System

Abstract

A system is provided.SOLUTION: The system includes an entry means for a AI developer to enter a generation AI, a battle means for the entered generation AI to compete with each other for a specific subject, and an evaluation and disclosure means for evaluating and disclosing a battle result.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Currently, the development of generative AI is progressing rapidly among individuals and companies, but there is no fair competition platform for evaluating and optimizing its performance. Furthermore, there is a problem that ordinary consumers have few opportunities to intuitively understand and experience technological advances in generative AI. As a result, the technological improvement of AI developers is hindered, and the understanding and adoption of AI technology among ordinary users is hindered. The purpose of this invention is to solve these problems. [Means for solving the problem]

[0005] To solve the above problems, the present invention provides the following means. First, an entry means is provided for AI developers to enter their own generated AIs. This entry means includes a transmission means for transmitting information about the generated AI from a user terminal to a server and a storage means for saving the transmitted information in a database. Next, a battle means is provided in which the entered generated AIs compete against each other on a specific challenge. This battle means includes a collection means for providing a challenge and collecting the solution results of each generated AI, and an evaluation means for analyzing the collected results and determining a winner based on evaluation criteria. Furthermore, an evaluation and publication means is provided for evaluating and publishing the battle results. This means includes a scoring means for scoring the battle results and creating a ranking, and a publication means for saving the rankings in a database and updating the publication page. This allows AI developers to compete fairly against each other in terms of the performance of their generated AIs and receive evaluations, while also providing a forum for general consumers to intuitively understand and experience the technological evolution of generated AIs.

[0006] "Entry means" refers to the means by which a generation AI developer registers their generation AI in the system.

[0007] "Transmission means" refers to a means for transmitting information about the generated AI from the user terminal to the server.

[0008] The "storage means" is a means by which the server stores the information of the generated AI received through the transmission means in a database.

[0009] "Battle methods" are the means by which entered generated AIs compete against each other on specific tasks.

[0010] The "collection means" is the means by which the server provides tasks to each generation AI and collects the results of their solutions.

[0011] The "evaluation means" is a means for analyzing the solution results of the generated AI collected by the server and determining the winner based on the evaluation criteria.

[0012] "Means for evaluation and disclosure" refers to means for evaluating and disclosing battle results.

[0013] The "scoring means" is a means by which the server calculates a score for the battle result based on an evaluation standard.

[0014] The "publication means" is a means for storing the scores calculated by the scoring means in a database in a ranking format and updating the public page.

[0015] "Generative AI" refers to algorithms and models generated using machine learning and artificial intelligence techniques.

[0016] "User terminal" refers to a device used by a generation AI developer to input and transmit information about the generation AI using an entry means. [Brief explanation of the drawings]

[0017] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0018] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0019] First, the terms used in the following description will be explained.

[0020] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0021] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0022] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0023] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0024] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0025] [First embodiment]

[0026] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0027] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0028] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0029] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0030] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0031] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0032] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0033] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0034] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0035] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0036] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0037] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0038] The present invention is a system for competing in problem-solving ability using generative AI, and is implemented in the following form.

[0039] Entry function

[0040] 1. User enters AI

[0041] Users use their device to enter their own generated AI. A form is displayed on the user device in which they can enter the name, description, executable file, developer information, etc. of the generated AI. Based on this information, they press the entry button to submit the information.

[0042] 2. The device sends the entry information to the server

[0043] The user terminal sends the entry information entered by the user to the server, where it is packaged in a standard format such as JSON.

[0044] 3. The server receives the entry information

[0045] The server receives the entry information sent from the user's device and stores it in a database, including the name of the generated AI, a description, the path to the executable file, and the developer's contact information.

[0046] 4. Notification of entry completion

[0047] The server notifies the user terminal that the entry has been completed successfully. An entry completion message is displayed on the user terminal, allowing the user to confirm that the entry has been completed successfully.

[0048] Battle Features

[0049] 1. The server starts the battle

[0050] The server will start a battle between the registered generated AIs at a pre-set time. At the start of the battle, specific tasks will be set for each generated AI.

[0051] 2. The server provides each AI with a task

[0052] The server provides the details of the challenge (e.g., input data and problem definition) to the submitted AI generators, who then generate a solution to the challenge.

[0053] 3. The server receives the results of each AI

[0054] Each generation AI outputs the solution to the problem and returns it to the server, which receives these results and temporarily stores them in a database.

[0055] 4. The server evaluates the results

[0056] The server evaluates the received results based on evaluation criteria (e.g., solution accuracy, speed, resource consumption, etc.) and assigns a score to each generated AI.

[0057] 5. The server decides the winner

[0058] Based on the evaluation results, the server determines the best performing AI generator as the winner, and the winner's information is stored in the database.

[0059] Result announcement and evaluation function

[0060] 1. The server evaluates the battle results

[0061] After the battle, the server will tally the scores of each AI generator and create a ranking, which is a fair evaluation of the AI ​​generator's ability to solve the problem.

[0062] 2. The server creates the rankings

[0063] The server creates a ranking of the AI ​​generators based on the collected evaluation results. This ranking is based on the performance of the AI ​​generators.

[0064] 3. The server stores the results in a database

[0065] The server saves the rankings it creates in a database, which also stores past battle results and the evaluation history of each generated AI.

[0066] 4. The server updates the public page

[0067] The server updates the public page based on the ranking information stored in the database, allowing general users to view the latest battle results and the rankings of the generated AI.

[0068] 5. User checks the results

[0069] Users can access the public page using their devices to check the battle results and the rankings of the generated AI, allowing general users to intuitively understand and experience the technological evolution of generated AI.

[0070] The above system provides AI developers with a forum where they can fairly compete and receive evaluations on the performance of their generative AI, and it also provides general consumers with a forum where they can intuitively understand and experience the technological evolution of generative AI.

[0071] The processing flow will be explained below.

[0072] Entry function

[0073] Step 1: User enters AI

[0074] The user enters information about their generated AI (name, description, executable file, developer information, etc.) into the user terminal and presses the entry button.

[0075] Step 2: The device sends the entry information to the server

[0076] The terminal sends the entry information entered by the user to the server in a standard format such as JSON.

[0077] Step 3: Server receives entry information

[0078] The server receives the entry information sent from the terminal.

[0079] Step 4: The server saves the entry information to a database

[0080] The server stores the received entry information in a database, including the name of the generated AI, its description, the path to the executable file, and developer information.

[0081] Step 5: The server notifies you that the entry is complete

[0082] The server notifies the user terminal that the entry has been successfully completed, and an entry completion message is displayed on the user terminal.

[0083] Battle Features

[0084] Step 1: The server starts the battle

[0085] The server will start a battle between the entered generated AIs at a pre-set time.

[0086] Step 2: The server provides each AI with a task

[0087] The server provides the details of the challenge (e.g., input data and problem definition) to the entered generated AI.

[0088] Step 3: The server receives the results of each AI.

[0089] Each generation AI generates a problem-solving result and returns it to the server, which receives the result and temporarily stores it in a database.

[0090] Step 4: The server evaluates the results

[0091] The server evaluates the received results based on criteria (e.g., accuracy, speed, resource consumption), and the evaluation results are scored.

[0092] Step 5: Server determines winner

[0093] Based on the evaluation results, the server determines the AI ​​generator with the best performance as the winner, and the winner's information is recorded in the database.

[0094] Result announcement and evaluation function

[0095] Step 1: The server evaluates the battle results

[0096] The server tally the scores of each generated AI and create a ranking based on the evaluation criteria.

[0097] Step 2: The server creates the rankings

[0098] The server uses the evaluation results to create a ranking of the generated AIs, which is ranked based on the performance of the generated AIs.

[0099] Step 3: The server stores the results in a database

[0100] The server saves the ranking information it creates in a database, which also stores past battle results and evaluation history.

[0101] Step 4: The server updates the public page

[0102] The server updates the public page with the latest information based on the rankings stored in the database.

[0103] Step 5: User confirms the results

[0104] Users can access the public page using their devices to check the battle results and the ranking of the generated AI, allowing them to intuitively understand and experience the technological evolution of the generated AI.

[0105] Example 1

[0106] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0107] Currently, many AI developers are developing generative AI, but there is a lack of venues for objectively evaluating the performance of these generative AIs and for them to compete against each other. This makes it difficult for generative AI developers to compare the capabilities of their own AI with those of other AIs and identify areas for improvement. It is also difficult for ordinary users to intuitively understand and experience the evolution and characteristics of generative AI. A new system is needed to solve this problem and promote technological advances in generative AI.

[0108] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0109] In this invention, the server includes a storage means for receiving entry information and storing it in a database, a management means for checking the start time of the battle and providing challenges to the AI ​​generators, and an evaluation means for evaluating the results of the challenges and determining the best AI generators. This allows the performance of AI generators to be fairly evaluated and allows developers to compete against each other.

[0110] "Entry means" refers to the means by which an AI developer enters a generated AI into the system, and includes information sent from the user terminal to the server.

[0111] "Battle means" refers to the means by which entered generated AIs compete against each other on specific challenges, and includes the function of providing challenges and collecting the results of their solutions.

[0112] "Means for evaluation and publication" refers to means for evaluating the results of battles and publishing the results, and includes functions for evaluating the generating AI based on evaluation criteria, storing the results in a database, and publishing them to the general public.

[0113] "Input means" refers to the means by which a user inputs and submits entry information for the generated AI, and refers to the input form provided on the user's terminal.

[0114] "Storage means" refers to the means by which the server stores the received entry information in a database, and includes temporary and permanent storage of data.

[0115] The "management means" is a means by which the server checks the start time of the battle and starts the battle at the set time, and includes a function for managing the progress of the battle.

[0116] "Task providing means" refers to the means by which the server provides specific tasks to the entered generation AI, and includes the generation and distribution of tasks.

[0117] "Evaluation means" refers to the means by which the server evaluates and scores the solution results of each generated AI based on the evaluation criteria, and includes analyzing the results and applying the evaluation criteria.

[0118] "Determination means" refers to the means by which the server determines the best generating AI and stores that information in a database, including determining the winner and storing the results.

[0119] "Server communication means" refers to the means by which the server provides challenges to each generation AI and each generation AI transmits a solution to the server, and includes establishing communication and sending and receiving data.

[0120] "Evaluation criteria setting means" means a means for the evaluation means to set criteria for evaluating each generative AI based on the accuracy, speed, and resource consumption of the solution, and includes the definition and application of the evaluation criteria.

[0121] This invention is a system for competing in problem-solving ability using generative AI, and it operates in cooperation with users, terminals, and a server. The system has the following functions.

[0122] Entry function

[0123] To enter a generated AI, a user first uses their device to enter the name, description, executable file path, and developer information into an entry form. Specific hardware used for this is a personal computer or smartphone, and the user accesses the form via a browser. A web browser (e.g., Google Chrome or Safari) is used as the software.

[0124] When the user presses the entry button, the terminal converts the entered information into JSON format and sends it to the server as an HTTP POST request. The server analyzes the received data and saves it in a database. This saving process uses the server's internal database system (e.g., MySQL, PostgreSQL). Once the entry is complete, the server sends a completion notification to the user's terminal. The user's terminal displays the message "Entry completed successfully."

[0125] As a specific example, a user enters a generation AI called "AI_Champion" and enters the description, "This is an AI that analyzes the emotions in English sentences."

[0126] Battle Features

[0127] The server starts a battle between the entered generated AIs at a pre-set time. It checks the start time of the battle and provides specific tasks to the entered generated AIs. For example, a task might be to "analyze emotions from English sentences and assign a score." The server packages and sends the task details to each generated AI.

[0128] Each generation AI generates a solution to the provided problem and sends the solution results to the server. The server temporarily stores the received solution results in a database. The server then evaluates and scores the results of each generation AI based on evaluation criteria (e.g., solution accuracy, speed, resource consumption, etc.). Various evaluation algorithms and evaluation criterion setting methods are used for the evaluation process.

[0129] As an example of specific operation, the server receives an emotion score of 0.95 assigned by "AI_Champion" and evaluates it as "AI_Champion's score is 95 points."

[0130] Result announcement and evaluation function

[0131] After the battle is over, the server tally the scores of each generated AI and create a ranking. This ranking is based on the performance of the generated AI and is stored in a database. The server updates a public page based on the ranking information, allowing general users to view the latest battle results and the rankings of the generated AI. Users can access the public page using their devices to check the results.

[0132] As a specific example, a user accesses a public page and sees the information "AI_Champion is in first place."

[0133] Prompt Sentence Examples

[0134] "Create a generative AI that takes an English sentence as input, analyzes its sentiment, and gives it a score."

[0135] This invention provides AI developers with a forum where they can fairly compete and have the performance of their generative AI evaluated, and it also provides general users with a forum where they can intuitively understand and experience the technological evolution of generative AI.

[0136] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0137] Entry function

[0138] Step 1: Display the entry form

[0139] The user opens the entry page in a browser and enters the name, description, executable file path, and developer information of the generated AI. HTML and JavaScript are used to display the required fields in this form.

[0140] Input: User-entered name, description, executable path, and developer information for the generated AI.

[0141] Output: The input form reflects the user's input.

[0142] Step 2: Prepare input information for submission

[0143] When the user clicks the entry button, the terminal converts the input information into JSON format and packages it as an HTTP POST request.

[0144] Enter: When the user clicks the enter button.

[0145] Output: Data package converted to JSON format.

[0146] Step 3: Submit your entry

[0147] The terminal sends the packaged entry information to the server, and this communication uses the HTTP protocol.

[0148] Input: A data package in JSON format.

[0149] Output: The HTTP POST request sent to the server.

[0150] Step 4: Receiving and analyzing data

[0151] The server receives the HTTP request and parses the JSON data to extract the name, description, executable path, and developer information of the generated AI.

[0152] Input: HTTP POST request.

[0153] Output: Extracted entry information.

[0154] Step 5: Save your data

[0155] The server stores the extracted entry information in a database using SQL queries.

[0156] Input: The extracted entry information.

[0157] Output: Entry information stored in a database.

[0158] Step 6: Notification of completed entry

[0159] The server generates a response to notify the user that the entry has been successfully completed, and sends it to the user terminal.

[0160] Input: Notification that saving is complete.

[0161] Output: Entry complete message as an HTTP response.

[0162] Step 7: Displaying the entry completion message

[0163] The user terminal receives the response from the server and displays the message "Entry completed successfully" in the browser.

[0164] Input: The response from the server.

[0165] Output: Entry successful message displayed in browser.

[0166] Battle Features

[0167] Step 1: Check the battle start time

[0168] The server checks the pre-set battle start time and sees if that time has come, using a timer or scheduling function.

[0169] Input: Pre-set battle start time.

[0170] Output: Trigger to start the battle.

[0171] Step 2: Prepare for assignment delivery

[0172] The server generates and packages challenges to initiate battles between AIs, including problem definitions and input data.

[0173] Input: Trigger to start the battle.

[0174] Output: Packaged assignment content.

[0175] Step 3: Submit your assignment

[0176] The server sends the packaged tasks to each generating AI, and this communication may use HTTP or WebSocket.

[0177] Input: Packaged assignment content.

[0178] Output: The challenges sent to each generated AI.

[0179] Step 4: Generate solutions

[0180] Each generative AI generates a solution to a problem, which is achieved by the AI ​​model performing internal data calculations and analysis.

[0181] Input: Received assignment content.

[0182] Output: The generated solution.

[0183] Step 5: Submit your solution

[0184] Each generating AI returns a solution to the server, again using the HTTP protocol for this communication.

[0185] Input: The generated solution.

[0186] Output: The solution sent to the server.

[0187] Step 6: Receive and save the solution

[0188] The server temporarily stores the received solutions in a database using SQL queries.

[0189] Input: The solutions submitted by each generating AI.

[0190] Output: The solution stored in the database.

[0191] Step 7: Evaluate the solution

[0192] The server evaluates the saved solutions based on evaluation criteria and assigns a score to each generated AI, including solution accuracy, speed, and resource consumption.

[0193] Input: Solutions stored in the database.

[0194] Output: The evaluated score.

[0195] Step 8: Determine the winner

[0196] The server will determine the generator AI that performed best based on the evaluation results as the winner.

[0197] Input: The assessed score.

[0198] Output: Determining the winner and storing that information in a database.

[0199] Result announcement and evaluation function

[0200] Step 1: Counting scores

[0201] The server compiles the scores of each generated AI and creates a ranking, which is done by processing and calculating the score data.

[0202] Input: The score of each generated AI.

[0203] Output: Aggregated scores and rankings.

[0204] Step 2: Save ranking information

[0205] The server stores the rankings in a database.

[0206] Input: Aggregated scores and rankings.

[0207] Output: Ranking information stored in a database.

[0208] Step 3: Update your public page

[0209] The server updates the public page based on the ranking information stored in the database, using a script on the web server.

[0210] Input: Ranking information stored in the database.

[0211] Output: The updated public page.

[0212] Step 4: User confirmation of results

[0213] Users can access the public page using their devices to check the battle results and the ranking of the generated AI. This operation is performed using a web browser.

[0214] Input: Access to public pages.

[0215] Output: Battle results and ranking information displayed on the user's device.

[0216] The above is a detailed explanation of each processing step.

[0217] (Application example 1)

[0218] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0219] Current generative AI competition systems allow AI developers to enter their generative AIs, battle them, and evaluate and publish the results. However, they lack the functionality to allow general users to watch generative AI battles in real time or view past battle results, making it difficult to provide opportunities for users to deepen their interest and understanding. For this reason, there is a need for a system that allows users to intuitively understand and experience the technological evolution of generative AI.

[0220] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0221] In this invention, the server includes an entry means for AI developers to enter generated AIs, a battle means in which the entered generated AIs compete against each other in specific tasks, an evaluation and publication means for evaluating and publishing the battle results, a real-time distribution means for distributing the battle results in real time, an archive viewing means for selecting and viewing past battle results from an archive, and a user evaluation means for users to evaluate the battle results and enter comments. This enables general users to intuitively understand and experience the technological evolution of generated AIs by watching battles between generated AIs in real time and viewing past battle results from the archive.

[0222] "Entry means" refers to a device, method, or system that allows an AI developer to enter a generated AI.

[0223] "Battle means" refers to a device, method, or system that allows entered generated AIs to compete against each other in a specific challenge.

[0224] The "evaluation and disclosure means" refers to a device, method, or system for evaluating the battle results and disclosing the results to users.

[0225] "Real-time distribution means" refers to a device, method, or system for distributing the battle results of the generated AI to users in real time.

[0226] "Archive viewing means" refers to a device, method, or system that allows a user to select past generated AI battle results from a database and view them.

[0227] The "user evaluation means" refers to a device, method, or system that allows users to evaluate the battle results and input comments.

[0228] "Transmission means" refers to a device, method, or system for transmitting information about the generated AI from a user terminal to a server.

[0229] "Storage means" means a device, method, or system for receiving and storing transmitted information in a database.

[0230] "Collection means" refers to a device, method, or system for providing challenges to the entered generation AI and collecting the results of its solutions.

[0231] "Evaluation means" refers to a device, method, or system for analyzing the collected solution results of the generative AI and determining a winner based on evaluation criteria.

[0232] This invention is a system for competing in problem-solving ability using generative AI, and includes as its components an entry means, a battle means, an evaluation and publication means, a real-time distribution means, an archive viewing means, and a user evaluation means. How this system is specifically implemented is explained below.

[0233] System configuration

[0234] The main components are a server, user devices (smartphones, smart glasses, head-mounted displays (HMDs)), and a database. The server uses a cloud service (AWS) and the database is MySQL. React Native is used for the front end, and Node.js and Express.js are used for the back end.

[0235] Entry Method

[0236] Using a user terminal, the AI ​​developer enters the generated AI. At this time, a form is displayed in which the name, description, executable file, developer information, etc. of the generated AI can be entered, and the information is submitted by pressing the "entry" button. The submitted information is received by the server, packaged in JSON format, etc., and stored in a database.

[0237] Battle Method

[0238] The server starts a battle between the AI ​​generators at a pre-set date and time. The AI ​​generators are given specific challenges, and each AI generates a solution to the challenge. The solutions output by the AI ​​generators are sent to the server, where they are analyzed and evaluated based on the evaluation criteria.

[0239] Evaluation and Publication Methods

[0240] The server evaluates the solutions of each AI generator and creates a ranking based on the results, which is then updated on a public page where users can access the rankings.

[0241] Real-time delivery methods

[0242] The server uses WebSocket to deliver the battle results of the generated AI in real time. The user device is built with React Native and receives and displays the real-time data, allowing users to watch the progress of the generated AI battle in real time.

[0243] Archive viewing method

[0244] The results of past generated AI battles are stored in a database and can be viewed by users using the archive function. The data is retrieved using a REST API and displayed on the user's device.

[0245] User evaluation method

[0246] Users can rate and comment on battle results. The entered ratings and comments are sent from the front end to the server and stored in a database, allowing other users to view the feedback.

[0247] Specific examples

[0248] For example, the user can operate the system by entering the following prompts:

[0249] Example viewing prompt for a generated AI battle:

[0250] "Watch the next generative AI battle: Battle ID 12345. Challenge: Natural Language Processing."

[0251] Example of an archive search prompt:

[0252] "View past generative AI battle results: Check how generative AI performed on the challenge."

[0253] In this way, ordinary users will be able to intuitively understand and experience the technological evolution of generative AI.

[0254] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0255] Step 1:

[0256] The user inputs information into the terminal to enter the generated AI. The input information includes the name of the generated AI, a description, an executable file, developer information, etc. The terminal collects this information, packages it in JSON format, etc., and sends it to the server. The input of the data sent to the server is the generated AI information, and the output is the packaged entry information.

[0257] Step 2:

[0258] The server receives the entry information sent by the user. The received information is stored in a database. The database stores the name of the generated AI, a description, the path to the executable file, and the developer's contact information. The input is the entry information sent from the terminal, and the output is the entry information stored in the database.

[0259] Step 3:

[0260] The server will start a battle between the entered generated AIs at a pre-set date and time. Each generated AI will be provided with a specific task. The task details include input data and a problem definition, which are sent to each generated AI. The input is the generated AI's entry information and task definition, and the output is the task provided to the generated AI.

