompt text. An example of a prompt text is "The story of a brave knight battling a dragon." The server uses the generative AI model to generate content based on this prompt text. The generated content is in the form of text, images, etc., and is displayed visually to the user.

[0833] The following hardware and software are required to implement this system. Hardware includes smartphones and servers. Software includes Python and the OpenAI API. Python is a programming language for implementing natural language processing techniques, and the OpenAI API is an interface for using generative AI models.

[0834] As a concrete example, if a user enters the prompt "A story about a brave knight fighting a dragon," the server will use a generative AI model to generate a story like the following.

[0835] Once upon a time, there was a brave knight named Arthur. Arthur decided to fight a fearsome dragon to protect his kingdom. Reaching the dragon's lair, Arthur drew his sword and confronted the dragon. After a fierce battle, Arthur defeated the dragon and brought peace to the kingdom.

[0836] In this way, users can enjoy generating and displaying content using generative AI models.

[0837] The flow of a specific process in Application Example 2 will be explained using Figure 18.

[0838] Step 1:

[0839] Users enter free-form responses to a survey using a device such as a smartphone. The entered responses are sent from the device to the server. The input data is in text format and is sent to the server.

[0840] Step 2:

[0841] The server analyzes the received open-ended response data using natural language processing techniques. Specifically, it extracts key keywords and phrases and summarizes the overall content concisely. The input data is the open-ended response text, and the output data is a summarized text containing the key keywords and phrases.

[0842] Step 3:

[0843] The server generates a report containing the aggregated and summarized results. The report is output in a visually easy-to-understand format, such as graphs and charts. The input data is a summarized text, while the output data is a visual report.

[0844] Step 4:

[0845] The user provides instructions to the generated AI model by entering prompts. These prompts are entered in text format and sent from the terminal to the server. The input data is the text of the prompts.

[0846] Step 5:

[0847] The server generates content based on prompt text using a generative AI model. The generative AI model automatically generates content such as text and images based on the input prompt text. The input data is the text of the prompt, and the output data is the generated content.

[0848] Step 6:

[0849] The server visually displays the generated content to the user. The generated content is sent to the terminal and displayed on the user's screen. The input data is the generated content, and the output data is the content displayed on the user's screen.

[0850] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0851] The data generation model 58 is a form of so-called generative AI (Artificial Intelligence). One example of the data generation model 58 is ChatGPT (Internet Search).<URL: https: / / openai.com / blog / chatgpt> Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0852] Other examples of generative AI include Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) are some examples.

[0853] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0854] [Third Embodiment]

[0855] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0856] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0857] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0858] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0859] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0860] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0861] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0862] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0863] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0864] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0865] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0866] Next, the identification process performed by the identification processing unit 290 of the data processing device 12 will be described.

[0867] "Example of form 1"

[0868] The system of the present invention includes means for collecting open-ended responses from a questionnaire, means for automatically aggregating and summarizing the collected open-ended responses, and means for outputting the aggregation and summarization results as a report. Specifically, the open-ended responses from the questionnaire are collected using an online questionnaire tool such as Google Forms. The collected open-ended responses are automatically aggregated and summarized using natural language processing technology. This natural language processing technology includes, for example, keyword extraction, topic modeling, and sentiment analysis. The aggregation and summarization results are output in a visually easy-to-understand format, such as graphs or charts. This reduces the workload on the person in charge and saves time.

[0869] The following describes the processing flow for each example of the form.

[0870] "Example of form 1"

[0871] Step 1: Collect open-ended responses from the survey. Specifically, use online survey tools such as Google Forms to collect open-ended responses from users.

[0872] Step 2: Automatically aggregate and summarize the collected open-ended responses. In this step, natural language processing techniques are used to analyze the content of the open-ended responses. Specifically, techniques such as keyword extraction, topic modeling, and sentiment analysis are used to extract the main keywords and phrases from the open-ended responses, and the responses are aggregated and summarized based on these.

[0873] Step 3: Output the aggregated and summarized results as a report. In this step, the aggregated and summarized results obtained in the previous step are output in a visually easy-to-understand format. Specifically, the aggregated and summarized results are displayed in graphs and charts and output as a report. This allows the person in charge to grasp the overview of the open-ended responses simply by looking at the report, thereby reducing the workload on the person in charge and saving time.

[0874] (Example 1)

[0875] Next, we will describe Embodiment 1 of Embodiment Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0876] Traditional systems had problems such as requiring users to have specialized knowledge when generating programs, and requiring a high level of technical understanding to understand the processing of the generated programs. Furthermore, there was often a lack of concrete examples to understand the specific operation of the generated programs, making it unclear how users should actually use the programs.

[0877] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0878] In this invention, the server includes means for the user to input a prompt, means for generating a program using a generative AI model, means for explaining the processing of the generated program in natural language, and means for adding concrete examples to the explanation. This makes it easier for the user to generate a program and understand its processing content without specialized knowledge, and furthermore, to clearly grasp how to use it in practice through concrete examples.

[0879] A "user" is an entity that uses a system to input prompt messages and receives program generation and explanations.

[0880] A "prompt" is a set of instructions that a user inputs to the system, and it is the text that the generative AI model uses to generate a program.

[0881] A "generative AI model" is an artificial intelligence model that generates program code based on prompt text entered by a user, such as a model that uses natural language processing technology.

[0882] A "program" is the code generated by a generative AI model, consisting of a set of instructions for performing a specific task.

[0883] A "server" is a computer system that receives prompt messages from users, generates programs using a generative AI model, and explains the processing details.

[0884] "Natural language" refers to the language that humans use on a daily basis, and is used to explain the processing content of generated programs in a way that is easy for users to understand.

[0885] A "concrete example" is a specific instance that demonstrates how to actually use the generated program, helping the user understand how the program works.

[0886] This invention relates to a system in which a user inputs a prompt, a generation AI model generates a program, explains the processing content in natural language, and adds concrete examples.

[0887] First, the user accesses the system interface using a terminal and enters a prompt. For example, they might enter a prompt such as, "Generate a program that analyzes the text entered by the user and determines its sentiment."

[0888] Next, the server receives this prompt and generates a program using a generative AI model. This generative AI model, for example, is a model that uses natural language processing technology and generates appropriate program code based on the prompt entered by the user.

[0889] The generated program is written in a programming language such as Python. The server analyzes the processing content of this generated program and explains it in natural language. Specifically, it clearly states what kind of data processing and calculations the program performs. For example, if it uses the NLTK library to analyze the sentiment of text, it will explain the details.

[0890] Furthermore, the server adds concrete examples to the description. For example, if the user enters "Today is a very good day," it specifically shows how the program determines the emotion. This makes it easier for the user to understand the actual behavior of the generated program.

[0891] This system allows users to generate programs without specialized knowledge, easily understand their processing, and clearly grasp their practical usage through concrete examples.

[0892] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0893] Step 1:

[0894] The user enters a prompt message.

[0895] The user accesses the system interface using a terminal and enters prompt messages. For example, they might enter a prompt message such as, "Generate a program that analyzes the text entered by the user and determines its sentiment." The entered prompt message is then sent to the server.

[0896] Step 2:

[0897] The server generates the program using the generated AI model.

[0898] The server inputs the prompt text received from the user into a generative AI model. The generative AI model generates appropriate program code based on the prompt text. For example, a model using natural language processing techniques might generate Python code. The generated program code is stored on the server.

[0899] Step 3:

[0900] The server explains the processing of the generated program in natural language.

[0901] The server analyzes the generated program code and explains its processing in natural language. Specifically, it clearly indicates what kind of data processing and calculations the program performs. For example, if it uses the NLTK library to analyze the sentiment of text, it will explain the details. The explanation is provided to the user.

[0902] Step 4:

[0903] The server adds specific examples to the description.

[0904] The server adds specific examples to the description of the generated program. For example, it specifically shows how the program determines emotion when the user inputs "Today is a very good day." Specific examples are important for demonstrating the actual operation of the program. The description, including the specific examples, is provided to the user.

[0905] The above describes the processing flow of this system's program.

[0906] (Application Example 1)

[0907] Next, we will describe Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0908] Traditional survey systems often require manual collection, tabulation, and summarization of open-ended responses, resulting in significant time and effort. Furthermore, outputting the results as reports in a visually easy-to-understand format can be challenging. Additionally, creating high-quality ad copy requires specialized knowledge, making it difficult to generate such content quickly. To address these issues, there is a need for a system that efficiently and automatically tabulates and summarizes open-ended responses, generates visually easy-to-understand reports, and further utilizes a generation AI model to generate ad copy based on prompts.

[0909] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0910] In this invention, the server includes means for collecting open-ended responses from a questionnaire, means for automatically aggregating and summarizing the collected open-ended responses, means for outputting the aggregation and summarization results as a report, means for generating ad copy based on prompt sentences using a generation AI model, and means for displaying the generated ad copy. This enables efficient aggregation and summarization of open-ended responses and the generation of visually easy-to-understand reports, as well as the rapid generation of high-quality ad copy.

[0911] "Methods for collecting open-ended responses in surveys" refers to functions for collecting opinions and comments freely written by respondents in surveys as data.

[0912] "Methods for automatically aggregating and summarizing collected open-ended responses" refers to functions that analyze collected open-ended responses, extract key keywords and phrases, and concisely summarize the overall content.

[0913] "Means for outputting aggregated and summarized results as a report" refers to a function for generating and outputting aggregated and summarized data as a report in a visually easy-to-understand format.

[0914] "Means for generating ad copy based on prompt text using a generative AI model" refers to a function that uses a generative AI model to automatically create ad copy based on prompt text entered by the user.

[0915] "Means for displaying generated ad copy" refers to a function for visually displaying ad copy created by a generation AI model to the user.

[0916] The system for implementing this invention includes functions for collecting open-ended responses from questionnaires, automatically aggregating and summarizing the collected responses, and outputting the aggregation and summarization results as a report. It also includes a function for generating advertisement text based on prompt text using a generation AI model and displaying the generated advertisement text.

[0917] Hardware and software configuration

[0918] Hardware:

[0919] server

[0920] User devices (smartphones, smart glasses, etc.)

[0921] software:

[0922] Analysis program using natural language processing technology

[0923] Generative AI model using the OpenAI API

[0924] Database Management System

[0925] Visualization tools for generating graphs and charts

[0926] Data processing and data calculation

[0927] server:

[0928] 1. The server collects the open-ended responses from the survey submitted by the user's terminal into a database.

[0929] 2. Using natural language processing techniques, the collected open-ended responses are analyzed to extract key keywords and phrases.

[0930] 3. Based on the extracted keywords and phrases, the content of the free-response answers is automatically compiled and summarized.

[0931] 4. Generate a report containing the aggregated and summarized results in a visually easy-to-understand format (e.g., graphs and charts) and send it to the user's terminal.

[0932] User terminal:

[0933] 1. The user enters a prompt and sends it to the generating AI model.

[0934] 2. The server uses the OpenAI API to generate ad text based on the input prompt.

[0935] 3. Display the generated ad copy on the user's device.

[0936] Specific example

[0937] Example of a prompt:

[0938] "Please create ad copy for a new smartphone. Its features include a high-resolution camera and long battery life."

[0939] Example of generated ad copy:

[0940] "A new smartphone has arrived! Capture every beautiful moment with its high-resolution camera. Plus, enjoy worry-free use all day long with its long-lasting battery. Get yours now and enjoy the best experience!"

[0941] In this way, users can simply input prompt text, and the AI ​​generation model will quickly generate high-quality ad copy. Furthermore, open-ended survey responses are efficiently compiled and summarized, and output as a visually easy-to-understand report, thus improving operational efficiency.

[0942] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0943] Step 1:

[0944] The user enters their free-response answers to the survey and sends them from their device to the server.

