System

Abstract

A system is provided.SOLUTION: A system, comprising: means for collecting previously submitted ideas; means for loading a generative AI model and setting evaluation criteria; means for inputting the collected ideas into the generative AI model and scoring the ideas; means for aggregating the scores and sorting by highest score; and means for feeding back high-scoring ideas to a user.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] In conventional generative AI contests, the evaluation of submitted ideas is highly subjective, making it prone to bias by the evaluators. Furthermore, there is a lack of efficient and objective methods for evaluating a large number of ideas, which can lead to excellent ideas being overlooked. This undermines the fairness and efficiency of the contest. [Means for solving the problem]

[0005] The system of the present invention includes a means for collecting previously submitted ideas, a means for loading a generative AI model and setting evaluation criteria, a means for inputting the collected ideas into the generative AI model and scoring the ideas, a means for aggregating the scores and sorting them by highest score, and a means for providing feedback on the ideas with the highest scores to the user. This makes it possible to utilize generative AI to objectively and efficiently evaluate ideas and accurately select superior ideas. Furthermore, by using a natural language processing algorithm as the generative AI model, it is possible to perform sophisticated analysis of text-based ideas and perform highly accurate evaluations based on evaluation criteria such as creativity, feasibility, and impact.

[0006] "Means for collecting" refers to a method or device for incorporating previously submitted ideas into the system.

[0007] A "generative AI model" refers to an algorithm or statistical model that uses natural language processing and machine learning techniques to understand and evaluate text data.

[0008] "Evaluation criteria" refers to a set of characteristics or conditions established to objectively and consistently evaluate ideas.

[0009] "Scoring means" refers to a method or device for using a generative AI model to provide a numerical rating for each idea.

[0010] "Means of aggregation" refers to the method or device for collecting and integrating the scores assigned to individual ideas to produce a final evaluation result.

[0011] "Feedback means" refers to a method or device for notifying or presenting the evaluation results to the user.

[0012] "Natural language processing algorithms" refer to computational techniques for analyzing and understanding text data.

[0013] "Creativity" refers to the originality and innovative elements of an idea.

[0014] "Feasibility" refers to the measure of whether an idea can actually be put into practice.

[0015] "Influence" refers to a measure of the impact an idea has on society and the market. [Brief explanation of the drawings]

[0016] [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

[0017] 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.

[0018] First, the terms used in the following description will be explained.

[0019] 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).

[0020] 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.

[0021] 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.

[0022] 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.

[0023] 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."

[0024] [First embodiment]

[0025] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0026] 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.

[0027] 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).

[0028] 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.

[0029] 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.

[0030] 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.

[0031] 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.

[0032] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0033] 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.

[0034] 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.

[0035] 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.

[0036] 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."

[0037] MODE FOR CARRYING OUT THE INVENTION

[0038] The present invention relates to a system that utilizes generative AI to collect and evaluate previously submitted ideas and select the best idea. This system is configured using the following means.

[0039] 1. Collecting idea data

[0040] The server accesses the database of the Generative AI Contest and queries and retrieves previously submitted idea data. The retrieved data is then converted into an internal dictionary format. This operation allows the system to efficiently manage ideas to be evaluated.

[0041] 2. Preparing the Generative AI

[0042] The server loads the generative AI model (e.g., natural language processing model) to be used for evaluation, including the appropriate tokenizer to convert the text data of the idea into a format that can be evaluated, and sets the evaluation criteria, such as creativity, feasibility, and impact.

[0043] 3. Evaluate ideas

[0044] The server converts the collected text of each idea into tokens using a tokenizer, then inputs them into a generative AI model to score the idea. Specifically, a score is generated for each idea based on the evaluation criteria of creativity, feasibility, and impact. This process allows for a relative evaluation of each idea.

[0045] 4. Score tallying

[0046] The server tally the scores of each idea and sort them by highest score. This allows the best ideas to be identified at a glance. The higher the score, the better each idea is compared to the set evaluation criteria.

[0047] 5. Feedback of results

[0048] The server then provides feedback to the user on the ideas with the highest scores. Specifically, the top few ideas are selected and notified to the user. This feedback allows the user to check the details of the highly rated ideas and refer to the evaluation results.

[0049] Natural language explanation of the process

[0050] First, the server accesses the database of generative AI contests to collect ideas that have been submitted in the past. The collected ideas are then converted into an internal dictionary format, which allows for efficient data management.

[0051] The server then loads the generative AI model for evaluation, which includes natural language processing algorithms and a tokenizer for properly processing text data, and sets evaluation criteria such as creativity, feasibility, and impact.

[0052] The server then converts each idea's text into tokens using a tokenizer and feeds them into a generative AI model, which scores the ideas and generates a score for each idea based on a set of evaluation criteria.

[0053] After each idea has a score generated, the server aggregates it and sorts it by highest score, clearly identifying the best ideas.

[0054] Finally, the server provides feedback on the top ideas to the user, who can then check the details of the highly rated ideas and use the evaluation results as a reference.

[0055] Specific examples

[0056] For example, the following idea is submitted:

[0057] 1. "AI-based automatic translation system"

[0058] 2. "An app that creates music using generative AI"

[0059] 3. "System for controlling self-driving cars using generative AI"

[0060] The server collects these ideas from a database and uses a generative AI model to score them based on evaluation criteria. For example, the following scores might be generated:

[0061] Idea 1: Creativity 80 points, Feasibility 85 points, Impact 90 points

[0062] Idea 2: Creativity 90 points, Feasibility 80 points, Impact 85 points

[0063] Idea 3: Creativity 85, Feasibility 90, Impact 80

[0064] The server aggregates these scores and sorts them by highest score, and the highest scoring idea is then fed back to the user.

[0065] In this way, the system of the present invention utilizes generative AI to efficiently and objectively evaluate ideas and select and present excellent ideas.

[0066] The processing flow will be explained below.

[0067] Specific processing flow of the program

[0068] Idea data collection

[0069] Step 1: Query the database

[0070] The server queries the generative AI contest database to retrieve all submitted ideas.

[0071] Step 2: Transform the data

[0072] The server converts the acquired idea data into an internal dictionary format so that it can be managed efficiently.

[0073] Preparing for generative AI

[0074] Step 1: Loading the Generative AI Model

[0075] The server loads generative AI models (natural language processing algorithms) for evaluation.

[0076] Step 2: Preparing the tokenizer

[0077] The server also loads the tokenizer corresponding to the generative AI model, preparing it to properly tokenize the idea text data.

[0078] Step 3: Set evaluation criteria

[0079] The server sets the criteria for creativity, feasibility, and impact.

[0080] Idea Evaluation

[0081] Step 1: Tokenize your ideas

[0082] The server converts the text of each idea into tokens using a tokenizer.

[0083] Step 2: Score with the model

[0084] The server inputs each tokenized idea into a generative AI model and calculates a score for each.

[0085] Step 3: Create a score list

[0086] The server stores the scores generated for each idea in the form of a list.

[0087] Score tally

[0088] Step 1: Sort the scores

[0089] The server sorts all ideas by their scores, from highest to lowest.

[0090] Step 2: Extracting top ideas

[0091] The server extracts the top few ideas (e.g., the top three).

[0092] Feedback of results

[0093] Step 1: Generate feedback data

[0094] The server generates feedback data including details of the top ranked ideas.

[0095] Step 2: Submit your feedback

[0096] The server transmits the generated feedback data to the user.

[0097] For example, the following processing is performed:

[0098] 1. The server retrieves three ideas from the database: an automatic translation system using AI, an app that creates music using generative AI, and a system that controls self-driving cars using generative AI.

[0099] 2. The server converts these ideas into tokens using a tokenizer, which are then fed into a generative AI model to generate the following score:

[0100] Idea 1: Creativity 80 points, Feasibility 85 points, Impact 90 points

[0101] Idea 2: Creativity 90 points, Feasibility 80 points, Impact 85 points

[0102] Idea 3: Creativity 85, Feasibility 90, Impact 80

[0103] 3. The server aggregates these scores, sorts them by highest score, and sends the top three ideas to the user as feedback data.

[0104] This series of processes creates a system that uses generative AI to objectively and efficiently evaluate ideas and select the best ones.

[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] Existing systems have had problems in efficiently and objectively evaluating previously submitted concepts and selecting the best ones. In particular, when processing large amounts of concept data, issues often arise in setting evaluation criteria and consistency of evaluation. It has also been difficult to provide users with prompt and accurate feedback.

[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 means for collecting previously submitted concepts, means for loading a generative AI model and setting evaluation criteria, means for inputting the collected concepts into the generative AI model and scoring the concepts, means for tokenizing the text of each concept, means for aggregating the scores and sorting them in descending order of score, and means for feeding back the concepts with high scores to the user. This enables efficient and objective evaluation of concepts and rapid selection and feedback of excellent concepts.

[0110] "Previously Submitted Ideas" means ideas or suggestions previously submitted by you or a third party.

[0111] A "generative artificial intelligence model" refers to an algorithm or system that uses artificial intelligence technology to perform natural language processing and predictive analysis.

[0112] "Evaluation criteria" refers to the specific measures or criteria for evaluating ideas, including creativity, feasibility, and impact.

[0113] "Scoring" refers to the process of assigning a score or rating to an initiative based on established evaluation criteria.

[0114] "Tokenization" refers to the process of breaking down text data into words and phrases for processing.

[0115] "Feedback" refers to the action or process of returning information such as evaluation results or notifications to users.

[0116] "User" refers to any person or entity that uses this system to receive the results of an initiative evaluation.

[0117] MODE FOR CARRYING OUT THE INVENTION

[0118] This invention relates to a system that utilizes a generative artificial intelligence model to collect and evaluate previously submitted concepts and select the best concept. This system operates in cooperation with a server, terminals, and users.

[0119] First, the server accesses the database of the Generative AI Contest and collects previously submitted concepts. The server retrieves the data using queries such as SQL and converts the retrieved concept data into an internal dictionary format. This data conversion allows the server to efficiently manage the data and perform subsequent processing.

[0120] The server then loads the generative AI model (e.g., BERT or GPT-3) to be used for evaluation. This generative AI model is an algorithm that uses generative artificial intelligence technology and includes a tokenizer to process the text data. The tokenizer breaks the text data down into tokens and converts them into a format that can be fed into the generative AI model. The server then sets the criteria for evaluating the ideas: creativity, feasibility, and impact.

[0121] Specifically, the server converts the text of each idea into tokens using a tokenizer, and then inputs these into a generative AI model. The generative AI model generates a score for each idea based on the evaluation criteria of creativity, feasibility, and impact. This scoring process provides a relative evaluation of each idea.

[0122] After the scores are generated, the server aggregates them and sorts them by highest score, clearly identifying the best ideas. The server then provides feedback to the user about the highly scored ideas. Specifically, the server notifies the user of the top few highly rated ideas. This feedback allows the user to view the details of the highly rated ideas and refer to the evaluation results.

[0123] Specific examples

[0124] For example, a user might submit the following idea to the database:

[0125] 1. "AI-based automatic translation system"

[0126] 2. "An app that creates music using generative AI"

[0127] 3. "System for controlling self-driving cars using generative AI"

[0128] The server collects these ideas from a database and uses a generative AI model to score them based on evaluation criteria, for example, the following scores are generated:

[0129] Idea 1: Creativity 80 points, Feasibility 85 points, Impact 90 points

[0130] Idea 2: Creativity 90 points, Feasibility 80 points, Impact 85 points

[0131] Idea 3: Creativity 85, Feasibility 90, Impact 80

[0132] The server aggregates these scores and sorts them by highest score, and the highest scoring ideas are then fed back to the user.

[0133] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0134] Program processing flow

[0135] Step 1: Collect idea data

[0136] The server accesses the database of the Generative AI Contest and queries and retrieves previously submitted idea data. The input is the query result from the database, and the output is the idea data converted into dictionary format.

[0137] Specifically, it executes the SQL query "SELECT FROM ideas WHERE submission_date > '2022-01-01'" and converts the results into a Python dictionary, where each idea is managed as a key-value pair. For example, you might get a data structure like this:

[0138] python

[0139] ideas = {

[0140] "idea1": {"title": "AI-based automatic translation system", "submitted_by": "user123", "date": "2023-05-10"},

[0141] "idea2": {"title": "App to create music with generative AI", "submitted_by": "user456", "date": "2023-06-12"},

[0142] }

[0143] Step 2: Prepare the generative AI

[0144] The server loads the generative AI model to be used for evaluation and sets the evaluation criteria. The input is the generative AI model and tokenizer parameters, and the output is the loaded model and the set evaluation criteria. Specifically, it imports libraries such as "from transformers import GPT-3" and creates an instance of the model. It also creates an instance of the tokenizer:

[0145] python

[0146] tokenizer = GPT3Tokenizer.from_pretrained('gpt-3')

[0147] model = GPT3Model.from_pretrained('gpt-3')

[0148] Set the criteria for evaluation: creativity, feasibility, and impact:

[0149] python

[0150] evaluation_criteria = ["creativity", "feasibility", "impact"]

[0151] Step 3: Tokenize your ideas

[0152] The server converts the text of each idea into tokens using a tokenizer. The input is the text data of each idea, and the output is the tokenized text data. Specifically, it converts into tokens as follows:

[0153] python

[0154] tokens = tokenizer.encode(ideas["idea1"]["title"], return_tensors='pt')

[0155] This converts the text of the idea into a format that can be input into a generative AI model.

[0156] Step 4: Evaluate your ideas

[0157] The server inputs the tokenized data into a generative AI model to score ideas. The input is tokenized text data, and the output is a score for each idea. Specifically, the data is input into the model as follows:

[0158] python

[0159] scores = model(tokens)

[0160] You will receive a score based on each criterion:

[0161] python

[0162] scores = {"creativity": 80, "feasibility": 85, "impact": 90}

[0163] Step 5: Counting the scores

[0164] The server aggregates the scores of each idea and sorts them in descending order. The input is the score of each idea, and the output is a list of ideas sorted in descending order. Specifically, the server aggregates and sorts the scores as follows:

[0165] python

[0166] sorted_ideas = sorted(ideas.items(), key=lambda x: x[1]['scores']['total'], reverse=True)

[0167] This allows the best ideas to rise to the top.

[0168] Step 6: Feedback on results

[0169] The server feeds back the highest-scoring ideas to the user. The input is the high-scoring ideas, and the output is a feedback message to the user. Specifically, it sends the following feedback message to the user's device:

[0170] python

[0171] feedback_message = f"The idea with the highest score is "{sorted_ideas[0]['title']}". Learn more here."

[0172] send_feedback_to_user(user_id, feedback_message)

[0173] In this way, the system of the present invention can efficiently and objectively evaluate ideas and quickly present excellent ideas to the user.

[0174] (Application example 1)

[0175] 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."

[0176] In existing factory automation systems, the evaluation and selection of new efficiency improvement measures and automation ideas is often done manually, making objective and efficient evaluation difficult. Furthermore, it is difficult to quickly find the most suitable idea from among many ideas and put it into practice. Given this background, there is a demand for a faster and more reliable idea evaluation system in workplaces seeking innovation and efficiency improvements in factory automation.