[0261] Step 4:

[0262] The generation AI generates a solution to the provided challenge. The generated solution is sent to the server. The input is the challenge provided by the server, and the output is the generation AI's solution.

[0263] Step 5:

[0264] The server receives the solution sent from the generation AI and temporarily stores it in a database. The input is the solution sent from the generation AI, and the output is the solution temporarily stored in the database.

[0265] Step 6:

[0266] The server analyzes the collected solutions of the generative AI and assigns a score based on evaluation criteria, such as solution accuracy, speed, and resource consumption. The input is the solution stored in the database, and the output is the evaluated score.

[0267] Step 7:

[0268] The server creates a ranking of the generative AI based on the evaluation results and stores it in a database. The input is the evaluated score, and the output is the ranking information.

[0269] Step 8:

[0270] The server distributes the battle results of the generated AI in real time. It uses WebSocket to send data sequentially to the user's device. The input is the evaluated score and the progress of the solution, and the output is the battle result data distributed in real time.

[0271] Step 9:

[0272] The user watches the progress of the generated AI battle in real time. The device receives data from the server and reflects it on the UI. The input is real-time data sent from the server, and the output is the progress of the battle displayed on the device.

[0273] Step 10:

[0274] Users can view past generated AI battle results using the archive function. The device sends a request for archive data to the server, which retrieves the relevant data from the database and sends it to the device. The input is the archive request from the user, and the output is the past battle results displayed on the device.

[0275] Step 11:

[0276] Users rate and comment on the battle results. The terminal sends the entered ratings and comments to the server, which stores them in a database. The input is the user's rating and comments, and the output is the rating and comments stored in the database.

[0277] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0278] This invention is a system that uses generative AI to compete in problem-solving ability, and combines it with an emotion engine that recognizes users' emotions. This system consists of the following elements: entry means, battle means, evaluation and publication means, and the emotion engine.

[0279] Entry function

[0280] Entry

[0281] Users use their device to enter their own generated AI. A form is displayed on the user device in which they can enter the name, description, executable file, developer information, etc. of the generated AI. Based on this information, they press the entry button to submit the information.

[0282] send

[0283] The user terminal sends the entry information entered by the user to the server, where it is packaged in a standard format such as JSON.

[0284] Receive and store

[0285] The server receives the entry information sent from the device and stores it in a database, including the name of the generated AI, a description, the path to the executable file, and the developer's contact information.

[0286] Completion notification

[0287] The server notifies the user terminal that the entry has been successfully completed, and an entry completion message is displayed on the user terminal.

[0288] Battle Features

[0289] Battle Begins

[0290] The server will start a battle between the registered generated AIs at a pre-set time, with specific challenges set for each generated AI.

[0291] Provide assignments

[0292] The server provides the generative AIs with the details of the challenge (e.g., input data and problem definition), and each generative AI generates a solution to this challenge.

[0293] Result collection

[0294] The generative AI outputs the results of the problem solving and returns them to the server, which temporarily stores these results in a database.

[0295] Results evaluation

[0296] The server evaluates and scores the results based on evaluation criteria (e.g., solution accuracy, speed, resource consumption).

[0297] Winner Decided

[0298] Based on the evaluation results, the server determines the best performing AI generator as the winner, and the winner's information is stored in the database.

[0299] Rating and publishing features

[0300] Evaluation and Scoring

[0301] The server evaluates the battle results and assigns a score to each generated AI based on the scoring criteria. A ranking is created based on the scoring results.

[0302] Save and publish results

[0303] The server saves the rankings in a database and updates a public page where users can view the latest battle results and the rankings of the generated AI.

[0304] Incorporating an emotion engine

[0305] emotion recognition

[0306] The user device is equipped with an emotion engine that analyzes the user's emotions from their facial expressions and voice. This emotion engine analyzes the user's emotions in real time as they view the battle results.

[0307] Sending emotional data

[0308] The emotion engine sends the analyzed emotion data to the server, which includes the user's emotional state (e.g., joy, surprise, frustration, etc.).

[0309] Collecting and storing emotional data

[0310] The server stores the transmitted emotion data in a database and uses it to improve the AI ​​generator and optimize battles. This emotion data can be used to provide feedback on the AI ​​generator's performance and user experience.

[0311] Feedback and Adjustments

[0312] The accumulated emotional data is analyzed and used to improve the generation AI, battle settings, and evaluation criteria, thereby improving the overall user experience of the system and the performance of the generation AI.

[0313] For example, if the emotion engine detects a smile when a user views the results of a battle between the generated AI, that data is sent to the server and stored in a database. This data is used in the feedback process of the generated AI and acts as part of its optimization. This allows the generated AI to compete more effectively in the next battle, taking into account the user's emotional reactions.

[0314] In this way, by incorporating an emotion engine in addition to entry means, battle means, evaluation and publication means, the present invention makes the generative AI competition and evaluation process more user-friendly and utilizes emotion data to improve and optimize the generative AI.

[0315] The processing flow will be explained below.

[0316] Entry function

[0317] Step 1: User enters AI

[0318] The user enters information about their generated AI (name, description, executable file, developer information, etc.) into the user terminal and presses the entry button.

[0319] Step 2: The device sends the entry information to the server

[0320] The terminal sends the entry information entered by the user to the server in a standard format such as JSON.

[0321] Step 3: The server receives the entry information

[0322] The server receives the entry information sent from the terminal.

[0323] Step 4: The server saves the entry information to a database

[0324] The server stores the received entry information in a database, including the name of the generated AI, its description, the path to the executable file, and developer information.

[0325] Step 5: The server notifies you that the entry is complete

[0326] The server notifies the user terminal that the entry has been successfully completed, and an entry completion message is displayed on the user terminal.

[0327] Battle Features

[0328] Step 1: The server starts the battle

[0329] The server will start a battle between the entered generated AIs at a pre-set time.

[0330] Step 2: The server provides each AI with a task

[0331] The server provides the submitted generated AI with details of the challenge (e.g., input data and problem definition).

[0332] Step 3: Generative AI solves the problem

[0333] The generative AI generates a solution to the problem and returns the result to the server.

[0334] Step 4: The server receives the results of each AI.

[0335] The server receives the problem-solving results returned by the generation AI and temporarily stores them in a database.

[0336] Step 5: The server evaluates the results

[0337] The server evaluates and scores the received results based on criteria (e.g., accuracy, speed, resource consumption, etc.).

[0338] Step 6: Server determines winner

[0339] Based on the evaluation results, the server determines the best performing AI generator as the winner, and the winner's information is stored in a database.

[0340] Rating and publishing features

[0341] Step 1: The server evaluates the battle results

[0342] The server tally the scores of each generated AI and create a ranking based on the evaluation criteria.

[0343] Step 2: The server creates the rankings

[0344] The server will create a ranking of the AI ​​generators based on the evaluation results. This ranking will be ranked based on the performance of the AI ​​generators.

[0345] Step 3: The server stores the results in a database

[0346] The server saves the ranking information it creates in a database, which also stores past battle results and evaluation history.

[0347] Step 4: Server updates public page

[0348] The server updates the public page with the latest information based on the ranking information stored in the database.

[0349] Step 5: User confirms the results

[0350] Users can access the public page using their devices to check the battle results and the ranking of the generated AI, allowing them to intuitively understand and experience the technological evolution of the generated AI.

[0351] Incorporating an emotion engine

[0352] Step 1: User visits public page

[0353] Users use their devices to access a public page that displays the battle results and rankings of the generated AI.

[0354] Step 2: The emotion engine analyzes the user's emotions

[0355] The emotion engine is built into the user's device and analyzes the user's facial expressions and voice in real time, for example, tracking the user's facial expressions through a camera and analyzing the user's tone of voice through a microphone.

[0356] Step 3: Send emotion data to the server

[0357] The emotion engine sends the analyzed emotion data to the server, which includes the user's emotional state (e.g., joy, surprise, dissatisfaction, etc.).

[0358] Step 4: The server receives the emotion data

[0359] The server receives the emotion data sent from the user terminal.

[0360] Step 5: The server stores the emotion data in a database

[0361] The server stores the received emotional data in a database, which includes the user's emotional response to each battle outcome.

[0362] Step 6: Analyze the sentiment data

[0363] The server analyzes the accumulated emotional data and uses it to improve the generation AI and optimize battles.

[0364] Step 7: Feedback and Adjustments

[0365] Feedback based on emotional data is provided to the developers of the AI ​​generator, allowing them to make adjustments to improve the AI's performance and user experience. For example, if a user is satisfied with the outcome of a particular battle, the AI ​​generator will use that strategy or approach in the next battle.

[0366] In this way, by incorporating an emotion engine in addition to entry means, battle means, evaluation and publication means, the present invention makes the generative AI competition and evaluation process more user-friendly and utilizes emotion data to improve and optimize the generative AI.

[0367] Example 2

[0368] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0369] Conventional generative algorithm competition systems did not take into account user emotional data when optimizing or evaluating generative algorithms, and therefore were unable to sufficiently improve the user experience. Furthermore, the evaluation criteria for generative algorithms were limited and lacked diversity. Furthermore, the publication and evaluation of battle results were not centrally managed, making it difficult for user feedback to be reflected in real time.

[0370] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a transmission means for transmitting information about the generation algorithm from the user terminal to the central processing unit, a storage means for receiving the transmitted information and storing it in a storage device, a provision means for providing challenges to the entered generation algorithms, a collection means for collecting the solution results of each generation algorithm, an evaluation means for analyzing the collected results and determining a winner based on evaluation criteria, an emotion engine means for recognizing the user's emotions, and a data collection and analysis means for collecting emotion data obtained by the emotion engine means and using it to optimize the system. This makes it possible to evaluate and optimize generation algorithms that reflect the user's emotion data, thereby improving the user experience and enabling diverse evaluation of generation algorithms.

[0371] A "generative algorithm" refers to a series of computational steps for generating new data or information from existing data.

[0372] "Entry Method" refers to the method or process by which a Generator Algorithm is registered with the System.

[0373] "Competition method" refers to the method or process by which the entered generative algorithms compete to provide solutions to a given challenge.

[0374] "Evaluation and disclosure measures" refers to the methods and processes for evaluating the results of the competition and disclosing those evaluation results to the public.

[0375] "Emotion engine means" refers to technology or tools for analyzing emotions from a user's facial expressions and voice.

[0376] "Data collection and analysis means" refers to the method or process of collecting data obtained by the emotion engine means and using it to optimize the system.

[0377] "Transmission means" refers to a method or process for transmitting information of the generation algorithm from the user terminal to the central processing unit.

[0378] "Storage" refers to the method or process for receiving transmitted information and storing it in a storage device.

[0379] "Provision means" refers to the method or process for providing a challenge to the entered generation algorithm.

[0380] "Collection means" refers to the method or process for collecting the solution results of each generation algorithm.

[0381] "Evaluation means" refers to a method or process for analyzing the collected solution results of the generative algorithms and determining a winner based on evaluation criteria.

[0382] "Central Processing Unit" refers to a central computer that performs multiple calculations.

[0383] "Storage" refers to a device or medium for storing information.

[0384] The present invention is a competition system using a generative AI model, which combines an entry means, a competition means, an evaluation and publication means, an emotion engine means, and a data collection and analysis means. In this system, the user, the terminal, and the server each play different roles. Specific embodiments of the system are described below.

[0385] Hardware and software used

[0386] This invention is composed of various hardware and software, including a user terminal, a server, a database, and an emotion engine. The user terminal is equipped with an emotion engine (e.g., facial recognition software, voice recognition software) for analyzing emotions from the user's facial expressions and voice. The server is a central processing unit that processes information sent from the user terminal and stores the information in a database (e.g., MySQL).

[0387] Entry function

[0388] Users use their device to enter their own generative AI model. A form is displayed on the user device to enter the name, description, executable file, developer information, etc. of the generative AI model. When the user enters the information and clicks the entry button, the entered information is converted to JSON format and sent to the server. The server receives the submitted information, verifies its validity, and stores it in a database. The information is encrypted before being stored and is kept securely.

[0389] Battle Features

[0390] At a pre-set time, the server will start a battle between the entered generative AI models. As the start time approaches, the server will prepare the execution environment for all relevant generative AI models. The server will provide each generative AI model with details of the challenge (e.g., input data and problem definition). Each generative AI model will generate a solution based on this challenge and return the results to the server in JSON format. The server will receive these results and temporarily store them in a database.

[0391] Rating and publishing features

[0392] The server applies an algorithm to score the collected results based on evaluation criteria (e.g., solution accuracy, speed, resource consumption). The resulting score is then assigned to the generative AI model. Furthermore, the server creates a ranking based on the scoring results and stores it in a database. It updates a public page with the latest ranking information, allowing the public to view the latest battle results through this public page.

[0393] Incorporating an emotion engine

[0394] The user device is equipped with an emotion engine that analyzes emotions from the user's facial expressions and voice. While the user is viewing the battle results, the emotion engine analyzes the user's emotional state (e.g., joy, surprise, dissatisfaction, etc.) in real time. The analyzed emotion data is sent to the server and stored in a database. This data is used to improve the generative AI model and optimize battles. By analyzing the accumulated emotion data and reflecting it in improving the generative AI model, battle settings, and adjusting evaluation criteria, the overall user experience of the system and the performance of the generative AI model are improved.

[0395] Specific examples

[0396] For example, when a user is viewing the results of a battle between generative AI models, the emotion engine will detect the user's smile. This data will be sent to the server and stored in the database. This data will then be analyzed and used in the generative AI model's feedback process. As a result, the next battle will take the user's emotional reactions into account, resulting in a more effective generative AI competition.

[0397] Prompt Sentence Examples

[0398] Below are some example prompts that can be used as input to a generative AI model:

[0399] Based on the reaction data when the user smiles, please suggest ways to optimize the performance of the generative AI.

[0400] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0401] Step 1:

[0402] The user uses a user terminal to enter their own generative AI model. A form is displayed on the user terminal in which the user can enter the name, description, executable file, developer information, etc. of the generative AI model. The user enters the required information in these fields and clicks the entry button. The input data is processed as metadata for the generative AI model and converted to JSON format. The output is the entry information (JSON format) that is sent to the server.

[0403] Step 2:

[0404] The terminal sends the entry information entered by the user to the server in JSON format. The transmitted data is encrypted and securely sent to the server. The input is the entry information entered by the user, and the output is the entry information securely sent to the server.

[0405] Step 3:

[0406] The server receives the entry information sent from the terminal. After receiving it, the server verifies the validity of the entry information, and if it is confirmed to be valid, it stores it in a database. The stored information includes the name of the generated AI model, a description, the path to the executable file, and the developer's contact information. The input is the received entry information, and the output is the entry information stored in the database.

[0407] Step 4:

[0408] The server notifies the user terminal that the entry has been completed. An entry completion message is displayed on the user terminal, allowing the user to confirm that the entry was successful. The input is the entry completion notification from the server, and the output is the entry completion message displayed on the user terminal.

[0409] Step 5:

[0410] At a pre-set time, the server will start a battle between the entered generative AI models. As the start time of the battle approaches, the server prepares the execution environments for all related generative AI models. The input is the battle start time, and the output is the execution environments for the prepared generative AI models.

[0411] Step 6:

[0412] The server provides each generative AI model with details of the challenge (e.g., input data and problem definition). Each generative AI model generates a solution based on this challenge. The input is the problem description from the server, and the output is the solution generated by the generative AI model.

[0413] Step 7:

[0414] The generative AI model outputs the results of problem solving and returns them to the server in JSON format. The server receives the returned results and temporarily stores them in a database. The input is the solution result of the generative AI model, and the output is the result stored in the database.

[0415] Step 8:

[0416] The server applies an algorithm to score the collected results based on evaluation criteria (e.g., solution accuracy, speed, resource consumption). The input is the collected results, and the output is a score assigned to each generative AI model.

[0417] Step 9:

[0418] The server determines the winner and stores the winner's information in a database. The input is the scored result, and the output is the information of the generative AI model that was determined as the winner.

[0419] Step 10:

[0420] The server creates a ranking based on the scoring results and saves it in a database. It then updates the public page with the latest ranking information. The input is the scoring results, and the output is the updated public page.

[0421] Step 11:

[0422] The user device is equipped with an emotion engine that analyzes emotions from the user's facial expressions and voice. While the user is viewing the battle results, the emotion engine analyzes the user's emotional state in real time. The input is the user's facial expressions and voice data, and the output is the analyzed emotional data.

[0423] Step 12:

[0424] The emotion engine sends the analyzed emotion data to the server. The input is the analyzed emotion data, and the output is the emotion data sent to the server.

[0425] Step 13:

[0426] The server receives the emotion data and stores it in a database. This data is used to improve the generative AI model and optimize battles. The input is emotion data, and the output is the emotion data stored in the database.

[0427] Step 14:

[0428] The server analyzes the accumulated emotional data and reflects it in improving the generative AI model, battle settings, and adjusting evaluation criteria. This improves the user experience of the entire system and the performance of the generative AI model. The input is the accumulated emotional data, and the output is the overall system performance improved through improvements and adjustments.

[0429] (Application example 2)

[0430] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0431] Conventional battle systems using generative models lacked evaluation and optimization that took user emotions into account, resulting in a limited user experience. Furthermore, because the performance of generative models was evaluated solely based on technical criteria, users were not provided with sufficient satisfaction. Furthermore, there was a lack of a mechanism for analyzing user emotion data in real time and using it as feedback.

[0432] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0433] In this invention, the server includes an entry means for AI developers to enter generative models, a battle means in which the entered generative models compete against each other in specific tasks, an evaluation and publication means for evaluating and publishing the battle results, an emotion engine that analyzes emotions from the user's facial expressions and voice, and an adjustment means for optimizing the performance of the generative model based on the user's emotion data. This enables evaluation and optimization that takes user emotions into consideration, providing a more satisfying user experience.

[0434] "Entry means" refers to the means by which an AI developer registers a generative model in the system, and includes a transmission means for inputting information about the generative model and sending it to a server, and a storage means for receiving the sent information and storing it in a database.

[0435] The "battle means" refers to a means by which entered generative models compete against each other over specific pre-set challenges, and includes a collection means for providing the challenges and collecting the solution results of each generative model, and an evaluation means for analyzing the collected results and determining a winner based on evaluation criteria.

[0436] The "evaluation and publication means" is a means for evaluating the battle results, scoring the generative models based on the scoring criteria, creating rankings, and storing and publishing them in a database.

[0437] The "emotion engine" is part of a system that analyzes emotions from the user's facial expressions and voice in real time and sends the analysis data to a server.

[0438] The "adjustment method" is a method for collecting and analyzing user emotional data, optimizing the performance of the generative model based on that data, and improving the user experience.

[0439] As a form for implementing this invention, a system using a smartphone application is constructed.

[0440] System Configuration

[0441] 1. Hardware and Software Configuration

[0442] User device: Smartphone

[0443] Emotion Engine: EmotionRecognizer

[0444] AI Battle System: AIBattle

[0445] Content Management System: ContentManager

[0446] Entry Method

[0447] Using a user device, an AI developer registers a generative model in the system. An entry form is provided, and the developer enters the name of the generative model, a description, an executable file, developer information, etc. The input information is packaged in JSON format or similar and sent to the server. The server then stores the submitted information in a database.

[0448] Battle Method

[0449] The server sets up battles in which the submitted generative models compete against each other for specific tasks. Each task is provided with detailed information, including user emotional data. The generative models generate solutions to the tasks and return the results to the server. The server analyzes the collected results and determines the winner based on evaluation criteria.

[0450] Evaluation and Publication Methods

[0451] The battle results are evaluated and the generative models are scored based on the scoring criteria. The rankings are created and saved in a database, and are made available to the general public by updating a public page.

[0452] Incorporating an emotion engine

[0453] The system incorporates an emotion engine that uses the smartphone's camera and microphone to analyze the user's facial expressions and voice. The analyzed emotion data is sent to a server in real time. The server accumulates the emotion data and makes adjustments to optimize the performance of the generative model based on the user's emotional responses.

[0454] Specific examples

[0455] If a user smiles while watching a drama, their emotional data is analyzed in real time and stored in a database. Based on this emotional data, the next time content is delivered, optimized content is provided, taking into account the user's emotional response.

[0456] Prompt Sentence Examples

[0457] For example, you might input the following prompt into a generative AI model:

[0458] "Collect emotional data when a user smiles while watching a drama, and generate a list of content to display next based on that."

[0459] In this way, the invention incorporates an emotion engine in addition to entry means, battle means, evaluation and publication means, making the generative model competition and evaluation process more user-friendly and utilizing emotion data to improve and optimize generative models.

[0460] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0461] Step 1:

[0462] Execution of entry method

[0463] The user enters information about the generative model into an input form on the device. Specifically, the user enters the name of the generative model, a description, an executable file, developer information, etc. The entered information is packaged in JSON format and sent to the server.

[0464] Input: Name of the generated model, description, executable file, developer information

[0465] Output: Generated model information packaged in JSON format

[0466] Specific operation: The user enters information into the form and presses the submit button to send the data to the server.

[0467] Step 2:

[0468] Saving entry information

[0469] The server receives the generated model information in JSON format sent from the user device and stores it in a database, including the name of the generated model, its description, the path to the executable file, and the developer's contact information.

[0470] Input: Generated model information in JSON format

[0471] Output: Generative model information stored in a database

[0472] Specific operation: The server receives the information and stores it in a database.

[0473] Step 3:

[0474] The start of the battle

[0475] The server will start a battle between the entered generative models at a pre-set time, providing specific challenges to the generative models.

[0476] Input: Generative model information stored in the database, task details

[0477] Output: A generative model that receives the task.

[0478] Specific operation: The server triggers the start of the battle and sends the details of the challenge to the generative model.

[0479] Step 4:

[0480] Collection of problem-solving results

[0481] The generative model generates a solution to the problem and returns the result to the server, which receives the result and temporarily stores it in a database.

[0482] Input: Results of solving the problem using the generative model

[0483] Output: Solution results stored in the database

[0484] Specific operation: The generative model executes the problem-solving process and returns the results to the server.