[0945] Input: User-generated free response

[0946] Output: Free-response data sent to the server

[0947] Specific operation: When a user enters free-form answers into a survey form and presses the submit button, the device sends that data to the server.

[0948] Step 2:

[0949] The server saves the received open-ended responses to a database.

[0950] Input: Free-response data sent from the device.

[0951] Output: Open-ended response data stored in the database

[0952] Specific operation: The server stores and saves the received open-ended response data in a database.

[0953] Step 3:

[0954] The server uses natural language processing technology to analyze the open-ended responses and extract key keywords and phrases.

[0955] Input: Open-ended response data stored in the database

[0956] Output: Extracted keywords and phrases

[0957] Specific operation: The server executes a natural language processing algorithm to extract key keywords and phrases from the open-ended text data.

[0958] Step 4:

[0959] The server automatically aggregates and summarizes the free-response content based on the extracted keywords and phrases.

[0960] Input: Extracted keywords or phrases

[0961] Output: Aggregated and summarized data

[0962] Specific operation: The server aggregates the extracted keywords and phrases and uses a summarization algorithm to concisely summarize the content of the open-ended responses.

[0963] Step 5:

[0964] The server generates a report containing the aggregated and summarized results in a visually easy-to-understand format and sends it to the user's terminal.

[0965] Input: Aggregated and summarized data

[0966] Output: A visually easy-to-understand report (in graph and chart format)

[0967] Specific operation: The server uses a visualization tool to generate aggregated and summarized results as a report in graph and chart format, and sends it to the user's terminal.

[0968] Step 6:

[0969] The user enters a prompt and sends it to the generating AI model.

[0970] Input: User-entered prompt message

[0971] Output: Prompt message sent to the server

[0972] Specific operation: When the user enters a prompt message and presses the send button, the terminal sends that data to the server.

[0973] Step 7:

[0974] The server uses the OpenAI API to generate ad copy based on the input prompt text.

[0975] Input: Prompt message sent by the user

[0976] Output: Generated ad copy

[0977] Specific operation: The server calls the OpenAI API, inputs the prompt text into the AI ​​model that generates the text, and retrieves the generated ad text.

[0978] Step 8:

[0979] The server sends the generated ad text to the user's device, and the user's device displays it.

[0980] Input: Generated ad copy

[0981] Output: Ad text displayed on the user's device

[0982] Specific operation: The server sends the generated ad text to the user's terminal, and the user's terminal displays the received ad text on the screen.

[0983] (Example 2)

[0984] Next, we will describe Example 2 of the Form Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0985] Conventional systems have struggled to provide appropriate responses quickly and accurately to user-inputted prompts. Furthermore, the lack of a visually easy-to-understand format for displaying generated responses resulted in low user convenience. This invention aims to solve these problems by providing a system that offers quick and accurate responses to user-inputted prompts and displays those responses in a visually easy-to-understand format.

[0986] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0987] In this invention, the server includes means for the user to input a prompt sentence, means for the terminal to send the prompt sentence to the server, means for the server to pass the prompt sentence to a generation AI model, means for the generation AI model to analyze the prompt sentence and generate a response, means for the server to return the generated response to the terminal, and means for the terminal to display the response to the user. This makes it possible to provide a quick and accurate response to a prompt sentence entered by the user and to display the response in a visually easy-to-understand format.

[0988] A "user" is an individual or group that uses a system to input prompt messages and receives the generated responses.

[0989] A "prompt message" is a sentence containing a question or instruction that a user enters into the system.

[0990] A "terminal" is a device used by a user to input prompt messages and display the generated response. Examples include personal computers and smartphones.

[0991] A "server" is a computer system that receives prompt messages sent from a terminal, passes them to a generation AI model, and returns the generated response to the terminal.

[0992] A "generative AI model" is an artificial intelligence model that analyzes prompt sentences and generates appropriate responses. Examples include models that utilize natural language processing techniques.

[0993] "Natural language processing technology" is a technique that enables computers to understand and analyze human language. This makes it possible to analyze the content of a prompt and generate an appropriate response.

[0994] A "response" refers to the answer or information that a generative AI model generates by analyzing a prompt sentence.

[0995] A "visually easy-to-understand format" is a format that displays the generated response in a way that is easy for the user to understand. Examples include text, graphs, and charts.

[0996] This invention provides a system that offers a rapid and accurate response to a user-inputted prompt and displays that response in a visually easy-to-understand format. The system includes a user, a terminal, a server, and a generative AI model.

[0997] The user enters a prompt message into the terminal's input field. Examples of prompt messages include "What's the weather like today?" or "Please tell me the latest news." The terminal then sends the user's prompt message to the server. At this time, the prompt message is packaged in JSON format.

[0998] The server parses the received prompt and sends a request to the generative AI model's API endpoint. The generative AI model used is, for example, an advanced model employing natural language processing techniques (e.g., GPT-3 or GPT-4). The server parses the response received from the generative AI model and returns it to the terminal. At this point, the response is again packaged in JSON format.

[0999] The terminal analyzes the response received from the server and displays it to the user. The display format is visually easy to understand, such as text, graphs, and charts. For example, responses such as "Today's weather is sunny" or "The latest news is as follows" may be displayed.

[1000] This system allows users to obtain appropriate responses to their input prompts using a generative AI model. Furthermore, the responses are displayed in a visually easy-to-understand format, improving user convenience.

[1001] As a concrete example, consider a case where a user enters the prompt "How do I write a Hello World program in Python?". In this case, the generative AI model will generate the response "Here's how to write a Hello World program in Python: python print('Hello, World!')" and display it on the terminal.

[1002] In this way, the invention provides a rapid and accurate response to a prompt message entered by the user and displays that response in a visually easy-to-understand format.

[1003] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1004] Step 1:

[1005] The user enters a prompt message.

[1006] The user enters a prompt message into the terminal's input field. For example, they might type, "What's the weather like today?". The entered prompt message is temporarily stored in the terminal's memory.

[1007] Step 2:

[1008] The terminal sends a prompt message to the server.

[1009] The terminal sends the prompt text entered by the user to the server as an HTTP request. The prompt text is packaged in JSON format. The input is the prompt text, and the output is the HTTP request to the server.

[1010] Step 3:

[1011] The server passes the prompt message to the AI ​​model that generates it.

[1012] The server parses the received prompt and sends a request to the generative AI model's API endpoint. For example, it might send the prompt along with the API key as a POST request. The input is the prompt, and the output is the API request to the generative AI model.

[1013] Step 4:

[1014] The generative AI model analyzes the prompt sentence and generates a response.

[1015] The generative AI model analyzes the received prompt and generates an appropriate response. For example, in response to the prompt "What's the weather like today?", it generates the response "The weather is sunny today." The input is the prompt, and the output is the generated response.

[1016] Step 5:

[1017] The server returns the generated response to the terminal.

[1018] The server analyzes the response received from the generated AI model and returns it to the terminal. At this point, the response is again packaged in JSON format. The input is the generated response, and the output is the HTTP response sent to the terminal.

[1019] Step 6:

[1020] The terminal displays a response to the user.

[1021] The terminal analyzes the response received from the server and displays it to the user. For example, "Today's weather is sunny." might be displayed on the screen. The input is the generated response, and the output is what is displayed to the user.

[1022] (Application Example 2)

[1023] Next, we will describe 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."

[1024] Traditional survey systems often required manual collection, tabulation, and summarization of open-ended responses, resulting in significant time and effort. Furthermore, outputting the results in a visually easy-to-understand report format was challenging. Additionally, automated ad generation lacked a robust mechanism for generating appropriate ads based on user input. This made ad creation cumbersome and hindered the rapid creation of effective advertisements.

[1025] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1026] In this invention, the server includes means for collecting open-ended responses from a questionnaire, means for automatically aggregating and summarizing the collected open-ended responses, means for outputting the aggregation and summarization results as a report, means for automatically generating advertisements based on prompt text using a generation AI model, means for previewing the generated advertisements, means for customizing the generated advertisements, and means for saving and sharing the generated advertisements. This enables efficient collection, aggregation, and summarization of open-ended responses from questionnaires, and allows for the output of reports in a visually easy-to-understand format. Furthermore, it enables the rapid and automatic generation of effective advertisements based on prompt text, and allows users to easily customize and share them.

[1027] "Methods for collecting open-ended responses in surveys" refers to functions for collecting free-form responses from users.

[1028] "Means for automatically aggregating and summarizing collected open-ended responses" refers to a function that automatically analyzes collected open-ended responses, extracts key information, and summarizes it.

[1029] "Means for outputting aggregated and summarized results as a report" refers to a function for outputting aggregated and summarized results in a report format.

[1030] "A means of automatically generating advertisements based on prompt text using a generative AI model" refers to a function that uses a generative AI model to automatically generate advertisements based on prompt text entered by the user.

[1031] "Means for previewing generated advertisements" refers to a function that allows users to preview the generated advertisements.

[1032] "Means for customizing generated ads" refers to features that allow users to edit and customize generated ads.

[1033] "Means for saving and sharing generated ads" refers to features for saving generated ads and sharing them with other users and the platform.

[1034] The system for carrying out this invention has the following configuration: The system includes means for collecting open-ended responses from a questionnaire, means for automatically aggregating and summarizing the collected open-ended responses, means for outputting the aggregation and summarization results as a report, means for automatically generating advertisements based on prompt text using a generation AI model, means for previewing the generated advertisements, means for customizing the generated advertisements, and means for saving and sharing the generated advertisements.

[1035] Hardware and software configuration

[1036] Hardware: Smartphones, servers

[1037] Software: Generative AI models (e.g., OpenAI's GPT-4), mobile app development frameworks (e.g., React Native), cloud storage services (e.g., AWS S3)

[1038] Data processing and data calculation

[1039] 1. Collection of open-ended survey responses: Users enter their responses via a smartphone app. These responses are sent to a server and stored in a database.

[1040] 2. Automatic aggregation and summarization of open-ended responses: The server analyzes open-ended responses using natural language processing technology and extracts key keywords and phrases. This allows for the aggregation and summarization of the responses.

[1041] 3. Report Output: The aggregated and summarized results are output as a report in a visually easy-to-understand format (e.g., graphs and charts). This report can be viewed by users on their smartphones.

[1042] 4. Automated ad generation: When a user enters a prompt, that prompt is sent to the server. The generation AI model then generates ad text and images based on the prompt.

[1043] 5. Ad preview display: The generated ads are previewed to the user in the smartphone app.

[1044] 6. Ad customization: Users can edit and customize the generated ads within the app.

[1045] 7. Saving and sharing ads: Completed ads are saved to cloud storage, and a link is generated that can be shared via social media or email.

[1046] Specific example

[1047] For example, if a user wants to create an advertisement to promote the opening of a new cafe, they would enter a prompt message like the following:

[1048] Example prompt: "Create an advertisement to promote the opening of a new cafe. The target audience is young people in their 20s and 30s, and the cafe features organic coffee and a relaxing atmosphere."

[1049] When this prompt is input into the AI ​​model, the model generates advertising text and images like the following.

[1050] Example ad text: "A new cafe, 'Relax Cafe,' has opened! With organic coffee and a relaxing atmosphere, it's perfect for young people in their 20s and 30s. Please stop by!"

[1051] Example of advertising images: A generative AI model generates images of cafe interiors and organic coffee, which are then incorporated into advertisements.

[1052] In this way, users can easily create compelling advertisements and effectively reach their target audience.

[1053] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1054] Step 1:

[1055] Users enter free-form answers to a survey via a smartphone app.

[1056] Input: The user enters free-form text.

[1057] Output: The entered free-response answers are sent to the server.

[1058] Specific operation: When a user enters a free-response answer in the app's text input field and presses the "Submit" button, that data is sent to the server.

[1059] Step 2:

[1060] The server collects open-ended responses and saves them to a database.

[1061] Input: Text data of open-ended responses submitted by users.

[1062] Output: Free-response answers stored in the database.