[0177] 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.

[0178] In this invention, the server includes means for collecting previously submitted ideas, means for loading a generative AI model and setting evaluation criteria, means for inputting the collected ideas into the generative AI model and scoring the ideas, means for aggregating the scores and sorting them in descending order of score, means for providing feedback on the ideas with the highest scores to the user, and means for applying the idea evaluation results to the factory automation system and providing optimal solutions for improving efficiency. This enables objective and efficient evaluation of new efficiency improvement measures and automation ideas in the factory automation system, and makes it possible to quickly find and implement the most suitable ideas.

[0179] A "server" is an information processing device connected to a computer network, which collects, analyzes, processes, and provides data.

[0180] "Means for collecting idea data" refers to devices or algorithms that retrieve previously submitted ideas from databases, etc., and manage them appropriately.

[0181] A "generative AI model" refers to an artificial intelligence program that uses natural language processing algorithms and other techniques to analyze input data and perform specific tasks.

[0182] "Means for setting evaluation criteria" refers to devices or methods for defining and setting criteria for evaluation based on the creativity, feasibility, impact, etc. of ideas.

[0183] "Idea scoring means" refers to a device or method that uses a generative AI model to evaluate collected ideas and assign scores based on each evaluation criterion.

[0184] "Means for aggregating scores and sorting ideas by highest score" refers to a device or algorithm that aggregates the scores assigned to ideas and orders the ideas based on that aggregated score.

[0185] "Means for providing feedback to the user" refers to a device or method for notifying the user of the ideas with high scores and communicating the results.

[0186] "Factory automation system" is a general term for hardware and software that automates manufacturing processes and operations within a factory to improve efficiency.

[0187] "Means for providing optimal solutions" refers to devices and methods that are proposed and introduced based on the evaluation results to improve factory efficiency and optimize automation.

[0188] The system for implementing the present invention involves a process of collecting previously submitted ideas, evaluating them using a generative AI model, and providing optimal solutions to a factory automation system based on the evaluation results. A specific embodiment of the system is shown below.

[0189] First, the server accesses the database and collects previously submitted ideas. This idea data is converted into an internal dictionary format and efficiently managed.

[0190] The server then loads a generative AI model for evaluation. The model includes a natural language processing algorithm and uses a tokenizer to convert the text data of the ideas into a format that can be evaluated. The server also sets evaluation criteria such as creativity, feasibility, and impact.

[0191] Each collected idea text is converted into tokens by a tokenizer and then input into a generative AI model. The generative AI model evaluates the input ideas and generates a score for each idea based on the evaluation criteria of creativity, feasibility, and impact. The server aggregates these scores and sorts them by highest score, allowing it to identify superior ideas.

[0192] The server then provides feedback on the highly scored ideas to users, such as factory managers. Specifically, the top few ideas are selected and notified to the users. This feedback allows users to check the details of the highly rated ideas and refer to the evaluation results.

[0193] Furthermore, the most highly evaluated ideas will be applied to factory automation systems to provide optimal solutions for improving efficiency, thereby optimizing factory automation processes and increasing productivity.

[0194] For example, the following prompts can be fed into a generative AI model to evaluate ideas:

[0195] Prompt statement:

[0196] "Please rate the following ideas:

[0197] Idea: "Inventory management system using unmanned robots"

[0198] Evaluation criteria: Creativity, feasibility, impact.”

[0199] The expected output is, for example, "80, 90, 85", which are scores based on each evaluation criterion.

[0200] Based on the evaluation results, the server provides optimal solutions to the factory automation system, improving factory efficiency.

[0201] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0202] Step 1:

[0203] The server accesses a database to collect previously submitted ideas. The input is a database query, and the output is a list of ideas. This list is converted into an internal dictionary format and efficiently managed.

[0204] Step 2:

[0205] The server loads the generative AI model to be used for evaluation. At this stage, the generative AI model containing a specific natural language processing algorithm is loaded. The input is an instruction to load the AI ​​model, and the output is an AI model ready for evaluation.

[0206] Step 3:

[0207] The server sets the evaluation criteria, specifically defining criteria such as creativity, feasibility, and impact. The input is the evaluation criteria setting, and the output is the set evaluation criteria.

[0208] Step 4:

[0209] Each piece of text in the collected idea data is converted into tokens by a tokenizer. The input to this step is the text data of the idea, and the output is the tokenized data. The tokenizer analyzes the text and divides it into small units that can be analyzed.

[0210] Step 5:

[0211] The server uses a generative AI model to evaluate the tokenized ideas. In this step, the tokenized ideas are input into the generative AI model, and a score based on each evaluation criterion is generated as output. Specifically, the AI ​​model performs semantic analysis of the text and calculates a score for each evaluation criterion.

[0212] Step 6:

[0213] The server aggregates the scores obtained from the evaluation and sorts them in descending order. The input is the score of each idea, and the output is a list of ideas sorted in descending order. The server adds up the scores and sorts the list in descending order.

[0214] Step 7:

[0215] The server feeds back the highly scored ideas to the user. Here, the top few ideas are selected and their details are notified to the user. The input is a list of highly scored ideas, and the output is a notification of the fed back ideas. Specifically, the server sends a notification to the user's device.

[0216] Step 8:

[0217] Based on the feedback, the user applies the optimal solution to the factory automation system. The input of this step is the feedback received by the user, and the output is the applied automation solution. Specifically, the user implements the selected idea to optimize the factory automation process.

[0218] 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.

[0219] MODE FOR CARRYING OUT THE INVENTION

[0220] This invention relates to a system that uses generative AI to collect and evaluate previously submitted ideas and select the best idea, and further combines it with an emotion engine that recognizes the user's emotions. This system is implemented as follows:

[0221] First, the server accesses the database of generative AI contests and collects data on ideas submitted in the past. The collected data is then converted into an internal dictionary format and efficiently managed.

[0222] Next, the server loads the generative AI model to be used for evaluation, prepares the tokenizer, which is used to properly tokenize the text data of the idea, and sets the evaluation criteria of creativity, feasibility, and impact.

[0223] The server then converts the collected text of each idea into tokens using a tokenizer and feeds them into a generative AI model, which scores each idea and generates a score for each idea based on the evaluation criteria set.

[0224] Furthermore, this system uses an emotion engine to recognize the user's emotions in real time. When the user provides feedback on an idea, the device inputs the feedback into the emotion engine for analysis. Emotion data based on the analysis results is reflected in the evaluation of the idea.

[0225] The server then recalculates a final score for each idea, taking into account the emotional data obtained by the emotion engine. The scores obtained in this way are aggregated and sorted by highest score.

[0226] Finally, the server provides feedback on the top ideas to the user. Feedback data containing details of the top ideas is generated and sent to the user. This allows the user to check the details of the highly rated ideas and to identify areas for future improvement based on the emotional data obtained from the emotion engine.

[0227] Natural language explanation of the process

[0228] First, the server accesses a database to retrieve previously submitted ideas, which are then converted into an internal dictionary format.

[0229] The server then loads the generative AI model and tokenizer, ready to evaluate each idea based on the set criteria, which include creativity, feasibility, and impact.

[0230] The server then converts the text of each idea into tokens using a tokenizer, feeds them into a generative AI model, and calculates a score, which indicates how good each idea is.

[0231] Next, when receiving feedback from users, they input the feedback into the emotion engine using their terminal. The emotion engine analyzes the user's emotions in real time and obtains the results. This emotion data is incorporated into the score so that it is also reflected in the evaluation of the idea.

[0232] The final scores are then tallied and sorted by highest score, thereby identifying the best ideas.

[0233] Finally, the server provides feedback on the top ideas to the user, who can then check the details of the highly rated ideas and refer to the evaluation results and the analysis results of the emotion engine.

[0234] Specific examples

[0235] For example, the following idea is submitted:

[0236] 1. "AI-based automatic translation system"

[0237] 2. "An app that creates music using generative AI"

[0238] 3. "System for controlling self-driving cars using generative AI"

[0239] The server collects these ideas from the database, converts them into tokens using a tokenizer, and feeds them into a generative AI model, which generates a score for each idea, such as:

[0240] Idea 1: Creativity 80 points, Feasibility 85 points, Impact 90 points

[0241] Idea 2: Creativity 90 points, Feasibility 80 points, Impact 85 points

[0242] Idea 3: Creativity 85, Feasibility 90, Impact 80

[0243] When users provide feedback on these ideas, the device uses an emotion engine to analyze their emotions in real time and reflects the analysis results in the score. For example, if a user shows strong interest in idea 2, the emotion analysis may increase the evaluation score.

[0244] Finally, the server recalculates the score taking into account the emotion data and sends the top three ideas as feedback to the user, realizing an advanced evaluation system that combines generative AI and emotion recognition.

[0245] The processing flow will be explained below.

[0246] MODE FOR CARRYING OUT THE INVENTION

[0247] Idea data collection

[0248] Step 1: Query the database

[0249] The server queries the generative AI contest database to retrieve all previously submitted idea data.

[0250] Step 2: Transform the data

[0251] The server converts the acquired idea data into an internal dictionary format and manages it efficiently.

[0252] Preparing for generative AI

[0253] Step 1: Loading the Generative AI Model

[0254] The server loads generative AI models (natural language processing algorithms) for evaluation.

[0255] Step 2: Preparing the tokenizer

[0256] The server loads the generative AI model and the corresponding tokenizer, preparing it to properly tokenize the idea text data.

[0257] Step 3: Set evaluation criteria

[0258] The server sets the criteria for evaluation: creativity, feasibility, and impact.

[0259] Idea Evaluation

[0260] Step 1: Tokenize your ideas

[0261] The server converts the text of each collected idea into tokens using a tokenizer.

[0262] Step 2: Score with the model

[0263] The server inputs each tokenized idea into a generative AI model and calculates a score for each.

[0264] Step 3: Create a score list

[0265] The server stores the scores generated for each idea in the form of a list.

[0266] Using the Emotion Engine

[0267] Step 1: Accepting user feedback

[0268] The device receives feedback from the user, including detailed thoughts on the ideas that the user has rated.

[0269] Step 2: Analysis by the emotion engine

[0270] The device inputs the received feedback into an emotion engine and analyzes the user's emotions in real time.

[0271] Step 3: Generate emotion data

[0272] The terminal generates emotion data of the user based on the analysis results.

[0273] Step 4: Reflecting emotional data

[0274] The server reflects the generated emotion data in the score of the idea and updates the evaluation result.

[0275] Score tally

[0276] Step 1: Sort the scores

[0277] The server sorts the final scores from highest to lowest.

[0278] Step 2: Extracting top ideas

[0279] The server extracts the top few ideas (e.g., the top three).

[0280] Feedback of results

[0281] Step 1: Generate feedback data

[0282] The server generates feedback data including details of the top ideas.

[0283] Step 2: Submit your feedback

[0284] The server transmits the generated feedback data to the user.

[0285] Specific examples

[0286] 1. The server retrieves ideas from the Generative AI Utilization Contest Database, such as "an automatic translation system using AI," "an app that creates music using generative AI," and "a system that controls self-driving cars using generative AI."

[0287] 2. The server converts these ideas into tokens using a tokenizer, which are then fed into a generative AI model to generate a score:

[0288] Idea 1: Creativity 80 points, Feasibility 85 points, Impact 90 points

[0289] Idea 2: Creativity 90 points, Feasibility 80 points, Impact 85 points

[0290] Idea 3: Creativity 85, Feasibility 90, Impact 80

[0291] 3. When users provide feedback on these ideas, the device uses an emotion engine to analyze the feedback in real time and generate emotion data based on the analysis results. For example, if the user shows strong interest in idea 2, the emotion analysis may increase the evaluation score.

[0292] 4. The server recalculates the score taking into account the emotional data and sends the top three ideas to the user as feedback data.

[0293] This series of processes realizes an advanced evaluation system that combines generative AI and emotion recognition.

[0294] Example 2

[0295] 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."

[0296] Conventional idea evaluation systems have limited evaluation criteria, making it difficult to provide flexible evaluations that reflect user sentiment. Furthermore, the accuracy and reliability of the evaluations are lacking, making it difficult to accurately identify the best ideas. This has led to the problem that many excellent ideas are not properly evaluated and end up being buried.

[0297] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0298] In this invention, the server includes means for collecting previously submitted information, means for loading an AI model and setting evaluation criteria, means for inputting the collected information into the AI ​​model and scoring the information, means for analyzing the user's emotions, means for recalculating the score based on the emotion data, means for aggregating the scores and sorting them in descending order, and means for feeding back information on the high scores to the user. This enables flexible and accurate evaluation that reflects the user's emotions.

[0299] "Previously Submitted Information" means data, such as ideas or suggestions, previously submitted by users or other interested parties.

[0300] An "artificial intelligence model" is a system that analyzes and scores data using machine learning algorithms trained for a specific purpose.

[0301] "Metrics" are the indicators or standards used to evaluate the quality or usefulness of information or ideas.

[0302] "Means for analyzing user emotions" refers to software or devices that analyze emotions and impressions based on feedback and reactions from users.

[0303] "Emotional data" refers to information about the analyzed user's emotions and reactions.

[0304] "Means for recalculating scores" means software or computational algorithms for reevaluating existing scores based on emotion data and generating updated scores.

[0305] "Means for aggregating scores and sorting by highest score" refers to software or algorithms that calculate the score for each piece of information and sort it in descending order.

[0306] The "means for feeding back information with high scores to the user" refers to a method or means for reporting information with high scores from among the sorted information to the user.

[0307] MODE FOR CARRYING OUT THE INVENTION

[0308] This invention relates to a system that uses generative AI to collect and evaluate previously submitted information and select the best information, and also combines it with an emotion engine that recognizes the user's emotions. This system can realize consistent evaluations that also reflect the user's emotions.

[0309] First, the server accesses the AI-powered contest database and collects previously submitted information. The software used here can be a database management system (e.g., MySQL or PostgreSQL). The collected information is internally converted into a dictionary format (e.g., Python dictionary data structure) for efficient management.

[0310] Next, the server loads the generative AI model and tokenizer to be used for evaluation. Specifically, it uses a natural language processing algorithm (e.g., GPT-3 or BERT). The tokenizer (e.g., Hugging Face's tokenizer) is used to properly tokenize the information text data. At this stage, the evaluation criteria are set: creativity, feasibility, and impact.

[0311] The server then converts the text of each piece of collected information into tokens using a tokenizer and feeds them into a generative AI model. The generative AI model scores each piece of information and generates a score based on the set evaluation criteria. These scores are stored in a temporary data store (e.g., Redis).

[0312] Furthermore, this system uses an emotion engine to recognize the user's emotions in real time. When the user provides feedback on information, the terminal inputs the feedback into the emotion engine (e.g., emotion analysis model Sentiment 140) for analysis. Emotion data based on the analysis results is sent back to the server and reflected in the evaluation of the information.