[0485] Step 5:

[0486] Evaluating the results

[0487] The server scores the collected solutions based on evaluation criteria, including solution accuracy, speed, and resource consumption.

[0488] Input: Solution results stored in the database

[0489] Output: Scoring results

[0490] Specific Operation: The server runs the evaluation algorithm and generates a score.

[0491] Step 6:

[0492] Determining the Winner

[0493] The server determines the best performing generative model as the winner based on the scoring results, and the winner's information is stored in a database.

[0494] Input: Scoring results

[0495] Output: Winner information

[0496] What happens: The server compares the scores and marks the generative model with the highest score as the winner.

[0497] Step 7:

[0498] Publication of ratings and rankings

[0499] The server creates a ranking based on the scoring results and updates the public page, allowing general users to view the latest battle results and the rankings of generative models.

[0500] Input: Scoring results

[0501] Output: Published rankings

[0502] Specific operation: The server updates the ranking page and publishes the latest information.

[0503] Step 8:

[0504] User sentiment analysis

[0505] When a user browses a public page, the smartphone's camera and microphone are used to analyze the user's facial expressions and voice. Emotional data is analyzed in real time and sent to the server.

[0506] Input: User's facial expression data, voice data

[0507] Output: Parsed emotion data

[0508] Specific operation: The smartphone runs the emotion analysis engine and sends the analysis data to the server.

[0509] Step 9:

[0510] Emotional Data Feedback

[0511] The server stores the received emotion data in a database and uses it to improve the generative model and optimize battles.

[0512] Input: Parsed emotion data

[0513] Output: Improved generative model performance

[0514] Specific operation: The server analyzes the emotion data and uses it as feedback to optimize the performance of the generative model.

[0515] In this way, through a series of processing steps, evaluation and optimization of the generative model taking into account the user's emotions is achieved.

[0516] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0517] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0518] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0519] [Second embodiment]

[0520] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0521] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0522] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0523] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0524] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0525] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0526] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0527] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0528] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0529] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0530] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0531] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0532] The present invention is a system for competing in problem-solving ability using generative AI, and is implemented in the following form.

[0533] Entry function

[0534] 1. User enters AI

[0535] Users use their device to enter their own generated AI. A form is displayed on the user device in which they can enter the name, description, executable file, developer information, etc. of the generated AI. Based on this information, they press the entry button to submit the information.

[0536] 2. The device sends the entry information to the server

[0537] The user terminal sends the entry information entered by the user to the server, where it is packaged in a standard format such as JSON.

[0538] 3. The server receives the entry information

[0539] The server receives the entry information sent from the user's device and stores it in a database, including the name of the generated AI, a description, the path to the executable file, and the developer's contact information.

[0540] 4. Notification of entry completion

[0541] The server notifies the user terminal that the entry has been completed successfully. An entry completion message is displayed on the user terminal, allowing the user to confirm that the entry has been completed successfully.

[0542] Battle Features

[0543] 1. The server starts the battle

[0544] The server will start a battle between the registered generated AIs at a pre-set time. At the start of the battle, specific tasks will be set for each generated AI.

[0545] 2. The server provides each AI with a task

[0546] The server provides the details of the challenge (e.g., input data and problem definition) to the submitted AI generators, who then generate a solution to the challenge.

[0547] 3. The server receives the results of each AI

[0548] Each generation AI outputs the solution to the problem and returns it to the server, which receives these results and temporarily stores them in a database.

[0549] 4. The server evaluates the results

[0550] The server evaluates the received results based on evaluation criteria (e.g., solution accuracy, speed, resource consumption, etc.) and assigns a score to each generated AI.

[0551] 5. The server decides the winner

[0552] Based on the evaluation results, the server determines the best performing AI generator as the winner, and the winner's information is stored in the database.

[0553] Result announcement and evaluation function

[0554] 1. The server evaluates the battle results

[0555] After the battle, the server will tally the scores of each AI generator and create a ranking, which is a fair evaluation of the AI ​​generator's ability to solve the problem.

[0556] 2. The server creates the rankings

[0557] The server creates a ranking of the AI ​​generators based on the collected evaluation results. This ranking is based on the performance of the AI ​​generators.

[0558] 3. The server stores the results in a database

[0559] The server saves the rankings it creates in a database, which also stores past battle results and the evaluation history of each generated AI.

[0560] 4. The server updates the public page

[0561] The server updates the public page based on the ranking information stored in the database, allowing general users to view the latest battle results and the rankings of the generated AI.

[0562] 5. User checks the results

[0563] Users can access the public page using their devices to check the battle results and the rankings of the generated AI, allowing general users to intuitively understand and experience the technological evolution of generated AI.

[0564] The above system provides AI developers with a forum where they can fairly compete and receive evaluations on the performance of their generative AI, and it also provides general consumers with a forum where they can intuitively understand and experience the technological evolution of generative AI.

[0565] The processing flow will be explained below.

[0566] Entry function

[0567] Step 1: User enters AI

[0568] The user enters information about their generated AI (name, description, executable file, developer information, etc.) into the user terminal and presses the entry button.

[0569] Step 2: The device sends the entry information to the server

[0570] The terminal sends the entry information entered by the user to the server in a standard format such as JSON.

[0571] Step 3: Server receives entry information

[0572] The server receives the entry information sent from the terminal.

[0573] Step 4: The server saves the entry information to a database

[0574] The server stores the received entry information in a database, including the name of the generated AI, its description, the path to the executable file, and developer information.

[0575] Step 5: The server notifies you that the entry is complete

[0576] The server notifies the user terminal that the entry has been successfully completed, and an entry completion message is displayed on the user terminal.

[0577] Battle Features

[0578] Step 1: The server starts the battle

[0579] The server will start a battle between the entered generated AIs at a pre-set time.

[0580] Step 2: The server provides each AI with a task

[0581] The server provides the details of the challenge (e.g., input data and problem definition) to the entered generated AI.

[0582] Step 3: The server receives the results of each AI.

[0583] Each generation AI generates a problem-solving result and returns it to the server, which receives the result and temporarily stores it in a database.

[0584] Step 4: The server evaluates the results

[0585] The server evaluates the received results based on criteria (e.g., accuracy, speed, resource consumption), and the evaluation results are scored.

[0586] Step 5: Server determines winner

[0587] Based on the evaluation results, the server determines the AI ​​generator with the best performance as the winner, and the winner's information is recorded in the database.

[0588] Result announcement and evaluation function

[0589] Step 1: The server evaluates the battle results

[0590] The server tally the scores of each generated AI and create a ranking based on the evaluation criteria.

[0591] Step 2: The server creates the rankings

[0592] The server uses the evaluation results to create a ranking of the generated AIs, which is ranked based on the performance of the generated AIs.

[0593] Step 3: The server stores the results in a database

[0594] The server saves the ranking information it creates in a database, which also stores past battle results and evaluation history.

[0595] Step 4: The server updates the public page

[0596] The server updates the public page with the latest information based on the rankings stored in the database.

[0597] Step 5: User confirms the results

[0598] Users can access the public page using their devices to check the battle results and the ranking of the generated AI, allowing them to intuitively understand and experience the technological evolution of the generated AI.

[0599] Example 1

[0600] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0601] Currently, many AI developers are developing generative AI, but there is a lack of venues for objectively evaluating the performance of these generative AIs and for them to compete against each other. This makes it difficult for generative AI developers to compare the capabilities of their own AI with those of other AIs and identify areas for improvement. It is also difficult for ordinary users to intuitively understand and experience the evolution and characteristics of generative AI. A new system is needed to solve this problem and promote technological advances in generative AI.

[0602] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0603] In this invention, the server includes a storage means for receiving entry information and storing it in a database, a management means for checking the start time of the battle and providing challenges to the AI ​​generators, and an evaluation means for evaluating the results of the challenges and determining the best AI generators. This allows the performance of AI generators to be fairly evaluated and allows developers to compete against each other.

[0604] "Entry means" refers to the means by which an AI developer enters a generated AI into the system, and includes information sent from the user terminal to the server.

[0605] "Battle means" refers to the means by which entered generated AIs compete against each other on specific challenges, and includes the function of providing challenges and collecting the results of their solutions.

[0606] "Means for evaluation and publication" refers to means for evaluating the results of battles and publishing the results, and includes functions for evaluating the generating AI based on evaluation criteria, storing the results in a database, and publishing them to the general public.

[0607] "Input means" refers to the means by which a user inputs and submits entry information for the generated AI, and refers to the input form provided on the user's terminal.

[0608] "Storage means" refers to the means by which the server stores the received entry information in a database, and includes temporary and permanent storage of data.

[0609] The "management means" is a means by which the server checks the start time of the battle and starts the battle at the set time, and includes a function for managing the progress of the battle.

[0610] "Task providing means" refers to the means by which the server provides specific tasks to the entered generation AI, and includes the generation and distribution of tasks.

[0611] "Evaluation means" refers to the means by which the server evaluates and scores the solution results of each generated AI based on the evaluation criteria, and includes analyzing the results and applying the evaluation criteria.

[0612] "Determination means" refers to the means by which the server determines the best generating AI and stores that information in a database, including determining the winner and storing the results.

[0613] "Server communication means" refers to the means by which the server provides challenges to each generation AI and each generation AI transmits a solution to the server, and includes establishing communication and sending and receiving data.

[0614] "Evaluation criteria setting means" means a means for the evaluation means to set criteria for evaluating each generative AI based on the accuracy, speed, and resource consumption of the solution, and includes the definition and application of the evaluation criteria.

[0615] This invention is a system for competing in problem-solving ability using generative AI, and it operates in cooperation with users, terminals, and a server. The system has the following functions.

[0616] Entry function

[0617] To enter a generated AI, a user first uses their device to enter the name, description, executable file path, and developer information into an entry form. Specific hardware used for this is a personal computer or smartphone, and the user accesses the form via a browser. A web browser (e.g., Google Chrome or Safari) is used as the software.

[0618] When the user presses the entry button, the terminal converts the entered information into JSON format and sends it to the server as an HTTP POST request. The server analyzes the received data and saves it in a database. This saving process uses the server's internal database system (e.g., MySQL, PostgreSQL). Once the entry is complete, the server sends a completion notification to the user's terminal. The user's terminal displays the message "Entry completed successfully."

[0619] As a specific example, a user enters a generation AI called "AI_Champion" and enters the description, "This is an AI that analyzes the emotions in English sentences."

[0620] Battle Features

[0621] The server starts a battle between the entered generated AIs at a pre-set time. It checks the start time of the battle and provides specific tasks to the entered generated AIs. For example, a task might be to "analyze emotions from English sentences and assign a score." The server packages and sends the task details to each generated AI.

[0622] Each generation AI generates a solution to the provided problem and sends the solution results to the server. The server temporarily stores the received solution results in a database. The server then evaluates and scores the results of each generation AI based on evaluation criteria (e.g., solution accuracy, speed, resource consumption, etc.). Various evaluation algorithms and evaluation criterion setting methods are used for the evaluation process.

[0623] As an example of specific operation, the server receives an emotion score of 0.95 assigned by "AI_Champion" and evaluates it as "AI_Champion's score is 95 points."

[0624] Result announcement and evaluation function

[0625] After the battle is over, the server tally the scores of each generated AI and create a ranking. This ranking is based on the performance of the generated AI and is stored in a database. The server updates a public page based on the ranking information, allowing general users to view the latest battle results and the rankings of the generated AI. Users can access the public page using their devices to check the results.

[0626] As a specific example, a user accesses a public page and sees the information "AI_Champion is in first place."

[0627] Prompt Sentence Examples

[0628] "Create a generative AI that takes an English sentence as input, analyzes its sentiment, and gives it a score."

[0629] This invention provides AI developers with a forum where they can fairly compete and have the performance of their generative AI evaluated, and it also provides general users with a forum where they can intuitively understand and experience the technological evolution of generative AI.

[0630] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0631] Entry function

[0632] Step 1: Display the entry form

[0633] The user opens the entry page in a browser and enters the name, description, executable file path, and developer information of the generated AI. HTML and JavaScript are used to display the required fields in this form.

[0634] Input: User-entered name, description, executable path, and developer information for the generated AI.

[0635] Output: The input form reflects the user's input.

[0636] Step 2: Prepare input information for submission

[0637] When the user clicks the entry button, the terminal converts the input information into JSON format and packages it as an HTTP POST request.

[0638] Enter: When the user clicks the enter button.

[0639] Output: Data package converted to JSON format.

[0640] Step 3: Submit your entry

[0641] The terminal sends the packaged entry information to the server, and this communication uses the HTTP protocol.

[0642] Input: A data package in JSON format.

[0643] Output: The HTTP POST request sent to the server.

[0644] Step 4: Receiving and analyzing data

[0645] The server receives the HTTP request and parses the JSON data to extract the name, description, executable path, and developer information of the generated AI.

[0646] Input: HTTP POST request.

[0647] Output: Extracted entry information.

[0648] Step 5: Save your data

[0649] The server stores the extracted entry information in a database using SQL queries.

[0650] Input: The extracted entry information.

[0651] Output: Entry information stored in a database.

[0652] Step 6: Notification of completed entry

[0653] The server generates a response to notify the user that the entry has been successfully completed, and sends it to the user terminal.

[0654] Input: Notification that saving is complete.

[0655] Output: Entry complete message as an HTTP response.

[0656] Step 7: Displaying the entry completion message

[0657] The user terminal receives the response from the server and displays the message "Entry completed successfully" in the browser.

[0658] Input: The response from the server.

[0659] Output: Entry successful message displayed in browser.

[0660] Battle Features

[0661] Step 1: Check the battle start time

[0662] The server checks the pre-set battle start time and sees if that time has come, using a timer or scheduling function.

[0663] Input: Pre-set battle start time.

[0664] Output: Trigger to start the battle.

[0665] Step 2: Prepare for assignment delivery

[0666] The server generates and packages challenges to initiate battles between AIs, including problem definitions and input data.

[0667] Input: Trigger to start the battle.

[0668] Output: Packaged assignment content.

[0669] Step 3: Submit your assignment

[0670] The server sends the packaged tasks to each generating AI, and this communication may use HTTP or WebSocket.

[0671] Input: Packaged assignment content.

[0672] Output: The challenges sent to each generated AI.

[0673] Step 4: Generate solutions

[0674] Each generative AI generates a solution to a problem, which is achieved by the AI ​​model performing internal data calculations and analysis.

[0675] Input: Received assignment content.

[0676] Output: The generated solution.

[0677] Step 5: Submit your solution

[0678] Each generating AI returns a solution to the server, again using the HTTP protocol for this communication.

[0679] Input: The generated solution.

[0680] Output: The solution sent to the server.

[0681] Step 6: Receive and save the solution

[0682] The server temporarily stores the received solutions in a database using SQL queries.

[0683] Input: The solutions submitted by each generating AI.

[0684] Output: The solution stored in the database.

[0685] Step 7: Evaluate the solution

[0686] The server evaluates the saved solutions based on evaluation criteria and assigns a score to each generated AI, including solution accuracy, speed, and resource consumption.

[0687] Input: Solutions stored in the database.

[0688] Output: The evaluated score.

[0689] Step 8: Determine the winner

[0690] The server will determine the generator AI that performed best based on the evaluation results as the winner.

[0691] Input: The assessed score.

[0692] Output: Determining the winner and storing that information in a database.

[0693] Result announcement and evaluation function

[0694] Step 1: Counting scores

[0695] The server compiles the scores of each generated AI and creates a ranking, which is done by processing and calculating the score data.

[0696] Input: The score of each generated AI.

[0697] Output: Aggregated scores and rankings.

[0698] Step 2: Save ranking information

[0699] The server stores the rankings in a database.

[0700] Input: Aggregated scores and rankings.

[0701] Output: Ranking information stored in a database.

[0702] Step 3: Update your public page

[0703] The server updates the public page based on the ranking information stored in the database, using a script on the web server.

[0704] Input: Ranking information stored in the database.

[0705] Output: The updated public page.

[0706] Step 4: User confirmation of results

[0707] Users can access the public page using their devices to check the battle results and the ranking of the generated AI. This operation is performed using a web browser.

[0708] Input: Access to public pages.

[0709] Output: Battle results and ranking information displayed on the user's device.

[0710] The above is a detailed explanation of each processing step.

[0711] (Application example 1)

[0712] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0713] Current generative AI competition systems allow AI developers to enter their generative AIs, battle them, and evaluate and publish the results. However, they lack the functionality to allow general users to watch generative AI battles in real time or view past battle results, making it difficult to provide opportunities for users to deepen their interest and understanding. For this reason, there is a need for a system that allows users to intuitively understand and experience the technological evolution of generative AI.

[0714] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0715] In this invention, the server includes an entry means for AI developers to enter generated AIs, a battle means in which the entered generated AIs compete against each other in specific tasks, an evaluation and publication means for evaluating and publishing the battle results, a real-time distribution means for distributing the battle results in real time, an archive viewing means for selecting and viewing past battle results from an archive, and a user evaluation means for users to evaluate the battle results and enter comments. This enables general users to intuitively understand and experience the technological evolution of generated AIs by watching battles between generated AIs in real time and viewing past battle results from the archive.

[0716] "Entry means" refers to a device, method, or system that allows an AI developer to enter a generated AI.

[0717] "Battle means" refers to a device, method, or system that allows entered generated AIs to compete against each other in a specific challenge.

[0718] The "evaluation and disclosure means" refers to a device, method, or system for evaluating the battle results and disclosing the results to users.

[0719] "Real-time distribution means" refers to a device, method, or system for distributing the battle results of the generated AI to users in real time.

[0720] "Archive viewing means" refers to a device, method, or system that allows a user to select past generated AI battle results from a database and view them.

[0721] The "user evaluation means" refers to a device, method, or system that allows users to evaluate the battle results and input comments.

[0722] "Transmission means" refers to a device, method, or system for transmitting information about the generated AI from a user terminal to a server.

[0723] "Storage means" means a device, method, or system for receiving and storing transmitted information in a database.

[0724] "Collection means" refers to a device, method, or system for providing challenges to the entered generation AI and collecting the results of its solutions.

[0725] "Evaluation means" refers to a device, method, or system for analyzing the collected solution results of the generative AI and determining a winner based on evaluation criteria.

[0726] This invention is a system for competing in problem-solving ability using generative AI, and includes as its components an entry means, a battle means, an evaluation and publication means, a real-time distribution means, an archive viewing means, and a user evaluation means. How this system is specifically implemented is explained below.

[0727] System configuration

[0728] The main components are a server, user devices (smartphones, smart glasses, head-mounted displays (HMDs)), and a database. The server uses a cloud service (AWS) and the database is MySQL. React Native is used for the front end, and Node.js and Express.js are used for the back end.

[0729] Entry Method

[0730] Using a user terminal, the AI ​​developer enters the generated AI. At this time, a form is displayed in which the name, description, executable file, developer information, etc. of the generated AI can be entered, and the information is submitted by pressing the "entry" button. The submitted information is received by the server, packaged in JSON format, etc., and stored in a database.

[0731] Battle Method

[0732] The server starts a battle between the AI ​​generators at a pre-set date and time. The AI ​​generators are given specific challenges, and each AI generates a solution to the challenge. The solutions output by the AI ​​generators are sent to the server, where they are analyzed and evaluated based on the evaluation criteria.

[0733] Evaluation and Publication Methods

[0734] The server evaluates the solutions of each AI generator and creates a ranking based on the results, which is then updated on a public page where users can access the rankings.

[0735] Real-time delivery methods

[0736] The server uses WebSocket to deliver the battle results of the generated AI in real time. The user device is built with React Native and receives and displays the real-time data, allowing users to watch the progress of the generated AI battle in real time.

[0737] Archive viewing method

[0738] The results of past generated AI battles are stored in a database and can be viewed by users using the archive function. The data is retrieved using a REST API and displayed on the user's device.

[0739] User evaluation method

[0740] Users can rate and comment on battle results. The entered ratings and comments are sent from the front end to the server and stored in a database, allowing other users to view the feedback.

[0741] Specific examples

[0742] For example, the user can operate the system by entering the following prompts:

[0743] Example viewing prompt for a generated AI battle:

[0744] "Watch the next generative AI battle: Battle ID 12345. Challenge: Natural Language Processing."

[0745] Example of an archive search prompt:

[0746] "View past generative AI battle results: Check how generative AI performed on the challenge."

[0747] In this way, ordinary users will be able to intuitively understand and experience the technological evolution of generative AI.

[0748] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0749] Step 1:

[0750] The user inputs information into the terminal to enter the generated AI. The input information includes the name of the generated AI, a description, an executable file, developer information, etc. The terminal collects this information, packages it in JSON format, etc., and sends it to the server. The input of the data sent to the server is the generated AI information, and the output is the packaged entry information.

[0751] Step 2:

[0752] The server receives the entry information sent by the user. The received information is stored in a database. The database stores the name of the generated AI, a description, the path to the executable file, and the developer's contact information. The input is the entry information sent from the terminal, and the output is the entry information stored in the database.

[0753] Step 3:

[0754] The server will start a battle between the entered generated AIs at a pre-set date and time. Each generated AI will be provided with a specific task. The task details include input data and a problem definition, which are sent to each generated AI. The input is the generated AI's entry information and task definition, and the output is the task provided to the generated AI.

[0755] Step 4:

[0756] The generation AI generates a solution to the provided challenge. The generated solution is sent to the server. The input is the challenge provided by the server, and the output is the generation AI's solution.

[0757] Step 5:

[0758] The server receives the solution sent from the generation AI and temporarily stores it in a database. The input is the solution sent from the generation AI, and the output is the solution temporarily stored in the database.

[0759] Step 6:

[0760] The server analyzes the collected solutions of the generative AI and assigns a score based on evaluation criteria, such as solution accuracy, speed, and resource consumption. The input is the solution stored in the database, and the output is the evaluated score.

[0761] Step 7:

[0762] The server creates a ranking of the generative AI based on the evaluation results and stores it in a database. The input is the evaluated score, and the output is the ranking information.

[0763] Step 8:

[0764] The server distributes the battle results of the generated AI in real time. It uses WebSocket to send data sequentially to the user's device. The input is the evaluated score and the progress of the solution, and the output is the battle result data distributed in real time.