[1063] Specific operation: The server saves the received open-ended responses to the database and generates a confirmation message when saving is complete.

[1064] Step 3:

[1065] The server uses natural language processing technology to analyze the open-ended responses and extract key keywords and phrases.

[1066] Input: Text data of open-ended responses stored in the database.

[1067] Output: Extracted main keywords and phrases.

[1068] Specific operation: The server executes a natural language processing algorithm to extract key keywords and phrases from the open-ended text.

[1069] Step 4:

[1070] The server performs aggregation and summarization based on the extracted keywords and phrases.

[1071] Input: The main keywords or phrases extracted.

[1072] Output: Aggregated and summarized data.

[1073] Specific operation: The server aggregates the extracted keywords and phrases and generates summary data using a summarization algorithm.

[1074] Step 5:

[1075] The server outputs the aggregated and summarized results as a report in a visually easy-to-understand format.

[1076] Input: Aggregated and summarized data.

[1077] Output: Reports in graph and chart format.

[1078] Specific operation: The server converts aggregated and summarized data into graphs and charts and generates them as reports.

[1079] Step 6:

[1080] The user enters a prompt and sends it to the generating AI model.

[1081] Input: The prompt text entered by the user.

[1082] Output: The prompt text sent to the generating AI model.

[1083] Specific operation: When the user enters a prompt in the app's text input field and presses the "Generate" button, that data is sent to the generating AI model.

[1084] Step 7:

[1085] The generative AI model generates ad text and images based on the prompt text.

[1086] Input: The prompt text sent to the generating AI model.

[1087] Output: Generated ad text and images.

[1088] Specific operation: The generative AI model analyzes the prompt text and generates advertising text and images.

[1089] Step 8:

[1090] The server displays a preview of the generated advertisement to the user.

[1091] Input: Generated ad text and images.

[1092] Output: Ad preview displayed on the user's smartphone.

[1093] Specific operation: The server sends the generated advertisement to the user's smartphone and displays a preview in the app.

[1094] Step 9:

[1095] Users can customize the ads that are generated.

[1096] Input: The ad displayed in preview.

[1097] Output: Customized ads.

[1098] Specific operation: Users can edit and customize ad text and images using the app's editing function.

[1099] Step 10:

[1100] The server saves the customized ad to cloud storage and generates a sharing link.

[1101] Input: Customized ad.

[1102] Output: Advertisements and shared links stored in cloud storage.

[1103] Specific operation: The server saves customized advertisements to cloud storage and generates links that can be shared via social media or email.

[1104] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1105] "Example of form 1"

[1106] This invention relates to a system that collects open-ended responses from questionnaires, automatically aggregates and summarizes the collected responses, and outputs the aggregated and summarized results as a report. As a means of automatically aggregating and summarizing the open-ended responses, this system uses natural language processing technology to analyze the content of the responses and extract key keywords and phrases. Furthermore, this system analyzes the content of the open-ended responses using an emotion engine that recognizes user emotions. This emotion engine extracts user emotions from the text data of the open-ended responses and reflects those emotions in the report. Specifically, if a user gives the open-ended response "This product is very good," the emotion engine extracts the emotion "good" from this response and reflects that emotion in the report. The report output means outputs the aggregated and summarized results along with the emotions extracted by the emotion engine in a visually easy-to-understand format, such as a graph or chart. This allows the person in charge to grasp the overview of the open-ended responses and user emotions simply by looking at the report, thereby reducing the workload and saving time.

[1107] The following describes the processing flow for each example of the form.

[1108] "Example of form 1"

[1109] Step 1: Collect open-ended responses from the survey. Specifically, use a survey tool such as Google Forms to collect open-ended responses from users.

[1110] Step 2: Automatically aggregate and summarize the collected open-ended responses. Specifically, natural language processing technology is used to analyze the content of the open-ended responses and extract key keywords and phrases.

[1111] Step 3: Recognize the user's emotions. Specifically, use an emotion engine to analyze the content of open-ended responses and extract the user's emotions.

[1112] Step 4: Reflect the extracted emotions in the report. Specifically, the emotions extracted by the emotion engine are reflected in the report and output in a visually easy-to-understand format, such as a graph or chart.

[1113] Step 5: Output the aggregated / summarized results and sentiment analysis results as a report. Specifically, the aggregated / summarized results and sentiment analysis results are integrated and output as a single report. This allows the person in charge to understand the overview of the open-ended responses and the users' sentiments simply by looking at the report.

[1114] (Example 1)

[1115] Next, we will describe Embodiment 1 of Embodiment Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1116] In conventional systems, the process of generating appropriate responses to user-inputted prompts was complex and difficult to perform efficiently. Furthermore, the quality and display format of the generated results were inconsistent, sometimes making them difficult for users to understand. This resulted in a degraded user experience and limited the system's usefulness.

[1117] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1118] In this invention, the server includes means for receiving prompt statements and performing preprocessing, means for inputting prompt statements into a generation AI model and performing inference, and means for post-processing the generated results. This makes it possible to consistently and efficiently perform everything from preprocessing prompt statements to post-processing of the generated results.

[1119] A "user" is the entity that operates the system and enters prompt messages.

[1120] A "terminal" is a device used by a user to operate something, and includes PCs, smartphones, and other similar devices.

[1121] A "prompt" is the text of instructions or questions that a user inputs to the generated AI model.

[1122] A "server" is a computing system that receives prompt messages, performs preprocessing, inputs data into a generative AI model, performs inference, performs postprocessing, and sends the results.

[1123] "Preprocessing" refers to the process of normalizing and tokenizing the text of a prompt message.

[1124] A "generative AI model" is an artificial intelligence model that performs inference based on input prompt sentences and generates results.

[1125] "Inference" is the process by which a generative AI model generates responses or results based on a prompt.

[1126] "Post-processing" refers to the process of formatting the generated results and removing unnecessary information.

[1127] "Results" refer to the responses and information generated by the generative AI model based on the prompt text.

[1128] "Display" refers to the act of providing the user with a visual representation of the results received by the terminal from the server.

[1129] This invention is a system in which a user inputs a prompt, and a generation AI model generates a response or result based on that prompt, which is then provided to the user. Specific embodiments of this system are described below.

[1130] First, the user enters a prompt using a device (such as a PC or smartphone). An example of a prompt is, "Please use AI to come up with a new recipe."

[1131] The terminal sends the entered prompt text to the server. HTTP requests are used for communication. The server is a computing system with high computing power, specifically a server equipped with a high-performance GPU.

[1132] The server receives prompt messages sent from the terminal and performs preprocessing. Preprocessing includes text normalization (e.g., converting full-width characters to half-width characters) and tokenization (e.g., splitting sentences into words or phrases). Python libraries (e.g., pandas, numpy) are used for preprocessing.

[1133] Next, the server inputs the pre-processed prompt sentences into the generative AI model. For example, GPT-4 is used as the generative AI model. The model performs inference based on the input prompt sentences and generates results. TensorFlow or PyTorch is used for the model's inference.

[1134] The generated results are post-processed on the server. Post-processing includes formatting the results (e.g., converting to JSON format) and removing unnecessary information.

[1135] The post-processed results are sent from the server to the terminal. HTTP responses are used for communication. The terminal displays the received results to the user. HTML and CSS are used for display.

[1136] As a concrete example, if a user enters the prompt "Please use AI to come up with a new recipe," the server receives this prompt, performs preprocessing, and inputs it into the AI ​​model. The model generates a new recipe, and the server processes the result and sends it to the terminal. The terminal then displays the generated recipe to the user.

[1137] In this way, users can obtain new recipes using the generative AI model. This system can efficiently handle everything from pre-processing of prompts to post-processing of the generated results, providing users with high-quality responses.

[1138] The flow of the specific processing in Example 1 will be explained using Figure 15.

[1139] Step 1:

[1140] The user enters a prompt message.

[1141] The user enters "Use AI to come up with a new recipe" into the terminal's input field. The entered prompt is saved to the terminal's memory.

[1142] Step 2:

[1143] The terminal sends a prompt message to the server.

[1144] The terminal sends the entered prompt text to the server as an HTTP request. The input is the prompt text, and the output is the request sent to the server.

[1145] Step 3:

[1146] The server receives the prompt message and performs preprocessing.

[1147] The server receives prompt messages sent from the terminal. It then performs text normalization (e.g., converting full-width characters to half-width characters) and tokenization (e.g., splitting sentences into words and phrases) on the received prompt messages. The input is the prompt message, and the output is pre-processed text data.

[1148] Step 4:

[1149] The server inputs prompt messages into the generated AI model and performs inference.

[1150] The server inputs pre-processed prompt sentences into a generating AI model (e.g., GPT-4). The model performs inference based on the input prompt sentences and generates results. The input is pre-processed text data, and the output is the generated result (e.g., a new recipe).

[1151] Step 5:

[1152] The server then performs post-processing on the generated results.

[1153] The server performs post-processing on the results obtained from the generated AI model. Post-processing includes formatting the results (e.g., converting to JSON format) and removing unnecessary information. The input is the generated result, and the output is the post-processed data.

[1154] Step 6:

[1155] The server sends the results to the terminal.

[1156] The server sends the post-processed results to the terminal as an HTTP response. The input is the post-processed data, and the output is the response sent to the terminal.

[1157] Step 7:

[1158] The device displays the results to the user.

[1159] The terminal displays the results received from the server to the user. HTML and CSS are used for the display. The input is the response data from the server, and the output is the information visually presented to the user.

[1160] (Application Example 1)

[1161] Next, we will describe Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1162] Traditional survey systems often involve manual collection, tabulation, summarization, and report generation of open-ended responses, resulting in time-consuming and labor-intensive processes. Furthermore, generating advertising content requires specialized knowledge, making it difficult to reach target audiences quickly and effectively. To address these challenges, a system integrating automated tabulation and summarization of open-ended responses with automated advertising content generation is needed.

[1163] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1164] In this invention, the server includes means for collecting open-ended responses to a questionnaire, prompts for aggregating and summarizing the collected open-ended responses and generating a report of the aggregated and summarized results, and means for generating the report using a generation AI model, prompts for generating content based on the generated report and means for generating the content using the generation AI model, means for previewing the generated content and accepting editing operations, means for publishing the edited content, and means for analyzing the performance of the published content. This enables efficient aggregation and summarization of open-ended responses, and automatic generation and publication of advertising content based on prompts.

[1165] "Means for collecting open-ended responses in surveys" refers to interfaces and functions for collecting responses from users in a free-form format.

[1166] "Means for automatically aggregating and summarizing collected open-ended responses" refers to algorithms or software that analyze collected open-ended response data, extract key information, and summarize it.

[1167] "Means for outputting aggregated and summarized results as a report" refers to a function for generating and outputting aggregated and summarized data as a report in a visually easy-to-understand format.

[1168] "Means for entering prompt text" refers to a text input interface for users to give instructions to the generated AI model.

[1169] "Means of generating content based on prompt text using a generative AI model" refers to algorithms and software that enable a generative AI model to automatically generate content such as text and images based on input prompt text.

[1170] "Means for previewing and editing generated content" refers to interfaces and functions that allow users to review generated content and edit it as needed.

[1171] "Means for publishing generated content" refers to functions for publishing generated and edited content on internet platforms, social networking services, etc.

[1172] "Means of analyzing the performance of published content" refers to tools and software used to collect and analyze performance data such as the number of views and clicks on published content.

[1173] The system for implementing this invention has the following configuration. First, the user inputs free-form answers to a questionnaire about a specific product using a terminal such as a smartphone. The terminal is equipped with means for collecting free-form answers to the questionnaire and transmits the free-form answers entered by the user to a server.