[0313] The server recalculates the scores based on the emotion data, aggregates these scores, and sorts them in order of highest score. Finally, feedback data is generated to provide the top ranking information to the user. This data is sent to the user's device via API or email. This process allows the user to check the details of the highly rated information and to identify areas for future improvement based on the emotion data obtained from the emotion engine.

[0314] Specific examples

[0315] For example, if the following information is submitted:

[0316] 1. "AI-based automatic translation system"

[0317] 2. "An app that creates music using generative AI"

[0318] 3. "System for controlling self-driving cars using generative AI"

[0319] The server collects this information from the database, converts it into tokens using a tokenizer, and inputs them into the generative AI model, which generates the following scores for each piece of information:

[0320] Information 1: Creativity 80 points, Feasibility 85 points, Impact 90 points

[0321] Information 2: Creativity 90 points, Feasibility 80 points, Impact 85 points

[0322] Information 3: Creativity 85 points, Feasibility 90 points, Impact 80 points

[0323] When the user provides feedback on this information, the device uses an emotion engine to analyze the user's emotions in real time. For example, if the user shows strong interest in information 2, the evaluation score for information 2 may increase based on the emotion analysis results.

[0324] Finally, the server recalculates the score taking into account the emotional data and sends the top three scores to the user as feedback data, realizing an advanced evaluation system that combines generative AI and emotional recognition.

[0325] Prompt Sentence Examples

[0326] To facilitate information evaluation in an AI-based machine translation system, the following prompts could be input to the generative AI model:

[0327] "AI-based automatic translation system" is a system that enables automatic translation between multiple languages. Please rate this idea in terms of creativity, feasibility, and impact.

[0328] For the emotion engine, the following prompt is an example:

[0329] Analyze user feedback on your AI-powered machine translation system to identify emotions such as interest, delight, and surprise.

[0330] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0331] Step 1: Obtaining collected information

[0332] The server collects previously submitted information. The input for this process is the database connection information and query. Specifically, the server executes an SQL query to extract past information, parses the retrieved information into JSON format, and then converts it into an internal dictionary format. The output is the information data converted into dictionary format.

[0333] Step 2: Load the generative AI model and tokenizer

[0334] The server loads the generative AI model and tokenizer. The input to this process is the path to the model and tokenizer. Specifically, the server imports a machine learning library (e.g., TensorFlow, PyTorch) and loads the model file and tokenizer. The output is the loaded generative AI model and tokenizer.

[0335] Step 3: Set the evaluation criteria

[0336] The server sets the criteria for evaluation. The input of this process is the setting information of the evaluation criteria (creativity, feasibility, impact). In concrete terms, the server reads the evaluation criteria information from a setting file or database and sets these criteria for the model. The output is the set evaluation criteria.

[0337] Step 4: Tokenize your idea data

[0338] The server converts the collected text data into tokens using a tokenizer. The inputs to this process are the information data converted into dictionary format and the loaded tokenizer. Specifically, each piece of text data is passed to the tokenizer and converted into tokens. The output is the tokenized data.

[0339] Step 5: Score your idea data

[0340] The server inputs the tokenized data into the generative AI model and calculates the score. The inputs for this process are the tokenized data and the loaded generative AI model. Specifically, it inputs a prompt sentence into the generative AI model and performs scoring. The output is the calculated score.

[0341] Step 6: User feedback analysis

[0342] The terminal analyzes the feedback provided by the user using an emotion engine. The input of this process is the user's feedback text. Specifically, the feedback text is input to the emotion engine (e.g., Sentiment 140) to obtain emotion data. The output is the analyzed emotion data.

[0343] Step 7: Recalculate the score using emotion data

[0344] The server recalculates the score based on the obtained emotion data. The inputs to this process are the original score and the analyzed emotion data. Specifically, the server updates the score using recalculation logic to generate a final score. The output is the recalculated final score.

[0345] Step 8: Count and sort the scores

[0346] The server tally the final scores of each piece of information and sorts them in descending order. The input to this process is the recalculated final scores. Specific operations include using an algorithm to sort the scores in descending order and listing the top pieces of information. The output is a sorted list of information.

[0347] Step 9: User Feedback

[0348] The server feeds back the top-level information to the user. The input to this process is a sorted information list. Specific operations include generating feedback data and sending it to the user's device via a communication method such as an API or email. The output is the feedback data sent to the user.

[0349] (Application example 2)

[0350] 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."

[0351] Conventional idea evaluation systems using generative AI have the problem of being unable to fully reflect user opinions and emotions, making it difficult to select truly valuable ideas. Particularly for ideas for advertising campaigns, where user emotions are an important evaluation factor, improvement in this area was required.

[0352] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting previously submitted ideas, means for loading a generative AI model and setting evaluation criteria, means for inputting the collected ideas into the generative AI model and scoring the ideas, means for receiving feedback from users and analyzing them with a sentiment analysis engine, means for re-evaluating ideas based on the sentiment analysis results and calculating a final score, and means for feeding back high-scoring ideas to the user. This enables evaluation that takes into account the user's emotions, making it possible to select more appropriate ideas.

[0353] "Means of collection" refers to the means of acquiring ideas that have been submitted in the past and storing them in a database.

[0354] A "generative AI model" is an AI model that uses natural language processing algorithms to analyze and evaluate text data.

[0355] "Evaluation criteria" are the standards set to evaluate ideas for their creativity, feasibility, and impact.

[0356] A "means for scoring" is a means for analyzing ideas input into a generative AI model and assigning a score based on set evaluation criteria.

[0357] The "emotion analysis engine" is an engine that analyzes feedback from users and generates emotion data based on that feedback.

[0358] The "re-evaluation means" is a means for adjusting the initial evaluation score and calculating the final score in consideration of the sentiment analysis results.

[0359] The "means of providing feedback" is a means of reporting high-scoring ideas and their details to the user.

[0360] A system for implementing this invention includes a server, a sentiment analysis engine, and a terminal, and uses a generative AI model to evaluate and select ideas for an advertising campaign.

[0361] System program generation

[0362] System Overview

[0363] The server collects previously submitted ideas and sets evaluation criteria. Next, the collected ideas are input into a generative AI model and scored. It also receives feedback from users, analyzes it in real time using a sentiment analysis engine, and generates emotional data. The ideas are then re-evaluated based on the emotional data and a final score is calculated. Finally, the ideas with the highest scores are fed back to the user.

[0364] Hardware and Software

[0365] Hardware: Smartphones, servers

[0366] software:

[0367] Hugging Face's Transformers library: uses a pre-trained BERT model and tokenizer

[0368] VADER (Valence Aware Dictionary and sEntiment Reasoner): for sentiment analysis

[0369] OpenAI API: Text analysis and evaluation using generative AI models

[0370] Data processing and calculation

[0371] 1. Tokenization of ideas

[0372] The server converts the text data of the idea into tokens using Hugging Face's tokenizer.

[0373] 2. Idea scoring

[0374] The server inputs the tokenized idea data into the BERT model and scores it based on the criteria of creativity, feasibility, and impact.

[0375] 3. Sentiment Analysis of Feedback

[0376] The device collects feedback from users in real time and performs sentiment analysis using VADER to generate a sentiment score.

[0377] 4. Calculation of Final Score

[0378] The server calculates a weighted average of the idea score and the emotion score to arrive at a final score.

[0379] 5. Providing Feedback

[0380] The server reports the ideas to the user sorted by highest score.

[0381] Specific examples

[0382] Below are some examples of specific ideas and feedback prompts:

[0383] Example ideas:

[0384] 1. "A video advertising campaign highlighting the features of a new smartphone"

[0385] 2. "Personalized advertising using generative AI"

[0386] 3. "Advertisements that recommend products that match the user's preferences"

[0387] Feedback example:

[0388] 1. "I think this is a really great idea!"

[0389] 2. "I'm not interested."

[0390] 3. "It's an interesting idea, but it seems difficult to implement."

[0391] Example prompt:

[0392] Prompt: Generate an advertising campaign based on the idea.

[0393] Example: A video ad campaign highlighting the features of a new smartphone

[0394] Feedback: I think this is a really great idea!

[0395] The above system makes it possible to evaluate ideas by incorporating the user's emotions, allowing for the selection of more appropriate ideas.

[0396] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0397] Step 1:

[0398] The server collects previously submitted ideas from a database. Specifically, the server accesses the database and retrieves idea data. The input at this stage is the idea data in the database, and the output is a list of collected ideas.

[0399] Step 2:

[0400] The server converts the collected ideas into tokens using Hugging Face's tokenizer. The tokenization process converts text data into a set of tokens. The input of this step is the collected idea list, and the output is the tokenized idea data.

[0401] Step 3:

[0402] The server inputs the tokenized idea data into the BERT model and scores them based on the criteria of creativity, feasibility, and impact. Here, the input is the tokenized idea data, and the output is a score for each idea.

[0403] Step 4:

[0404] The server collects feedback from users through their terminals. Users provide feedback on ideas, and the data is sent to the server. The input of this step is the feedback from users, and the output is the collected feedback data.

[0405] Step 5:

[0406] The device uses VADER to perform real-time sentiment analysis of the collected feedback data and generate a sentiment score. Specifically, the feedback text is input into the sentiment analysis engine, and the analysis results are output as a positive, negative, or neutral score. The input of this step is the user's feedback data, and the output is a sentiment score.

[0407] Step 6:

[0408] The server calculates the final score by weighting the initial evaluation score and the sentiment score. The inputs at this stage are the idea score and the sentiment score, and the output is the re-evaluated final score.

[0409] Step 7:

[0410] The server sorts the ideas by final score and feeds back the ideas with the highest scores to the user. Specifically, the server selects the top sorted ideas and notifies the user of their details. The input of this step is the re-evaluated final score, and the output is the feedback data of the highly rated ideas.

[0411] 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.

[0412] 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.

[0413] 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.

[0414] [Second embodiment]

[0415] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0416] 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.

[0417] 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).

[0418] 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.

[0419] 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.

[0420] 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).

[0421] 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.

[0422] 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.

[0423] 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.

[0424] 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.

[0425] 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.

[0426] 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."

[0427] MODE FOR CARRYING OUT THE INVENTION

[0428] The present invention relates to a system that utilizes generative AI to collect and evaluate previously submitted ideas and select the best idea. This system is configured using the following means.

[0429] 1. Collecting idea data

[0430] The server accesses the database of the Generative AI Contest and queries and retrieves previously submitted idea data. The retrieved data is then converted into an internal dictionary format. This operation allows the system to efficiently manage ideas to be evaluated.

[0431] 2. Preparing the Generative AI

[0432] The server loads the generative AI model (e.g., natural language processing model) to be used for evaluation, including the appropriate tokenizer to convert the text data of the idea into a format that can be evaluated, and sets the evaluation criteria, such as creativity, feasibility, and impact.

[0433] 3. Evaluate ideas

[0434] The server converts the collected text of each idea into tokens using a tokenizer, then inputs them into a generative AI model to score the idea. Specifically, a score is generated for each idea based on the evaluation criteria of creativity, feasibility, and impact. This process allows for a relative evaluation of each idea.

[0435] 4. Score tallying

[0436] The server tally the scores of each idea and sort them by highest score. This allows the best ideas to be identified at a glance. The higher the score, the better each idea is compared to the set evaluation criteria.

[0437] 5. Feedback of results

[0438] The server then provides feedback to the user on the ideas with the highest scores. Specifically, the top few ideas are selected and notified to the user. This feedback allows the user to check the details of the highly rated ideas and refer to the evaluation results.

[0439] Natural language explanation of the process

[0440] First, the server accesses the database of generative AI contests to collect ideas that have been submitted in the past. The collected ideas are then converted into an internal dictionary format, which allows for efficient data management.

[0441] The server then loads the generative AI model for evaluation, which includes natural language processing algorithms and a tokenizer for properly processing text data, and sets evaluation criteria such as creativity, feasibility, and impact.

[0442] The server then converts each idea's text into tokens using a tokenizer and feeds them into a generative AI model, which scores the ideas and generates a score for each idea based on a set of evaluation criteria.

[0443] After each idea has a score generated, the server aggregates it and sorts it by highest score, clearly identifying the best ideas.

[0444] Finally, the server provides feedback on the top ideas to the user, who can then check the details of the highly rated ideas and use the evaluation results as a reference.

[0445] Specific examples

[0446] For example, the following idea is submitted:

[0447] 1. "AI-based automatic translation system"

[0448] 2. "An app that creates music using generative AI"

[0449] 3. "System for controlling self-driving cars using generative AI"

[0450] The server collects these ideas from a database and uses a generative AI model to score them based on evaluation criteria. For example, the following scores might be generated:

[0451] Idea 1: Creativity 80 points, Feasibility 85 points, Impact 90 points

[0452] Idea 2: Creativity 90 points, Feasibility 80 points, Impact 85 points

[0453] Idea 3: Creativity 85, Feasibility 90, Impact 80

[0454] The server aggregates these scores and sorts them by highest score, and the highest scoring idea is then fed back to the user.

[0455] In this way, the system of the present invention utilizes generative AI to efficiently and objectively evaluate ideas and select and present excellent ideas.

[0456] The processing flow will be explained below.

[0457] Specific processing flow of the program

[0458] Idea data collection

[0459] Step 1: Query the database

[0460] The server queries the generative AI contest database to retrieve all submitted ideas.

[0461] Step 2: Transform the data

[0462] The server converts the acquired idea data into an internal dictionary format so that it can be managed efficiently.

[0463] Preparing for generative AI

[0464] Step 1: Loading the Generative AI Model

[0465] The server loads generative AI models (natural language processing algorithms) for evaluation.

[0466] Step 2: Preparing the tokenizer

[0467] The server also loads the tokenizer corresponding to the generative AI model, preparing it to properly tokenize the idea text data.

[0468] Step 3: Set evaluation criteria

[0469] The server sets the criteria for creativity, feasibility, and impact.

[0470] Idea Evaluation

[0471] Step 1: Tokenize your ideas

[0472] The server converts the text of each idea into tokens using a tokenizer.

[0473] Step 2: Score with the model

[0474] The server inputs each tokenized idea into a generative AI model and calculates a score for each.

[0475] Step 3: Create a score list

[0476] The server stores the scores generated for each idea in the form of a list.

[0477] Score tally

[0478] Step 1: Sort the scores

[0479] The server sorts all ideas by their scores, from highest to lowest.

[0480] Step 2: Extracting top ideas

[0481] The server extracts the top few ideas (e.g., the top three).

[0482] Feedback of results

[0483] Step 1: Generate feedback data

[0484] The server generates feedback data including details of the top ranked ideas.

[0485] Step 2: Submit your feedback

[0486] The server transmits the generated feedback data to the user.

[0487] For example, the following processing is performed:

[0488] 1. The server retrieves three ideas from the database: an automatic translation system using AI, an app that creates music using generative AI, and a system that controls self-driving cars using generative AI.