[0765] Step 9:

[0766] The user watches the progress of the generated AI battle in real time. The device receives data from the server and reflects it on the UI. The input is real-time data sent from the server, and the output is the progress of the battle displayed on the device.

[0767] Step 10:

[0768] Users can view past generated AI battle results using the archive function. The device sends a request for archive data to the server, which retrieves the relevant data from the database and sends it to the device. The input is the archive request from the user, and the output is the past battle results displayed on the device.

[0769] Step 11:

[0770] Users rate and comment on the battle results. The terminal sends the entered ratings and comments to the server, which stores them in a database. The input is the user's rating and comments, and the output is the rating and comments stored in the database.

[0771] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0772] This invention is a system that uses generative AI to compete in problem-solving ability, and combines it with an emotion engine that recognizes users' emotions. This system consists of the following elements: entry means, battle means, evaluation and publication means, and the emotion engine.

[0773] Entry function

[0774] Entry

[0775] Users use their device to enter their own generated AI. A form is displayed on the user device in which they can enter the name, description, executable file, developer information, etc. of the generated AI. Based on this information, they press the entry button to submit the information.

[0776] send

[0777] The user terminal sends the entry information entered by the user to the server, where it is packaged in a standard format such as JSON.

[0778] Receive and store

[0779] The server receives the entry information sent from the device and stores it in a database, including the name of the generated AI, a description, the path to the executable file, and the developer's contact information.

[0780] Completion notification

[0781] The server notifies the user terminal that the entry has been successfully completed, and an entry completion message is displayed on the user terminal.

[0782] Battle Features

[0783] Battle Begins

[0784] The server will start a battle between the registered generated AIs at a pre-set time, with specific challenges set for each generated AI.

[0785] Provide assignments

[0786] The server provides the generative AIs with the details of the challenge (e.g., input data and problem definition), and each generative AI generates a solution to this challenge.

[0787] Result collection

[0788] The generative AI outputs the results of the problem solving and returns them to the server, which temporarily stores these results in a database.

[0789] Results evaluation

[0790] The server evaluates and scores the results based on evaluation criteria (e.g., solution accuracy, speed, resource consumption).

[0791] Winner Decided

[0792] Based on the evaluation results, the server determines the best performing AI generator as the winner, and the winner's information is stored in the database.

[0793] Rating and publishing features

[0794] Evaluation and Scoring

[0795] The server evaluates the battle results and assigns a score to each generated AI based on the scoring criteria. A ranking is created based on the scoring results.

[0796] Save and publish results

[0797] The server saves the rankings in a database and updates a public page where users can view the latest battle results and the rankings of the generated AI.

[0798] Incorporating an emotion engine

[0799] emotion recognition

[0800] The user device is equipped with an emotion engine that analyzes the user's emotions from their facial expressions and voice. This emotion engine analyzes the user's emotions in real time as they view the battle results.

[0801] Sending emotional data

[0802] The emotion engine sends the analyzed emotion data to the server, which includes the user's emotional state (e.g., joy, surprise, frustration, etc.).

[0803] Collecting and storing emotional data

[0804] The server stores the transmitted emotion data in a database and uses it to improve the AI ​​generator and optimize battles. This emotion data can be used to provide feedback on the AI ​​generator's performance and user experience.

[0805] Feedback and Adjustments

[0806] The accumulated emotional data is analyzed and used to improve the generation AI, battle settings, and evaluation criteria, thereby improving the overall user experience of the system and the performance of the generation AI.

[0807] For example, if the emotion engine detects a smile when a user views the results of a battle between the generated AI, that data is sent to the server and stored in a database. This data is used in the feedback process of the generated AI and acts as part of its optimization. This allows the generated AI to compete more effectively in the next battle, taking into account the user's emotional reactions.

[0808] In this way, by incorporating an emotion engine in addition to entry means, battle means, evaluation and publication means, the present invention makes the generative AI competition and evaluation process more user-friendly and utilizes emotion data to improve and optimize the generative AI.

[0809] The processing flow will be explained below.

[0810] Entry function

[0811] Step 1: User enters AI

[0812] The user enters information about their generated AI (name, description, executable file, developer information, etc.) into the user terminal and presses the entry button.

[0813] Step 2: The device sends the entry information to the server

[0814] The terminal sends the entry information entered by the user to the server in a standard format such as JSON.

[0815] Step 3: The server receives the entry information

[0816] The server receives the entry information sent from the terminal.

[0817] Step 4: The server saves the entry information to a database

[0818] The server stores the received entry information in a database, including the name of the generated AI, its description, the path to the executable file, and developer information.

[0819] Step 5: The server notifies you that the entry is complete

[0820] The server notifies the user terminal that the entry has been successfully completed, and an entry completion message is displayed on the user terminal.

[0821] Battle Features

[0822] Step 1: The server starts the battle

[0823] The server will start a battle between the entered generated AIs at a pre-set time.

[0824] Step 2: The server provides each AI with a task

[0825] The server provides the details of the challenge (e.g., input data and problem definition) to the entered generated AI.

[0826] Step 3: Generative AI solves the problem

[0827] The generative AI generates a solution to the problem and returns the result to the server.

[0828] Step 4: The server receives the results of each AI.

[0829] The server receives the problem-solving results returned by the generation AI and temporarily stores them in a database.

[0830] Step 5: The server evaluates the results

[0831] The server evaluates and scores the received results based on criteria (e.g., accuracy, speed, resource consumption, etc.).

[0832] Step 6: Server determines winner

[0833] Based on the evaluation results, the server determines the best performing AI generator as the winner, and the winner's information is stored in a database.

[0834] Rating and publishing features

[0835] Step 1: The server evaluates the battle results

[0836] The server tally the scores of each generated AI and create a ranking based on the evaluation criteria.

[0837] Step 2: The server creates the rankings

[0838] The server will create a ranking of the AI ​​generators based on the evaluation results. This ranking will be ranked based on the performance of the AI ​​generators.

[0839] Step 3: The server stores the results in a database

[0840] The server saves the ranking information it creates in a database, which also stores past battle results and evaluation history.

[0841] Step 4: Server updates public page

[0842] The server updates the public page with the latest information based on the ranking information stored in the database.

[0843] Step 5: User confirms the results

[0844] Users can access the public page using their devices to check the battle results and the ranking of the generated AI, allowing them to intuitively understand and experience the technological evolution of the generated AI.

[0845] Incorporating an emotion engine

[0846] Step 1: User visits public page

[0847] Users use their devices to access a public page that displays the battle results and rankings of the generated AI.

[0848] Step 2: The emotion engine analyzes the user's emotions

[0849] The emotion engine is built into the user's device and analyzes the user's facial expressions and voice in real time, for example, tracking the user's facial expressions through a camera and analyzing the user's tone of voice through a microphone.

[0850] Step 3: Send emotion data to the server

[0851] The emotion engine sends the analyzed emotion data to the server, which includes the user's emotional state (e.g., joy, surprise, dissatisfaction, etc.).

[0852] Step 4: The server receives the emotion data

[0853] The server receives the emotion data sent from the user terminal.

[0854] Step 5: The server stores the emotion data in a database

[0855] The server stores the received emotional data in a database, which includes the user's emotional response to each battle outcome.

[0856] Step 6: Analyze the sentiment data

[0857] The server analyzes the accumulated emotional data and uses it to improve the generation AI and optimize battles.

[0858] Step 7: Feedback and Adjustments

[0859] Feedback based on emotional data is provided to the developers of the AI ​​generator, allowing them to make adjustments to improve the AI's performance and user experience. For example, if a user is satisfied with the outcome of a particular battle, the AI ​​generator will use that strategy or approach in the next battle.

[0860] In this way, by incorporating an emotion engine in addition to entry means, battle means, evaluation and publication means, the present invention makes the generative AI competition and evaluation process more user-friendly and utilizes emotion data to improve and optimize the generative AI.

[0861] Example 2

[0862] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0863] Conventional generative algorithm competition systems did not take into account user emotional data when optimizing or evaluating generative algorithms, and therefore were unable to sufficiently improve the user experience. Furthermore, the evaluation criteria for generative algorithms were limited and lacked diversity. Furthermore, the publication and evaluation of battle results were not centrally managed, making it difficult for user feedback to be reflected in real time.

[0864] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a transmission means for transmitting information about the generation algorithm from the user terminal to the central processing unit, a storage means for receiving the transmitted information and storing it in a storage device, a provision means for providing challenges to the entered generation algorithms, a collection means for collecting the solution results of each generation algorithm, an evaluation means for analyzing the collected results and determining a winner based on evaluation criteria, an emotion engine means for recognizing the user's emotions, and a data collection and analysis means for collecting emotion data obtained by the emotion engine means and using it to optimize the system. This makes it possible to evaluate and optimize generation algorithms that reflect the user's emotion data, thereby improving the user experience and enabling diverse evaluation of generation algorithms.

[0865] A "generative algorithm" refers to a series of computational steps for generating new data or information from existing data.

[0866] "Entry Method" refers to the method or process by which a Generator Algorithm is registered with the System.

[0867] "Competition method" refers to the method or process by which the entered generative algorithms compete to provide solutions to a given challenge.

[0868] "Evaluation and disclosure measures" refers to the methods and processes for evaluating the results of the competition and disclosing those evaluation results to the public.

[0869] "Emotion engine means" refers to technology or tools for analyzing emotions from a user's facial expressions and voice.

[0870] "Data collection and analysis means" refers to the method or process of collecting data obtained by the emotion engine means and using it to optimize the system.

[0871] "Transmission means" refers to a method or process for transmitting information of the generation algorithm from the user terminal to the central processing unit.

[0872] "Storage" refers to the method or process for receiving transmitted information and storing it in a storage device.

[0873] "Provision means" refers to the method or process for providing a challenge to the entered generation algorithm.

[0874] "Collection means" refers to the method or process for collecting the solution results of each generation algorithm.

[0875] "Evaluation means" refers to a method or process for analyzing the collected solution results of the generative algorithms and determining a winner based on evaluation criteria.

[0876] "Central Processing Unit" refers to a central computer that performs multiple calculations.

[0877] "Storage" refers to a device or medium for storing information.

[0878] The present invention is a competition system using a generative AI model, which combines an entry means, a competition means, an evaluation and publication means, an emotion engine means, and a data collection and analysis means. In this system, the user, the terminal, and the server each play different roles. Specific embodiments of the system are described below.

[0879] Hardware and software used

[0880] This invention is composed of various hardware and software, including a user terminal, a server, a database, and an emotion engine. The user terminal is equipped with an emotion engine (e.g., facial recognition software, voice recognition software) for analyzing emotions from the user's facial expressions and voice. The server is a central processing unit that processes information sent from the user terminal and stores the information in a database (e.g., MySQL).

[0881] Entry function

[0882] Users use their device to enter their own generative AI model. A form is displayed on the user device to enter the name, description, executable file, developer information, etc. of the generative AI model. When the user enters the information and clicks the entry button, the entered information is converted to JSON format and sent to the server. The server receives the submitted information, verifies its validity, and stores it in a database. The information is encrypted before being stored and is kept securely.

[0883] Battle Features

[0884] At a pre-set time, the server will start a battle between the entered generative AI models. As the start time approaches, the server will prepare the execution environment for all relevant generative AI models. The server will provide each generative AI model with details of the challenge (e.g., input data and problem definition). Each generative AI model will generate a solution based on this challenge and return the results to the server in JSON format. The server will receive these results and temporarily store them in a database.

[0885] Rating and publishing features

[0886] The server applies an algorithm to score the collected results based on evaluation criteria (e.g., solution accuracy, speed, resource consumption). The resulting score is then assigned to the generative AI model. Furthermore, the server creates a ranking based on the scoring results and stores it in a database. It updates a public page with the latest ranking information, allowing the public to view the latest battle results through this public page.

[0887] Incorporating an emotion engine

[0888] The user device is equipped with an emotion engine that analyzes emotions from the user's facial expressions and voice. While the user is viewing the battle results, the emotion engine analyzes the user's emotional state (e.g., joy, surprise, dissatisfaction, etc.) in real time. The analyzed emotion data is sent to the server and stored in a database. This data is used to improve the generative AI model and optimize battles. By analyzing the accumulated emotion data and reflecting it in improving the generative AI model, battle settings, and adjusting evaluation criteria, the overall user experience of the system and the performance of the generative AI model are improved.

[0889] Specific examples

[0890] For example, when a user is viewing the results of a battle between generative AI models, the emotion engine will detect the user's smile. This data will be sent to the server and stored in the database. This data will then be analyzed and used in the generative AI model's feedback process. As a result, the next battle will take the user's emotional reactions into account, resulting in a more effective generative AI competition.

[0891] Prompt Sentence Examples

[0892] Below are some example prompts that can be used as input to a generative AI model:

[0893] Based on the reaction data when the user smiles, please suggest ways to optimize the performance of the generative AI.

[0894] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0895] Step 1:

[0896] The user uses a user terminal to enter their own generative AI model. A form is displayed on the user terminal in which the user can enter the name, description, executable file, developer information, etc. of the generative AI model. The user enters the required information in these fields and clicks the entry button. The input data is processed as metadata for the generative AI model and converted to JSON format. The output is the entry information (JSON format) that is sent to the server.

[0897] Step 2:

[0898] The terminal sends the entry information entered by the user to the server in JSON format. The transmitted data is encrypted and securely sent to the server. The input is the entry information entered by the user, and the output is the entry information securely sent to the server.

[0899] Step 3:

[0900] The server receives the entry information sent from the terminal. After receiving it, the server verifies the validity of the entry information, and if it is confirmed to be valid, it stores it in a database. The stored information includes the name of the generated AI model, a description, the path to the executable file, and the developer's contact information. The input is the received entry information, and the output is the entry information stored in the database.

[0901] Step 4:

[0902] The server notifies the user terminal that the entry has been completed. An entry completion message is displayed on the user terminal, allowing the user to confirm that the entry was successful. The input is the entry completion notification from the server, and the output is the entry completion message displayed on the user terminal.

[0903] Step 5:

[0904] At a pre-set time, the server will start a battle between the entered generative AI models. As the start time of the battle approaches, the server prepares the execution environments for all related generative AI models. The input is the battle start time, and the output is the execution environments for the prepared generative AI models.

[0905] Step 6:

[0906] The server provides each generative AI model with details of the challenge (e.g., input data and problem definition). Each generative AI model generates a solution based on this challenge. The input is the problem description from the server, and the output is the solution generated by the generative AI model.

[0907] Step 7:

[0908] The generative AI model outputs the results of problem solving and returns them to the server in JSON format. The server receives the returned results and temporarily stores them in a database. The input is the solution result of the generative AI model, and the output is the result stored in the database.

[0909] Step 8:

[0910] The server applies an algorithm to score the collected results based on evaluation criteria (e.g., solution accuracy, speed, resource consumption). The input is the collected results, and the output is a score assigned to each generative AI model.

[0911] Step 9:

[0912] The server determines the winner and stores the winner's information in a database. The input is the scored result, and the output is the information of the generative AI model that was determined as the winner.

[0913] Step 10:

[0914] The server creates a ranking based on the scoring results and saves it in a database. It then updates the public page with the latest ranking information. The input is the scoring results, and the output is the updated public page.

[0915] Step 11:

[0916] The user device is equipped with an emotion engine that analyzes emotions from the user's facial expressions and voice. While the user is viewing the battle results, the emotion engine analyzes the user's emotional state in real time. The input is the user's facial expressions and voice data, and the output is the analyzed emotional data.

[0917] Step 12:

[0918] The emotion engine sends the analyzed emotion data to the server. The input is the analyzed emotion data, and the output is the emotion data sent to the server.

[0919] Step 13:

[0920] The server receives the emotion data and stores it in a database. This data is used to improve the generative AI model and optimize battles. The input is emotion data, and the output is the emotion data stored in the database.

[0921] Step 14:

[0922] The server analyzes the accumulated emotional data and reflects it in improving the generative AI model, battle settings, and adjusting evaluation criteria. This improves the user experience of the entire system and the performance of the generative AI model. The input is the accumulated emotional data, and the output is the overall system performance improved through improvements and adjustments.

[0923] (Application example 2)

[0924] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0925] Conventional battle systems using generative models lacked evaluation and optimization that took user emotions into account, resulting in a limited user experience. Furthermore, because the performance of generative models was evaluated solely based on technical criteria, users were not provided with sufficient satisfaction. Furthermore, there was a lack of a mechanism for analyzing user emotion data in real time and using it as feedback.

[0926] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0927] In this invention, the server includes an entry means for AI developers to enter generative models, a battle means in which the entered generative models compete against each other in specific tasks, an evaluation and publication means for evaluating and publishing the battle results, an emotion engine that analyzes emotions from the user's facial expressions and voice, and an adjustment means for optimizing the performance of the generative model based on the user's emotion data. This enables evaluation and optimization that takes user emotions into consideration, providing a more satisfying user experience.

[0928] "Entry means" refers to the means by which an AI developer registers a generative model in the system, and includes a transmission means for inputting information about the generative model and sending it to a server, and a storage means for receiving the sent information and storing it in a database.

[0929] The "battle means" refers to a means by which entered generative models compete against each other over specific pre-set challenges, and includes a collection means for providing the challenges and collecting the solution results of each generative model, and an evaluation means for analyzing the collected results and determining a winner based on evaluation criteria.

[0930] The "evaluation and publication means" is a means for evaluating the battle results, scoring the generative models based on the scoring criteria, creating rankings, and storing and publishing them in a database.

[0931] The "emotion engine" is part of a system that analyzes emotions from the user's facial expressions and voice in real time and sends the analysis data to a server.

[0932] The "adjustment method" is a method for collecting and analyzing user emotional data, optimizing the performance of the generative model based on that data, and improving the user experience.

[0933] As a form for implementing this invention, a system using a smartphone application is constructed.

[0934] System Configuration

[0935] 1. Hardware and Software Configuration

[0936] User device: Smartphone

[0937] Emotion Engine: EmotionRecognizer

[0938] AI Battle System: AIBattle

[0939] Content Management System: ContentManager

[0940] Entry Method

[0941] Using a user device, an AI developer registers a generative model in the system. An entry form is provided, and the developer enters the name of the generative model, a description, an executable file, developer information, etc. The input information is packaged in JSON format or similar and sent to the server. The server then stores the submitted information in a database.

[0942] Battle Method

[0943] The server sets up battles in which the submitted generative models compete against each other for specific tasks. Each task is provided with detailed information, including user emotional data. The generative models generate solutions to the tasks and return the results to the server. The server analyzes the collected results and determines the winner based on evaluation criteria.

[0944] Evaluation and Publication Methods

[0945] The battle results are evaluated and the generative models are scored based on the scoring criteria. The rankings are created and saved in a database, and are made available to the general public by updating a public page.

[0946] Incorporating an emotion engine

[0947] The system incorporates an emotion engine that uses the smartphone's camera and microphone to analyze the user's facial expressions and voice. The analyzed emotion data is sent to a server in real time. The server accumulates the emotion data and makes adjustments to optimize the performance of the generative model based on the user's emotional responses.

[0948] Specific examples

[0949] If a user smiles while watching a drama, their emotional data is analyzed in real time and stored in a database. Based on this emotional data, the next time content is delivered, optimized content is provided, taking into account the user's emotional response.

[0950] Prompt Sentence Examples

[0951] For example, you might input the following prompt into a generative AI model:

[0952] "Collect emotional data when a user smiles while watching a drama, and generate a list of content to display next based on that."

[0953] In this way, the invention incorporates an emotion engine in addition to entry means, battle means, evaluation and publication means, making the generative model competition and evaluation process more user-friendly and utilizing emotion data to improve and optimize generative models.

[0954] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0955] Step 1:

[0956] Execution of entry method

[0957] The user enters information about the generative model into an input form on the device. Specifically, the user enters the name of the generative model, a description, an executable file, developer information, etc. The entered information is packaged in JSON format and sent to the server.

[0958] Input: Name of the generated model, description, executable file, developer information

[0959] Output: Generated model information packaged in JSON format

[0960] Specific operation: The user enters information into the form and presses the submit button to send the data to the server.

[0961] Step 2:

[0962] Saving entry information

[0963] The server receives the generated model information in JSON format sent from the user device and stores it in a database, including the name of the generated model, its description, the path to the executable file, and the developer's contact information.

[0964] Input: Generated model information in JSON format

[0965] Output: Generative model information stored in a database

[0966] Specific operation: The server receives the information and stores it in a database.

[0967] Step 3:

[0968] The start of the battle

[0969] The server will start a battle between the entered generative models at a pre-set time, providing specific challenges to the generative models.

[0970] Input: Generative model information stored in the database, task details

[0971] Output: A generative model that receives the task.

[0972] Specific operation: The server triggers the start of the battle and sends the details of the challenge to the generative model.

[0973] Step 4:

[0974] Collection of problem-solving results

[0975] The generative model generates a solution to the problem and returns the result to the server, which receives the result and temporarily stores it in a database.

[0976] Input: Results of solving the problem using the generative model

[0977] Output: Solution results stored in the database

[0978] Specific operation: The generative model executes the problem-solving process and returns the results to the server.

[0979] Step 5:

[0980] Evaluating the results

[0981] The server scores the collected solutions based on evaluation criteria, including solution accuracy, speed, and resource consumption.

[0982] Input: Solution results stored in the database

[0983] Output: Scoring results

[0984] Specific Operation: The server runs the evaluation algorithm and generates a score.

[0985] Step 6:

[0986] Determining the Winner

[0987] The server determines the best performing generative model as the winner based on the scoring results, and the winner's information is stored in a database.

[0988] Input: Scoring results

[0989] Output: Winner information

[0990] What happens: The server compares the scores and marks the generative model with the highest score as the winner.

[0991] Step 7:

[0992] Publication of ratings and rankings

[0993] The server creates a ranking based on the scoring results and updates the public page, allowing general users to view the latest battle results and the rankings of generative models.

[0994] Input: Scoring results

[0995] Output: Published rankings

[0996] Specific operation: The server updates the ranking page and publishes the latest information.