[1174] The server is equipped with a means to automatically aggregate and summarize the collected open-ended responses. This means analyzes the content of the open-ended responses using natural language processing techniques and extracts key keywords and phrases. Specifically, it uses natural language processing libraries (e.g., NLTK, spaCy) to analyze the text data and extract important information. The server is also equipped with a means to output the aggregated and summarized results as a report. This means outputs the aggregated and summarized results in a visually easy-to-understand format, such as graphs and charts. This uses data visualization tools (e.g., Matplotlib, D3.js). In this embodiment, the server generates a report using a prompt message that instructs the server to aggregate and summarize the collected open-ended responses and generate the aggregated and summarized results as a report, and a generation AI model.

[1175] Furthermore, the user provides instructions to the generating AI model using prompts that instruct it to generate content based on the generated report, thereby generating the content. An example of a prompt message is: "Enter the generated report and, using this report as a reference, create an advertisement promoting a new eco-friendly detergent to housewives in their 30s." In this example, the specific product is a detergent.

[1176] The server is equipped with a means for generating content based on prompt text using a generative AI model. This means uses a generative AI model (e.g., OpenAI GPT-4) to generate advertising text and images based on the input prompt text.

[1177] The generated content is provided to the user through means of previewing and editing on their device. Users can review the generated content and edit it as needed. The edited content is then published to internet platforms and social media via the server.

[1178] Finally, the server has a means to analyze the performance of the published content. This means uses tools (e.g., Google Analytics) to collect and analyze performance data such as the number of views and clicks on the published content.

[1179] In this way, efficient aggregation and summarization of open-ended responses, as well as the automatic generation and publication of advertising content based on prompt text, become possible. Alternatively, the content may be modified using a prompt message instructing it to change based on the analyzed performance, along with a generative AI model. This allows the content to be modified in order to improve its performance.

[1180] The flow of a specific process in Application Example 1 will be explained using Figure 16.

[1181] Step 1:

[1182] Users input free-response answers to a survey using a smartphone or other device. The entered free-response answers are sent from the device to the server. The input data is free-response text, and the output is the free-response data sent to the server.

[1183] Step 2:

[1184] The server automatically aggregates and summarizes the collected open-ended responses. It analyzes the content of the open-ended responses using natural language processing techniques and extracts key keywords and phrases. The input data is the open-ended response text, and the output is summarized data containing the key keywords and phrases. Specifically, it analyzes the text data using natural language processing libraries (e.g., NLTK, spaCy). Alternatively, it generates aggregated and summarized results of the collected open-ended responses using a prompt message instructing the server to aggregate and summarize the collected open-ended responses, along with a generative AI model.

[1185] Step 3:

[1186] The server outputs the aggregated and summarized results as a report. The aggregated and summarized results are output in a visually easy-to-understand format, such as graphs or charts. The input data is summarized data, and the output is a visually represented report. Data visualization tools (e.g., Matplotlib, D3.js) are used. Specifically, the report is generated using a prompt message that instructs the server to generate a report based on the generated aggregated and summarized results, and a generating AI model.

[1187] Step 4:

[1188] The user provides instructions to the generating AI model by entering prompts. The input data consists of reports and prompts, and the output consists of reports and prompts sent to the server. An example of a prompt is to input a generated report along with the instruction, "Using this report as a reference, create an advertisement promoting a new eco-friendly detergent to housewives in their 30s."

[1189] Step 5:

[1190] The server generates content based on prompt text using a generative AI model. The input data is the prompt text, and the output is the generated advertisement text and images. A generative AI model (e.g., OpenAI GPT-4) is used to generate content based on the input prompt text.

[1191] Step 6:

[1192] The generated content is provided to the user through means of previewing and editing on the device. The user reviews the generated content and edits it as needed. The input data is the generated content, and the output is the edited content.

[1193] Step 7:

[1194] The edited content is published to internet platforms and social media via a server. The input data is the edited content, and the output is the published content.

[1195] Step 8:

[1196] The server analyzes the performance of published content. It collects and analyzes performance data such as the number of views and clicks on published content. The input data is performance data, and the output is the analysis results. Data analysis tools (e.g., Google Analytics) are used.

[1197] (Example 2)

[1198] Next, we will describe Example 2 of the Form Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1199] Traditional survey systems often involved manual collection, tabulation, and summarization of open-ended responses, resulting in significant time and effort. Furthermore, generating specific programs required programming knowledge, making it difficult for users without specialized expertise. Additionally, the quality and efficiency of the generated program code were often not guaranteed.

[1200] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting free-response answers from a questionnaire, means for automatically aggregating and summarizing the collected free-response answers, means for outputting the aggregation and summarization results as a report, means for the user to input prompt sentences, means for the generating AI model to generate program code based on the prompt sentences, and means for returning the generated program code to the user. This enables efficient aggregation and summarization of free-response answers and allows the user to easily generate high-quality program code.

[1201] A "survey" is a set of questions designed to collect specific information.

[1202] An "open-ended response" is a type of answer in which the respondent writes freely in their own words.

[1203] "Means of collection" refers to the methods and devices used to collect data.

[1204] "Means of automatic aggregation and summarization" refers to methods and devices for mechanically organizing collected data and extracting important information.

[1205] "Means of outputting as a report" refers to methods or devices for providing aggregated and summarized results in document format.

[1206] "User" refers to an individual or group that uses the system.

[1207] A "prompt message" is text used to input specific instructions or questions to a generative AI model.

[1208] A "generative AI model" is an artificial intelligence model that generates program code or other output based on input prompts.

[1209] "Program code" is a set of instructions that a computer executes.

[1210] A "server" is a computer system that provides data and services over a network.

[1211] This invention relates to a system that efficiently collects, compiles, and summarizes open-ended responses from questionnaires, and further enables users to easily generate program code. Specific embodiments of this system are described below.

[1212] First, users enter free-form responses to a survey through the system interface. These responses are collected by the server. The server uses natural language processing technology to automatically aggregate and summarize the collected responses. Specifically, the server extracts key keywords and phrases and then aggregates and summarizes the data based on these.

[1213] Next, the server outputs the aggregated and summarized results as a report. This report is provided in a visually easy-to-understand format, such as graphs and charts. This allows users to intuitively grasp the content of the open-ended responses.

[1214] Furthermore, this system provides a means for the user to input prompts. The user inputs specific instructions or questions to the generating AI model as prompts. For example, the user might input a prompt such as, "Generate a program that counts specific words from a text file."

[1215] The server receives prompt messages from the user and passes them to the generative AI model. The generative AI model generates program code based on the prompt messages. The generated program code is returned to the user via the server. The user can then execute this program code on their own device.

[1216] The following hardware and software will be used to implement this system. The server is a computer system equipped with a high-performance processor and sufficient memory, and features an NVIDIA GPU. For software, deep learning frameworks such as TensorFlow and PyTorch will be used. This will allow the generative AI model to operate efficiently and perform data processing and calculations based on prompts entered by the user.

[1217] As a concrete example, consider the case where a user inputs the following prompt into the AI ​​model:

[1218] Example of a prompt:

[1219] "Generate a program that counts specific words from a text file."

[1220] Based on this prompt, the generative AI model generates Python code and returns it to the user via the server. The user can then run this code on their device to check the frequency of specific words.

[1221] As described above, this system provides an efficient way to process open-ended responses in surveys and a means for users to easily generate program code.

[1222] The flow of the specific processing in Example 2 will be explained using Figure 17.

[1223] Step 1:

[1224] The user enters a prompt message.

[1225] The user inputs prompts for the generating AI model through the system interface. For example, they might input, "Generate a program that counts specific words from a text file." The entered prompts are then sent to the server.

[1226] Step 2:

[1227] The server receives the prompt message and passes it to the generating AI model.

[1228] The server receives prompt messages sent by the user. It then makes appropriate API calls to pass the received prompt messages to the generative AI model. The server may also convert the prompt messages into a format that the generative AI model can understand. The input is a prompt message, and the output is a request to the generative AI model.

[1229] Step 3:

[1230] The generative AI model generates program code based on the prompt statement.

[1231] The generative AI model analyzes the received prompt and generates program code based on it. For example, if a user inputs "Generate a program that counts specific words from a text file," the generative AI model will generate Python code. The input is the prompt, and the output is the generated program code.

[1232] Step 4:

[1233] The server returns the generated program code to the user.

[1234] The server receives the program code returned from the generated AI model and returns it to the user. It provides the code in a user-friendly format (e.g., a text file or text in a code editor). The input is the generated program code, and the output is the transmission of the program code to the user.

[1235] Step 5:

[1236] The user executes the generated program code.

[1237] The user executes program code received from the server on their own terminal. For example, if Python code is received, the user executes the code in their Python environment and checks the result. The input is the generated program code, and the output is the result of the program execution. Specifically, the user opens their Python environment, copies the received code, and executes it. As a result, the frequency of occurrence of a particular word is displayed.

[1238] (Application Example 2)

[1239] Next, we will describe 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."

[1240] Traditional survey systems suffered from the problem of requiring a great deal of time and effort to collect, compile, and summarize open-ended responses. Furthermore, despite advancements in content generation technology using generative AI models, methods for applying this to survey systems had not been established. Additionally, there was a lack of functionality to automatically generate and display content based on user-inputted prompts.

[1241] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1242] In this invention, the server includes means for collecting open-ended responses from a questionnaire, means for automatically aggregating and summarizing the collected open-ended responses, means for outputting the aggregation and summarization results as a report, means for inputting prompt sentences, means for generating content based on the prompt sentences using a generative AI model, and means for displaying the generated content. This enables efficient collection, aggregation, and summarization of open-ended responses from questionnaires, and further enables content generation and display using a generative AI model.

[1243] "Means for collecting open-ended responses from surveys" refers to devices or software that allow users to input answers in a free-form format and collect that information as data.

[1244] "Means for automatically aggregating and summarizing collected open-ended responses" refers to a device or software that analyzes collected open-ended response data, extracts key keywords and phrases, and concisely summarizes the overall content.

[1245] "Means for outputting aggregated and summarized results as a report" refers to a device or software for generating and outputting aggregated and summarized data as a report in a visually easy-to-understand format.

[1246] "Means for inputting prompt text" refers to a device or software for a user to input text to give instructions to a generated AI model.

[1247] "Means for generating content based on prompt text using a generative AI model" refers to a device or software that uses a generative AI model to automatically generate content such as text and images based on input prompt text.

[1248] "Means for displaying generated content" refers to a device or software for visually displaying content generated by a generative AI model to a user.

[1249] The system for implementing this invention has the following configuration. First, the user inputs free-response answers to a questionnaire using a terminal such as a smartphone. The terminal sends the inputted free-response answers to the server. The server uses natural language processing technology to automatically aggregate and summarize the collected free-response answers. Specifically, it extracts key keywords and phrases and concisely summarizes the overall content.

[1250] Next, the server generates a report containing the aggregated and summarized results, outputting it in a visually easy-to-understand format, such as graphs and charts. This allows users to intuitively grasp the survey results.

[1251] Furthermore, the user provides instructions to the generative AI model by entering prompt text. An example of a prompt text is "The story of a brave knight battling a dragon." The server uses the generative AI model to generate content based on this prompt text. The generated content is in the form of text, images, etc., and is displayed visually to the user.

[1252] The following hardware and software are required to implement this system. Hardware includes smartphones and servers. Software includes Python and the OpenAI API. Python is a programming language for implementing natural language processing techniques, and the OpenAI API is an interface for using generative AI models.

[1253] As a concrete example, if a user enters the prompt "A story about a brave knight fighting a dragon," the server will use a generative AI model to generate a story like the following.

[1254] Once upon a time, there was a brave knight named Arthur. Arthur decided to fight a fearsome dragon to protect his kingdom. Reaching the dragon's lair, Arthur drew his sword and confronted the dragon. After a fierce battle, Arthur defeated the dragon and brought peace to the kingdom.

[1255] In this way, users can enjoy generating and displaying content using generative AI models.