[0489] 2. The server converts these ideas into tokens using a tokenizer, which are then fed into a generative AI model to generate the following score:

[0490] Idea 1: Creativity 80 points, Feasibility 85 points, Impact 90 points

[0491] Idea 2: Creativity 90 points, Feasibility 80 points, Impact 85 points

[0492] Idea 3: Creativity 85, Feasibility 90, Impact 80

[0493] 3. The server aggregates these scores, sorts them by highest score, and sends the top three ideas to the user as feedback data.

[0494] This series of processes creates a system that uses generative AI to objectively and efficiently evaluate ideas and select the best ones.

[0495] Example 1

[0496] 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."

[0497] Existing systems have had problems in efficiently and objectively evaluating previously submitted concepts and selecting the best ones. In particular, when processing large amounts of concept data, issues often arise in setting evaluation criteria and consistency of evaluation. It has also been difficult to provide users with prompt and accurate feedback.

[0498] 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.

[0499] In this invention, the server includes means for collecting previously submitted concepts, means for loading a generative AI model and setting evaluation criteria, means for inputting the collected concepts into the generative AI model and scoring the concepts, means for tokenizing the text of each concept, means for aggregating the scores and sorting them in descending order of score, and means for feeding back the concepts with high scores to the user. This enables efficient and objective evaluation of concepts and rapid selection and feedback of excellent concepts.

[0500] "Previously Submitted Ideas" means ideas or suggestions previously submitted by you or a third party.

[0501] A "generative artificial intelligence model" refers to an algorithm or system that uses artificial intelligence technology to perform natural language processing and predictive analysis.

[0502] "Evaluation criteria" refers to the specific measures or criteria for evaluating ideas, including creativity, feasibility, and impact.

[0503] "Scoring" refers to the process of assigning a score or rating to an initiative based on established evaluation criteria.

[0504] "Tokenization" refers to the process of breaking down text data into words and phrases for processing.

[0505] "Feedback" refers to the action or process of returning information such as evaluation results or notifications to users.

[0506] "User" refers to any person or entity that uses this system to receive the results of an initiative evaluation.

[0507] MODE FOR CARRYING OUT THE INVENTION

[0508] This invention relates to a system that utilizes a generative artificial intelligence model to collect and evaluate previously submitted concepts and select the best concept. This system operates in cooperation with a server, terminals, and users.

[0509] First, the server accesses the database of the Generative AI Contest and collects previously submitted concepts. The server retrieves the data using queries such as SQL and converts the retrieved concept data into an internal dictionary format. This data conversion allows the server to efficiently manage the data and perform subsequent processing.

[0510] The server then loads the generative AI model (e.g., BERT or GPT-3) to be used for evaluation. This generative AI model is an algorithm that uses generative artificial intelligence technology and includes a tokenizer to process the text data. The tokenizer breaks the text data down into tokens and converts them into a format that can be fed into the generative AI model. The server then sets the criteria for evaluating the ideas: creativity, feasibility, and impact.

[0511] Specifically, the server converts the text of each idea into tokens using a tokenizer, and then inputs these into a generative AI model. The generative AI model generates a score for each idea based on the evaluation criteria of creativity, feasibility, and impact. This scoring process provides a relative evaluation of each idea.

[0512] After the scores are generated, the server aggregates them and sorts them by highest score, clearly identifying the best ideas. The server then provides feedback to the user about the highly scored ideas. Specifically, the server notifies the user of the top few highly rated ideas. This feedback allows the user to view the details of the highly rated ideas and refer to the evaluation results.

[0513] Specific examples

[0514] For example, a user might submit the following idea to the database:

[0515] 1. "AI-based automatic translation system"

[0516] 2. "An app that creates music using generative AI"

[0517] 3. "System for controlling self-driving cars using generative AI"

[0518] The server collects these ideas from a database and uses a generative AI model to score them based on evaluation criteria, for example, the following scores are generated:

[0519] Idea 1: Creativity 80 points, Feasibility 85 points, Impact 90 points

[0520] Idea 2: Creativity 90 points, Feasibility 80 points, Impact 85 points

[0521] Idea 3: Creativity 85, Feasibility 90, Impact 80

[0522] The server aggregates these scores and sorts them by highest score, and the highest scoring ideas are then fed back to the user.

[0523] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0524] Program processing flow

[0525] Step 1: Collect idea data

[0526] The server accesses the database of the Generative AI Contest and queries and retrieves previously submitted idea data. The input is the query result from the database, and the output is the idea data converted into dictionary format.

[0527] Specifically, it executes the SQL query "SELECT FROM ideas WHERE submission_date > '2022-01-01'" and converts the results into a Python dictionary, where each idea is managed as a key-value pair. For example, you might get a data structure like this:

[0528] python

[0529] ideas = {

[0530] "idea1": {"title": "AI-based automatic translation system", "submitted_by": "user123", "date": "2023-05-10"},

[0531] "idea2": {"title": "App to create music with generative AI", "submitted_by": "user456", "date": "2023-06-12"},

[0532] }

[0533] Step 2: Prepare the generative AI

[0534] The server loads the generative AI model to be used for evaluation and sets the evaluation criteria. The input is the generative AI model and tokenizer parameters, and the output is the loaded model and the set evaluation criteria. Specifically, it imports libraries such as "from transformers import GPT-3" and creates an instance of the model. It also creates an instance of the tokenizer:

[0535] python

[0536] tokenizer = GPT3Tokenizer.from_pretrained('gpt-3')

[0537] model = GPT3Model.from_pretrained('gpt-3')

[0538] Set the criteria for evaluation: creativity, feasibility, and impact:

[0539] python

[0540] evaluation_criteria = ["creativity", "feasibility", "impact"]

[0541] Step 3: Tokenize your ideas

[0542] The server converts the text of each idea into tokens using a tokenizer. The input is the text data of each idea, and the output is the tokenized text data. Specifically, it converts into tokens as follows:

[0543] python

[0544] tokens = tokenizer.encode(ideas["idea1"]["title"], return_tensors='pt')

[0545] This converts the text of the idea into a format that can be input into a generative AI model.

[0546] Step 4: Evaluate your ideas

[0547] The server inputs the tokenized data into a generative AI model to score ideas. The input is tokenized text data, and the output is a score for each idea. Specifically, the data is input into the model as follows:

[0548] python

[0549] scores = model(tokens)

[0550] You will receive a score based on each criterion:

[0551] python

[0552] scores = {"creativity": 80, "feasibility": 85, "impact": 90}

[0553] Step 5: Counting the scores

[0554] The server aggregates the scores of each idea and sorts them in descending order. The input is the score of each idea, and the output is a list of ideas sorted in descending order. Specifically, the server aggregates and sorts the scores as follows:

[0555] python

[0556] sorted_ideas = sorted(ideas.items(), key=lambda x: x[1]['scores']['total'], reverse=True)

[0557] This allows the best ideas to rise to the top.

[0558] Step 6: Feedback on results

[0559] The server feeds back the highest-scoring ideas to the user. The input is the high-scoring ideas, and the output is a feedback message to the user. Specifically, it sends the following feedback message to the user's device:

[0560] python

[0561] feedback_message = f"The idea with the highest score is "{sorted_ideas[0]['title']}". Learn more here."

[0562] send_feedback_to_user(user_id, feedback_message)

[0563] In this way, the system of the present invention can efficiently and objectively evaluate ideas and quickly present excellent ideas to the user.

[0564] (Application example 1)

[0565] 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."

[0566] In existing factory automation systems, the evaluation and selection of new efficiency improvement measures and automation ideas is often done manually, making objective and efficient evaluation difficult. Furthermore, it is difficult to quickly find the most suitable idea from among many ideas and put it into practice. Given this background, there is a demand for a faster and more reliable idea evaluation system in workplaces seeking innovation and efficiency improvements in factory automation.

[0567] 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.

[0568] In this invention, the server includes means for collecting previously submitted ideas, means for loading a generative AI model and setting evaluation criteria, means for inputting the collected ideas into the generative AI model and scoring the ideas, means for aggregating the scores and sorting them in descending order of score, means for providing feedback on the ideas with the highest scores to the user, and means for applying the idea evaluation results to the factory automation system and providing optimal solutions for improving efficiency. This enables objective and efficient evaluation of new efficiency improvement measures and automation ideas in the factory automation system, and makes it possible to quickly find and implement the most suitable ideas.

[0569] A "server" is an information processing device connected to a computer network, which collects, analyzes, processes, and provides data.

[0570] "Means for collecting idea data" refers to devices or algorithms that retrieve previously submitted ideas from databases, etc., and manage them appropriately.

[0571] A "generative AI model" refers to an artificial intelligence program that uses natural language processing algorithms and other techniques to analyze input data and perform specific tasks.

[0572] "Means for setting evaluation criteria" refers to devices or methods for defining and setting criteria for evaluation based on the creativity, feasibility, impact, etc. of ideas.

[0573] "Idea scoring means" refers to a device or method that uses a generative AI model to evaluate collected ideas and assign scores based on each evaluation criterion.

[0574] "Means for aggregating scores and sorting ideas by highest score" refers to a device or algorithm that aggregates the scores assigned to ideas and orders the ideas based on that aggregated score.

[0575] "Means for providing feedback to the user" refers to a device or method for notifying the user of the ideas with high scores and communicating the results.

[0576] "Factory automation system" is a general term for hardware and software that automates manufacturing processes and operations within a factory to improve efficiency.

[0577] "Means for providing optimal solutions" refers to devices and methods that are proposed and introduced based on the evaluation results to improve factory efficiency and optimize automation.

[0578] The system for implementing the present invention involves a process of collecting previously submitted ideas, evaluating them using a generative AI model, and providing optimal solutions to a factory automation system based on the evaluation results. A specific embodiment of the system is shown below.

[0579] First, the server accesses the database and collects previously submitted ideas. This idea data is converted into an internal dictionary format and efficiently managed.

[0580] The server then loads a generative AI model for evaluation. The model includes a natural language processing algorithm and uses a tokenizer to convert the text data of the ideas into a format that can be evaluated. The server also sets evaluation criteria such as creativity, feasibility, and impact.

[0581] Each collected idea text is converted into tokens by a tokenizer and then input into a generative AI model. The generative AI model evaluates the input ideas and generates a score for each idea based on the evaluation criteria of creativity, feasibility, and impact. The server aggregates these scores and sorts them by highest score, allowing it to identify superior ideas.

[0582] The server then provides feedback on the highly scored ideas to users, such as factory managers. Specifically, the top few ideas are selected and notified to the users. This feedback allows users to check the details of the highly rated ideas and refer to the evaluation results.

[0583] Furthermore, the most highly evaluated ideas will be applied to factory automation systems to provide optimal solutions for improving efficiency, thereby optimizing factory automation processes and increasing productivity.

[0584] For example, the following prompts can be fed into a generative AI model to evaluate ideas:

[0585] Prompt statement:

[0586] "Please rate the following ideas:

[0587] Idea: "Inventory management system using unmanned robots"

[0588] Evaluation criteria: Creativity, feasibility, impact.”

[0589] The expected output is, for example, "80, 90, 85", which are scores based on each evaluation criterion.

[0590] Based on the evaluation results, the server provides optimal solutions to the factory automation system, improving factory efficiency.

[0591] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0592] Step 1:

[0593] The server accesses a database to collect previously submitted ideas. The input is a database query, and the output is a list of ideas. This list is converted into an internal dictionary format and efficiently managed.

[0594] Step 2:

[0595] The server loads the generative AI model to be used for evaluation. At this stage, the generative AI model containing a specific natural language processing algorithm is loaded. The input is an instruction to load the AI ​​model, and the output is an AI model ready for evaluation.

[0596] Step 3:

[0597] The server sets the evaluation criteria, specifically defining criteria such as creativity, feasibility, and impact. The input is the evaluation criteria setting, and the output is the set evaluation criteria.

[0598] Step 4:

[0599] Each piece of text in the collected idea data is converted into tokens by a tokenizer. The input to this step is the text data of the idea, and the output is the tokenized data. The tokenizer analyzes the text and divides it into small units that can be analyzed.

[0600] Step 5:

[0601] The server uses a generative AI model to evaluate the tokenized ideas. In this step, the tokenized ideas are input into the generative AI model, and a score based on each evaluation criterion is generated as output. Specifically, the AI ​​model performs semantic analysis of the text and calculates a score for each evaluation criterion.

[0602] Step 6:

[0603] The server aggregates the scores obtained from the evaluation and sorts them in descending order. The input is the score of each idea, and the output is a list of ideas sorted in descending order. The server adds up the scores and sorts the list in descending order.

[0604] Step 7:

[0605] The server feeds back the highly scored ideas to the user. Here, the top few ideas are selected and their details are notified to the user. The input is a list of highly scored ideas, and the output is a notification of the fed back ideas. Specifically, the server sends a notification to the user's device.

[0606] Step 8:

[0607] Based on the feedback, the user applies the optimal solution to the factory automation system. The input of this step is the feedback received by the user, and the output is the applied automation solution. Specifically, the user implements the selected idea to optimize the factory automation process.

[0608] 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.

[0609] MODE FOR CARRYING OUT THE INVENTION

[0610] This invention relates to a system that uses generative AI to collect and evaluate previously submitted ideas and select the best idea, and further combines it with an emotion engine that recognizes the user's emotions. This system is implemented as follows:

[0611] First, the server accesses the database of generative AI contests and collects data on ideas submitted in the past. The collected data is then converted into an internal dictionary format and efficiently managed.

[0612] Next, the server loads the generative AI model to be used for evaluation, prepares the tokenizer, which is used to properly tokenize the text data of the idea, and sets the evaluation criteria of creativity, feasibility, and impact.

[0613] The server then converts the collected text of each idea into tokens using a tokenizer and feeds them into a generative AI model, which scores each idea and generates a score for each idea based on the evaluation criteria set.

[0614] Furthermore, this system uses an emotion engine to recognize the user's emotions in real time. When the user provides feedback on an idea, the device inputs the feedback into the emotion engine for analysis. Emotion data based on the analysis results is reflected in the evaluation of the idea.

[0615] The server then recalculates a final score for each idea, taking into account the emotional data obtained by the emotion engine. The scores obtained in this way are aggregated and sorted by highest score.

[0616] Finally, the server provides feedback on the top ideas to the user. Feedback data containing details of the top ideas is generated and sent to the user. This allows the user to check the details of the highly rated ideas and to identify areas for future improvement based on the emotional data obtained from the emotion engine.

[0617] Natural language explanation of the process

[0618] First, the server accesses a database to retrieve previously submitted ideas, which are then converted into an internal dictionary format.

[0619] The server then loads the generative AI model and tokenizer, ready to evaluate each idea based on the set criteria, which include creativity, feasibility, and impact.

[0620] The server then converts the text of each idea into tokens using a tokenizer, feeds them into a generative AI model, and calculates a score, which indicates how good each idea is.

[0621] Next, when receiving feedback from users, they input the feedback into the emotion engine using their terminal. The emotion engine analyzes the user's emotions in real time and obtains the results. This emotion data is incorporated into the score so that it is also reflected in the evaluation of the idea.