[0997] Step 8:

[0998] User sentiment analysis

[0999] When a user browses a public page, the smartphone's camera and microphone are used to analyze the user's facial expressions and voice. Emotional data is analyzed in real time and sent to the server.

[1000] Input: User's facial expression data, voice data

[1001] Output: Parsed emotion data

[1002] Specific operation: The smartphone runs the emotion analysis engine and sends the analysis data to the server.

[1003] Step 9:

[1004] Emotional Data Feedback

[1005] The server stores the received emotion data in a database and uses it to improve the generative model and optimize battles.

[1006] Input: Parsed emotion data

[1007] Output: Improved generative model performance

[1008] Specific operation: The server analyzes the emotion data and uses it as feedback to optimize the performance of the generative model.

[1009] In this way, through a series of processing steps, evaluation and optimization of the generative model taking into account the user's emotions is achieved.

[1010] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1011] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1012] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[1013] [Third embodiment]

[1014] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1015] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[1016] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1017] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[1018] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1019] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1020] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1021] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1022] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1023] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1024] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1025] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[1026] The present invention is a system for competing in problem-solving ability using generative AI, and is implemented in the following form.

[1027] Entry function

[1028] 1. User enters AI

[1029] Users use their device to enter their own generated AI. A form is displayed on the user device in which they can enter the name, description, executable file, developer information, etc. of the generated AI. Based on this information, they press the entry button to submit the information.

[1030] 2. The device sends the entry information to the server

[1031] The user terminal sends the entry information entered by the user to the server, where it is packaged in a standard format such as JSON.

[1032] 3. The server receives the entry information

[1033] The server receives the entry information sent from the user's device and stores it in a database, including the name of the generated AI, a description, the path to the executable file, and the developer's contact information.

[1034] 4. Notification of entry completion

[1035] The server notifies the user terminal that the entry has been completed successfully. An entry completion message is displayed on the user terminal, allowing the user to confirm that the entry has been completed successfully.

[1036] Battle Features

[1037] 1. The server starts the battle

[1038] The server will start a battle between the registered generated AIs at a pre-set time. At the start of the battle, specific tasks will be set for each generated AI.

[1039] 2. The server provides each AI with a task

[1040] The server provides the details of the challenge (e.g., input data and problem definition) to the submitted AI generators, who then generate a solution to the challenge.

[1041] 3. The server receives the results of each AI

[1042] Each generation AI outputs the solution to the problem and returns it to the server, which receives these results and temporarily stores them in a database.

[1043] 4. The server evaluates the results

[1044] The server evaluates the received results based on evaluation criteria (e.g., solution accuracy, speed, resource consumption, etc.) and assigns a score to each generated AI.

[1045] 5. The server decides the winner

[1046] Based on the evaluation results, the server determines the best performing AI generator as the winner, and the winner's information is stored in the database.

[1047] Result announcement and evaluation function

[1048] 1. The server evaluates the battle results

[1049] After the battle, the server will tally the scores of each AI generator and create a ranking, which is a fair evaluation of the AI ​​generator's ability to solve the problem.

[1050] 2. The server creates the rankings

[1051] The server creates a ranking of the AI ​​generators based on the collected evaluation results. This ranking is based on the performance of the AI ​​generators.

[1052] 3. The server stores the results in a database

[1053] The server saves the rankings it creates in a database, which also stores past battle results and the evaluation history of each generated AI.

[1054] 4. The server updates the public page

[1055] The server updates the public page based on the ranking information stored in the database, allowing general users to view the latest battle results and the rankings of the generated AI.

[1056] 5. User checks the results

[1057] Users can access the public page using their devices to check the battle results and the rankings of the generated AI, allowing general users to intuitively understand and experience the technological evolution of generated AI.

[1058] The above system provides AI developers with a forum where they can fairly compete and receive evaluations on the performance of their generative AI, and it also provides general consumers with a forum where they can intuitively understand and experience the technological evolution of generative AI.

[1059] The processing flow will be explained below.

[1060] Entry function

[1061] Step 1: User enters AI

[1062] The user enters information about their generated AI (name, description, executable file, developer information, etc.) into the user terminal and presses the entry button.

[1063] Step 2: The device sends the entry information to the server

[1064] The terminal sends the entry information entered by the user to the server in a standard format such as JSON.

[1065] Step 3: Server receives entry information

[1066] The server receives the entry information sent from the terminal.

[1067] Step 4: The server saves the entry information to a database

[1068] The server stores the received entry information in a database, including the name of the generated AI, its description, the path to the executable file, and developer information.

[1069] Step 5: The server notifies you that the entry is complete

[1070] The server notifies the user terminal that the entry has been successfully completed, and an entry completion message is displayed on the user terminal.

[1071] Battle Features

[1072] Step 1: The server starts the battle

[1073] The server will start a battle between the entered generated AIs at a pre-set time.

[1074] Step 2: The server provides each AI with a task

[1075] The server provides the details of the challenge (e.g., input data and problem definition) to the entered generated AI.

[1076] Step 3: The server receives the results of each AI.

[1077] Each generation AI generates a problem-solving result and returns it to the server, which receives the result and temporarily stores it in a database.

[1078] Step 4: The server evaluates the results

[1079] The server evaluates the received results based on criteria (e.g., accuracy, speed, resource consumption), and the evaluation results are scored.

[1080] Step 5: Server determines winner

[1081] Based on the evaluation results, the server determines the AI ​​generator with the best performance as the winner, and the winner's information is recorded in the database.

[1082] Result announcement and evaluation function

[1083] Step 1: The server evaluates the battle results

[1084] The server tally the scores of each generated AI and create a ranking based on the evaluation criteria.

[1085] Step 2: The server creates the rankings

[1086] The server uses the evaluation results to create a ranking of the generated AIs, which is ranked based on the performance of the generated AIs.

[1087] Step 3: The server stores the results in a database

[1088] The server saves the ranking information it creates in a database, which also stores past battle results and evaluation history.

[1089] Step 4: The server updates the public page

[1090] The server updates the public page with the latest information based on the rankings stored in the database.

[1091] Step 5: User confirms the results

[1092] Users can access the public page using their devices to check the battle results and the ranking of the generated AI, allowing them to intuitively understand and experience the technological evolution of the generated AI.

[1093] Example 1

[1094] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1095] Currently, many AI developers are developing generative AI, but there is a lack of venues for objectively evaluating the performance of these generative AIs and for them to compete against each other. This makes it difficult for generative AI developers to compare the capabilities of their own AI with those of other AIs and identify areas for improvement. It is also difficult for ordinary users to intuitively understand and experience the evolution and characteristics of generative AI. A new system is needed to solve this problem and promote technological advances in generative AI.

[1096] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1097] In this invention, the server includes a storage means for receiving entry information and storing it in a database, a management means for checking the start time of the battle and providing challenges to the AI ​​generators, and an evaluation means for evaluating the results of the challenges and determining the best AI generators. This allows the performance of AI generators to be fairly evaluated and allows developers to compete against each other.

[1098] "Entry means" refers to the means by which an AI developer enters a generated AI into the system, and includes information sent from the user terminal to the server.

[1099] "Battle means" refers to the means by which entered generated AIs compete against each other on specific challenges, and includes the function of providing challenges and collecting the results of their solutions.

[1100] "Means for evaluation and publication" refers to means for evaluating the results of battles and publishing the results, and includes functions for evaluating the generating AI based on evaluation criteria, storing the results in a database, and publishing them to the general public.

[1101] "Input means" refers to the means by which a user inputs and submits entry information for the generated AI, and refers to the input form provided on the user's terminal.

[1102] "Storage means" refers to the means by which the server stores the received entry information in a database, and includes temporary and permanent storage of data.

[1103] The "management means" is a means by which the server checks the start time of the battle and starts the battle at the set time, and includes a function for managing the progress of the battle.

[1104] "Task providing means" refers to the means by which the server provides specific tasks to the entered generation AI, and includes the generation and distribution of tasks.

[1105] "Evaluation means" refers to the means by which the server evaluates and scores the solution results of each generated AI based on the evaluation criteria, and includes analyzing the results and applying the evaluation criteria.

[1106] "Determination means" refers to the means by which the server determines the best generating AI and stores that information in a database, including determining the winner and storing the results.

[1107] "Server communication means" refers to the means by which the server provides challenges to each generation AI and each generation AI transmits a solution to the server, and includes establishing communication and sending and receiving data.

[1108] "Evaluation criteria setting means" means a means for the evaluation means to set criteria for evaluating each generative AI based on the accuracy, speed, and resource consumption of the solution, and includes the definition and application of the evaluation criteria.

[1109] This invention is a system for competing in problem-solving ability using generative AI, and it operates in cooperation with users, terminals, and a server. The system has the following functions.

[1110] Entry function

[1111] To enter a generated AI, a user first uses their device to enter the name, description, executable file path, and developer information into an entry form. Specific hardware used for this is a personal computer or smartphone, and the user accesses the form via a browser. A web browser (e.g., Google Chrome or Safari) is used as the software.

[1112] When the user presses the entry button, the terminal converts the entered information into JSON format and sends it to the server as an HTTP POST request. The server analyzes the received data and saves it in a database. This saving process uses the server's internal database system (e.g., MySQL, PostgreSQL). Once the entry is complete, the server sends a completion notification to the user's terminal. The user's terminal displays the message "Entry completed successfully."

[1113] As a specific example, a user enters a generation AI called "AI_Champion" and enters the description, "This is an AI that analyzes the emotions in English sentences."

[1114] Battle Features

[1115] The server starts a battle between the entered generated AIs at a pre-set time. It checks the start time of the battle and provides specific tasks to the entered generated AIs. For example, a task might be to "analyze emotions from English sentences and assign a score." The server packages and sends the task details to each generated AI.

[1116] Each generation AI generates a solution to the provided problem and sends the solution results to the server. The server temporarily stores the received solution results in a database. The server then evaluates and scores the results of each generation AI based on evaluation criteria (e.g., solution accuracy, speed, resource consumption, etc.). Various evaluation algorithms and evaluation criterion setting methods are used for the evaluation process.

[1117] As an example of specific operation, the server receives an emotion score of 0.95 assigned by "AI_Champion" and evaluates it as "AI_Champion's score is 95 points."

[1118] Result announcement and evaluation function

[1119] After the battle is over, the server tally the scores of each generated AI and create a ranking. This ranking is based on the performance of the generated AI and is stored in a database. The server updates a public page based on the ranking information, allowing general users to view the latest battle results and the rankings of the generated AI. Users can access the public page using their devices to check the results.

[1120] As a specific example, a user accesses a public page and sees the information "AI_Champion is in first place."

[1121] Prompt Sentence Examples

[1122] "Create a generative AI that takes an English sentence as input, analyzes its sentiment, and gives it a score."

[1123] This invention provides AI developers with a forum where they can fairly compete and have the performance of their generative AI evaluated, and it also provides general users with a forum where they can intuitively understand and experience the technological evolution of generative AI.

[1124] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1125] Entry function

[1126] Step 1: Display the entry form

[1127] The user opens the entry page in a browser and enters the name, description, executable file path, and developer information of the generated AI. HTML and JavaScript are used to display the required fields in this form.

[1128] Input: User-entered name, description, executable path, and developer information for the generated AI.

[1129] Output: The input form reflects the user's input.

[1130] Step 2: Prepare input information for submission

[1131] When the user clicks the entry button, the terminal converts the input information into JSON format and packages it as an HTTP POST request.

[1132] Enter: When the user clicks the enter button.

[1133] Output: Data package converted to JSON format.

[1134] Step 3: Submit your entry

[1135] The terminal sends the packaged entry information to the server, and this communication uses the HTTP protocol.

[1136] Input: A data package in JSON format.

[1137] Output: The HTTP POST request sent to the server.

[1138] Step 4: Receiving and analyzing data

[1139] The server receives the HTTP request and parses the JSON data to extract the name, description, executable path, and developer information of the generated AI.

[1140] Input: HTTP POST request.

[1141] Output: Extracted entry information.

[1142] Step 5: Save your data

[1143] The server stores the extracted entry information in a database using SQL queries.

[1144] Input: The extracted entry information.

[1145] Output: Entry information stored in a database.

[1146] Step 6: Notification of completed entry

[1147] The server generates a response to notify the user that the entry has been successfully completed, and sends it to the user terminal.

[1148] Input: Notification that saving is complete.

[1149] Output: Entry complete message as an HTTP response.

[1150] Step 7: Displaying the entry completion message

[1151] The user terminal receives the response from the server and displays the message "Entry completed successfully" in the browser.

[1152] Input: The response from the server.

[1153] Output: Entry successful message displayed in browser.

[1154] Battle Features

[1155] Step 1: Check the battle start time

[1156] The server checks the pre-set battle start time and sees if that time has come, using a timer or scheduling function.

[1157] Input: Pre-set battle start time.

[1158] Output: Trigger to start the battle.

[1159] Step 2: Prepare for assignment delivery

[1160] The server generates and packages challenges to initiate battles between AIs, including problem definitions and input data.

[1161] Input: Trigger to start the battle.

[1162] Output: Packaged assignment content.

[1163] Step 3: Submit your assignment

[1164] The server sends the packaged tasks to each generating AI, and this communication may use HTTP or WebSocket.

[1165] Input: Packaged assignment content.

[1166] Output: The challenges sent to each generated AI.

[1167] Step 4: Generate solutions

[1168] Each generative AI generates a solution to a problem, which is achieved by the AI ​​model performing internal data calculations and analysis.

[1169] Input: Received assignment content.

[1170] Output: The generated solution.

[1171] Step 5: Submit your solution

[1172] Each generating AI returns a solution to the server, again using the HTTP protocol for this communication.

[1173] Input: The generated solution.

[1174] Output: The solution sent to the server.

[1175] Step 6: Receive and save the solution

[1176] The server temporarily stores the received solutions in a database using SQL queries.

[1177] Input: The solutions submitted by each generating AI.

[1178] Output: The solution stored in the database.

[1179] Step 7: Evaluate the solution

[1180] The server evaluates the saved solutions based on evaluation criteria and assigns a score to each generated AI, including solution accuracy, speed, and resource consumption.

[1181] Input: Solutions stored in the database.

[1182] Output: The evaluated score.

[1183] Step 8: Determine the winner

[1184] The server will determine the generator AI that performed best based on the evaluation results as the winner.

[1185] Input: The assessed score.

[1186] Output: Determining the winner and storing that information in a database.

[1187] Result announcement and evaluation function

[1188] Step 1: Counting scores

[1189] The server compiles the scores of each generated AI and creates a ranking, which is done by processing and calculating the score data.

[1190] Input: The score of each generated AI.

[1191] Output: Aggregated scores and rankings.

[1192] Step 2: Save ranking information

[1193] The server stores the rankings in a database.

[1194] Input: Aggregated scores and rankings.

[1195] Output: Ranking information stored in a database.

[1196] Step 3: Update your public page

[1197] The server updates the public page based on the ranking information stored in the database, using a script on the web server.

[1198] Input: Ranking information stored in the database.

[1199] Output: The updated public page.

[1200] Step 4: User confirmation of results

[1201] Users can access the public page using their devices to check the battle results and the ranking of the generated AI. This operation is performed using a web browser.

[1202] Input: Access to public pages.

[1203] Output: Battle results and ranking information displayed on the user's device.

[1204] The above is a detailed explanation of each processing step.

[1205] (Application example 1)

[1206] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1207] Current generative AI competition systems allow AI developers to enter their generative AIs, battle them, and evaluate and publish the results. However, they lack the functionality to allow general users to watch generative AI battles in real time or view past battle results, making it difficult to provide opportunities for users to deepen their interest and understanding. For this reason, there is a need for a system that allows users to intuitively understand and experience the technological evolution of generative AI.

[1208] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1209] In this invention, the server includes an entry means for AI developers to enter generated AIs, a battle means in which the entered generated AIs compete against each other in specific tasks, an evaluation and publication means for evaluating and publishing the battle results, a real-time distribution means for distributing the battle results in real time, an archive viewing means for selecting and viewing past battle results from an archive, and a user evaluation means for users to evaluate the battle results and enter comments. This enables general users to intuitively understand and experience the technological evolution of generated AIs by watching battles between generated AIs in real time and viewing past battle results from the archive.

[1210] "Entry means" refers to a device, method, or system that allows an AI developer to enter a generated AI.

[1211] "Battle means" refers to a device, method, or system that allows entered generated AIs to compete against each other in a specific challenge.

[1212] The "evaluation and disclosure means" refers to a device, method, or system for evaluating the battle results and disclosing the results to users.

[1213] "Real-time distribution means" refers to a device, method, or system for distributing the battle results of the generated AI to users in real time.

[1214] "Archive viewing means" refers to a device, method, or system that allows a user to select past generated AI battle results from a database and view them.

[1215] The "user evaluation means" refers to a device, method, or system that allows users to evaluate the battle results and input comments.

[1216] "Transmission means" refers to a device, method, or system for transmitting information about the generated AI from a user terminal to a server.

[1217] "Storage means" means a device, method, or system for receiving and storing transmitted information in a database.

[1218] "Collection means" refers to a device, method, or system for providing challenges to the entered generation AI and collecting the results of its solutions.

[1219] "Evaluation means" refers to a device, method, or system for analyzing the collected solution results of the generative AI and determining a winner based on evaluation criteria.

[1220] This invention is a system for competing in problem-solving ability using generative AI, and includes as its components an entry means, a battle means, an evaluation and publication means, a real-time distribution means, an archive viewing means, and a user evaluation means. How this system is specifically implemented is explained below.

[1221] System configuration

[1222] The main components are a server, user devices (smartphones, smart glasses, head-mounted displays (HMDs)), and a database. The server uses a cloud service (AWS) and the database is MySQL. React Native is used for the front end, and Node.js and Express.js are used for the back end.

[1223] Entry Method

[1224] Using a user terminal, the AI ​​developer enters the generated AI. At this time, a form is displayed in which the name, description, executable file, developer information, etc. of the generated AI can be entered, and the information is submitted by pressing the "entry" button. The submitted information is received by the server, packaged in JSON format, etc., and stored in a database.

[1225] Battle Method

[1226] The server starts a battle between the AI ​​generators at a pre-set date and time. The AI ​​generators are given specific challenges, and each AI generates a solution to the challenge. The solutions output by the AI ​​generators are sent to the server, where they are analyzed and evaluated based on the evaluation criteria.

[1227] Evaluation and Publication Methods

[1228] The server evaluates the solutions of each AI generator and creates a ranking based on the results, which is then updated on a public page where users can access the rankings.

[1229] Real-time delivery methods

[1230] The server uses WebSocket to deliver the battle results of the generated AI in real time. The user device is built with React Native and receives and displays the real-time data, allowing users to watch the progress of the generated AI battle in real time.

[1231] Archive viewing method

[1232] The results of past generated AI battles are stored in a database and can be viewed by users using the archive function. The data is retrieved using a REST API and displayed on the user's device.

[1233] User evaluation method

[1234] Users can rate and comment on battle results. The entered ratings and comments are sent from the front end to the server and stored in a database, allowing other users to view the feedback.

[1235] Specific examples

[1236] For example, the user can operate the system by entering the following prompts:

[1237] Example viewing prompt for a generated AI battle:

[1238] "Watch the next generative AI battle: Battle ID 12345. Challenge: Natural Language Processing."

[1239] Example of an archive search prompt:

[1240] "View past generative AI battle results: Check how generative AI performed on the challenge."

[1241] In this way, ordinary users will be able to intuitively understand and experience the technological evolution of generative AI.

[1242] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1243] Step 1:

[1244] The user inputs information into the terminal to enter the generated AI. The input information includes the name of the generated AI, a description, an executable file, developer information, etc. The terminal collects this information, packages it in JSON format, etc., and sends it to the server. The input of the data sent to the server is the generated AI information, and the output is the packaged entry information.

[1245] Step 2:

[1246] The server receives the entry information sent by the user. The received information is stored in a database. The database stores the name of the generated AI, a description, the path to the executable file, and the developer's contact information. The input is the entry information sent from the terminal, and the output is the entry information stored in the database.

[1247] Step 3:

[1248] The server will start a battle between the entered generated AIs at a pre-set date and time. Each generated AI will be provided with a specific task. The task details include input data and a problem definition, which are sent to each generated AI. The input is the generated AI's entry information and task definition, and the output is the task provided to the generated AI.

[1249] Step 4:

[1250] The generation AI generates a solution to the provided challenge. The generated solution is sent to the server. The input is the challenge provided by the server, and the output is the generation AI's solution.

[1251] Step 5:

[1252] The server receives the solution sent from the generation AI and temporarily stores it in a database. The input is the solution sent from the generation AI, and the output is the solution temporarily stored in the database.

[1253] Step 6:

[1254] The server analyzes the collected solutions of the generative AI and assigns a score based on evaluation criteria, such as solution accuracy, speed, and resource consumption. The input is the solution stored in the database, and the output is the evaluated score.

[1255] Step 7:

[1256] The server creates a ranking of the generative AI based on the evaluation results and stores it in a database. The input is the evaluated score, and the output is the ranking information.

[1257] Step 8:

[1258] The server distributes the battle results of the generated AI in real time. It uses WebSocket to send data sequentially to the user's device. The input is the evaluated score and the progress of the solution, and the output is the battle result data distributed in real time.

[1259] Step 9:

[1260] The user watches the progress of the generated AI battle in real time. The device receives data from the server and reflects it on the UI. The input is real-time data sent from the server, and the output is the progress of the battle displayed on the device.

[1261] Step 10:

[1262] Users can view past generated AI battle results using the archive function. The device sends a request for archive data to the server, which retrieves the relevant data from the database and sends it to the device. The input is the archive request from the user, and the output is the past battle results displayed on the device.

[1263] Step 11:

[1264] Users rate and comment on the battle results. The terminal sends the entered ratings and comments to the server, which stores them in a database. The input is the user's rating and comments, and the output is the rating and comments stored in the database.

[1265] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1266] This invention is a system that uses generative AI to compete in problem-solving ability, and combines it with an emotion engine that recognizes users' emotions. This system consists of the following elements: entry means, battle means, evaluation and publication means, and the emotion engine.