[1256] The flow of a specific process in Application Example 2 will be explained using Figure 18.

[1257] Step 1:

[1258] Users enter free-form responses to a survey using a device such as a smartphone. The entered responses are sent from the device to the server. The input data is in text format and is sent to the server.

[1259] Step 2:

[1260] The server analyzes the received open-ended response data using natural language processing techniques. Specifically, it extracts key keywords and phrases and summarizes the overall content concisely. The input data is the open-ended response text, and the output data is a summarized text containing the key keywords and phrases.

[1261] Step 3:

[1262] The server generates a report containing the aggregated and summarized results. The report is output in a visually easy-to-understand format, such as graphs and charts. The input data is a summarized text, while the output data is a visual report.

[1263] Step 4:

[1264] The user provides instructions to the generated AI model by entering prompts. These prompts are entered in text format and sent from the terminal to the server. The input data is the text of the prompts.

[1265] Step 5:

[1266] The server generates content based on prompt text using a generative AI model. The generative AI model automatically generates content such as text and images based on the input prompt text. The input data is the text of the prompt, and the output data is the generated content.

[1267] Step 6:

[1268] The server visually displays the generated content to the user. The generated content is sent to the terminal and displayed on the user's screen. The input data is the generated content, and the output data is the content displayed on the user's screen.

[1269] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1270] The data generation model 58 is a form of so-called generative AI (Artificial Intelligence). One example of the data generation model 58 is ChatGPT (Internet Search).<URL: https: / / openai.com / blog / chatgpt> Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1271] Other examples of generative AI include Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) are some examples.

[1272] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[1273] [Fourth Embodiment]

[1274] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1275] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1276] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1277] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[1278] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1279] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1280] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1281] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[1282] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414.

[1283] As shown in Figure 8, in the data processing device 12, specific processing is performed by the processor 28. The storage 32 stores the specific processing program 56.

[1284] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1285] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1286] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1287] Next, the identification process performed by the identification processing unit 290 of the data processing device 12 will be described.

[1288] "Example of form 1"

[1289] The system of the present invention includes means for collecting open-ended responses from a questionnaire, means for automatically aggregating and summarizing the collected open-ended responses, and means for outputting the aggregation and summarization results as a report. Specifically, the open-ended responses from the questionnaire are collected using an online questionnaire tool such as Google Forms. The collected open-ended responses are automatically aggregated and summarized using natural language processing technology. This natural language processing technology includes, for example, keyword extraction, topic modeling, and sentiment analysis. The aggregation and summarization results are output in a visually easy-to-understand format, such as graphs or charts. This reduces the workload on the person in charge and saves time.

[1290] The following describes the processing flow for each example of the form.

[1291] "Example of form 1"

[1292] Step 1: Collect open-ended responses from the survey. Specifically, use online survey tools such as Google Forms to collect open-ended responses from users.

[1293] Step 2: Automatically aggregate and summarize the collected open-ended responses. In this step, natural language processing techniques are used to analyze the content of the open-ended responses. Specifically, techniques such as keyword extraction, topic modeling, and sentiment analysis are used to extract the main keywords and phrases from the open-ended responses, and the responses are aggregated and summarized based on these.

[1294] Step 3: Output the aggregated and summarized results as a report. In this step, the aggregated and summarized results obtained in the previous step are output in a visually easy-to-understand format. Specifically, the aggregated and summarized results are displayed in graphs and charts and output as a report. This allows the person in charge to grasp the overview of the open-ended responses simply by looking at the report, thereby reducing the workload on the person in charge and saving time.

[1295] (Example 1)

[1296] Next, we will describe Embodiment 1 of Example Form 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1297] Traditional systems had problems such as requiring users to have specialized knowledge when generating programs, and requiring a high level of technical understanding to understand the processing of the generated programs. Furthermore, there was often a lack of concrete examples to understand the specific operation of the generated programs, making it unclear how users should actually use the programs.

[1298] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1299] In this invention, the server includes means for the user to input a prompt, means for generating a program using a generative AI model, means for explaining the processing of the generated program in natural language, and means for adding concrete examples to the explanation. This makes it easier for the user to generate a program and understand its processing content without specialized knowledge, and furthermore, to clearly grasp how to use it in practice through concrete examples.

[1300] A "user" is an entity that uses a system to input prompt messages and receives program generation and explanations.

[1301] A "prompt" is a set of instructions that a user inputs to the system, and it is the text that the generative AI model uses to generate a program.

[1302] A "generative AI model" is an artificial intelligence model that generates program code based on prompt text entered by a user, such as a model that uses natural language processing technology.

[1303] A "program" is the code generated by a generative AI model, consisting of a set of instructions for performing a specific task.

[1304] A "server" is a computer system that receives prompt messages from users, generates programs using a generative AI model, and explains the processing details.

[1305] "Natural language" refers to the language that humans use on a daily basis, and is used to explain the processing content of generated programs in a way that is easy for users to understand.

[1306] A "concrete example" is a specific instance that demonstrates how to actually use the generated program, helping the user understand how the program works.

[1307] This invention relates to a system in which a user inputs a prompt, a generation AI model generates a program, explains the processing content in natural language, and adds concrete examples.

[1308] First, the user accesses the system interface using a terminal and enters a prompt. For example, they might enter a prompt such as, "Generate a program that analyzes the text entered by the user and determines its sentiment."

[1309] Next, the server receives this prompt and generates a program using a generative AI model. This generative AI model, for example, is a model that uses natural language processing technology and generates appropriate program code based on the prompt entered by the user.

[1310] The generated program is written in a programming language such as Python. The server analyzes the processing content of this generated program and explains it in natural language. Specifically, it clearly states what kind of data processing and calculations the program performs. For example, if it uses the NLTK library to analyze the sentiment of text, it will explain the details.

[1311] Furthermore, the server adds concrete examples to the description. For example, if the user enters "Today is a very good day," it specifically shows how the program determines the emotion. This makes it easier for the user to understand the actual behavior of the generated program.

[1312] This system allows users to generate programs without specialized knowledge, easily understand their processing, and clearly grasp their practical usage through concrete examples.

[1313] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1314] Step 1:

[1315] The user enters a prompt message.

[1316] The user accesses the system interface using a terminal and enters prompt messages. For example, they might enter a prompt message such as, "Generate a program that analyzes the text entered by the user and determines its sentiment." The entered prompt message is then sent to the server.

[1317] Step 2:

[1318] The server generates the program using the generated AI model.

[1319] The server inputs the prompt text received from the user into a generative AI model. The generative AI model generates appropriate program code based on the prompt text. For example, a model using natural language processing techniques might generate Python code. The generated program code is stored on the server.

[1320] Step 3:

[1321] The server explains the processing of the generated program in natural language.

[1322] The server analyzes the generated program code and explains its processing in natural language. Specifically, it clearly indicates what kind of data processing and calculations the program performs. For example, if it uses the NLTK library to analyze the sentiment of text, it will explain the details. The explanation is provided to the user.

[1323] Step 4:

[1324] The server adds specific examples to the description.

[1325] The server adds specific examples to the description of the generated program. For example, it specifically shows how the program determines emotion when the user inputs "Today is a very good day." Specific examples are important for demonstrating the actual operation of the program. The description, including the specific examples, is provided to the user.

[1326] The above describes the processing flow of this system's program.

[1327] (Application Example 1)

[1328] Next, we will describe Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1329] Traditional survey systems often require manual collection, tabulation, and summarization of open-ended responses, resulting in significant time and effort. Furthermore, outputting the results as reports in a visually easy-to-understand format can be challenging. Additionally, creating high-quality ad copy requires specialized knowledge, making it difficult to generate such content quickly. To address these issues, there is a need for a system that efficiently and automatically tabulates and summarizes open-ended responses, generates visually easy-to-understand reports, and further utilizes a generation AI model to generate ad copy based on prompts.

[1330] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1331] In this invention, the server includes means for collecting open-ended responses from a questionnaire, means for automatically aggregating and summarizing the collected open-ended responses, means for outputting the aggregation and summarization results as a report, means for generating ad copy based on prompt sentences using a generation AI model, and means for displaying the generated ad copy. This enables efficient aggregation and summarization of open-ended responses and the generation of visually easy-to-understand reports, as well as the rapid generation of high-quality ad copy.

[1332] "Methods for collecting open-ended responses in surveys" refers to functions for collecting opinions and comments freely written by respondents in surveys as data.

[1333] "Methods for automatically aggregating and summarizing collected open-ended responses" refers to functions that analyze collected open-ended responses, extract key keywords and phrases, and concisely summarize the overall content.

[1334] "Means for outputting aggregated and summarized results as a report" refers to a function for generating and outputting aggregated and summarized data as a report in a visually easy-to-understand format.

[1335] "Means for generating ad copy based on prompt text using a generative AI model" refers to a function that uses a generative AI model to automatically create ad copy based on prompt text entered by the user.

[1336] "Means for displaying generated ad copy" refers to a function for visually displaying ad copy created by a generation AI model to the user.

[1337] The system for implementing this invention includes functions for collecting open-ended responses from questionnaires, automatically aggregating and summarizing the collected responses, and outputting the aggregation and summarization results as a report. It also includes a function for generating advertisement text based on prompt text using a generation AI model and displaying the generated advertisement text.

[1338] Hardware and software configuration

[1339] Hardware:

[1340] server

[1341] User devices (smartphones, smart glasses, etc.)

[1342] software:

[1343] Analysis program using natural language processing technology

[1344] Generative AI model using the OpenAI API

[1345] Database Management System

[1346] Visualization tools for generating graphs and charts

[1347] Data processing and data calculation

[1348] server:

[1349] 1. The server collects the open-ended responses from the survey submitted by the user's terminal into a database.

[1350] 2. Using natural language processing techniques, the collected open-ended responses are analyzed to extract key keywords and phrases.

[1351] 3. Based on the extracted keywords and phrases, the content of the free-response answers is automatically compiled and summarized.

[1352] 4. Generate a report containing the aggregated and summarized results in a visually easy-to-understand format (e.g., graphs and charts) and send it to the user's terminal.

[1353] User terminal:

[1354] 1. The user enters a prompt and sends it to the generating AI model.

[1355] 2. The server uses the OpenAI API to generate ad text based on the input prompt.

[1356] 3. Display the generated ad copy on the user's device.

[1357] Specific example

[1358] Example of a prompt:

[1359] "Please create ad copy for a new smartphone. Its features include a high-resolution camera and long battery life."

[1360] Example of generated ad copy:

[1361] "A new smartphone has arrived! Capture every beautiful moment with its high-resolution camera. Plus, enjoy worry-free use all day long with its long-lasting battery. Get yours now and enjoy the best experience!"

[1362] In this way, users can simply input prompt text, and the AI ​​generation model will quickly generate high-quality ad copy. Furthermore, open-ended survey responses are efficiently compiled and summarized, and output as a visually easy-to-understand report, thus improving operational efficiency.

[1363] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1364] Step 1:

[1365] The user enters their free-response answers to the survey and sends them from their device to the server.

[1366] Input: User-generated free response

[1367] Output: Free-response data sent to the server

[1368] Specific operation: When a user enters free-form answers into a survey form and presses the submit button, the device sends that data to the server.

[1369] Step 2:

[1370] The server saves the received open-ended responses to a database.

[1371] Input: Free-response data sent from the device.

[1372] Output: Open-ended response data stored in the database

[1373] Specific operation: The server stores and saves the received open-ended response data in a database.

[1374] Step 3:

[1375] The server uses natural language processing technology to analyze the open-ended responses and extract key keywords and phrases.

[1376] Input: Open-ended response data stored in the database

[1377] Output: Extracted keywords and phrases

[1378] Specific operation: The server executes a natural language processing algorithm to extract key keywords and phrases from the open-ended text data.

[1379] Step 4:

[1380] The server automatically aggregates and summarizes the free-response content based on the extracted keywords and phrases.