[0622] The final scores are then tallied and sorted by highest score, thereby identifying the best ideas.

[0623] Finally, the server provides feedback on the top ideas to the user, who can then check the details of the highly rated ideas and refer to the evaluation results and the analysis results of the emotion engine.

[0624] Specific examples

[0625] For example, the following idea is submitted:

[0626] 1. "AI-based automatic translation system"

[0627] 2. "An app that creates music using generative AI"

[0628] 3. "System for controlling self-driving cars using generative AI"

[0629] The server collects these ideas from the database, converts them into tokens using a tokenizer, and feeds them into a generative AI model, which generates a score for each idea, such as:

[0630] Idea 1: Creativity 80 points, Feasibility 85 points, Impact 90 points

[0631] Idea 2: Creativity 90 points, Feasibility 80 points, Impact 85 points

[0632] Idea 3: Creativity 85, Feasibility 90, Impact 80

[0633] When users provide feedback on these ideas, the device uses an emotion engine to analyze their emotions in real time and reflects the analysis results in the score. For example, if a user shows strong interest in idea 2, the emotion analysis may increase the evaluation score.

[0634] Finally, the server recalculates the score taking into account the emotion data and sends the top three ideas as feedback to the user, realizing an advanced evaluation system that combines generative AI and emotion recognition.

[0635] The processing flow will be explained below.

[0636] MODE FOR CARRYING OUT THE INVENTION

[0637] Idea data collection

[0638] Step 1: Query the database

[0639] The server queries the generative AI contest database to retrieve all previously submitted idea data.

[0640] Step 2: Transform the data

[0641] The server converts the acquired idea data into an internal dictionary format and manages it efficiently.

[0642] Preparing for generative AI

[0643] Step 1: Loading the Generative AI Model

[0644] The server loads generative AI models (natural language processing algorithms) for evaluation.

[0645] Step 2: Preparing the tokenizer

[0646] The server loads the generative AI model and the corresponding tokenizer, preparing it to properly tokenize the idea text data.

[0647] Step 3: Set evaluation criteria

[0648] The server sets the criteria for evaluation: creativity, feasibility, and impact.

[0649] Idea Evaluation

[0650] Step 1: Tokenize your ideas

[0651] The server converts the text of each collected idea into tokens using a tokenizer.

[0652] Step 2: Score with the model

[0653] The server inputs each tokenized idea into a generative AI model and calculates a score for each.

[0654] Step 3: Create a score list

[0655] The server stores the scores generated for each idea in the form of a list.

[0656] Using the Emotion Engine

[0657] Step 1: Accepting user feedback

[0658] The device receives feedback from the user, including detailed thoughts on the ideas that the user has rated.

[0659] Step 2: Analysis by the emotion engine

[0660] The device inputs the received feedback into an emotion engine and analyzes the user's emotions in real time.

[0661] Step 3: Generate emotion data

[0662] The terminal generates emotion data of the user based on the analysis results.

[0663] Step 4: Reflecting emotional data

[0664] The server reflects the generated emotion data in the score of the idea and updates the evaluation result.

[0665] Score tally

[0666] Step 1: Sort the scores

[0667] The server sorts the final scores from highest to lowest.

[0668] Step 2: Extracting top ideas

[0669] The server extracts the top few ideas (e.g., the top three).

[0670] Feedback of results

[0671] Step 1: Generate feedback data

[0672] The server generates feedback data including details of the top ideas.

[0673] Step 2: Submit your feedback

[0674] The server transmits the generated feedback data to the user.

[0675] Specific examples

[0676] 1. The server retrieves ideas from the Generative AI Utilization Contest Database, such as "an automatic translation system using AI," "an app that creates music using generative AI," and "a system that controls self-driving cars using generative AI."

[0677] 2. The server converts these ideas into tokens using a tokenizer, which are then fed into a generative AI model to generate a score:

[0678] Idea 1: Creativity 80 points, Feasibility 85 points, Impact 90 points

[0679] Idea 2: Creativity 90 points, Feasibility 80 points, Impact 85 points

[0680] Idea 3: Creativity 85, Feasibility 90, Impact 80

[0681] 3. When users provide feedback on these ideas, the device uses an emotion engine to analyze the feedback in real time and generate emotion data based on the analysis results. For example, if the user shows strong interest in idea 2, the emotion analysis may increase the evaluation score.

[0682] 4. The server recalculates the score taking into account the emotional data and sends the top three ideas to the user as feedback data.

[0683] This series of processes realizes an advanced evaluation system that combines generative AI and emotion recognition.

[0684] Example 2

[0685] 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."

[0686] Conventional idea evaluation systems have limited evaluation criteria, making it difficult to provide flexible evaluations that reflect user sentiment. Furthermore, the accuracy and reliability of the evaluations are lacking, making it difficult to accurately identify the best ideas. This has led to the problem that many excellent ideas are not properly evaluated and end up being buried.

[0687] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0688] In this invention, the server includes means for collecting previously submitted information, means for loading an AI model and setting evaluation criteria, means for inputting the collected information into the AI ​​model and scoring the information, means for analyzing the user's emotions, means for recalculating the score based on the emotion data, means for aggregating the scores and sorting them in descending order, and means for feeding back information on the high scores to the user. This enables flexible and accurate evaluation that reflects the user's emotions.

[0689] "Previously Submitted Information" means data, such as ideas or suggestions, previously submitted by users or other interested parties.

[0690] An "artificial intelligence model" is a system that analyzes and scores data using machine learning algorithms trained for a specific purpose.

[0691] "Metrics" are the indicators or standards used to evaluate the quality or usefulness of information or ideas.

[0692] "Means for analyzing user emotions" refers to software or devices that analyze emotions and impressions based on feedback and reactions from users.

[0693] "Emotional data" refers to information about the analyzed user's emotions and reactions.

[0694] "Means for recalculating scores" means software or computational algorithms for reevaluating existing scores based on emotion data and generating updated scores.

[0695] "Means for aggregating scores and sorting by highest score" refers to software or algorithms that calculate the score for each piece of information and sort it in descending order.

[0696] The "means for feeding back information with high scores to the user" refers to a method or means for reporting information with high scores from among the sorted information to the user.

[0697] MODE FOR CARRYING OUT THE INVENTION

[0698] This invention relates to a system that uses generative AI to collect and evaluate previously submitted information and select the best information, and also combines it with an emotion engine that recognizes the user's emotions. This system can realize consistent evaluations that also reflect the user's emotions.

[0699] First, the server accesses the AI-powered contest database and collects previously submitted information. The software used here can be a database management system (e.g., MySQL or PostgreSQL). The collected information is internally converted into a dictionary format (e.g., Python dictionary data structure) for efficient management.

[0700] Next, the server loads the generative AI model and tokenizer to be used for evaluation. Specifically, it uses a natural language processing algorithm (e.g., GPT-3 or BERT). The tokenizer (e.g., Hugging Face's tokenizer) is used to properly tokenize the information text data. At this stage, the evaluation criteria are set: creativity, feasibility, and impact.

[0701] The server then converts the text of each piece of collected information into tokens using a tokenizer and feeds them into a generative AI model. The generative AI model scores each piece of information and generates a score based on the set evaluation criteria. These scores are stored in a temporary data store (e.g., Redis).

[0702] Furthermore, this system uses an emotion engine to recognize the user's emotions in real time. When the user provides feedback on information, the terminal inputs the feedback into the emotion engine (e.g., emotion analysis model Sentiment 140) for analysis. Emotion data based on the analysis results is sent back to the server and reflected in the evaluation of the information.

[0703] The server recalculates the scores based on the emotion data, aggregates these scores, and sorts them in order of highest score. Finally, feedback data is generated to provide the top ranking information to the user. This data is sent to the user's device via API or email. This process allows the user to check the details of the highly rated information and to identify areas for future improvement based on the emotion data obtained from the emotion engine.

[0704] Specific examples

[0705] For example, if the following information is submitted:

[0706] 1. "AI-based automatic translation system"

[0707] 2. "An app that creates music using generative AI"

[0708] 3. "System for controlling self-driving cars using generative AI"

[0709] The server collects this information from the database, converts it into tokens using a tokenizer, and inputs them into the generative AI model, which generates the following scores for each piece of information:

[0710] Information 1: Creativity 80 points, Feasibility 85 points, Impact 90 points

[0711] Information 2: Creativity 90 points, Feasibility 80 points, Impact 85 points

[0712] Information 3: Creativity 85 points, Feasibility 90 points, Impact 80 points

[0713] When the user provides feedback on this information, the device uses an emotion engine to analyze the user's emotions in real time. For example, if the user shows strong interest in information 2, the evaluation score for information 2 may increase based on the emotion analysis results.

[0714] Finally, the server recalculates the score taking into account the emotional data and sends the top three scores to the user as feedback data, realizing an advanced evaluation system that combines generative AI and emotional recognition.

[0715] Prompt Sentence Examples

[0716] To facilitate information evaluation in an AI-based machine translation system, the following prompts could be input to the generative AI model:

[0717] "AI-based automatic translation system" is a system that enables automatic translation between multiple languages. Please rate this idea in terms of creativity, feasibility, and impact.

[0718] For the emotion engine, the following prompt is an example:

[0719] Analyze user feedback on your AI-powered machine translation system to identify emotions such as interest, delight, and surprise.

[0720] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0721] Step 1: Obtaining collected information

[0722] The server collects previously submitted information. The input for this process is the database connection information and query. Specifically, the server executes an SQL query to extract past information, parses the retrieved information into JSON format, and then converts it into an internal dictionary format. The output is the information data converted into dictionary format.

[0723] Step 2: Load the generative AI model and tokenizer

[0724] The server loads the generative AI model and tokenizer. The input to this process is the path to the model and tokenizer. Specifically, the server imports a machine learning library (e.g., TensorFlow, PyTorch) and loads the model file and tokenizer. The output is the loaded generative AI model and tokenizer.

[0725] Step 3: Set the evaluation criteria

[0726] The server sets the criteria for evaluation. The input of this process is the setting information of the evaluation criteria (creativity, feasibility, impact). In concrete terms, the server reads the evaluation criteria information from a setting file or database and sets these criteria for the model. The output is the set evaluation criteria.

[0727] Step 4: Tokenize your idea data

[0728] The server converts the collected text data into tokens using a tokenizer. The inputs to this process are the information data converted into dictionary format and the loaded tokenizer. Specifically, each piece of text data is passed to the tokenizer and converted into tokens. The output is the tokenized data.

[0729] Step 5: Score your idea data

[0730] The server inputs the tokenized data into the generative AI model and calculates the score. The inputs for this process are the tokenized data and the loaded generative AI model. Specifically, it inputs a prompt sentence into the generative AI model and performs scoring. The output is the calculated score.

[0731] Step 6: User feedback analysis

[0732] The terminal analyzes the feedback provided by the user using an emotion engine. The input of this process is the user's feedback text. Specifically, the feedback text is input to the emotion engine (e.g., Sentiment 140) to obtain emotion data. The output is the analyzed emotion data.

[0733] Step 7: Recalculate the score using emotion data

[0734] The server recalculates the score based on the obtained emotion data. The inputs to this process are the original score and the analyzed emotion data. Specifically, the server updates the score using recalculation logic to generate a final score. The output is the recalculated final score.

[0735] Step 8: Count and sort the scores

[0736] The server tally the final scores of each piece of information and sorts them in descending order. The input to this process is the recalculated final scores. Specific operations include using an algorithm to sort the scores in descending order and listing the top pieces of information. The output is a sorted list of information.

[0737] Step 9: User Feedback

[0738] The server feeds back the top-level information to the user. The input to this process is a sorted information list. Specific operations include generating feedback data and sending it to the user's device via a communication method such as an API or email. The output is the feedback data sent to the user.

[0739] (Application example 2)

[0740] 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."

[0741] Conventional idea evaluation systems using generative AI have the problem of being unable to fully reflect user opinions and emotions, making it difficult to select truly valuable ideas. Particularly for ideas for advertising campaigns, where user emotions are an important evaluation factor, improvement in this area was required.

[0742] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting previously submitted ideas, means for loading a generative AI model and setting evaluation criteria, means for inputting the collected ideas into the generative AI model and scoring the ideas, means for receiving feedback from users and analyzing them with a sentiment analysis engine, means for re-evaluating ideas based on the sentiment analysis results and calculating a final score, and means for feeding back high-scoring ideas to the user. This enables evaluation that takes into account the user's emotions, making it possible to select more appropriate ideas.

[0743] "Means of collection" refers to the means of acquiring ideas that have been submitted in the past and storing them in a database.

[0744] A "generative AI model" is an AI model that uses natural language processing algorithms to analyze and evaluate text data.

[0745] "Evaluation criteria" are the standards set to evaluate ideas for their creativity, feasibility, and impact.

[0746] A "means for scoring" is a means for analyzing ideas input into a generative AI model and assigning a score based on set evaluation criteria.

[0747] The "emotion analysis engine" is an engine that analyzes feedback from users and generates emotion data based on that feedback.

[0748] The "re-evaluation means" is a means for adjusting the initial evaluation score and calculating the final score in consideration of the sentiment analysis results.

[0749] The "means of providing feedback" is a means of reporting high-scoring ideas and their details to the user.

[0750] A system for implementing this invention includes a server, a sentiment analysis engine, and a terminal, and uses a generative AI model to evaluate and select ideas for an advertising campaign.

[0751] System program generation

[0752] System Overview

[0753] The server collects previously submitted ideas and sets evaluation criteria. Next, the collected ideas are input into a generative AI model and scored. It also receives feedback from users, analyzes it in real time using a sentiment analysis engine, and generates emotional data. The ideas are then re-evaluated based on the emotional data and a final score is calculated. Finally, the ideas with the highest scores are fed back to the user.

[0754] Hardware and Software

[0755] Hardware: Smartphones, servers

[0756] software:

[0757] Hugging Face's Transformers library: uses a pre-trained BERT model and tokenizer

[0758] VADER (Valence Aware Dictionary and sEntiment Reasoner): for sentiment analysis

[0759] OpenAI API: Text analysis and evaluation using generative AI models

[0760] Data processing and calculation

[0761] 1. Tokenization of ideas

[0762] The server converts the text data of the idea into tokens using Hugging Face's tokenizer.

[0763] 2. Idea scoring

[0764] The server inputs the tokenized idea data into the BERT model and scores it based on the criteria of creativity, feasibility, and impact.

[0765] 3. Sentiment Analysis of Feedback

[0766] The device collects feedback from users in real time and performs sentiment analysis using VADER to generate a sentiment score.

[0767] 4. Calculation of Final Score

[0768] The server calculates a weighted average of the idea score and the emotion score to arrive at a final score.

[0769] 5. Providing Feedback

[0770] The server reports the ideas to the user sorted by highest score.