[1267] Entry function

[1268] Entry

[1269] Users use their device to enter their own generated AI. A form is displayed on the user device in which they can enter the name, description, executable file, developer information, etc. of the generated AI. Based on this information, they press the entry button to submit the information.

[1270] send

[1271] The user terminal sends the entry information entered by the user to the server, where it is packaged in a standard format such as JSON.

[1272] Receive and store

[1273] The server receives the entry information sent from the device and stores it in a database, including the name of the generated AI, a description, the path to the executable file, and the developer's contact information.

[1274] Completion notification

[1275] The server notifies the user terminal that the entry has been successfully completed, and an entry completion message is displayed on the user terminal.

[1276] Battle Features

[1277] Battle Begins

[1278] The server will start a battle between the registered generated AIs at a pre-set time, with specific challenges set for each generated AI.

[1279] Provide assignments

[1280] The server provides the generative AIs with the details of the challenge (e.g., input data and problem definition), and each generative AI generates a solution to this challenge.

[1281] Result collection

[1282] The generative AI outputs the results of the problem solving and returns them to the server, which temporarily stores these results in a database.

[1283] Results evaluation

[1284] The server evaluates and scores the results based on evaluation criteria (e.g., solution accuracy, speed, resource consumption).

[1285] Winner Decided

[1286] Based on the evaluation results, the server determines the best performing AI generator as the winner, and the winner's information is stored in the database.

[1287] Rating and publishing features

[1288] Evaluation and Scoring

[1289] The server evaluates the battle results and assigns a score to each generated AI based on the scoring criteria. A ranking is created based on the scoring results.

[1290] Save and publish results

[1291] The server saves the rankings in a database and updates a public page where users can view the latest battle results and the rankings of the generated AI.

[1292] Incorporating an emotion engine

[1293] emotion recognition

[1294] The user device is equipped with an emotion engine that analyzes the user's emotions from their facial expressions and voice. This emotion engine analyzes the user's emotions in real time as they view the battle results.

[1295] Sending emotional data

[1296] The emotion engine sends the analyzed emotion data to the server, which includes the user's emotional state (e.g., joy, surprise, frustration, etc.).

[1297] Collecting and storing emotional data

[1298] The server stores the transmitted emotion data in a database and uses it to improve the AI ​​generator and optimize battles. This emotion data can be used to provide feedback on the AI ​​generator's performance and user experience.

[1299] Feedback and Adjustments

[1300] The accumulated emotional data is analyzed and used to improve the generation AI, battle settings, and evaluation criteria, thereby improving the overall user experience of the system and the performance of the generation AI.

[1301] For example, if the emotion engine detects a smile when a user views the results of a battle between the generated AI, that data is sent to the server and stored in a database. This data is used in the feedback process of the generated AI and acts as part of its optimization. This allows the generated AI to compete more effectively in the next battle, taking into account the user's emotional reactions.

[1302] In this way, by incorporating an emotion engine in addition to entry means, battle means, evaluation and publication means, the present invention makes the generative AI competition and evaluation process more user-friendly and utilizes emotion data to improve and optimize the generative AI.

[1303] The processing flow will be explained below.

[1304] Entry function

[1305] Step 1: User enters AI

[1306] The user enters information about their generated AI (name, description, executable file, developer information, etc.) into the user terminal and presses the entry button.

[1307] Step 2: The device sends the entry information to the server

[1308] The terminal sends the entry information entered by the user to the server in a standard format such as JSON.

[1309] Step 3: The server receives the entry information

[1310] The server receives the entry information sent from the terminal.

[1311] Step 4: The server saves the entry information to a database

[1312] The server stores the received entry information in a database, including the name of the generated AI, its description, the path to the executable file, and developer information.

[1313] Step 5: The server notifies you that the entry is complete

[1314] The server notifies the user terminal that the entry has been successfully completed, and an entry completion message is displayed on the user terminal.

[1315] Battle Features

[1316] Step 1: The server starts the battle

[1317] The server will start a battle between the entered generated AIs at a pre-set time.

[1318] Step 2: The server provides each AI with a task

[1319] The server provides the details of the challenge (e.g., input data and problem definition) to the entered generated AI.

[1320] Step 3: Generative AI solves the problem

[1321] The generative AI generates a solution to the problem and returns the result to the server.

[1322] Step 4: The server receives the results of each AI.

[1323] The server receives the problem-solving results returned by the generation AI and temporarily stores them in a database.

[1324] Step 5: The server evaluates the results

[1325] The server evaluates and scores the received results based on criteria (e.g., accuracy, speed, resource consumption, etc.).

[1326] Step 6: Server determines winner

[1327] Based on the evaluation results, the server determines the best performing AI generator as the winner, and the winner's information is stored in a database.

[1328] Rating and publishing features

[1329] Step 1: The server evaluates the battle results

[1330] The server tally the scores of each generated AI and create a ranking based on the evaluation criteria.

[1331] Step 2: The server creates the rankings

[1332] The server will create a ranking of the AI ​​generators based on the evaluation results. This ranking will be ranked based on the performance of the AI ​​generators.

[1333] Step 3: The server stores the results in a database

[1334] The server saves the ranking information it creates in a database, which also stores past battle results and evaluation history.

[1335] Step 4: Server updates public page

[1336] The server updates the public page with the latest information based on the ranking information stored in the database.

[1337] Step 5: User confirms the results

[1338] Users can access the public page using their devices to check the battle results and the ranking of the generated AI, allowing them to intuitively understand and experience the technological evolution of the generated AI.

[1339] Incorporating an emotion engine

[1340] Step 1: User visits public page

[1341] Users use their devices to access a public page that displays the battle results and rankings of the generated AI.

[1342] Step 2: The emotion engine analyzes the user's emotions

[1343] The emotion engine is built into the user's device and analyzes the user's facial expressions and voice in real time, for example, tracking the user's facial expressions through a camera and analyzing the user's tone of voice through a microphone.

[1344] Step 3: Send emotion data to the server

[1345] The emotion engine sends the analyzed emotion data to the server, which includes the user's emotional state (e.g., joy, surprise, dissatisfaction, etc.).

[1346] Step 4: The server receives the emotion data

[1347] The server receives the emotion data sent from the user terminal.

[1348] Step 5: The server stores the emotion data in a database

[1349] The server stores the received emotional data in a database, which includes the user's emotional response to each battle outcome.

[1350] Step 6: Analyze the sentiment data

[1351] The server analyzes the accumulated emotional data and uses it to improve the generation AI and optimize battles.

[1352] Step 7: Feedback and Adjustments

[1353] Feedback based on emotional data is provided to the developers of the AI ​​generator, allowing them to make adjustments to improve the AI's performance and user experience. For example, if a user is satisfied with the outcome of a particular battle, the AI ​​generator will use that strategy or approach in the next battle.

[1354] In this way, by incorporating an emotion engine in addition to entry means, battle means, evaluation and publication means, the present invention makes the generative AI competition and evaluation process more user-friendly and utilizes emotion data to improve and optimize the generative AI.

[1355] Example 2

[1356] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1357] Conventional generative algorithm competition systems did not take into account user emotional data when optimizing or evaluating generative algorithms, and therefore were unable to sufficiently improve the user experience. Furthermore, the evaluation criteria for generative algorithms were limited and lacked diversity. Furthermore, the publication and evaluation of battle results were not centrally managed, making it difficult for user feedback to be reflected in real time.

[1358] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a transmission means for transmitting information about the generation algorithm from the user terminal to the central processing unit, a storage means for receiving the transmitted information and storing it in a storage device, a provision means for providing challenges to the entered generation algorithms, a collection means for collecting the solution results of each generation algorithm, an evaluation means for analyzing the collected results and determining a winner based on evaluation criteria, an emotion engine means for recognizing the user's emotions, and a data collection and analysis means for collecting emotion data obtained by the emotion engine means and using it to optimize the system. This makes it possible to evaluate and optimize generation algorithms that reflect the user's emotion data, thereby improving the user experience and enabling diverse evaluation of generation algorithms.

[1359] A "generative algorithm" refers to a series of computational steps for generating new data or information from existing data.

[1360] "Entry Method" refers to the method or process by which a Generator Algorithm is registered with the System.

[1361] "Competition method" refers to the method or process by which the entered generative algorithms compete to provide solutions to a given challenge.

[1362] "Evaluation and disclosure measures" refers to the methods and processes for evaluating the results of the competition and disclosing those evaluation results to the public.

[1363] "Emotion engine means" refers to technology or tools for analyzing emotions from a user's facial expressions and voice.

[1364] "Data collection and analysis means" refers to the method or process of collecting data obtained by the emotion engine means and using it to optimize the system.

[1365] "Transmission means" refers to a method or process for transmitting information of the generation algorithm from the user terminal to the central processing unit.

[1366] "Storage" refers to the method or process for receiving transmitted information and storing it in a storage device.

[1367] "Provision means" refers to the method or process for providing a challenge to the entered generation algorithm.

[1368] "Collection means" refers to the method or process for collecting the solution results of each generation algorithm.

[1369] "Evaluation means" refers to a method or process for analyzing the collected solution results of the generative algorithms and determining a winner based on evaluation criteria.

[1370] "Central Processing Unit" refers to a central computer that performs multiple calculations.

[1371] "Storage" refers to a device or medium for storing information.

[1372] The present invention is a competition system using a generative AI model, which combines an entry means, a competition means, an evaluation and publication means, an emotion engine means, and a data collection and analysis means. In this system, the user, the terminal, and the server each play different roles. Specific embodiments of the system are described below.

[1373] Hardware and software used

[1374] This invention is composed of various hardware and software, including a user terminal, a server, a database, and an emotion engine. The user terminal is equipped with an emotion engine (e.g., facial recognition software, voice recognition software) for analyzing emotions from the user's facial expressions and voice. The server is a central processing unit that processes information sent from the user terminal and stores the information in a database (e.g., MySQL).

[1375] Entry function

[1376] Users use their device to enter their own generative AI model. A form is displayed on the user device to enter the name, description, executable file, developer information, etc. of the generative AI model. When the user enters the information and clicks the entry button, the entered information is converted to JSON format and sent to the server. The server receives the submitted information, verifies its validity, and stores it in a database. The information is encrypted before being stored and is kept securely.

[1377] Battle Features

[1378] At a pre-set time, the server will start a battle between the entered generative AI models. As the start time approaches, the server will prepare the execution environment for all relevant generative AI models. The server will provide each generative AI model with details of the challenge (e.g., input data and problem definition). Each generative AI model will generate a solution based on this challenge and return the results to the server in JSON format. The server will receive these results and temporarily store them in a database.

[1379] Rating and publishing features

[1380] The server applies an algorithm to score the collected results based on evaluation criteria (e.g., solution accuracy, speed, resource consumption). The resulting score is then assigned to the generative AI model. Furthermore, the server creates a ranking based on the scoring results and stores it in a database. It updates a public page with the latest ranking information, allowing the public to view the latest battle results through this public page.

[1381] Incorporating an emotion engine

[1382] The user device is equipped with an emotion engine that analyzes emotions from the user's facial expressions and voice. While the user is viewing the battle results, the emotion engine analyzes the user's emotional state (e.g., joy, surprise, dissatisfaction, etc.) in real time. The analyzed emotion data is sent to the server and stored in a database. This data is used to improve the generative AI model and optimize battles. By analyzing the accumulated emotion data and reflecting it in improving the generative AI model, battle settings, and adjusting evaluation criteria, the overall user experience of the system and the performance of the generative AI model are improved.

[1383] Specific examples

[1384] For example, when a user is viewing the results of a battle between generative AI models, the emotion engine will detect the user's smile. This data will be sent to the server and stored in the database. This data will then be analyzed and used in the generative AI model's feedback process. As a result, the next battle will take the user's emotional reactions into account, resulting in a more effective generative AI competition.

[1385] Prompt Sentence Examples

[1386] Below are some example prompts that can be used as input to a generative AI model:

[1387] Based on the reaction data when the user smiles, please suggest ways to optimize the performance of the generative AI.

[1388] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1389] Step 1:

[1390] The user uses a user terminal to enter their own generative AI model. A form is displayed on the user terminal in which the user can enter the name, description, executable file, developer information, etc. of the generative AI model. The user enters the required information in these fields and clicks the entry button. The input data is processed as metadata for the generative AI model and converted to JSON format. The output is the entry information (JSON format) that is sent to the server.

[1391] Step 2:

[1392] The terminal sends the entry information entered by the user to the server in JSON format. The transmitted data is encrypted and securely sent to the server. The input is the entry information entered by the user, and the output is the entry information securely sent to the server.

[1393] Step 3:

[1394] The server receives the entry information sent from the terminal. After receiving it, the server verifies the validity of the entry information, and if it is confirmed to be valid, it stores it in a database. The stored information includes the name of the generated AI model, a description, the path to the executable file, and the developer's contact information. The input is the received entry information, and the output is the entry information stored in the database.

[1395] Step 4:

[1396] The server notifies the user terminal that the entry has been completed. An entry completion message is displayed on the user terminal, allowing the user to confirm that the entry was successful. The input is the entry completion notification from the server, and the output is the entry completion message displayed on the user terminal.

[1397] Step 5:

[1398] At a pre-set time, the server will start a battle between the entered generative AI models. As the start time of the battle approaches, the server prepares the execution environments for all related generative AI models. The input is the battle start time, and the output is the execution environments for the prepared generative AI models.

[1399] Step 6:

[1400] The server provides each generative AI model with details of the challenge (e.g., input data and problem definition). Each generative AI model generates a solution based on this challenge. The input is the problem description from the server, and the output is the solution generated by the generative AI model.

[1401] Step 7:

[1402] The generative AI model outputs the results of problem solving and returns them to the server in JSON format. The server receives the returned results and temporarily stores them in a database. The input is the solution result of the generative AI model, and the output is the result stored in the database.

[1403] Step 8:

[1404] The server applies an algorithm to score the collected results based on evaluation criteria (e.g., solution accuracy, speed, resource consumption). The input is the collected results, and the output is a score assigned to each generative AI model.

[1405] Step 9:

[1406] The server determines the winner and stores the winner's information in a database. The input is the scored result, and the output is the information of the generative AI model that was determined as the winner.

[1407] Step 10:

[1408] The server creates a ranking based on the scoring results and saves it in a database. It then updates the public page with the latest ranking information. The input is the scoring results, and the output is the updated public page.

[1409] Step 11:

[1410] The user device is equipped with an emotion engine that analyzes emotions from the user's facial expressions and voice. While the user is viewing the battle results, the emotion engine analyzes the user's emotional state in real time. The input is the user's facial expressions and voice data, and the output is the analyzed emotional data.

[1411] Step 12:

[1412] The emotion engine sends the analyzed emotion data to the server. The input is the analyzed emotion data, and the output is the emotion data sent to the server.

[1413] Step 13:

[1414] The server receives the emotion data and stores it in a database. This data is used to improve the generative AI model and optimize battles. The input is emotion data, and the output is the emotion data stored in the database.

[1415] Step 14:

[1416] The server analyzes the accumulated emotional data and reflects it in improving the generative AI model, battle settings, and adjusting evaluation criteria. This improves the user experience of the entire system and the performance of the generative AI model. The input is the accumulated emotional data, and the output is the overall system performance improved through improvements and adjustments.

[1417] (Application example 2)

[1418] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1419] Conventional battle systems using generative models lacked evaluation and optimization that took user emotions into account, resulting in a limited user experience. Furthermore, because the performance of generative models was evaluated solely based on technical criteria, users were not provided with sufficient satisfaction. Furthermore, there was a lack of a mechanism for analyzing user emotion data in real time and using it as feedback.

[1420] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1421] In this invention, the server includes an entry means for AI developers to enter generative models, a battle means in which the entered generative models compete against each other in specific tasks, an evaluation and publication means for evaluating and publishing the battle results, an emotion engine that analyzes emotions from the user's facial expressions and voice, and an adjustment means for optimizing the performance of the generative model based on the user's emotion data. This enables evaluation and optimization that takes user emotions into consideration, providing a more satisfying user experience.

[1422] "Entry means" refers to the means by which an AI developer registers a generative model in the system, and includes a transmission means for inputting information about the generative model and sending it to a server, and a storage means for receiving the sent information and storing it in a database.

[1423] The "battle means" refers to a means by which entered generative models compete against each other over specific pre-set challenges, and includes a collection means for providing the challenges and collecting the solution results of each generative model, and an evaluation means for analyzing the collected results and determining a winner based on evaluation criteria.

[1424] The "evaluation and publication means" is a means for evaluating the battle results, scoring the generative models based on the scoring criteria, creating rankings, and storing and publishing them in a database.

[1425] The "emotion engine" is part of a system that analyzes emotions from the user's facial expressions and voice in real time and sends the analysis data to a server.

[1426] The "adjustment method" is a method for collecting and analyzing user emotional data, optimizing the performance of the generative model based on that data, and improving the user experience.

[1427] As a form for implementing this invention, a system using a smartphone application is constructed.

[1428] System Configuration

[1429] 1. Hardware and Software Configuration

[1430] User device: Smartphone

[1431] Emotion Engine: EmotionRecognizer

[1432] AI Battle System: AIBattle

[1433] Content Management System: ContentManager

[1434] Entry Method

[1435] Using a user device, an AI developer registers a generative model in the system. An entry form is provided, and the developer enters the name of the generative model, a description, an executable file, developer information, etc. The input information is packaged in JSON format or similar and sent to the server. The server then stores the submitted information in a database.

[1436] Battle Method

[1437] The server sets up battles in which the submitted generative models compete against each other for specific tasks. Each task is provided with detailed information, including user emotional data. The generative models generate solutions to the tasks and return the results to the server. The server analyzes the collected results and determines the winner based on evaluation criteria.

[1438] Evaluation and Publication Methods

[1439] The battle results are evaluated and the generative models are scored based on the scoring criteria. The rankings are created and saved in a database, and are made available to the general public by updating a public page.

[1440] Incorporating an emotion engine

[1441] The system incorporates an emotion engine that uses the smartphone's camera and microphone to analyze the user's facial expressions and voice. The analyzed emotion data is sent to a server in real time. The server accumulates the emotion data and makes adjustments to optimize the performance of the generative model based on the user's emotional responses.

[1442] Specific examples

[1443] If a user smiles while watching a drama, their emotional data is analyzed in real time and stored in a database. Based on this emotional data, the next time content is delivered, optimized content is provided, taking into account the user's emotional response.

[1444] Prompt Sentence Examples

[1445] For example, you might input the following prompt into a generative AI model:

[1446] "Collect emotional data when a user smiles while watching a drama, and generate a list of content to display next based on that."

[1447] In this way, the invention incorporates an emotion engine in addition to entry means, battle means, evaluation and publication means, making the generative model competition and evaluation process more user-friendly and utilizing emotion data to improve and optimize generative models.

[1448] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1449] Step 1:

[1450] Execution of entry method

[1451] The user enters information about the generative model into an input form on the device. Specifically, the user enters the name of the generative model, a description, an executable file, developer information, etc. The entered information is packaged in JSON format and sent to the server.

[1452] Input: Name of the generated model, description, executable file, developer information

[1453] Output: Generated model information packaged in JSON format

[1454] Specific operation: The user enters information into the form and presses the submit button to send the data to the server.

[1455] Step 2:

[1456] Saving entry information

[1457] The server receives the generated model information in JSON format sent from the user device and stores it in a database, including the name of the generated model, its description, the path to the executable file, and the developer's contact information.

[1458] Input: Generated model information in JSON format

[1459] Output: Generative model information stored in a database

[1460] Specific operation: The server receives the information and stores it in a database.

[1461] Step 3:

[1462] The start of the battle

[1463] The server will start a battle between the entered generative models at a pre-set time, providing specific challenges to the generative models.

[1464] Input: Generative model information stored in the database, task details

[1465] Output: A generative model that receives the task.

[1466] Specific operation: The server triggers the start of the battle and sends the details of the challenge to the generative model.

[1467] Step 4:

[1468] Collection of problem-solving results

[1469] The generative model generates a solution to the problem and returns the result to the server, which receives the result and temporarily stores it in a database.

[1470] Input: Results of solving the problem using the generative model

[1471] Output: Solution results stored in the database

[1472] Specific operation: The generative model executes the problem-solving process and returns the results to the server.

[1473] Step 5:

[1474] Evaluating the results

[1475] The server scores the collected solutions based on evaluation criteria, including solution accuracy, speed, and resource consumption.

[1476] Input: Solution results stored in the database

[1477] Output: Scoring results

[1478] Specific Operation: The server runs the evaluation algorithm and generates a score.

[1479] Step 6:

[1480] Determining the Winner

[1481] The server determines the best performing generative model as the winner based on the scoring results, and the winner's information is stored in a database.

[1482] Input: Scoring results

[1483] Output: Winner information

[1484] What happens: The server compares the scores and marks the generative model with the highest score as the winner.

[1485] Step 7:

[1486] Publication of ratings and rankings

[1487] The server creates a ranking based on the scoring results and updates the public page, allowing general users to view the latest battle results and the rankings of generative models.

[1488] Input: Scoring results

[1489] Output: Published rankings

[1490] Specific operation: The server updates the ranking page and publishes the latest information.

[1491] Step 8:

[1492] User sentiment analysis

[1493] When a user browses a public page, the smartphone's camera and microphone are used to analyze the user's facial expressions and voice. Emotional data is analyzed in real time and sent to the server.

[1494] Input: User's facial expression data, voice data

[1495] Output: Parsed emotion data

[1496] Specific operation: The smartphone runs the emotion analysis engine and sends the analysis data to the server.

[1497] Step 9:

[1498] Emotional Data Feedback

[1499] The server stores the received emotion data in a database and uses it to improve the generative model and optimize battles.

[1500] Input: Parsed emotion data

[1501] Output: Improved generative model performance

[1502] Specific operation: The server analyzes the emotion data and uses it as feedback to optimize the performance of the generative model.

[1503] In this way, through a series of processing steps, evaluation and optimization of the generative model taking into account the user's emotions is achieved.