[1381] Input: Extracted keywords or phrases

[1382] Output: Aggregated and summarized data

[1383] Specific operation: The server aggregates the extracted keywords and phrases and uses a summarization algorithm to concisely summarize the content of the open-ended responses.

[1384] Step 5:

[1385] The server generates a report containing the aggregated and summarized results in a visually easy-to-understand format and sends it to the user's terminal.

[1386] Input: Aggregated and summarized data

[1387] Output: A visually easy-to-understand report (in graph and chart format)

[1388] Specific operation: The server uses a visualization tool to generate aggregated and summarized results as a report in graph and chart format, and sends it to the user's terminal.

[1389] Step 6:

[1390] The user enters a prompt and sends it to the generating AI model.

[1391] Input: User-entered prompt message

[1392] Output: Prompt message sent to the server

[1393] Specific operation: When the user enters a prompt message and presses the send button, the terminal sends that data to the server.

[1394] Step 7:

[1395] The server uses the OpenAI API to generate ad copy based on the input prompt text.

[1396] Input: Prompt message sent by the user

[1397] Output: Generated ad copy

[1398] Specific operation: The server calls the OpenAI API, inputs the prompt text into the AI ​​model that generates the text, and retrieves the generated ad text.

[1399] Step 8:

[1400] The server sends the generated ad text to the user's device, and the user's device displays it.

[1401] Input: Generated ad copy

[1402] Output: Ad text displayed on the user's device

[1403] Specific operation: The server sends the generated ad text to the user's terminal, and the user's terminal displays the received ad text on the screen.

[1404] (Example 2)

[1405] Next, we will describe Example 2 of the morphological example. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1406] Conventional systems have struggled to provide appropriate responses quickly and accurately to user-inputted prompts. Furthermore, the lack of a visually easy-to-understand format for displaying generated responses resulted in low user convenience. This invention aims to solve these problems by providing a system that offers quick and accurate responses to user-inputted prompts and displays those responses in a visually easy-to-understand format.

[1407] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1408] In this invention, the server includes means for the user to input a prompt sentence, means for the terminal to send the prompt sentence to the server, means for the server to pass the prompt sentence to a generation AI model, means for the generation AI model to analyze the prompt sentence and generate a response, means for the server to return the generated response to the terminal, and means for the terminal to display the response to the user. This makes it possible to provide a quick and accurate response to a prompt sentence entered by the user and to display the response in a visually easy-to-understand format.

[1409] A "user" is an individual or group that uses a system to input prompt messages and receives the generated responses.

[1410] A "prompt message" is a sentence containing a question or instruction that a user enters into the system.

[1411] A "terminal" is a device used by a user to input prompt messages and display the generated response. Examples include personal computers and smartphones.

[1412] A "server" is a computer system that receives prompt messages sent from a terminal, passes them to a generation AI model, and returns the generated response to the terminal.

[1413] A "generative AI model" is an artificial intelligence model that analyzes prompt sentences and generates appropriate responses. Examples include models that utilize natural language processing techniques.

[1414] "Natural language processing technology" is a technique that enables computers to understand and analyze human language. This makes it possible to analyze the content of a prompt and generate an appropriate response.

[1415] A "response" refers to the answer or information that a generative AI model generates by analyzing a prompt sentence.

[1416] A "visually easy-to-understand format" is a format that displays the generated response in a way that is easy for the user to understand. Examples include text, graphs, and charts.

[1417] This invention provides a system that offers a rapid and accurate response to a user-inputted prompt and displays that response in a visually easy-to-understand format. The system includes a user, a terminal, a server, and a generative AI model.

[1418] The user enters a prompt message into the terminal's input field. Examples of prompt messages include "What's the weather like today?" or "Please tell me the latest news." The terminal then sends the user's prompt message to the server. At this time, the prompt message is packaged in JSON format.

[1419] The server parses the received prompt and sends a request to the generative AI model's API endpoint. The generative AI model used is, for example, an advanced model employing natural language processing techniques (e.g., GPT-3 or GPT-4). The server parses the response received from the generative AI model and returns it to the terminal. At this point, the response is again packaged in JSON format.

[1420] The terminal analyzes the response received from the server and displays it to the user. The display format is visually easy to understand, such as text, graphs, and charts. For example, responses such as "Today's weather is sunny" or "The latest news is as follows" may be displayed.

[1421] This system allows users to obtain appropriate responses to their input prompts using a generative AI model. Furthermore, the responses are displayed in a visually easy-to-understand format, improving user convenience.

[1422] As a concrete example, consider a case where a user enters the prompt "How do I write a Hello World program in Python?". In this case, the generative AI model will generate the response "Here's how to write a Hello World program in Python: python print('Hello, World!')" and display it on the terminal.

[1423] In this way, the invention provides a rapid and accurate response to a prompt message entered by the user and displays that response in a visually easy-to-understand format.

[1424] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1425] Step 1:

[1426] The user enters a prompt message.

[1427] The user enters a prompt message into the terminal's input field. For example, they might type, "What's the weather like today?". The entered prompt message is temporarily stored in the terminal's memory.

[1428] Step 2:

[1429] The terminal sends a prompt message to the server.

[1430] The terminal sends the prompt text entered by the user to the server as an HTTP request. The prompt text is packaged in JSON format. The input is the prompt text, and the output is the HTTP request to the server.

[1431] Step 3:

[1432] The server passes the prompt message to the AI ​​model that generates it.

[1433] The server parses the received prompt and sends a request to the generative AI model's API endpoint. For example, it might send the prompt along with the API key as a POST request. The input is the prompt, and the output is the API request to the generative AI model.

[1434] Step 4:

[1435] The generative AI model analyzes the prompt sentence and generates a response.

[1436] The generative AI model analyzes the received prompt and generates an appropriate response. For example, in response to the prompt "What's the weather like today?", it generates the response "The weather is sunny today." The input is the prompt, and the output is the generated response.

[1437] Step 5:

[1438] The server returns the generated response to the terminal.

[1439] The server analyzes the response received from the generated AI model and returns it to the terminal. At this point, the response is again packaged in JSON format. The input is the generated response, and the output is the HTTP response sent to the terminal.

[1440] Step 6:

[1441] The terminal displays a response to the user.

[1442] The terminal analyzes the response received from the server and displays it to the user. For example, "Today's weather is sunny." might be displayed on the screen. The input is the generated response, and the output is what is displayed to the user.

[1443] (Application Example 2)

[1444] Next, we will describe application example 2 of form example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1445] Traditional survey systems often required manual collection, tabulation, and summarization of open-ended responses, resulting in significant time and effort. Furthermore, outputting the results in a visually easy-to-understand report format was challenging. Additionally, automated ad generation lacked a robust mechanism for generating appropriate ads based on user input. This made ad creation cumbersome and hindered the rapid creation of effective advertisements.

[1446] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1447] In this invention, the server includes means for collecting open-ended responses from a questionnaire, means for automatically aggregating and summarizing the collected open-ended responses, means for outputting the aggregation and summarization results as a report, means for automatically generating advertisements based on prompt text using a generation AI model, means for previewing the generated advertisements, means for customizing the generated advertisements, and means for saving and sharing the generated advertisements. This enables efficient collection, aggregation, and summarization of open-ended responses from questionnaires, and allows for the output of reports in a visually easy-to-understand format. Furthermore, it enables the rapid and automatic generation of effective advertisements based on prompt text, and allows users to easily customize and share them.

[1448] "Methods for collecting open-ended responses in surveys" refers to functions for collecting free-form responses from users.

[1449] "Means for automatically aggregating and summarizing collected open-ended responses" refers to a function that automatically analyzes collected open-ended responses, extracts key information, and summarizes it.

[1450] "Means for outputting aggregated and summarized results as a report" refers to a function for outputting aggregated and summarized results in a report format.

[1451] "A means of automatically generating advertisements based on prompt text using a generative AI model" refers to a function that uses a generative AI model to automatically generate advertisements based on prompt text entered by the user.

[1452] "Means for previewing generated advertisements" refers to a function that allows users to preview the generated advertisements.

[1453] "Means for customizing generated ads" refers to features that allow users to edit and customize generated ads.

[1454] "Means for saving and sharing generated ads" refers to features for saving generated ads and sharing them with other users and the platform.

[1455] The system for carrying out this invention has the following configuration: The system includes means for collecting open-ended responses from a questionnaire, means for automatically aggregating and summarizing the collected open-ended responses, means for outputting the aggregation and summarization results as a report, means for automatically generating advertisements based on prompt text using a generation AI model, means for previewing the generated advertisements, means for customizing the generated advertisements, and means for saving and sharing the generated advertisements.

[1456] Hardware and software configuration

[1457] Hardware: Smartphones, servers

[1458] Software: Generative AI models (e.g., OpenAI's GPT-4), mobile app development frameworks (e.g., React Native), cloud storage services (e.g., AWS S3)

[1459] Data processing and data calculation

[1460] 1. Collection of open-ended survey responses: Users enter their responses via a smartphone app. These responses are sent to a server and stored in a database.

[1461] 2. Automatic aggregation and summarization of open-ended responses: The server analyzes open-ended responses using natural language processing technology and extracts key keywords and phrases. This allows for the aggregation and summarization of the responses.

[1462] 3. Report Output: The aggregated and summarized results are output as a report in a visually easy-to-understand format (e.g., graphs and charts). This report can be viewed by users on their smartphones.

[1463] 4. Automated ad generation: When a user enters a prompt, that prompt is sent to the server. The generation AI model then generates ad text and images based on the prompt.

[1464] 5. Ad preview display: The generated ads are previewed to the user in the smartphone app.

[1465] 6. Ad customization: Users can edit and customize the generated ads within the app.

[1466] 7. Saving and sharing ads: Completed ads are saved to cloud storage, and a link is generated that can be shared via social media or email.

[1467] Specific example

[1468] For example, if a user wants to create an advertisement to promote the opening of a new cafe, they would enter a prompt message like the following:

[1469] Example prompt: "Create an advertisement to promote the opening of a new cafe. The target audience is young people in their 20s and 30s, and the cafe features organic coffee and a relaxing atmosphere."

[1470] When this prompt is input into the AI ​​model, the model generates advertising text and images like the following.

[1471] Example ad text: "A new cafe, 'Relax Cafe,' has opened! With organic coffee and a relaxing atmosphere, it's perfect for young people in their 20s and 30s. Please stop by!"

[1472] Example of advertising images: A generative AI model generates images of cafe interiors and organic coffee, which are then incorporated into advertisements.

[1473] In this way, users can easily create compelling advertisements and effectively reach their target audience.

[1474] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1475] Step 1:

[1476] Users enter free-form answers to a survey via a smartphone app.

[1477] Input: The user enters free-form text.

[1478] Output: The entered free-response answers are sent to the server.

[1479] Specific operation: When a user enters a free-response answer in the app's text input field and presses the "Submit" button, that data is sent to the server.

[1480] Step 2:

[1481] The server collects open-ended responses and saves them to a database.

[1482] Input: Text data of open-ended responses submitted by users.

[1483] Output: Free-response answers stored in the database.

[1484] Specific operation: The server saves the received open-ended responses to the database and generates a confirmation message when saving is complete.

[1485] Step 3:

[1486] The server uses natural language processing technology to analyze the open-ended responses and extract key keywords and phrases.

[1487] Input: Text data of open-ended responses stored in the database.

[1488] Output: Extracted main keywords and phrases.

[1489] Specific operation: The server executes a natural language processing algorithm to extract key keywords and phrases from the open-ended text.

[1490] Step 4:

[1491] The server performs aggregation and summarization based on the extracted keywords and phrases.

[1492] Input: The main keywords or phrases extracted.

[1493] Output: Aggregated and summarized data.

[1494] Specific operation: The server aggregates the extracted keywords and phrases and generates summary data using a summarization algorithm.