[0771] Specific examples

[0772] Below are some examples of specific ideas and feedback prompts:

[0773] Example ideas:

[0774] 1. "A video advertising campaign highlighting the features of a new smartphone"

[0775] 2. "Personalized advertising using generative AI"

[0776] 3. "Advertisements that recommend products that match the user's preferences"

[0777] Feedback example:

[0778] 1. "I think this is a really great idea!"

[0779] 2. "I'm not interested."

[0780] 3. "It's an interesting idea, but it seems difficult to implement."

[0781] Example prompt:

[0782] Prompt: Generate an advertising campaign based on the idea.

[0783] Example: A video ad campaign highlighting the features of a new smartphone

[0784] Feedback: I think this is a really great idea!

[0785] The above system makes it possible to evaluate ideas by incorporating the user's emotions, allowing for the selection of more appropriate ideas.

[0786] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0787] Step 1:

[0788] The server collects previously submitted ideas from a database. Specifically, the server accesses the database and retrieves idea data. The input at this stage is the idea data in the database, and the output is a list of collected ideas.

[0789] Step 2:

[0790] The server converts the collected ideas into tokens using Hugging Face's tokenizer. The tokenization process converts text data into a set of tokens. The input of this step is the collected idea list, and the output is the tokenized idea data.

[0791] Step 3:

[0792] The server inputs the tokenized idea data into the BERT model and scores them based on the criteria of creativity, feasibility, and impact. Here, the input is the tokenized idea data, and the output is a score for each idea.

[0793] Step 4:

[0794] The server collects feedback from users through their terminals. Users provide feedback on ideas, and the data is sent to the server. The input of this step is the feedback from users, and the output is the collected feedback data.

[0795] Step 5:

[0796] The device uses VADER to perform real-time sentiment analysis of the collected feedback data and generate a sentiment score. Specifically, the feedback text is input into the sentiment analysis engine, and the analysis results are output as a positive, negative, or neutral score. The input of this step is the user's feedback data, and the output is a sentiment score.

[0797] Step 6:

[0798] The server calculates the final score by weighting the initial evaluation score and the sentiment score. The inputs at this stage are the idea score and the sentiment score, and the output is the re-evaluated final score.

[0799] Step 7:

[0800] The server sorts the ideas by final score and feeds back the ideas with the highest scores to the user. Specifically, the server selects the top sorted ideas and notifies the user of their details. The input of this step is the re-evaluated final score, and the output is the feedback data of the highly rated ideas.

[0801] 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.

[0802] 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.

[0803] 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.

[0804] [Third embodiment]

[0805] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0806] 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.

[0807] 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).

[0808] 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.

[0809] 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.

[0810] 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).

[0811] 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.

[0812] 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.

[0813] 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.

[0814] 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.

[0815] 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.

[0816] 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."

[0817] MODE FOR CARRYING OUT THE INVENTION

[0818] The present invention relates to a system that utilizes generative AI to collect and evaluate previously submitted ideas and select the best idea. This system is configured using the following means.

[0819] 1. Collecting idea data

[0820] The server accesses the database of the Generative AI Contest and queries and retrieves previously submitted idea data. The retrieved data is then converted into an internal dictionary format. This operation allows the system to efficiently manage ideas to be evaluated.

[0821] 2. Preparing the Generative AI

[0822] The server loads the generative AI model (e.g., natural language processing model) to be used for evaluation, including the appropriate tokenizer to convert the text data of the idea into a format that can be evaluated, and sets the evaluation criteria, such as creativity, feasibility, and impact.

[0823] 3. Evaluate ideas

[0824] The server converts the collected text of each idea into tokens using a tokenizer, then inputs them into a generative AI model to score the idea. Specifically, a score is generated for each idea based on the evaluation criteria of creativity, feasibility, and impact. This process allows for a relative evaluation of each idea.

[0825] 4. Score tallying

[0826] The server tally the scores of each idea and sort them by highest score. This allows the best ideas to be identified at a glance. The higher the score, the better each idea is compared to the set evaluation criteria.

[0827] 5. Feedback of results

[0828] The server then provides feedback to the user on the ideas with the highest scores. Specifically, the top few ideas are selected and notified to the user. This feedback allows the user to check the details of the highly rated ideas and refer to the evaluation results.

[0829] Natural language explanation of the process

[0830] First, the server accesses the database of generative AI contests to collect ideas that have been submitted in the past. The collected ideas are then converted into an internal dictionary format, which allows for efficient data management.

[0831] The server then loads the generative AI model for evaluation, which includes natural language processing algorithms and a tokenizer for properly processing text data, and sets evaluation criteria such as creativity, feasibility, and impact.

[0832] The server then converts each idea's text into tokens using a tokenizer and feeds them into a generative AI model, which scores the ideas and generates a score for each idea based on a set of evaluation criteria.

[0833] After each idea has a score generated, the server aggregates it and sorts it by highest score, clearly identifying the best ideas.

[0834] Finally, the server provides feedback on the top ideas to the user, who can then check the details of the highly rated ideas and use the evaluation results as a reference.

[0835] Specific examples

[0836] For example, the following idea is submitted:

[0837] 1. "AI-based automatic translation system"

[0838] 2. "An app that creates music using generative AI"

[0839] 3. "System for controlling self-driving cars using generative AI"

[0840] The server collects these ideas from a database and uses a generative AI model to score them based on evaluation criteria. For example, the following scores might be generated:

[0841] Idea 1: Creativity 80 points, Feasibility 85 points, Impact 90 points

[0842] Idea 2: Creativity 90 points, Feasibility 80 points, Impact 85 points

[0843] Idea 3: Creativity 85, Feasibility 90, Impact 80

[0844] The server aggregates these scores and sorts them by highest score, and the highest scoring idea is then fed back to the user.

[0845] In this way, the system of the present invention utilizes generative AI to efficiently and objectively evaluate ideas and select and present excellent ideas.

[0846] The processing flow will be explained below.

[0847] Specific processing flow of the program

[0848] Idea data collection

[0849] Step 1: Query the database

[0850] The server queries the generative AI contest database to retrieve all submitted ideas.

[0851] Step 2: Transform the data

[0852] The server converts the acquired idea data into an internal dictionary format so that it can be managed efficiently.

[0853] Preparing for generative AI

[0854] Step 1: Loading the Generative AI Model

[0855] The server loads generative AI models (natural language processing algorithms) for evaluation.

[0856] Step 2: Preparing the tokenizer

[0857] The server also loads the tokenizer corresponding to the generative AI model, preparing it to properly tokenize the idea text data.

[0858] Step 3: Set evaluation criteria

[0859] The server sets the criteria for creativity, feasibility, and impact.

[0860] Idea Evaluation

[0861] Step 1: Tokenize your ideas

[0862] The server converts the text of each idea into tokens using a tokenizer.

[0863] Step 2: Score with the model

[0864] The server inputs each tokenized idea into a generative AI model and calculates a score for each.

[0865] Step 3: Create a score list

[0866] The server stores the scores generated for each idea in the form of a list.

[0867] Score tally

[0868] Step 1: Sort the scores

[0869] The server sorts all ideas by their scores, from highest to lowest.

[0870] Step 2: Extracting top ideas

[0871] The server extracts the top few ideas (e.g., the top three).

[0872] Feedback of results

[0873] Step 1: Generate feedback data

[0874] The server generates feedback data including details of the top ranked ideas.

[0875] Step 2: Submit your feedback

[0876] The server transmits the generated feedback data to the user.

[0877] For example, the following processing is performed:

[0878] 1. The server retrieves three ideas from the database: an automatic translation system using AI, an app that creates music using generative AI, and a system that controls self-driving cars using generative AI.

[0879] 2. The server converts these ideas into tokens using a tokenizer, which are then fed into a generative AI model to generate the following score:

[0880] Idea 1: Creativity 80 points, Feasibility 85 points, Impact 90 points

[0881] Idea 2: Creativity 90 points, Feasibility 80 points, Impact 85 points

[0882] Idea 3: Creativity 85, Feasibility 90, Impact 80

[0883] 3. The server aggregates these scores, sorts them by highest score, and sends the top three ideas to the user as feedback data.

[0884] This series of processes creates a system that uses generative AI to objectively and efficiently evaluate ideas and select the best ones.

[0885] Example 1

[0886] 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."

[0887] Existing systems have had problems in efficiently and objectively evaluating previously submitted concepts and selecting the best ones. In particular, when processing large amounts of concept data, issues often arise in setting evaluation criteria and consistency of evaluation. It has also been difficult to provide users with prompt and accurate feedback.

[0888] 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.

[0889] In this invention, the server includes means for collecting previously submitted concepts, means for loading a generative AI model and setting evaluation criteria, means for inputting the collected concepts into the generative AI model and scoring the concepts, means for tokenizing the text of each concept, means for aggregating the scores and sorting them in descending order of score, and means for feeding back the concepts with high scores to the user. This enables efficient and objective evaluation of concepts and rapid selection and feedback of excellent concepts.

[0890] "Previously Submitted Ideas" means ideas or suggestions previously submitted by you or a third party.

[0891] A "generative artificial intelligence model" refers to an algorithm or system that uses artificial intelligence technology to perform natural language processing and predictive analysis.

[0892] "Evaluation criteria" refers to the specific measures or criteria for evaluating ideas, including creativity, feasibility, and impact.

[0893] "Scoring" refers to the process of assigning a score or rating to an initiative based on established evaluation criteria.

[0894] "Tokenization" refers to the process of breaking down text data into words and phrases for processing.

[0895] "Feedback" refers to the action or process of returning information such as evaluation results or notifications to users.

[0896] "User" refers to any person or entity that uses this system to receive the results of an initiative evaluation.

[0897] MODE FOR CARRYING OUT THE INVENTION

[0898] This invention relates to a system that utilizes a generative artificial intelligence model to collect and evaluate previously submitted concepts and select the best concept. This system operates in cooperation with a server, terminals, and users.

[0899] First, the server accesses the database of the Generative AI Contest and collects previously submitted concepts. The server retrieves the data using queries such as SQL and converts the retrieved concept data into an internal dictionary format. This data conversion allows the server to efficiently manage the data and perform subsequent processing.

[0900] The server then loads the generative AI model (e.g., BERT or GPT-3) to be used for evaluation. This generative AI model is an algorithm that uses generative artificial intelligence technology and includes a tokenizer to process the text data. The tokenizer breaks the text data down into tokens and converts them into a format that can be fed into the generative AI model. The server then sets the criteria for evaluating the ideas: creativity, feasibility, and impact.

[0901] Specifically, the server converts the text of each idea into tokens using a tokenizer, and then inputs these into a generative AI model. The generative AI model generates a score for each idea based on the evaluation criteria of creativity, feasibility, and impact. This scoring process provides a relative evaluation of each idea.

[0902] After the scores are generated, the server aggregates them and sorts them by highest score, clearly identifying the best ideas. The server then provides feedback to the user about the highly scored ideas. Specifically, the server notifies the user of the top few highly rated ideas. This feedback allows the user to view the details of the highly rated ideas and refer to the evaluation results.

[0903] Specific examples

[0904] For example, a user might submit the following idea to the database:

[0905] 1. "AI-based automatic translation system"

[0906] 2. "An app that creates music using generative AI"

[0907] 3. "System for controlling self-driving cars using generative AI"

[0908] The server collects these ideas from a database and uses a generative AI model to score them based on evaluation criteria, for example, the following scores are generated:

[0909] Idea 1: Creativity 80 points, Feasibility 85 points, Impact 90 points

[0910] Idea 2: Creativity 90 points, Feasibility 80 points, Impact 85 points

[0911] Idea 3: Creativity 85, Feasibility 90, Impact 80

[0912] The server aggregates these scores and sorts them by highest score, and the highest scoring ideas are then fed back to the user.

[0913] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0914] Program processing flow

[0915] Step 1: Collect idea data

[0916] The server accesses the database of the Generative AI Contest and queries and retrieves previously submitted idea data. The input is the query result from the database, and the output is the idea data converted into dictionary format.

[0917] Specifically, it executes the SQL query "SELECT FROM ideas WHERE submission_date > '2022-01-01'" and converts the results into a Python dictionary, where each idea is managed as a key-value pair. For example, you might get a data structure like this:

[0918] python

[0919] ideas = {

[0920] "idea1": {"title": "AI-based automatic translation system", "submitted_by": "user123", "date": "2023-05-10"},

[0921] "idea2": {"title": "App to create music with generative AI", "submitted_by": "user456", "date": "2023-06-12"},

[0922] }

[0923] Step 2: Prepare the generative AI

[0924] The server loads the generative AI model to be used for evaluation and sets the evaluation criteria. The input is the generative AI model and tokenizer parameters, and the output is the loaded model and the set evaluation criteria. Specifically, it imports libraries such as "from transformers import GPT-3" and creates an instance of the model. It also creates an instance of the tokenizer:

[0925] python

[0926] tokenizer = GPT3Tokenizer.from_pretrained('gpt-3')

[0927] model = GPT3Model.from_pretrained('gpt-3')

[0928] Set the criteria for evaluation: creativity, feasibility, and impact:

[0929] python

[0930] evaluation_criteria = ["creativity", "feasibility", "impact"]

[0931] Step 3: Tokenize your ideas

[0932] The server converts the text of each idea into tokens using a tokenizer. The input is the text data of each idea, and the output is the tokenized text data. Specifically, it converts into tokens as follows:

[0933] python

[0934] tokens = tokenizer.encode(ideas["idea1"]["title"], return_tensors='pt')

[0935] This converts the text of the idea into a format that can be input into a generative AI model.

[0936] Step 4: Evaluate your ideas

[0937] The server inputs the tokenized data into a generative AI model to score ideas. The input is tokenized text data, and the output is a score for each idea. Specifically, the data is input into the model as follows:

[0938] python

[0939] scores = model(tokens)

[0940] You will receive a score based on each criterion:

[0941] python

[0942] scores = {"creativity": 80, "feasibility": 85, "impact": 90}

[0943] Step 5: Counting the scores

[0944] The server aggregates the scores of each idea and sorts them in descending order. The input is the score of each idea, and the output is a list of ideas sorted in descending order. Specifically, the server aggregates and sorts the scores as follows:

[0945] python

[0946] sorted_ideas = sorted(ideas.items(), key=lambda x: x[1]['scores']['total'], reverse=True)

[0947] This allows the best ideas to rise to the top.

[0948] Step 6: Feedback on results

[0949] The server feeds back the highest-scoring ideas to the user. The input is the high-scoring ideas, and the output is a feedback message to the user. Specifically, it sends the following feedback message to the user's device:

[0950] python

[0951] feedback_message = f"The idea with the highest score is "{sorted_ideas[0]['title']}". Learn more here."

[0952] send_feedback_to_user(user_id, feedback_message)

[0953] In this way, the system of the present invention can efficiently and objectively evaluate ideas and quickly present excellent ideas to the user.

[0954] (Application example 1)

[0955] 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."