[1504] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1505] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1506] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1507] [Fourth embodiment]

[1508] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1509] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1510] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1511] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1512] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1513] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1514] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1515] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1516] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1517] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1518] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1519] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1520] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1521] The present invention is a system for competing in problem-solving ability using generative AI, and is implemented in the following form.

[1522] Entry function

[1523] 1. User enters AI

[1524] Users use their device to enter their own generated AI. A form is displayed on the user device in which they can enter the name, description, executable file, developer information, etc. of the generated AI. Based on this information, they press the entry button to submit the information.

[1525] 2. The device sends the entry information to the server

[1526] The user terminal sends the entry information entered by the user to the server, where it is packaged in a standard format such as JSON.

[1527] 3. The server receives the entry information

[1528] The server receives the entry information sent from the user's device and stores it in a database, including the name of the generated AI, a description, the path to the executable file, and the developer's contact information.

[1529] 4. Notification of entry completion

[1530] The server notifies the user terminal that the entry has been completed successfully. An entry completion message is displayed on the user terminal, allowing the user to confirm that the entry has been completed successfully.

[1531] Battle Features

[1532] 1. The server starts the battle

[1533] The server will start a battle between the registered generated AIs at a pre-set time. At the start of the battle, specific tasks will be set for each generated AI.

[1534] 2. The server provides each AI with a task

[1535] The server provides the details of the challenge (e.g., input data and problem definition) to the submitted AI generators, who then generate a solution to the challenge.

[1536] 3. The server receives the results of each AI

[1537] Each generation AI outputs the solution to the problem and returns it to the server, which receives these results and temporarily stores them in a database.

[1538] 4. The server evaluates the results

[1539] The server evaluates the received results based on evaluation criteria (e.g., solution accuracy, speed, resource consumption, etc.) and assigns a score to each generated AI.

[1540] 5. The server decides the winner

[1541] Based on the evaluation results, the server determines the best performing AI generator as the winner, and the winner's information is stored in the database.

[1542] Result announcement and evaluation function

[1543] 1. The server evaluates the battle results

[1544] After the battle, the server will tally the scores of each AI generator and create a ranking, which is a fair evaluation of the AI ​​generator's ability to solve the problem.

[1545] 2. The server creates the rankings

[1546] The server creates a ranking of the AI ​​generators based on the collected evaluation results. This ranking is based on the performance of the AI ​​generators.

[1547] 3. The server stores the results in a database

[1548] The server saves the rankings it creates in a database, which also stores past battle results and the evaluation history of each generated AI.

[1549] 4. The server updates the public page

[1550] The server updates the public page based on the ranking information stored in the database, allowing general users to view the latest battle results and the rankings of the generated AI.

[1551] 5. User checks the results

[1552] Users can access the public page using their devices to check the battle results and the rankings of the generated AI, allowing general users to intuitively understand and experience the technological evolution of generated AI.

[1553] The above system provides AI developers with a forum where they can fairly compete and receive evaluations on the performance of their generative AI, and it also provides general consumers with a forum where they can intuitively understand and experience the technological evolution of generative AI.

[1554] The processing flow will be explained below.

[1555] Entry function

[1556] Step 1: User enters AI

[1557] The user enters information about their generated AI (name, description, executable file, developer information, etc.) into the user terminal and presses the entry button.

[1558] Step 2: The device sends the entry information to the server

[1559] The terminal sends the entry information entered by the user to the server in a standard format such as JSON.

[1560] Step 3: Server receives entry information

[1561] The server receives the entry information sent from the terminal.

[1562] Step 4: The server saves the entry information to a database

[1563] The server stores the received entry information in a database, including the name of the generated AI, its description, the path to the executable file, and developer information.

[1564] Step 5: The server notifies you that the entry is complete

[1565] The server notifies the user terminal that the entry has been successfully completed, and an entry completion message is displayed on the user terminal.

[1566] Battle Features

[1567] Step 1: The server starts the battle

[1568] The server will start a battle between the entered generated AIs at a pre-set time.

[1569] Step 2: The server provides each AI with a task

[1570] The server provides the details of the challenge (e.g., input data and problem definition) to the entered generated AI.

[1571] Step 3: The server receives the results of each AI.

[1572] Each generation AI generates a problem-solving result and returns it to the server, which receives the result and temporarily stores it in a database.

[1573] Step 4: The server evaluates the results

[1574] The server evaluates the received results based on criteria (e.g., accuracy, speed, resource consumption), and the evaluation results are scored.

[1575] Step 5: Server determines winner

[1576] Based on the evaluation results, the server determines the AI ​​generator with the best performance as the winner, and the winner's information is recorded in the database.

[1577] Result announcement and evaluation function

[1578] Step 1: The server evaluates the battle results

[1579] The server tally the scores of each generated AI and create a ranking based on the evaluation criteria.

[1580] Step 2: The server creates the rankings

[1581] The server uses the evaluation results to create a ranking of the generated AIs, which is ranked based on the performance of the generated AIs.

[1582] Step 3: The server stores the results in a database

[1583] The server saves the ranking information it creates in a database, which also stores past battle results and evaluation history.

[1584] Step 4: The server updates the public page

[1585] The server updates the public page with the latest information based on the rankings stored in the database.

[1586] Step 5: User confirms the results

[1587] Users can access the public page using their devices to check the battle results and the ranking of the generated AI, allowing them to intuitively understand and experience the technological evolution of the generated AI.

[1588] Example 1

[1589] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1590] Currently, many AI developers are developing generative AI, but there is a lack of venues for objectively evaluating the performance of these generative AIs and for them to compete against each other. This makes it difficult for generative AI developers to compare the capabilities of their own AI with those of other AIs and identify areas for improvement. It is also difficult for ordinary users to intuitively understand and experience the evolution and characteristics of generative AI. A new system is needed to solve this problem and promote technological advances in generative AI.

[1591] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1592] In this invention, the server includes a storage means for receiving entry information and storing it in a database, a management means for checking the start time of the battle and providing challenges to the AI ​​generators, and an evaluation means for evaluating the results of the challenges and determining the best AI generators. This allows the performance of AI generators to be fairly evaluated and allows developers to compete against each other.

[1593] "Entry means" refers to the means by which an AI developer enters a generated AI into the system, and includes information sent from the user terminal to the server.

[1594] "Battle means" refers to the means by which entered generated AIs compete against each other on specific challenges, and includes the function of providing challenges and collecting the results of their solutions.

[1595] "Means for evaluation and publication" refers to means for evaluating the results of battles and publishing the results, and includes functions for evaluating the generating AI based on evaluation criteria, storing the results in a database, and publishing them to the general public.

[1596] "Input means" refers to the means by which a user inputs and submits entry information for the generated AI, and refers to the input form provided on the user's terminal.

[1597] "Storage means" refers to the means by which the server stores the received entry information in a database, and includes temporary and permanent storage of data.

[1598] The "management means" is a means by which the server checks the start time of the battle and starts the battle at the set time, and includes a function for managing the progress of the battle.

[1599] "Task providing means" refers to the means by which the server provides specific tasks to the entered generation AI, and includes the generation and distribution of tasks.

[1600] "Evaluation means" refers to the means by which the server evaluates and scores the solution results of each generated AI based on the evaluation criteria, and includes analyzing the results and applying the evaluation criteria.

[1601] "Determination means" refers to the means by which the server determines the best generating AI and stores that information in a database, including determining the winner and storing the results.

[1602] "Server communication means" refers to the means by which the server provides challenges to each generation AI and each generation AI transmits a solution to the server, and includes establishing communication and sending and receiving data.

[1603] "Evaluation criteria setting means" means a means for the evaluation means to set criteria for evaluating each generative AI based on the accuracy, speed, and resource consumption of the solution, and includes the definition and application of the evaluation criteria.

[1604] This invention is a system for competing in problem-solving ability using generative AI, and it operates in cooperation with users, terminals, and a server. The system has the following functions.

[1605] Entry function

[1606] To enter a generated AI, a user first uses their device to enter the name, description, executable file path, and developer information into an entry form. Specific hardware used for this is a personal computer or smartphone, and the user accesses the form via a browser. A web browser (e.g., Google Chrome or Safari) is used as the software.

[1607] When the user presses the entry button, the terminal converts the entered information into JSON format and sends it to the server as an HTTP POST request. The server analyzes the received data and saves it in a database. This saving process uses the server's internal database system (e.g., MySQL, PostgreSQL). Once the entry is complete, the server sends a completion notification to the user's terminal. The user's terminal displays the message "Entry completed successfully."

[1608] As a specific example, a user enters a generation AI called "AI_Champion" and enters the description, "This is an AI that analyzes the emotions in English sentences."

[1609] Battle Features

[1610] The server starts a battle between the entered generated AIs at a pre-set time. It checks the start time of the battle and provides specific tasks to the entered generated AIs. For example, a task might be to "analyze emotions from English sentences and assign a score." The server packages and sends the task details to each generated AI.

[1611] Each generation AI generates a solution to the provided problem and sends the solution results to the server. The server temporarily stores the received solution results in a database. The server then evaluates and scores the results of each generation AI based on evaluation criteria (e.g., solution accuracy, speed, resource consumption, etc.). Various evaluation algorithms and evaluation criterion setting methods are used for the evaluation process.

[1612] As an example of specific operation, the server receives an emotion score of 0.95 assigned by "AI_Champion" and evaluates it as "AI_Champion's score is 95 points."

[1613] Result announcement and evaluation function

[1614] After the battle is over, the server tally the scores of each generated AI and create a ranking. This ranking is based on the performance of the generated AI and is stored in a database. The server updates a public page based on the ranking information, allowing general users to view the latest battle results and the rankings of the generated AI. Users can access the public page using their devices to check the results.

[1615] As a specific example, a user accesses a public page and sees the information "AI_Champion is in first place."

[1616] Prompt Sentence Examples

[1617] "Create a generative AI that takes an English sentence as input, analyzes its sentiment, and gives it a score."

[1618] This invention provides AI developers with a forum where they can fairly compete and have the performance of their generative AI evaluated, and it also provides general users with a forum where they can intuitively understand and experience the technological evolution of generative AI.

[1619] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1620] Entry function

[1621] Step 1: Display the entry form

[1622] The user opens the entry page in a browser and enters the name, description, executable file path, and developer information of the generated AI. HTML and JavaScript are used to display the required fields in this form.

[1623] Input: User-entered name, description, executable path, and developer information for the generated AI.

[1624] Output: The input form reflects the user's input.

[1625] Step 2: Prepare input information for submission

[1626] When the user clicks the entry button, the terminal converts the input information into JSON format and packages it as an HTTP POST request.

[1627] Enter: When the user clicks the enter button.

[1628] Output: Data package converted to JSON format.

[1629] Step 3: Submit your entry

[1630] The terminal sends the packaged entry information to the server, and this communication uses the HTTP protocol.

[1631] Input: A data package in JSON format.

[1632] Output: The HTTP POST request sent to the server.

[1633] Step 4: Receiving and analyzing data

[1634] The server receives the HTTP request and parses the JSON data to extract the name, description, executable path, and developer information of the generated AI.

[1635] Input: HTTP POST request.

[1636] Output: Extracted entry information.

[1637] Step 5: Save your data

[1638] The server stores the extracted entry information in a database using SQL queries.

[1639] Input: The extracted entry information.

[1640] Output: Entry information stored in a database.

[1641] Step 6: Notification of completed entry

[1642] The server generates a response to notify the user that the entry has been successfully completed, and sends it to the user terminal.

[1643] Input: Notification that saving is complete.

[1644] Output: Entry complete message as an HTTP response.

[1645] Step 7: Displaying the entry completion message

[1646] The user terminal receives the response from the server and displays the message "Entry completed successfully" in the browser.

[1647] Input: The response from the server.

[1648] Output: Entry successful message displayed in browser.

[1649] Battle Features

[1650] Step 1: Check the battle start time

[1651] The server checks the pre-set battle start time and sees if that time has come, using a timer or scheduling function.

[1652] Input: Pre-set battle start time.

[1653] Output: Trigger to start the battle.

[1654] Step 2: Prepare for assignment delivery

[1655] The server generates and packages challenges to initiate battles between AIs, including problem definitions and input data.

[1656] Input: Trigger to start the battle.

[1657] Output: Packaged assignment content.

[1658] Step 3: Submit your assignment

[1659] The server sends the packaged tasks to each generating AI, and this communication may use HTTP or WebSocket.

[1660] Input: Packaged assignment content.

[1661] Output: The challenges sent to each generated AI.

[1662] Step 4: Generate solutions

[1663] Each generative AI generates a solution to a problem, which is achieved by the AI ​​model performing internal data calculations and analysis.

[1664] Input: Received assignment content.

[1665] Output: The generated solution.

[1666] Step 5: Submit your solution

[1667] Each generating AI returns a solution to the server, again using the HTTP protocol for this communication.

[1668] Input: The generated solution.

[1669] Output: The solution sent to the server.

[1670] Step 6: Receive and save the solution

[1671] The server temporarily stores the received solutions in a database using SQL queries.

[1672] Input: The solutions submitted by each generating AI.

[1673] Output: The solution stored in the database.

[1674] Step 7: Evaluate the solution

[1675] The server evaluates the saved solutions based on evaluation criteria and assigns a score to each generated AI, including solution accuracy, speed, and resource consumption.

[1676] Input: Solutions stored in the database.

[1677] Output: The evaluated score.

[1678] Step 8: Determine the winner

[1679] The server will determine the generator AI that performed best based on the evaluation results as the winner.

[1680] Input: The assessed score.

[1681] Output: Determining the winner and storing that information in a database.

[1682] Result announcement and evaluation function

[1683] Step 1: Counting scores

[1684] The server compiles the scores of each generated AI and creates a ranking, which is done by processing and calculating the score data.

[1685] Input: The score of each generated AI.

[1686] Output: Aggregated scores and rankings.

[1687] Step 2: Save ranking information

[1688] The server stores the rankings in a database.

[1689] Input: Aggregated scores and rankings.

[1690] Output: Ranking information stored in a database.

[1691] Step 3: Update your public page

[1692] The server updates the public page based on the ranking information stored in the database, using a script on the web server.

[1693] Input: Ranking information stored in the database.

[1694] Output: The updated public page.

[1695] Step 4: User confirmation of results

[1696] Users can access the public page using their devices to check the battle results and the ranking of the generated AI. This operation is performed using a web browser.

[1697] Input: Access to public pages.

[1698] Output: Battle results and ranking information displayed on the user's device.

[1699] The above is a detailed explanation of each processing step.

[1700] (Application example 1)

[1701] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1702] Current generative AI competition systems allow AI developers to enter their generative AIs, battle them, and evaluate and publish the results. However, they lack the functionality to allow general users to watch generative AI battles in real time or view past battle results, making it difficult to provide opportunities for users to deepen their interest and understanding. For this reason, there is a need for a system that allows users to intuitively understand and experience the technological evolution of generative AI.

[1703] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1704] In this invention, the server includes an entry means for AI developers to enter generated AIs, a battle means in which the entered generated AIs compete against each other in specific tasks, an evaluation and publication means for evaluating and publishing the battle results, a real-time distribution means for distributing the battle results in real time, an archive viewing means for selecting and viewing past battle results from an archive, and a user evaluation means for users to evaluate the battle results and enter comments. This enables general users to intuitively understand and experience the technological evolution of generated AIs by watching battles between generated AIs in real time and viewing past battle results from the archive.

[1705] "Entry means" refers to a device, method, or system that allows an AI developer to enter a generated AI.

[1706] "Battle means" refers to a device, method, or system that allows entered generated AIs to compete against each other in a specific challenge.

[1707] The "evaluation and disclosure means" refers to a device, method, or system for evaluating the battle results and disclosing the results to users.

[1708] "Real-time distribution means" refers to a device, method, or system for distributing the battle results of the generated AI to users in real time.

[1709] "Archive viewing means" refers to a device, method, or system that allows a user to select past generated AI battle results from a database and view them.

[1710] The "user evaluation means" refers to a device, method, or system that allows users to evaluate the battle results and input comments.

[1711] "Transmission means" refers to a device, method, or system for transmitting information about the generated AI from a user terminal to a server.

[1712] "Storage means" means a device, method, or system for receiving and storing transmitted information in a database.

[1713] "Collection means" refers to a device, method, or system for providing challenges to the entered generation AI and collecting the results of its solutions.

[1714] "Evaluation means" refers to a device, method, or system for analyzing the collected solution results of the generative AI and determining a winner based on evaluation criteria.

[1715] This invention is a system for competing in problem-solving ability using generative AI, and includes as its components an entry means, a battle means, an evaluation and publication means, a real-time distribution means, an archive viewing means, and a user evaluation means. How this system is specifically implemented is explained below.

[1716] System configuration

[1717] The main components are a server, user devices (smartphones, smart glasses, head-mounted displays (HMDs)), and a database. The server uses a cloud service (AWS) and the database is MySQL. React Native is used for the front end, and Node.js and Express.js are used for the back end.

[1718] Entry Method

[1719] Using a user terminal, the AI ​​developer enters the generated AI. At this time, a form is displayed in which the name, description, executable file, developer information, etc. of the generated AI can be entered, and the information is submitted by pressing the "entry" button. The submitted information is received by the server, packaged in JSON format, etc., and stored in a database.

[1720] Battle Method

[1721] The server starts a battle between the AI ​​generators at a pre-set date and time. The AI ​​generators are given specific challenges, and each AI generates a solution to the challenge. The solutions output by the AI ​​generators are sent to the server, where they are analyzed and evaluated based on the evaluation criteria.

[1722] Evaluation and Publication Methods

[1723] The server evaluates the solutions of each AI generator and creates a ranking based on the results, which is then updated on a public page where users can access the rankings.

[1724] Real-time delivery methods

[1725] The server uses WebSocket to deliver the battle results of the generated AI in real time. The user device is built with React Native and receives and displays the real-time data, allowing users to watch the progress of the generated AI battle in real time.

[1726] Archive viewing method

[1727] The results of past generated AI battles are stored in a database and can be viewed by users using the archive function. The data is retrieved using a REST API and displayed on the user's device.

[1728] User evaluation method

[1729] Users can rate and comment on battle results. The entered ratings and comments are sent from the front end to the server and stored in a database, allowing other users to view the feedback.

[1730] Specific examples

[1731] For example, the user can operate the system by entering the following prompts:

[1732] Example viewing prompt for a generated AI battle:

[1733] "Watch the next generative AI battle: Battle ID 12345. Challenge: Natural Language Processing."

[1734] Example of an archive search prompt:

[1735] "View past generative AI battle results: Check how generative AI performed on the challenge."

[1736] In this way, ordinary users will be able to intuitively understand and experience the technological evolution of generative AI.

[1737] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1738] Step 1:

[1739] The user inputs information into the terminal to enter the generated AI. The input information includes the name of the generated AI, a description, an executable file, developer information, etc. The terminal collects this information, packages it in JSON format, etc., and sends it to the server. The input of the data sent to the server is the generated AI information, and the output is the packaged entry information.

[1740] Step 2:

[1741] The server receives the entry information sent by the user. The received information is stored in a database. The database stores the name of the generated AI, a description, the path to the executable file, and the developer's contact information. The input is the entry information sent from the terminal, and the output is the entry information stored in the database.

[1742] Step 3:

[1743] The server will start a battle between the entered generated AIs at a pre-set date and time. Each generated AI will be provided with a specific task. The task details include input data and a problem definition, which are sent to each generated AI. The input is the generated AI's entry information and task definition, and the output is the task provided to the generated AI.

[1744] Step 4:

[1745] The generation AI generates a solution to the provided challenge. The generated solution is sent to the server. The input is the challenge provided by the server, and the output is the generation AI's solution.

[1746] Step 5:

[1747] The server receives the solution sent from the generation AI and temporarily stores it in a database. The input is the solution sent from the generation AI, and the output is the solution temporarily stored in the database.

[1748] Step 6:

[1749] The server analyzes the collected solutions of the generative AI and assigns a score based on evaluation criteria, such as solution accuracy, speed, and resource consumption. The input is the solution stored in the database, and the output is the evaluated score.

[1750] Step 7:

[1751] The server creates a ranking of the generative AI based on the evaluation results and stores it in a database. The input is the evaluated score, and the output is the ranking information.

[1752] Step 8:

[1753] The server distributes the battle results of the generated AI in real time. It uses WebSocket to send data sequentially to the user's device. The input is the evaluated score and the progress of the solution, and the output is the battle result data distributed in real time.

[1754] Step 9:

[1755] The user watches the progress of the generated AI battle in real time. The device receives data from the server and reflects it on the UI. The input is real-time data sent from the server, and the output is the progress of the battle displayed on the device.

[1756] Step 10:

[1757] Users can view past generated AI battle results using the archive function. The device sends a request for archive data to the server, which retrieves the relevant data from the database and sends it to the device. The input is the archive request from the user, and the output is the past battle results displayed on the device.

[1758] Step 11:

[1759] Users rate and comment on the battle results. The terminal sends the entered ratings and comments to the server, which stores them in a database. The input is the user's rating and comments, and the output is the rating and comments stored in the database.

[1760] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1761] This invention is a system that uses generative AI to compete in problem-solving ability, and combines it with an emotion engine that recognizes users' emotions. This system consists of the following elements: entry means, battle means, evaluation and publication means, and the emotion engine.

[1762] Entry function

[1763] Entry

[1764] Users use their device to enter their own generated AI. A form is displayed on the user device in which they can enter the name, description, executable file, developer information, etc. of the generated AI. Based on this information, they press the entry button to submit the information.

[1765] send

[1766] The user terminal sends the entry information entered by the user to the server, where it is packaged in a standard format such as JSON.

[1767] Receive and store

[1768] The server receives the entry information sent from the device and stores it in a database, including the name of the generated AI, a description, the path to the executable file, and the developer's contact information.

[1769] Completion notification

[1770] The server notifies the user terminal that the entry has been successfully completed, and an entry completion message is displayed on the user terminal.

[1771] Battle Features

[1772] Battle Begins

[1773] The server will start a battle between the registered generated AIs at a pre-set time, with specific challenges set for each generated AI.

[1774] Provide assignments

[1775] The server provides the generative AIs with the details of the challenge (e.g., input data and problem definition), and each generative AI generates a solution to this challenge.