[1495] Step 5:

[1496] The server outputs the aggregated and summarized results as a report in a visually easy-to-understand format.

[1497] Input: Aggregated and summarized data.

[1498] Output: Reports in graph and chart format.

[1499] Specific operation: The server converts aggregated and summarized data into graphs and charts and generates them as reports.

[1500] Step 6:

[1501] The user enters a prompt and sends it to the generating AI model.

[1502] Input: The prompt text entered by the user.

[1503] Output: The prompt text sent to the generating AI model.

[1504] Specific operation: When the user enters a prompt in the app's text input field and presses the "Generate" button, that data is sent to the generating AI model.

[1505] Step 7:

[1506] The generative AI model generates ad text and images based on the prompt text.

[1507] Input: The prompt text sent to the generating AI model.

[1508] Output: Generated ad text and images.

[1509] Specific operation: The generative AI model analyzes the prompt text and generates advertising text and images.

[1510] Step 8:

[1511] The server displays a preview of the generated advertisement to the user.

[1512] Input: Generated ad text and images.

[1513] Output: Ad preview displayed on the user's smartphone.

[1514] Specific operation: The server sends the generated advertisement to the user's smartphone and displays a preview in the app.

[1515] Step 9:

[1516] Users can customize the ads that are generated.

[1517] Input: The ad displayed in preview.

[1518] Output: Customized ads.

[1519] Specific operation: Users can edit and customize ad text and images using the app's editing function.

[1520] Step 10:

[1521] The server saves the customized ad to cloud storage and generates a sharing link.

[1522] Input: Customized ad.

[1523] Output: Advertisements and shared links stored in cloud storage.

[1524] Specific operation: The server saves customized advertisements to cloud storage and generates links that can be shared via social media or email.

[1525] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1526] "Example of form 1"

[1527] This invention relates to a system that collects open-ended responses from questionnaires, automatically aggregates and summarizes the collected responses, and outputs the aggregated and summarized results as a report. As a means of automatically aggregating and summarizing the open-ended responses, this system uses natural language processing technology to analyze the content of the responses and extract key keywords and phrases. Furthermore, this system analyzes the content of the open-ended responses using an emotion engine that recognizes user emotions. This emotion engine extracts user emotions from the text data of the open-ended responses and reflects those emotions in the report. Specifically, if a user gives the open-ended response "This product is very good," the emotion engine extracts the emotion "good" from this response and reflects that emotion in the report. The report output means outputs the aggregated and summarized results along with the emotions extracted by the emotion engine in a visually easy-to-understand format, such as a graph or chart. This allows the person in charge to grasp the overview of the open-ended responses and user emotions simply by looking at the report, thereby reducing the workload and saving time.

[1528] The following describes the processing flow for each example of the form.

[1529] "Example of form 1"

[1530] Step 1: Collect open-ended responses from the survey. Specifically, use a survey tool such as Google Forms to collect open-ended responses from users.

[1531] Step 2: Automatically aggregate and summarize the collected open-ended responses. Specifically, natural language processing technology is used to analyze the content of the open-ended responses and extract key keywords and phrases.

[1532] Step 3: Recognize the user's emotions. Specifically, use an emotion engine to analyze the content of open-ended responses and extract the user's emotions.

[1533] Step 4: Reflect the extracted emotions in the report. Specifically, the emotions extracted by the emotion engine are reflected in the report and output in a visually easy-to-understand format, such as a graph or chart.

[1534] Step 5: Output the aggregated / summarized results and sentiment analysis results as a report. Specifically, the aggregated / summarized results and sentiment analysis results are integrated and output as a single report. This allows the person in charge to understand the overview of the open-ended responses and the users' sentiments simply by looking at the report.

[1535] (Example 1)

[1536] Next, we will describe Embodiment 1 of Example Form 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1537] In conventional systems, the process of generating appropriate responses to user-inputted prompts was complex and difficult to perform efficiently. Furthermore, the quality and display format of the generated results were inconsistent, sometimes making them difficult for users to understand. This resulted in a degraded user experience and limited the system's usefulness.

[1538] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1539] In this invention, the server includes means for receiving prompt statements and performing preprocessing, means for inputting prompt statements into a generation AI model and performing inference, and means for post-processing the generated results. This makes it possible to consistently and efficiently perform everything from preprocessing prompt statements to post-processing of the generated results.

[1540] A "user" is the entity that operates the system and enters prompt messages.

[1541] A "terminal" is a device used by a user to operate something, and includes PCs, smartphones, and other similar devices.

[1542] A "prompt" is the text of instructions or questions that a user inputs to the generated AI model.

[1543] A "server" is a computing system that receives prompt messages, performs preprocessing, inputs data into a generative AI model, performs inference, performs postprocessing, and sends the results.

[1544] "Preprocessing" refers to the process of normalizing and tokenizing the text of a prompt message.

[1545] A "generative AI model" is an artificial intelligence model that performs inference based on input prompt sentences and generates results.

[1546] "Inference" is the process by which a generative AI model generates responses or results based on a prompt.

[1547] "Post-processing" refers to the process of formatting the generated results and removing unnecessary information.

[1548] "Results" refer to the responses and information generated by the generative AI model based on the prompt text.

[1549] "Display" refers to the act of providing the user with a visual representation of the results received by the terminal from the server.

[1550] This invention is a system in which a user inputs a prompt, and a generation AI model generates a response or result based on that prompt, which is then provided to the user. Specific embodiments of this system are described below.

[1551] First, the user enters a prompt using a device (such as a PC or smartphone). An example of a prompt is, "Please use AI to come up with a new recipe."

[1552] The terminal sends the entered prompt text to the server. HTTP requests are used for communication. The server is a computing system with high computing power, specifically a server equipped with a high-performance GPU.

[1553] The server receives prompt messages sent from the terminal and performs preprocessing. Preprocessing includes text normalization (e.g., converting full-width characters to half-width characters) and tokenization (e.g., splitting sentences into words or phrases). Python libraries (e.g., pandas, numpy) are used for preprocessing.

[1554] Next, the server inputs the pre-processed prompt sentences into the generative AI model. For example, GPT-4 is used as the generative AI model. The model performs inference based on the input prompt sentences and generates results. TensorFlow or PyTorch is used for the model's inference.

[1555] The generated results are post-processed on the server. Post-processing includes formatting the results (e.g., converting to JSON format) and removing unnecessary information.

[1556] The post-processed results are sent from the server to the terminal. HTTP responses are used for communication. The terminal displays the received results to the user. HTML and CSS are used for display.

[1557] As a concrete example, if a user enters the prompt "Please use AI to come up with a new recipe," the server receives this prompt, performs preprocessing, and inputs it into the AI ​​model. The model generates a new recipe, and the server processes the result and sends it to the terminal. The terminal then displays the generated recipe to the user.

[1558] In this way, users can obtain new recipes using the generative AI model. This system can efficiently handle everything from pre-processing of prompts to post-processing of the generated results, providing users with high-quality responses.

[1559] The flow of the specific processing in Example 1 will be explained using Figure 15.

[1560] Step 1:

[1561] The user enters a prompt message.

[1562] The user enters "Use AI to come up with a new recipe" into the terminal's input field. The entered prompt is saved to the terminal's memory.

[1563] Step 2:

[1564] The terminal sends a prompt message to the server.

[1565] The terminal sends the entered prompt text to the server as an HTTP request. The input is the prompt text, and the output is the request sent to the server.

[1566] Step 3:

[1567] The server receives the prompt message and performs preprocessing.

[1568] The server receives prompt messages sent from the terminal. It then performs text normalization (e.g., converting full-width characters to half-width characters) and tokenization (e.g., splitting sentences into words and phrases) on the received prompt messages. The input is the prompt message, and the output is pre-processed text data.

[1569] Step 4:

[1570] The server inputs prompt messages into the generated AI model and performs inference.

[1571] The server inputs pre-processed prompt sentences into a generating AI model (e.g., GPT-4). The model performs inference based on the input prompt sentences and generates results. The input is pre-processed text data, and the output is the generated result (e.g., a new recipe).

[1572] Step 5:

[1573] The server then performs post-processing on the generated results.

[1574] The server performs post-processing on the results obtained from the generated AI model. Post-processing includes formatting the results (e.g., converting to JSON format) and removing unnecessary information. The input is the generated result, and the output is the post-processed data.

[1575] Step 6:

[1576] The server sends the results to the terminal.

[1577] The server sends the post-processed results to the terminal as an HTTP response. The input is the post-processed data, and the output is the response sent to the terminal.

[1578] Step 7:

[1579] The device displays the results to the user.

[1580] The terminal displays the results received from the server to the user. HTML and CSS are used for the display. The input is the response data from the server, and the output is the information visually presented to the user.

[1581] (Application Example 1)

[1582] Next, we will describe Application Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1583] Traditional survey systems often involve manual collection, tabulation, summarization, and report generation of open-ended responses, resulting in time-consuming and labor-intensive processes. Furthermore, generating advertising content requires specialized knowledge, making it difficult to reach target audiences quickly and effectively. To address these challenges, a system integrating automated tabulation and summarization of open-ended responses with automated advertising content generation is needed.

[1584] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1585] In this invention, the server includes means for collecting open-ended responses to a questionnaire, prompts for aggregating and summarizing the collected open-ended responses and generating a report of the aggregated and summarized results, and means for generating the report using a generation AI model, prompts for generating content based on the generated report and means for generating the content using the generation AI model, means for previewing the generated content and accepting editing operations, means for publishing the edited content, and means for analyzing the performance of the published content. This enables efficient aggregation and summarization of open-ended responses, and automatic generation and publication of advertising content based on prompts.

[1586] "Means for collecting open-ended responses in surveys" refers to interfaces and functions for collecting responses from users in a free-form format.

[1587] "Means for automatically aggregating and summarizing collected open-ended responses" refers to algorithms or software that analyze collected open-ended response data, extract key information, and summarize it.

[1588] "Means for outputting aggregated and summarized results as a report" refers to a function for generating and outputting aggregated and summarized data as a report in a visually easy-to-understand format.

[1589] "Means for entering prompt text" refers to a text input interface for users to give instructions to the generated AI model.

[1590] "Means of generating content based on prompt text using a generative AI model" refers to algorithms and software that enable a generative AI model to automatically generate content such as text and images based on input prompt text.

[1591] "Means for previewing and editing generated content" refers to interfaces and functions that allow users to review generated content and edit it as needed.

[1592] "Means for publishing generated content" refers to functions for publishing generated and edited content on internet platforms, social networking services, etc.

[1593] "Means of analyzing the performance of published content" refers to tools and software used to collect and analyze performance data such as the number of views and clicks on published content.

[1594] The system for implementing this invention has the following configuration. First, the user inputs free-form answers to a questionnaire about a specific product using a terminal such as a smartphone. The terminal is equipped with means for collecting free-form answers to the questionnaire and transmits the free-form answers entered by the user to a server.

[1595] The server is equipped with a means to automatically aggregate and summarize the collected open-ended responses. This means analyzes the content of the open-ended responses using natural language processing techniques and extracts key keywords and phrases. Specifically, it uses natural language processing libraries (e.g., NLTK, spaCy) to analyze the text data and extract important information. The server is also equipped with a means to output the aggregated and summarized results as a report. This means outputs the aggregated and summarized results in a visually easy-to-understand format, such as graphs and charts. This uses data visualization tools (e.g., Matplotlib, D3.js). In this embodiment, the server generates a report using a prompt message that instructs the server to aggregate and summarize the collected open-ended responses and generate the aggregated and summarized results as a report, and a generation AI model.

[1596] Furthermore, the user provides instructions to the generating AI model using prompts that instruct it to generate content based on the generated report, thereby generating the content. An example of a prompt message is: "Enter the generated report and, using this report as a reference, create an advertisement promoting a new eco-friendly detergent to housewives in their 30s." In this example, the specific product is a detergent.