[0956] In existing factory automation systems, the evaluation and selection of new efficiency improvement measures and automation ideas is often done manually, making objective and efficient evaluation difficult. Furthermore, it is difficult to quickly find the most suitable idea from among many ideas and put it into practice. Given this background, there is a demand for a faster and more reliable idea evaluation system in workplaces seeking innovation and efficiency improvements in factory automation.

[0957] 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.

[0958] In this invention, the server includes means for collecting previously submitted ideas, means for loading a generative AI model and setting evaluation criteria, means for inputting the collected ideas into the generative AI model and scoring the ideas, means for aggregating the scores and sorting them in descending order of score, means for providing feedback on the ideas with the highest scores to the user, and means for applying the idea evaluation results to the factory automation system and providing optimal solutions for improving efficiency. This enables objective and efficient evaluation of new efficiency improvement measures and automation ideas in the factory automation system, and makes it possible to quickly find and implement the most suitable ideas.

[0959] A "server" is an information processing device connected to a computer network, which collects, analyzes, processes, and provides data.

[0960] "Means for collecting idea data" refers to devices or algorithms that retrieve previously submitted ideas from databases, etc., and manage them appropriately.

[0961] A "generative AI model" refers to an artificial intelligence program that uses natural language processing algorithms and other techniques to analyze input data and perform specific tasks.

[0962] "Means for setting evaluation criteria" refers to devices or methods for defining and setting criteria for evaluation based on the creativity, feasibility, impact, etc. of ideas.

[0963] "Idea scoring means" refers to a device or method that uses a generative AI model to evaluate collected ideas and assign scores based on each evaluation criterion.

[0964] "Means for aggregating scores and sorting ideas by highest score" refers to a device or algorithm that aggregates the scores assigned to ideas and orders the ideas based on that aggregated score.

[0965] "Means for providing feedback to the user" refers to a device or method for notifying the user of the ideas with high scores and communicating the results.

[0966] "Factory automation system" is a general term for hardware and software that automates manufacturing processes and operations within a factory to improve efficiency.

[0967] "Means for providing optimal solutions" refers to devices and methods that are proposed and introduced based on the evaluation results to improve factory efficiency and optimize automation.

[0968] The system for implementing the present invention involves a process of collecting previously submitted ideas, evaluating them using a generative AI model, and providing optimal solutions to a factory automation system based on the evaluation results. A specific embodiment of the system is shown below.

[0969] First, the server accesses the database and collects previously submitted ideas. This idea data is converted into an internal dictionary format and efficiently managed.

[0970] The server then loads a generative AI model for evaluation. The model includes a natural language processing algorithm and uses a tokenizer to convert the text data of the ideas into a format that can be evaluated. The server also sets evaluation criteria such as creativity, feasibility, and impact.

[0971] Each collected idea text is converted into tokens by a tokenizer and then input into a generative AI model. The generative AI model evaluates the input ideas and generates a score for each idea based on the evaluation criteria of creativity, feasibility, and impact. The server aggregates these scores and sorts them by highest score, allowing it to identify superior ideas.

[0972] The server then provides feedback on the highly scored ideas to users, such as factory managers. Specifically, the top few ideas are selected and notified to the users. This feedback allows users to check the details of the highly rated ideas and refer to the evaluation results.

[0973] Furthermore, the most highly evaluated ideas will be applied to factory automation systems to provide optimal solutions for improving efficiency, thereby optimizing factory automation processes and increasing productivity.

[0974] For example, the following prompts can be fed into a generative AI model to evaluate ideas:

[0975] Prompt statement:

[0976] "Please rate the following ideas:

[0977] Idea: "Inventory management system using unmanned robots"

[0978] Evaluation criteria: Creativity, feasibility, impact.”

[0979] The expected output is, for example, "80, 90, 85", which are scores based on each evaluation criterion.

[0980] Based on the evaluation results, the server provides optimal solutions to the factory automation system, improving factory efficiency.

[0981] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0982] Step 1:

[0983] The server accesses a database to collect previously submitted ideas. The input is a database query, and the output is a list of ideas. This list is converted into an internal dictionary format and efficiently managed.

[0984] Step 2:

[0985] The server loads the generative AI model to be used for evaluation. At this stage, the generative AI model containing a specific natural language processing algorithm is loaded. The input is an instruction to load the AI ​​model, and the output is an AI model ready for evaluation.

[0986] Step 3:

[0987] The server sets the evaluation criteria, specifically defining criteria such as creativity, feasibility, and impact. The input is the evaluation criteria setting, and the output is the set evaluation criteria.

[0988] Step 4:

[0989] Each piece of text in the collected idea data is converted into tokens by a tokenizer. The input to this step is the text data of the idea, and the output is the tokenized data. The tokenizer analyzes the text and divides it into small units that can be analyzed.

[0990] Step 5:

[0991] The server uses a generative AI model to evaluate the tokenized ideas. In this step, the tokenized ideas are input into the generative AI model, and a score based on each evaluation criterion is generated as output. Specifically, the AI ​​model performs semantic analysis of the text and calculates a score for each evaluation criterion.

[0992] Step 6:

[0993] The server aggregates the scores obtained from the evaluation and sorts them in descending order. The input is the score of each idea, and the output is a list of ideas sorted in descending order. The server adds up the scores and sorts the list in descending order.

[0994] Step 7:

[0995] The server feeds back the highly scored ideas to the user. Here, the top few ideas are selected and their details are notified to the user. The input is a list of highly scored ideas, and the output is a notification of the fed back ideas. Specifically, the server sends a notification to the user's device.

[0996] Step 8:

[0997] Based on the feedback, the user applies the optimal solution to the factory automation system. The input of this step is the feedback received by the user, and the output is the applied automation solution. Specifically, the user implements the selected idea to optimize the factory automation process.

[0998] 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.

[0999] MODE FOR CARRYING OUT THE INVENTION

[1000] This invention relates to a system that uses generative AI to collect and evaluate previously submitted ideas and select the best idea, and further combines it with an emotion engine that recognizes the user's emotions. This system is implemented as follows:

[1001] First, the server accesses the database of generative AI contests and collects data on ideas submitted in the past. The collected data is then converted into an internal dictionary format and efficiently managed.

[1002] Next, the server loads the generative AI model to be used for evaluation, prepares the tokenizer, which is used to properly tokenize the text data of the idea, and sets the evaluation criteria of creativity, feasibility, and impact.

[1003] The server then converts the collected text of each idea into tokens using a tokenizer and feeds them into a generative AI model, which scores each idea and generates a score for each idea based on the evaluation criteria set.

[1004] Furthermore, this system uses an emotion engine to recognize the user's emotions in real time. When the user provides feedback on an idea, the device inputs the feedback into the emotion engine for analysis. Emotion data based on the analysis results is reflected in the evaluation of the idea.

[1005] The server then recalculates a final score for each idea, taking into account the emotional data obtained by the emotion engine. The scores obtained in this way are aggregated and sorted by highest score.

[1006] Finally, the server provides feedback on the top ideas to the user. Feedback data containing details of the top ideas is generated and sent to the user. This allows the user to check the details of the highly rated ideas and to identify areas for future improvement based on the emotional data obtained from the emotion engine.

[1007] Natural language explanation of the process

[1008] First, the server accesses a database to retrieve previously submitted ideas, which are then converted into an internal dictionary format.

[1009] The server then loads the generative AI model and tokenizer, ready to evaluate each idea based on the set criteria, which include creativity, feasibility, and impact.

[1010] The server then converts the text of each idea into tokens using a tokenizer, feeds them into a generative AI model, and calculates a score, which indicates how good each idea is.

[1011] Next, when receiving feedback from users, they input the feedback into the emotion engine using their terminal. The emotion engine analyzes the user's emotions in real time and obtains the results. This emotion data is incorporated into the score so that it is also reflected in the evaluation of the idea.

[1012] The final scores are then tallied and sorted by highest score, thereby identifying the best ideas.

[1013] Finally, the server provides feedback on the top ideas to the user, who can then check the details of the highly rated ideas and refer to the evaluation results and the analysis results of the emotion engine.

[1014] Specific examples

[1015] For example, the following idea is submitted:

[1016] 1. "AI-based automatic translation system"

[1017] 2. "An app that creates music using generative AI"

[1018] 3. "System for controlling self-driving cars using generative AI"

[1019] The server collects these ideas from the database, converts them into tokens using a tokenizer, and feeds them into a generative AI model, which generates a score for each idea, such as:

[1020] Idea 1: Creativity 80 points, Feasibility 85 points, Impact 90 points

[1021] Idea 2: Creativity 90 points, Feasibility 80 points, Impact 85 points

[1022] Idea 3: Creativity 85, Feasibility 90, Impact 80

[1023] When users provide feedback on these ideas, the device uses an emotion engine to analyze their emotions in real time and reflects the analysis results in the score. For example, if a user shows strong interest in idea 2, the emotion analysis may increase the evaluation score.

[1024] Finally, the server recalculates the score taking into account the emotion data and sends the top three ideas as feedback to the user, realizing an advanced evaluation system that combines generative AI and emotion recognition.

[1025] The processing flow will be explained below.

[1026] MODE FOR CARRYING OUT THE INVENTION

[1027] Idea data collection

[1028] Step 1: Query the database

[1029] The server queries the generative AI contest database to retrieve all previously submitted idea data.

[1030] Step 2: Transform the data

[1031] The server converts the acquired idea data into an internal dictionary format and manages it efficiently.

[1032] Preparing for generative AI

[1033] Step 1: Loading the Generative AI Model

[1034] The server loads generative AI models (natural language processing algorithms) for evaluation.

[1035] Step 2: Preparing the tokenizer

[1036] The server loads the generative AI model and the corresponding tokenizer, preparing it to properly tokenize the idea text data.

[1037] Step 3: Set evaluation criteria

[1038] The server sets the criteria for evaluation: creativity, feasibility, and impact.

[1039] Idea Evaluation

[1040] Step 1: Tokenize your ideas

[1041] The server converts the text of each collected idea into tokens using a tokenizer.

[1042] Step 2: Score with the model

[1043] The server inputs each tokenized idea into a generative AI model and calculates a score for each.

[1044] Step 3: Create a score list

[1045] The server stores the scores generated for each idea in the form of a list.

[1046] Using the Emotion Engine

[1047] Step 1: Accepting user feedback

[1048] The device receives feedback from the user, including detailed thoughts on the ideas that the user has rated.

[1049] Step 2: Analysis by the emotion engine

[1050] The device inputs the received feedback into an emotion engine and analyzes the user's emotions in real time.

[1051] Step 3: Generate emotion data

[1052] The terminal generates emotion data of the user based on the analysis results.

[1053] Step 4: Reflecting emotional data

[1054] The server reflects the generated emotion data in the score of the idea and updates the evaluation result.

[1055] Score tally

[1056] Step 1: Sort the scores

[1057] The server sorts the final scores from highest to lowest.

[1058] Step 2: Extracting top ideas

[1059] The server extracts the top few ideas (e.g., the top three).

[1060] Feedback of results

[1061] Step 1: Generate feedback data

[1062] The server generates feedback data including details of the top ideas.

[1063] Step 2: Submit your feedback

[1064] The server transmits the generated feedback data to the user.

[1065] Specific examples

[1066] 1. The server retrieves ideas from the Generative AI Utilization Contest Database, such as "an automatic translation system using AI," "an app that creates music using generative AI," and "a system that controls self-driving cars using generative AI."

[1067] 2. The server converts these ideas into tokens using a tokenizer, which are then fed into a generative AI model to generate a score:

[1068] Idea 1: Creativity 80 points, Feasibility 85 points, Impact 90 points

[1069] Idea 2: Creativity 90 points, Feasibility 80 points, Impact 85 points

[1070] Idea 3: Creativity 85, Feasibility 90, Impact 80

[1071] 3. When users provide feedback on these ideas, the device uses an emotion engine to analyze the feedback in real time and generate emotion data based on the analysis results. For example, if the user shows strong interest in idea 2, the emotion analysis may increase the evaluation score.

[1072] 4. The server recalculates the score taking into account the emotional data and sends the top three ideas to the user as feedback data.

[1073] This series of processes realizes an advanced evaluation system that combines generative AI and emotion recognition.

[1074] Example 2

[1075] 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."

[1076] Conventional idea evaluation systems have limited evaluation criteria, making it difficult to provide flexible evaluations that reflect user sentiment. Furthermore, the accuracy and reliability of the evaluations are lacking, making it difficult to accurately identify the best ideas. This has led to the problem that many excellent ideas are not properly evaluated and end up being buried.

[1077] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1078] In this invention, the server includes means for collecting previously submitted information, means for loading an AI model and setting evaluation criteria, means for inputting the collected information into the AI ​​model and scoring the information, means for analyzing the user's emotions, means for recalculating the score based on the emotion data, means for aggregating the scores and sorting them in descending order, and means for feeding back information on the high scores to the user. This enables flexible and accurate evaluation that reflects the user's emotions.

[1079] "Previously Submitted Information" means data, such as ideas or suggestions, previously submitted by users or other interested parties.

[1080] An "artificial intelligence model" is a system that analyzes and scores data using machine learning algorithms trained for a specific purpose.

[1081] "Metrics" are the indicators or standards used to evaluate the quality or usefulness of information or ideas.

[1082] "Means for analyzing user emotions" refers to software or devices that analyze emotions and impressions based on feedback and reactions from users.

[1083] "Emotional data" refers to information about the analyzed user's emotions and reactions.

[1084] "Means for recalculating scores" means software or computational algorithms for reevaluating existing scores based on emotion data and generating updated scores.

[1085] "Means for aggregating scores and sorting by highest score" refers to software or algorithms that calculate the score for each piece of information and sort it in descending order.

[1086] The "means for feeding back information with high scores to the user" refers to a method or means for reporting information with high scores from among the sorted information to the user.

[1087] MODE FOR CARRYING OUT THE INVENTION

[1088] This invention relates to a system that uses generative AI to collect and evaluate previously submitted information and select the best information, and also combines it with an emotion engine that recognizes the user's emotions. This system can realize consistent evaluations that also reflect the user's emotions.

[1089] First, the server accesses the AI-powered contest database and collects previously submitted information. The software used here can be a database management system (e.g., MySQL or PostgreSQL). The collected information is internally converted into a dictionary format (e.g., Python dictionary data structure) for efficient management.

[1090] Next, the server loads the generative AI model and tokenizer to be used for evaluation. Specifically, it uses a natural language processing algorithm (e.g., GPT-3 or BERT). The tokenizer (e.g., Hugging Face's tokenizer) is used to properly tokenize the information text data. At this stage, the evaluation criteria are set: creativity, feasibility, and impact.

[1091] The server then converts the text of each piece of collected information into tokens using a tokenizer and feeds them into a generative AI model. The generative AI model scores each piece of information and generates a score based on the set evaluation criteria. These scores are stored in a temporary data store (e.g., Redis).

[1092] Furthermore, this system uses an emotion engine to recognize the user's emotions in real time. When the user provides feedback on information, the terminal inputs the feedback into the emotion engine (e.g., emotion analysis model Sentiment 140) for analysis. Emotion data based on the analysis results is sent back to the server and reflected in the evaluation of the information.