[1776] Result collection

[1777] The generative AI outputs the results of the problem solving and returns them to the server, which temporarily stores these results in a database.

[1778] Results evaluation

[1779] The server evaluates and scores the results based on evaluation criteria (e.g., solution accuracy, speed, resource consumption).

[1780] Winner Decided

[1781] Based on the evaluation results, the server determines the best performing AI generator as the winner, and the winner's information is stored in the database.

[1782] Rating and publishing features

[1783] Evaluation and Scoring

[1784] The server evaluates the battle results and assigns a score to each generated AI based on the scoring criteria. A ranking is created based on the scoring results.

[1785] Save and publish results

[1786] The server saves the rankings in a database and updates a public page where users can view the latest battle results and the rankings of the generated AI.

[1787] Incorporating an emotion engine

[1788] emotion recognition

[1789] The user device is equipped with an emotion engine that analyzes the user's emotions from their facial expressions and voice. This emotion engine analyzes the user's emotions in real time as they view the battle results.

[1790] Sending emotional data

[1791] The emotion engine sends the analyzed emotion data to the server, which includes the user's emotional state (e.g., joy, surprise, frustration, etc.).

[1792] Collecting and storing emotional data

[1793] The server stores the transmitted emotion data in a database and uses it to improve the AI ​​generator and optimize battles. This emotion data can be used to provide feedback on the AI ​​generator's performance and user experience.

[1794] Feedback and Adjustments

[1795] The accumulated emotional data is analyzed and used to improve the generation AI, battle settings, and evaluation criteria, thereby improving the overall user experience of the system and the performance of the generation AI.

[1796] For example, if the emotion engine detects a smile when a user views the results of a battle between the generated AI, that data is sent to the server and stored in a database. This data is used in the feedback process of the generated AI and acts as part of its optimization. This allows the generated AI to compete more effectively in the next battle, taking into account the user's emotional reactions.

[1797] In this way, by incorporating an emotion engine in addition to entry means, battle means, evaluation and publication means, the present invention makes the generative AI competition and evaluation process more user-friendly and utilizes emotion data to improve and optimize the generative AI.

[1798] The processing flow will be explained below.

[1799] Entry function

[1800] Step 1: User enters AI

[1801] The user enters information about their generated AI (name, description, executable file, developer information, etc.) into the user terminal and presses the entry button.

[1802] Step 2: The device sends the entry information to the server

[1803] The terminal sends the entry information entered by the user to the server in a standard format such as JSON.

[1804] Step 3: The server receives the entry information

[1805] The server receives the entry information sent from the terminal.

[1806] Step 4: The server saves the entry information to a database

[1807] The server stores the received entry information in a database, including the name of the generated AI, its description, the path to the executable file, and developer information.

[1808] Step 5: The server notifies you that the entry is complete

[1809] The server notifies the user terminal that the entry has been successfully completed, and an entry completion message is displayed on the user terminal.

[1810] Battle Features

[1811] Step 1: The server starts the battle

[1812] The server will start a battle between the entered generated AIs at a pre-set time.

[1813] Step 2: The server provides each AI with a task

[1814] The server provides the details of the challenge (e.g., input data and problem definition) to the entered generated AI.

[1815] Step 3: Generative AI solves the problem

[1816] The generative AI generates a solution to the problem and returns the result to the server.

[1817] Step 4: The server receives the results of each AI.

[1818] The server receives the problem-solving results returned by the generation AI and temporarily stores them in a database.

[1819] Step 5: The server evaluates the results

[1820] The server evaluates and scores the received results based on criteria (e.g., accuracy, speed, resource consumption, etc.).

[1821] Step 6: Server determines winner

[1822] Based on the evaluation results, the server determines the best performing AI generator as the winner, and the winner's information is stored in a database.

[1823] Rating and publishing features

[1824] Step 1: The server evaluates the battle results

[1825] The server tally the scores of each generated AI and create a ranking based on the evaluation criteria.

[1826] Step 2: The server creates the rankings

[1827] The server will create a ranking of the AI ​​generators based on the evaluation results. This ranking will be ranked based on the performance of the AI ​​generators.

[1828] Step 3: The server stores the results in a database

[1829] The server saves the ranking information it creates in a database, which also stores past battle results and evaluation history.

[1830] Step 4: Server updates public page

[1831] The server updates the public page with the latest information based on the ranking information stored in the database.

[1832] Step 5: User confirms the results

[1833] Users can access the public page using their devices to check the battle results and the ranking of the generated AI, allowing them to intuitively understand and experience the technological evolution of the generated AI.

[1834] Incorporating an emotion engine

[1835] Step 1: User visits public page

[1836] Users use their devices to access a public page that displays the battle results and rankings of the generated AI.

[1837] Step 2: The emotion engine analyzes the user's emotions

[1838] The emotion engine is built into the user's device and analyzes the user's facial expressions and voice in real time, for example, tracking the user's facial expressions through a camera and analyzing the user's tone of voice through a microphone.

[1839] Step 3: Send emotion data to the server

[1840] The emotion engine sends the analyzed emotion data to the server, which includes the user's emotional state (e.g., joy, surprise, dissatisfaction, etc.).

[1841] Step 4: The server receives the emotion data

[1842] The server receives the emotion data sent from the user terminal.

[1843] Step 5: The server stores the emotion data in a database

[1844] The server stores the received emotional data in a database, which includes the user's emotional response to each battle outcome.

[1845] Step 6: Analyze the sentiment data

[1846] The server analyzes the accumulated emotional data and uses it to improve the generation AI and optimize battles.

[1847] Step 7: Feedback and Adjustments

[1848] Feedback based on emotional data is provided to the developers of the AI ​​generator, allowing them to make adjustments to improve the AI's performance and user experience. For example, if a user is satisfied with the outcome of a particular battle, the AI ​​generator will use that strategy or approach in the next battle.

[1849] In this way, by incorporating an emotion engine in addition to entry means, battle means, evaluation and publication means, the present invention makes the generative AI competition and evaluation process more user-friendly and utilizes emotion data to improve and optimize the generative AI.

[1850] Example 2

[1851] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1852] Conventional generative algorithm competition systems did not take into account user emotional data when optimizing or evaluating generative algorithms, and therefore were unable to sufficiently improve the user experience. Furthermore, the evaluation criteria for generative algorithms were limited and lacked diversity. Furthermore, the publication and evaluation of battle results were not centrally managed, making it difficult for user feedback to be reflected in real time.

[1853] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a transmission means for transmitting information about the generation algorithm from the user terminal to the central processing unit, a storage means for receiving the transmitted information and storing it in a storage device, a provision means for providing challenges to the entered generation algorithms, a collection means for collecting the solution results of each generation algorithm, an evaluation means for analyzing the collected results and determining a winner based on evaluation criteria, an emotion engine means for recognizing the user's emotions, and a data collection and analysis means for collecting emotion data obtained by the emotion engine means and using it to optimize the system. This makes it possible to evaluate and optimize generation algorithms that reflect the user's emotion data, thereby improving the user experience and enabling diverse evaluation of generation algorithms.

[1854] A "generative algorithm" refers to a series of computational steps for generating new data or information from existing data.

[1855] "Entry Method" refers to the method or process by which a Generator Algorithm is registered with the System.

[1856] "Competition method" refers to the method or process by which the entered generative algorithms compete to provide solutions to a given challenge.

[1857] "Evaluation and disclosure measures" refers to the methods and processes for evaluating the results of the competition and disclosing those evaluation results to the public.

[1858] "Emotion engine means" refers to technology or tools for analyzing emotions from a user's facial expressions and voice.

[1859] "Data collection and analysis means" refers to the method or process of collecting data obtained by the emotion engine means and using it to optimize the system.

[1860] "Transmission means" refers to a method or process for transmitting information of the generation algorithm from the user terminal to the central processing unit.

[1861] "Storage" refers to the method or process for receiving transmitted information and storing it in a storage device.

[1862] "Provision means" refers to the method or process for providing a challenge to the entered generation algorithm.

[1863] "Collection means" refers to the method or process for collecting the solution results of each generation algorithm.

[1864] "Evaluation means" refers to a method or process for analyzing the collected solution results of the generative algorithms and determining a winner based on evaluation criteria.

[1865] "Central Processing Unit" refers to a central computer that performs multiple calculations.

[1866] "Storage" refers to a device or medium for storing information.

[1867] The present invention is a competition system using a generative AI model, which combines an entry means, a competition means, an evaluation and publication means, an emotion engine means, and a data collection and analysis means. In this system, the user, the terminal, and the server each play different roles. Specific embodiments of the system are described below.

[1868] Hardware and software used

[1869] This invention is composed of various hardware and software, including a user terminal, a server, a database, and an emotion engine. The user terminal is equipped with an emotion engine (e.g., facial recognition software, voice recognition software) for analyzing emotions from the user's facial expressions and voice. The server is a central processing unit that processes information sent from the user terminal and stores the information in a database (e.g., MySQL).

[1870] Entry function

[1871] Users use their device to enter their own generative AI model. A form is displayed on the user device to enter the name, description, executable file, developer information, etc. of the generative AI model. When the user enters the information and clicks the entry button, the entered information is converted to JSON format and sent to the server. The server receives the submitted information, verifies its validity, and stores it in a database. The information is encrypted before being stored and is kept securely.

[1872] Battle Features

[1873] At a pre-set time, the server will start a battle between the entered generative AI models. As the start time approaches, the server will prepare the execution environment for all relevant generative AI models. The server will provide each generative AI model with details of the challenge (e.g., input data and problem definition). Each generative AI model will generate a solution based on this challenge and return the results to the server in JSON format. The server will receive these results and temporarily store them in a database.

[1874] Rating and publishing features

[1875] The server applies an algorithm to score the collected results based on evaluation criteria (e.g., solution accuracy, speed, resource consumption). The resulting score is then assigned to the generative AI model. Furthermore, the server creates a ranking based on the scoring results and stores it in a database. It updates a public page with the latest ranking information, allowing the public to view the latest battle results through this public page.

[1876] Incorporating an emotion engine

[1877] The user device is equipped with an emotion engine that analyzes emotions from the user's facial expressions and voice. While the user is viewing the battle results, the emotion engine analyzes the user's emotional state (e.g., joy, surprise, dissatisfaction, etc.) in real time. The analyzed emotion data is sent to the server and stored in a database. This data is used to improve the generative AI model and optimize battles. By analyzing the accumulated emotion data and reflecting it in improving the generative AI model, battle settings, and adjusting evaluation criteria, the overall user experience of the system and the performance of the generative AI model are improved.

[1878] Specific examples

[1879] For example, when a user is viewing the results of a battle between generative AI models, the emotion engine will detect the user's smile. This data will be sent to the server and stored in the database. This data will then be analyzed and used in the generative AI model's feedback process. As a result, the next battle will take the user's emotional reactions into account, resulting in a more effective generative AI competition.

[1880] Prompt Sentence Examples

[1881] Below are some example prompts that can be used as input to a generative AI model:

[1882] Based on the reaction data when the user smiles, please suggest ways to optimize the performance of the generative AI.

[1883] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1884] Step 1:

[1885] The user uses a user terminal to enter their own generative AI model. A form is displayed on the user terminal in which the user can enter the name, description, executable file, developer information, etc. of the generative AI model. The user enters the required information in these fields and clicks the entry button. The input data is processed as metadata for the generative AI model and converted to JSON format. The output is the entry information (JSON format) that is sent to the server.

[1886] Step 2:

[1887] The terminal sends the entry information entered by the user to the server in JSON format. The transmitted data is encrypted and securely sent to the server. The input is the entry information entered by the user, and the output is the entry information securely sent to the server.

[1888] Step 3:

[1889] The server receives the entry information sent from the terminal. After receiving it, the server verifies the validity of the entry information, and if it is confirmed to be valid, it stores it in a database. The stored information includes the name of the generated AI model, a description, the path to the executable file, and the developer's contact information. The input is the received entry information, and the output is the entry information stored in the database.

[1890] Step 4:

[1891] The server notifies the user terminal that the entry has been completed. An entry completion message is displayed on the user terminal, allowing the user to confirm that the entry was successful. The input is the entry completion notification from the server, and the output is the entry completion message displayed on the user terminal.

[1892] Step 5:

[1893] At a pre-set time, the server will start a battle between the entered generative AI models. As the start time of the battle approaches, the server prepares the execution environments for all related generative AI models. The input is the battle start time, and the output is the execution environments for the prepared generative AI models.

[1894] Step 6:

[1895] The server provides each generative AI model with details of the challenge (e.g., input data and problem definition). Each generative AI model generates a solution based on this challenge. The input is the problem description from the server, and the output is the solution generated by the generative AI model.

[1896] Step 7:

[1897] The generative AI model outputs the results of problem solving and returns them to the server in JSON format. The server receives the returned results and temporarily stores them in a database. The input is the solution result of the generative AI model, and the output is the result stored in the database.

[1898] Step 8:

[1899] The server applies an algorithm to score the collected results based on evaluation criteria (e.g., solution accuracy, speed, resource consumption). The input is the collected results, and the output is a score assigned to each generative AI model.

[1900] Step 9:

[1901] The server determines the winner and stores the winner's information in a database. The input is the scored result, and the output is the information of the generative AI model that was determined as the winner.

[1902] Step 10:

[1903] The server creates a ranking based on the scoring results and saves it in a database. It then updates the public page with the latest ranking information. The input is the scoring results, and the output is the updated public page.

[1904] Step 11:

[1905] The user device is equipped with an emotion engine that analyzes emotions from the user's facial expressions and voice. While the user is viewing the battle results, the emotion engine analyzes the user's emotional state in real time. The input is the user's facial expressions and voice data, and the output is the analyzed emotional data.

[1906] Step 12:

[1907] The emotion engine sends the analyzed emotion data to the server. The input is the analyzed emotion data, and the output is the emotion data sent to the server.

[1908] Step 13:

[1909] The server receives the emotion data and stores it in a database. This data is used to improve the generative AI model and optimize battles. The input is emotion data, and the output is the emotion data stored in the database.

[1910] Step 14:

[1911] The server analyzes the accumulated emotional data and reflects it in improving the generative AI model, battle settings, and adjusting evaluation criteria. This improves the user experience of the entire system and the performance of the generative AI model. The input is the accumulated emotional data, and the output is the overall system performance improved through improvements and adjustments.

[1912] (Application example 2)

[1913] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1914] Conventional battle systems using generative models lacked evaluation and optimization that took user emotions into account, resulting in a limited user experience. Furthermore, because the performance of generative models was evaluated solely based on technical criteria, users were not provided with sufficient satisfaction. Furthermore, there was a lack of a mechanism for analyzing user emotion data in real time and using it as feedback.

[1915] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1916] In this invention, the server includes an entry means for AI developers to enter generative models, a battle means in which the entered generative models compete against each other in specific tasks, an evaluation and publication means for evaluating and publishing the battle results, an emotion engine that analyzes emotions from the user's facial expressions and voice, and an adjustment means for optimizing the performance of the generative model based on the user's emotion data. This enables evaluation and optimization that takes user emotions into consideration, providing a more satisfying user experience.

[1917] "Entry means" refers to the means by which an AI developer registers a generative model in the system, and includes a transmission means for inputting information about the generative model and sending it to a server, and a storage means for receiving the sent information and storing it in a database.

[1918] The "battle means" refers to a means by which entered generative models compete against each other over specific pre-set challenges, and includes a collection means for providing the challenges and collecting the solution results of each generative model, and an evaluation means for analyzing the collected results and determining a winner based on evaluation criteria.

[1919] The "evaluation and publication means" is a means for evaluating the battle results, scoring the generative models based on the scoring criteria, creating rankings, and storing and publishing them in a database.

[1920] The "emotion engine" is part of a system that analyzes emotions from the user's facial expressions and voice in real time and sends the analysis data to a server.

[1921] The "adjustment method" is a method for collecting and analyzing user emotional data, optimizing the performance of the generative model based on that data, and improving the user experience.

[1922] As a form for implementing this invention, a system using a smartphone application is constructed.

[1923] System Configuration

[1924] 1. Hardware and Software Configuration

[1925] User device: Smartphone

[1926] Emotion Engine: EmotionRecognizer

[1927] AI Battle System: AIBattle

[1928] Content Management System: ContentManager

[1929] Entry Method

[1930] Using a user device, an AI developer registers a generative model in the system. An entry form is provided, and the developer enters the name of the generative model, a description, an executable file, developer information, etc. The input information is packaged in JSON format or similar and sent to the server. The server then stores the submitted information in a database.

[1931] Battle Method

[1932] The server sets up battles in which the submitted generative models compete against each other for specific tasks. Each task is provided with detailed information, including user emotional data. The generative models generate solutions to the tasks and return the results to the server. The server analyzes the collected results and determines the winner based on evaluation criteria.

[1933] Evaluation and Publication Methods

[1934] The battle results are evaluated and the generative models are scored based on the scoring criteria. The rankings are created and saved in a database, and are made available to the general public by updating a public page.

[1935] Incorporating an emotion engine

[1936] The system incorporates an emotion engine that uses the smartphone's camera and microphone to analyze the user's facial expressions and voice. The analyzed emotion data is sent to a server in real time. The server accumulates the emotion data and makes adjustments to optimize the performance of the generative model based on the user's emotional responses.

[1937] Specific examples

[1938] If a user smiles while watching a drama, their emotional data is analyzed in real time and stored in a database. Based on this emotional data, the next time content is delivered, optimized content is provided, taking into account the user's emotional response.

[1939] Prompt Sentence Examples

[1940] For example, you might input the following prompt into a generative AI model:

[1941] "Collect emotional data when a user smiles while watching a drama, and generate a list of content to display next based on that."

[1942] In this way, the invention incorporates an emotion engine in addition to entry means, battle means, evaluation and publication means, making the generative model competition and evaluation process more user-friendly and utilizing emotion data to improve and optimize generative models.

[1943] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1944] Step 1:

[1945] Execution of entry method

[1946] The user enters information about the generative model into an input form on the device. Specifically, the user enters the name of the generative model, a description, an executable file, developer information, etc. The entered information is packaged in JSON format and sent to the server.

[1947] Input: Name of the generated model, description, executable file, developer information

[1948] Output: Generated model information packaged in JSON format

[1949] Specific operation: The user enters information into the form and presses the submit button to send the data to the server.

[1950] Step 2:

[1951] Saving entry information

[1952] The server receives the generated model information in JSON format sent from the user device and stores it in a database, including the name of the generated model, its description, the path to the executable file, and the developer's contact information.

[1953] Input: Generated model information in JSON format

[1954] Output: Generative model information stored in a database

[1955] Specific operation: The server receives the information and stores it in a database.

[1956] Step 3:

[1957] The start of the battle

[1958] The server will start a battle between the entered generative models at a pre-set time, providing specific challenges to the generative models.

[1959] Input: Generative model information stored in the database, task details

[1960] Output: A generative model that receives the task.

[1961] Specific operation: The server triggers the start of the battle and sends the details of the challenge to the generative model.

[1962] Step 4:

[1963] Collection of problem-solving results

[1964] The generative model generates a solution to the problem and returns the result to the server, which receives the result and temporarily stores it in a database.

[1965] Input: Results of solving the problem using the generative model

[1966] Output: Solution results stored in the database

[1967] Specific operation: The generative model executes the problem-solving process and returns the results to the server.

[1968] Step 5:

[1969] Evaluating the results

[1970] The server scores the collected solutions based on evaluation criteria, including solution accuracy, speed, and resource consumption.

[1971] Input: Solution results stored in the database

[1972] Output: Scoring results

[1973] Specific Operation: The server runs the evaluation algorithm and generates a score.

[1974] Step 6:

[1975] Determining the Winner

[1976] The server determines the best performing generative model as the winner based on the scoring results, and the winner's information is stored in a database.

[1977] Input: Scoring results

[1978] Output: Winner information

[1979] What happens: The server compares the scores and marks the generative model with the highest score as the winner.

[1980] Step 7:

[1981] Publication of ratings and rankings

[1982] The server creates a ranking based on the scoring results and updates the public page, allowing general users to view the latest battle results and the rankings of generative models.

[1983] Input: Scoring results

[1984] Output: Published rankings

[1985] Specific operation: The server updates the ranking page and publishes the latest information.

[1986] Step 8:

[1987] User sentiment analysis

[1988] When a user browses a public page, the smartphone's camera and microphone are used to analyze the user's facial expressions and voice. Emotional data is analyzed in real time and sent to the server.

[1989] Input: User's facial expression data, voice data

[1990] Output: Parsed emotion data

[1991] Specific operation: The smartphone runs the emotion analysis engine and sends the analysis data to the server.

[1992] Step 9:

[1993] Emotional Data Feedback

[1994] The server stores the received emotion data in a database and uses it to improve the generative model and optimize battles.

[1995] Input: Parsed emotion data

[1996] Output: Improved generative model performance

[1997] Specific operation: The server analyzes the emotion data and uses it as feedback to optimize the performance of the generative model.

[1998] In this way, through a series of processing steps, evaluation and optimization of the generative model taking into account the user's emotions is achieved.

[1999] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[2000] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[2001] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[2002] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[2003] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[2004] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[2005] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[2006] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[2007] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[2008] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[2009] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[2010] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[2011] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[2012] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[2013] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[2014] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[2015] The hardware resource that executes the specific processing ...

Claims

1. An entry method for AI developers to enter generated AI; A battle method in which the entered generated AIs compete against each other on specific tasks, A system including an evaluation and publication means for evaluating and publishing battle results.

2. The entry means includes a transmission means for transmitting information about the generated AI from the user terminal to the server; 2. The system of claim 1, further comprising storage means for receiving the transmitted information and storing it in a database.

3. The battle means provides a challenge to the entered generated AI, A collection means for collecting the solution results of each generation AI; 10. The system of claim 1, further comprising evaluation means for analyzing the collected results and determining a winner based on evaluation criteria.

4. The evaluation and publication means includes a scoring means for scoring battle results and creating rankings; 2. The system of claim 1, further comprising publishing means for storing the rankings in a database and updating the publication page.

Images