[1597] The server is equipped with a means for generating content based on prompt text using a generative AI model. This means uses a generative AI model (e.g., OpenAI GPT-4) to generate advertising text and images based on the input prompt text.

[1598] The generated content is provided to the user through means of previewing and editing on their device. Users can review the generated content and edit it as needed. The edited content is then published to internet platforms and social media via the server.

[1599] Finally, the server has a means to analyze the performance of the published content. This means uses tools (e.g., Google Analytics) to collect and analyze performance data such as the number of views and clicks on the published content.

[1600] In this way, efficient aggregation and summarization of open-ended responses, as well as the automatic generation and publication of advertising content based on prompt text, become possible. Alternatively, the content may be modified using a prompt message instructing it to change based on the analyzed performance, along with a generative AI model. This allows the content to be modified in order to improve its performance.

[1601] The flow of a specific process in Application Example 1 will be explained using Figure 16.

[1602] Step 1:

[1603] Users input free-response answers to a survey using a smartphone or other device. The entered free-response answers are sent from the device to the server. The input data is free-response text, and the output is the free-response data sent to the server.

[1604] Step 2:

[1605] The server automatically aggregates and summarizes the collected open-ended responses. It analyzes the content of the open-ended responses using natural language processing techniques and extracts key keywords and phrases. The input data is the open-ended response text, and the output is summarized data containing the key keywords and phrases. Specifically, it analyzes the text data using natural language processing libraries (e.g., NLTK, spaCy). Alternatively, it generates aggregated and summarized results of the collected open-ended responses using a prompt message instructing the server to aggregate and summarize the collected open-ended responses, along with a generative AI model.

[1606] Step 3:

[1607] The server outputs the aggregated and summarized results as a report. The aggregated and summarized results are output in a visually easy-to-understand format, such as graphs or charts. The input data is summarized data, and the output is a visually represented report. Data visualization tools (e.g., Matplotlib, D3.js) are used. Specifically, the report is generated using a prompt message that instructs the server to generate a report based on the generated aggregated and summarized results, and a generating AI model.

[1608] Step 4:

[1609] The user provides instructions to the generating AI model by entering prompts. The input data consists of reports and prompts, and the output consists of reports and prompts sent to the server. An example of a prompt is to input a generated report along with the instruction, "Using this report as a reference, create an advertisement promoting a new eco-friendly detergent to housewives in their 30s."

[1610] Step 5:

[1611] The server generates content based on prompt text using a generative AI model. The input data is the prompt text, and the output is the generated advertisement text and images. A generative AI model (e.g., OpenAI GPT-4) is used to generate content based on the input prompt text.

[1612] Step 6:

[1613] The generated content is provided to the user through means of previewing and editing on the device. The user reviews the generated content and edits it as needed. The input data is the generated content, and the output is the edited content.

[1614] Step 7:

[1615] The edited content is published to internet platforms and social media via a server. The input data is the edited content, and the output is the published content.

[1616] Step 8:

[1617] The server analyzes the performance of published content. It collects and analyzes performance data such as the number of views and clicks on published content. The input data is performance data, and the output is the analysis results. Data analysis tools (e.g., Google Analytics) are used.

[1618] (Example 2)

[1619] Next, we will describe Example 2 of the morphological example. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1620] Traditional survey systems often involved manual collection, tabulation, and summarization of open-ended responses, resulting in significant time and effort. Furthermore, generating specific programs required programming knowledge, making it difficult for users without specialized expertise. Additionally, the quality and efficiency of the generated program code were often not guaranteed.

[1621] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting free-response answers from a questionnaire, means for automatically aggregating and summarizing the collected free-response answers, means for outputting the aggregation and summarization results as a report, means for the user to input prompt sentences, means for the generating AI model to generate program code based on the prompt sentences, and means for returning the generated program code to the user. This enables efficient aggregation and summarization of free-response answers and allows the user to easily generate high-quality program code.

[1622] A "survey" is a set of questions designed to collect specific information.

[1623] An "open-ended response" is a type of answer in which the respondent writes freely in their own words.

[1624] "Means of collection" refers to the methods and devices used to collect data.

[1625] "Means of automatic aggregation and summarization" refers to methods and devices for mechanically organizing collected data and extracting important information.

[1626] "Means of outputting as a report" refers to methods or devices for providing aggregated and summarized results in document format.

[1627] "User" refers to an individual or group that uses the system.

[1628] A "prompt message" is text used to input specific instructions or questions to a generative AI model.

[1629] A "generative AI model" is an artificial intelligence model that generates program code or other output based on input prompts.

[1630] "Program code" is a set of instructions that a computer executes.

[1631] A "server" is a computer system that provides data and services over a network.

[1632] This invention relates to a system that efficiently collects, compiles, and summarizes open-ended responses from questionnaires, and further enables users to easily generate program code. Specific embodiments of this system are described below.

[1633] First, users enter free-form responses to a survey through the system interface. These responses are collected by the server. The server uses natural language processing technology to automatically aggregate and summarize the collected responses. Specifically, the server extracts key keywords and phrases and then aggregates and summarizes the data based on these.

[1634] Next, the server outputs the aggregated and summarized results as a report. This report is provided in a visually easy-to-understand format, such as graphs and charts. This allows users to intuitively grasp the content of the open-ended responses.

[1635] Furthermore, this system provides a means for the user to input prompts. The user inputs specific instructions or questions to the generating AI model as prompts. For example, the user might input a prompt such as, "Generate a program that counts specific words from a text file."

[1636] The server receives prompt messages from the user and passes them to the generative AI model. The generative AI model generates program code based on the prompt messages. The generated program code is returned to the user via the server. The user can then execute this program code on their own device.

[1637] The following hardware and software will be used to implement this system. The server is a computer system equipped with a high-performance processor and sufficient memory, and features an NVIDIA GPU. For software, deep learning frameworks such as TensorFlow and PyTorch will be used. This will allow the generative AI model to operate efficiently and perform data processing and calculations based on prompts entered by the user.

[1638] As a concrete example, consider the case where a user inputs the following prompt into the AI ​​model:

[1639] Example of a prompt:

[1640] "Generate a program that counts specific words from a text file."

[1641] Based on this prompt, the generative AI model generates Python code and returns it to the user via the server. The user can then run this code on their device to check the frequency of specific words.

[1642] As described above, this system provides an efficient way to process open-ended responses in surveys and a means for users to easily generate program code.

[1643] The flow of the specific processing in Example 2 will be explained using Figure 17.

[1644] Step 1:

[1645] The user enters a prompt message.

[1646] The user inputs prompts for the generating AI model through the system interface. For example, they might input, "Generate a program that counts specific words from a text file." The entered prompts are then sent to the server.

[1647] Step 2:

[1648] The server receives the prompt message and passes it to the generating AI model.

[1649] The server receives prompt messages sent by the user. It then makes appropriate API calls to pass the received prompt messages to the generative AI model. The server may also convert the prompt messages into a format that the generative AI model can understand. The input is a prompt message, and the output is a request to the generative AI model.

[1650] Step 3:

[1651] The generative AI model generates program code based on the prompt statement.

[1652] The generative AI model analyzes the received prompt and generates program code based on it. For example, if a user inputs "Generate a program that counts specific words from a text file," the generative AI model will generate Python code. The input is the prompt, and the output is the generated program code.

[1653] Step 4:

[1654] The server returns the generated program code to the user.

[1655] The server receives the program code returned from the generated AI model and returns it to the user. It provides the code in a user-friendly format (e.g., a text file or text in a code editor). The input is the generated program code, and the output is the transmission of the program code to the user.

[1656] Step 5:

[1657] The user executes the generated program code.

[1658] The user executes program code received from the server on their own terminal. For example, if Python code is received, the user executes the code in their Python environment and checks the result. The input is the generated program code, and the output is the result of the program execution. Specifically, the user opens their Python environment, copies the received code, and executes it. As a result, the frequency of occurrence of a particular word is displayed.

[1659] (Application Example 2)

[1660] Next, we will describe application example 2 of form example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1661] Traditional survey systems suffered from the problem of requiring a great deal of time and effort to collect, compile, and summarize open-ended responses. Furthermore, despite advancements in content generation technology using generative AI models, methods for applying this to survey systems had not been established. Additionally, there was a lack of functionality to automatically generate and display content based on user-inputted prompts.

[1662] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1663] In this invention, the server includes means for collecting open-ended responses from a questionnaire, means for automatically aggregating and summarizing the collected open-ended responses, means for outputting the aggregation and summarization results as a report, means for inputting prompt sentences, means for generating content based on the prompt sentences using a generative AI model, and means for displaying the generated content. This enables efficient collection, aggregation, and summarization of open-ended responses from questionnaires, and further enables content generation and display using a generative AI model.

[1664] "Means for collecting open-ended responses from surveys" refers to devices or software that allow users to input answers in a free-form format and collect that information as data.

[1665] "Means for automatically aggregating and summarizing collected open-ended responses" refers to a device or software that analyzes collected open-ended response data, extracts key keywords and phrases, and concisely summarizes the overall content.

[1666] "Means for outputting aggregated and summarized results as a report" refers to a device or software for generating and outputting aggregated and summarized data as a report in a visually easy-to-understand format.

[1667] "Means for inputting prompt text" refers to a device or software for a user to input text to give instructions to a generated AI model.

[1668] "Means for generating content based on prompt text using a generative AI model" refers to a device or software that uses a generative AI model to automatically generate content such as text and images based on input prompt text.

[1669] "Means for displaying generated content" refers to a device or software for visually displaying content generated by a generative AI model to a user.

[1670] The system for implementing this invention has the following configuration. First, the user inputs free-response answers to a questionnaire using a terminal such as a smartphone. The terminal sends the inputted free-response answers to the server. The server uses natural language processing technology to automatically aggregate and summarize the collected free-response answers. Specifically, it extracts key keywords and phrases and concisely summarizes the overall content.

[1671] Next, the server generates a report containing the aggregated and summarized results, outputting it in a visually easy-to-understand format, such as graphs and charts. This allows users to intuitively grasp the survey results.

[1672] Furthermore, the user provides instructions to the generative AI model by entering prompt text. An example of a prompt text is "The story of a brave knight battling a dragon." The server uses the generative AI model to generate content based on this prompt text. The generated content is in the form of text, images, etc., and is displayed visually to the user.

[1673] The following hardware and software are required to implement this system. Hardware includes smartphones and servers. Software includes Python and the OpenAI API. Python is a programming language for implementing natural language processing techniques, and the OpenAI API is an interface for using generative AI models.

[1674] As a concrete example, if a user enters the prompt "A story about a brave knight fighting a dragon," the server will use a generative AI model to generate a story like the following.

[1675] Once upon a time, there was a brave knight named Arthur. Arthur decided to fight a fearsome dragon to protect his kingdom. Reaching the dragon's lair, Arthur drew his sword and confronted the dragon. After a fierce battle, Arthur defeated the dragon and brought peace to the kingdom.

[1676] In this way, users can enjoy generating and displaying content using generative AI models.

[1677] The fl...

Claims

[Claim 1] Methods for collecting open-ended responses from surveys, A means for automatically aggregating and summarizing the collected open-ended responses using natural language processing technology, A means for generating the report, using a generating AI model instructed by a prompt statement that instructs the generation of a report including the results of the aggregation and the results of the summary, A means for generating content, using a generation AI model instructed by a prompt statement that instructs the model to generate advertising text and images as content based on the generated report, A means for previewing the generated content and accepting editing operations, Means for publishing the edited content, A means for analyzing the performance of the aforementioned published content, including the number of views and clicks, A means for receiving a prompt message from the user indicating the content to be changed based on the analyzed performance, and for automatically changing the content using a generative AI model indicated by the prompt message indicating the content to be changed, A system that includes this.

Images