[1093] The server recalculates the scores based on the emotion data, aggregates these scores, and sorts them in order of highest score. Finally, feedback data is generated to provide the top ranking information to the user. This data is sent to the user's device via API or email. This process allows the user to check the details of the highly rated information and to identify areas for future improvement based on the emotion data obtained from the emotion engine.

[1094] Specific examples

[1095] For example, if the following information is submitted:

[1096] 1. "AI-based automatic translation system"

[1097] 2. "An app that creates music using generative AI"

[1098] 3. "System for controlling self-driving cars using generative AI"

[1099] The server collects this information from the database, converts it into tokens using a tokenizer, and inputs them into the generative AI model, which generates the following scores for each piece of information:

[1100] Information 1: Creativity 80 points, Feasibility 85 points, Impact 90 points

[1101] Information 2: Creativity 90 points, Feasibility 80 points, Impact 85 points

[1102] Information 3: Creativity 85 points, Feasibility 90 points, Impact 80 points

[1103] When the user provides feedback on this information, the device uses an emotion engine to analyze the user's emotions in real time. For example, if the user shows strong interest in information 2, the evaluation score for information 2 may increase based on the emotion analysis results.

[1104] Finally, the server recalculates the score taking into account the emotional data and sends the top three scores to the user as feedback data, realizing an advanced evaluation system that combines generative AI and emotional recognition.

[1105] Prompt Sentence Examples

[1106] To facilitate information evaluation in an AI-based machine translation system, the following prompts could be input to the generative AI model:

[1107] "AI-based automatic translation system" is a system that enables automatic translation between multiple languages. Please rate this idea in terms of creativity, feasibility, and impact.

[1108] For the emotion engine, the following prompt is an example:

[1109] Analyze user feedback on your AI-powered machine translation system to identify emotions such as interest, delight, and surprise.

[1110] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1111] Step 1: Obtaining collected information

[1112] The server collects previously submitted information. The input for this process is the database connection information and query. Specifically, the server executes an SQL query to extract past information, parses the retrieved information into JSON format, and then converts it into an internal dictionary format. The output is the information data converted into dictionary format.

[1113] Step 2: Load the generative AI model and tokenizer

[1114] The server loads the generative AI model and tokenizer. The input to this process is the path to the model and tokenizer. Specifically, the server imports a machine learning library (e.g., TensorFlow, PyTorch) and loads the model file and tokenizer. The output is the loaded generative AI model and tokenizer.

[1115] Step 3: Set the evaluation criteria

[1116] The server sets the criteria for evaluation. The input of this process is the setting information of the evaluation criteria (creativity, feasibility, impact). In concrete terms, the server reads the evaluation criteria information from a setting file or database and sets these criteria for the model. The output is the set evaluation criteria.

[1117] Step 4: Tokenize your idea data

[1118] The server converts the collected text data into tokens using a tokenizer. The inputs to this process are the information data converted into dictionary format and the loaded tokenizer. Specifically, each piece of text data is passed to the tokenizer and converted into tokens. The output is the tokenized data.

[1119] Step 5: Score your idea data

[1120] The server inputs the tokenized data into the generative AI model and calculates the score. The inputs for this process are the tokenized data and the loaded generative AI model. Specifically, it inputs a prompt sentence into the generative AI model and performs scoring. The output is the calculated score.

[1121] Step 6: User feedback analysis

[1122] The terminal analyzes the feedback provided by the user using an emotion engine. The input of this process is the user's feedback text. Specifically, the feedback text is input to the emotion engine (e.g., Sentiment 140) to obtain emotion data. The output is the analyzed emotion data.

[1123] Step 7: Recalculate the score using emotion data

[1124] The server recalculates the score based on the obtained emotion data. The inputs to this process are the original score and the analyzed emotion data. Specifically, the server updates the score using recalculation logic to generate a final score. The output is the recalculated final score.

[1125] Step 8: Count and sort the scores

[1126] The server tally the final scores of each piece of information and sorts them in descending order. The input to this process is the recalculated final scores. Specific operations include using an algorithm to sort the scores in descending order and listing the top pieces of information. The output is a sorted list of information.

[1127] Step 9: User Feedback

[1128] The server feeds back the top-level information to the user. The input to this process is a sorted information list. Specific operations include generating feedback data and sending it to the user's device via a communication method such as an API or email. The output is the feedback data sent to the user.

[1129] (Application example 2)

[1130] 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."

[1131] Conventional idea evaluation systems using generative AI have the problem of being unable to fully reflect user opinions and emotions, making it difficult to select truly valuable ideas. Particularly for ideas for advertising campaigns, where user emotions are an important evaluation factor, improvement in this area was required.

[1132] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting previously submitted ideas, means for loading a generative AI model and setting evaluation criteria, means for inputting the collected ideas into the generative AI model and scoring the ideas, means for receiving feedback from users and analyzing them with a sentiment analysis engine, means for re-evaluating ideas based on the sentiment analysis results and calculating a final score, and means for feeding back high-scoring ideas to the user. This enables evaluation that takes into account the user's emotions, making it possible to select more appropriate ideas.

[1133] "Means of collection" refers to the means of acquiring ideas that have been submitted in the past and storing them in a database.

[1134] A "generative AI model" is an AI model that uses natural language processing algorithms to analyze and evaluate text data.

[1135] "Evaluation criteria" are the standards set to evaluate ideas for their creativity, feasibility, and impact.

[1136] A "means for scoring" is a means for analyzing ideas input into a generative AI model and assigning a score based on set evaluation criteria.

[1137] The "emotion analysis engine" is an engine that analyzes feedback from users and generates emotion data based on that feedback.

[1138] The "re-evaluation means" is a means for adjusting the initial evaluation score and calculating the final score in consideration of the sentiment analysis results.

[1139] The "means of providing feedback" is a means of reporting high-scoring ideas and their details to the user.

[1140] A system for implementing this invention includes a server, a sentiment analysis engine, and a terminal, and uses a generative AI model to evaluate and select ideas for an advertising campaign.

[1141] System program generation

[1142] System Overview

[1143] The server collects previously submitted ideas and sets evaluation criteria. Next, the collected ideas are input into a generative AI model and scored. It also receives feedback from users, analyzes it in real time using a sentiment analysis engine, and generates emotional data. The ideas are then re-evaluated based on the emotional data and a final score is calculated. Finally, the ideas with the highest scores are fed back to the user.

[1144] Hardware and Software

[1145] Hardware: Smartphones, servers

[1146] software:

[1147] Hugging Face's Transformers library: uses a pre-trained BERT model and tokenizer

[1148] VADER (Valence Aware Dictionary and sEntiment Reasoner): for sentiment analysis

[1149] OpenAI API: Text analysis and evaluation using generative AI models

[1150] Data processing and calculation

[1151] 1. Tokenization of ideas

[1152] The server converts the text data of the idea into tokens using Hugging Face's tokenizer.

[1153] 2. Idea scoring

[1154] The server inputs the tokenized idea data into the BERT model and scores it based on the criteria of creativity, feasibility, and impact.

[1155] 3. Sentiment Analysis of Feedback

[1156] The device collects feedback from users in real time and performs sentiment analysis using VADER to generate a sentiment score.

[1157] 4. Calculation of Final Score

[1158] The server calculates a weighted average of the idea score and the emotion score to arrive at a final score.

[1159] 5. Providing Feedback

[1160] The server reports the ideas to the user sorted by highest score.

[1161] Specific examples

[1162] Below are some examples of specific ideas and feedback prompts:

[1163] Example ideas:

[1164] 1. "A video advertising campaign highlighting the features of a new smartphone"

[1165] 2. "Personalized advertising using generative AI"

[1166] 3. "Advertisements that recommend products that match the user's preferences"

[1167] Feedback example:

[1168] 1. "I think this is a really great idea!"

[1169] 2. "I'm not interested."

[1170] 3. "It's an interesting idea, but it seems difficult to implement."

[1171] Example prompt:

[1172] Prompt: Generate an advertising campaign based on the idea.

[1173] Example: A video ad campaign highlighting the features of a new smartphone

[1174] Feedback: I think this is a really great idea!

[1175] The above system makes it possible to evaluate ideas by incorporating the user's emotions, allowing for the selection of more appropriate ideas.

[1176] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1177] Step 1:

[1178] The server collects previously submitted ideas from a database. Specifically, the server accesses the database and retrieves idea data. The input at this stage is the idea data in the database, and the output is a list of collected ideas.

[1179] Step 2:

[1180] The server converts the collected ideas into tokens using Hugging Face's tokenizer. The tokenization process converts text data into a set of tokens. The input of this step is the collected idea list, and the output is the tokenized idea data.

[1181] Step 3:

[1182] The server inputs the tokenized idea data into the BERT model and scores them based on the criteria of creativity, feasibility, and impact. Here, the input is the tokenized idea data, and the output is a score for each idea.

[1183] Step 4:

[1184] The server collects feedback from users through their terminals. Users provide feedback on ideas, and the data is sent to the server. The input of this step is the feedback from users, and the output is the collected feedback data.

[1185] Step 5:

[1186] The device uses VADER to perform real-time sentiment analysis of the collected feedback data and generate a sentiment score. Specifically, the feedback text is input into the sentiment analysis engine, and the analysis results are output as a positive, negative, or neutral score. The input of this step is the user's feedback data, and the output is a sentiment score.

[1187] Step 6:

[1188] The server calculates the final score by weighting the initial evaluation score and the sentiment score. The inputs at this stage are the idea score and the sentiment score, and the output is the re-evaluated final score.

[1189] Step 7:

[1190] The server sorts the ideas by final score and feeds back the ideas with the highest scores to the user. Specifically, the server selects the top sorted ideas and notifies the user of their details. The input of this step is the re-evaluated final score, and the output is the feedback data of the highly rated ideas.

[1191] 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.

[1192] 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.

[1193] 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.

[1194] [Fourth embodiment]

[1195] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1196] 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.

[1197] 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).

[1198] 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.

[1199] 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.

[1200] 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).

[1201] 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.

[1202] 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.

[1203] 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.

[1204] 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.

[1205] 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.

[1206] 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.

[1207] 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."

[1208] MODE FOR CARRYING OUT THE INVENTION

[1209] The present invention relates to a system that utilizes generative AI to collect and evaluate previously submitted ideas and select the best idea. This system is configured using the following means.

[1210] 1. Collecting idea data

[1211] The server accesses the database of the Generative AI Contest and queries and retrieves previously submitted idea data. The retrieved data is then converted into an internal dictionary format. This operation allows the system to efficiently manage ideas to be evaluated.

[1212] 2. Preparing the Generative AI

[1213] The server loads the generative AI model (e.g., natural language processing model) to be used for evaluation, including the appropriate tokenizer to convert the text data of the idea into a format that can be evaluated, and sets the evaluation criteria, such as creativity, feasibility, and impact.

[1214] 3. Evaluate ideas

[1215] The server converts the collected text of each idea into tokens using a tokenizer, then inputs them into a generative AI model to score the idea. Specifically, a score is generated for each idea based on the evaluation criteria of creativity, feasibility, and impact. This process allows for a relative evaluation of each idea.

[1216] 4. Score tallying

[1217] The server tally the scores of each idea and sort them by highest score. This allows the best ideas to be identified at a glance. The higher the score, the better each idea is compared to the set evaluation criteria.

[1218] 5. Feedback of results

[1219] The server then provides feedback to the user on the ideas with the highest scores. Specifically, the top few ideas are selected and notified to the user. This feedback allows the user to check the details of the highly rated ideas and refer to the evaluation results.

[1220] Natural language explanation of the process

[1221] First, the server accesses the database of generative AI contests to collect ideas that have been submitted in the past. The collected ideas are then converted into an internal dictionary format, which allows for efficient data management.

[1222] The server then loads the generative AI model for evaluation, which includes natural language processing algorithms and a tokenizer for properly processing text data, and sets evaluation criteria such as creativity, feasibility, and impact.

[1223] The server then converts each idea's text into tokens using a tokenizer and feeds them into a generative AI model, which scores the ideas and generates a score for each idea based on a set of evaluation criteria.

[1224] After each idea has a score generated, the server aggregates it and sorts it by highest score, clearly identifying the best ideas.

[1225] Finally, the server provides feedback on the top ideas to the user, who can then check the details of the highly rated ideas and use the evaluation results as a reference.

[1226] Specific examples

[1227] For example, the following idea is submitted:

[1228] 1. "AI-based automatic translation system"

[1229] 2. "An app that creates music using generative AI"

[1230] 3. "System for controlling self-driving cars using generative AI"

[1231] The server collects these ideas from a database and uses a generative AI model to score them based on evaluation criteria. For example, the following scores might be generated:

[1232] Idea 1: Creativity 80 points, Feasibility 85 points, Impact 90 points

[1233] Idea 2: Creativity 90 points, Feasibility 80 points, Impact 85 points

[1234] Idea 3: Creativity 85, Feasibility 90, Impact 80

[1235] The server aggregates these scores and sorts them by highest score, and the highest scoring idea is then fed back to the user.

[1236] In this way, the system of the present invention utilizes generative AI to efficiently and objectively evaluate ideas and select and present excellent ideas.

[1237] The processing flow will be explained below.

[1238] Specific processing flow of the program

[1239] Idea data collection

[1240] Step 1: Query the database

[1241] The server queries the generative AI contest database to retrieve all submitted ideas.

[1242] Step 2: Transform the data

[1243] The server converts the acquired idea data into an internal dictionary format so that it can be managed efficiently.

[1244] Preparing for generative AI

[1245] Step 1: Loading the Generative AI Model

[1246] The server loads generative AI models (natural language processing algorithms) for evaluation.

[1247] Step 2: Preparing the tokenizer

[1248] The server also loads the tokenizer corresponding to the generative AI model, preparing it to properly tokenize the idea text data.

[1249] Step 3: Set evaluation criteria

[1250] The server sets the criteria for creativity, feasibility, and impact.

[1251] Idea Evaluation

[1252] Step 1: Tokenize your ideas

[1253] The server converts the text of each idea into tokens using a tokenizer.

[1254] Step 2: Score with the model

[1255] The server inputs each tokenized idea into a generative AI model and calculates a score for each.

[1256] Step 3: Create a score list

[1257] The server stores the scores generated for each idea in the form of a list.

[1258] Score tally

[1259] Step 1: Sort the scores

[1260] The server sorts all ideas by their scores, from highest to lowest.

[1261] Step 2: Extracting top ideas

[1262] The server extracts the top few ideas (e.g., the top three).

[1263] Feedback of results

[1264] Step 1: Generate feedback data

[1265] The server generates feedback data including details of the top ranked ideas.

[1266] Step 2: Submit your feedback

[1267] The server transmits the generated feedback data to the user.

[1268] For example, the following processing is performed:

[1269] 1. The server retrieves three ideas from the database: an automatic translation system using AI, an app that creates music using generative AI, and a system that controls self-driving cars using generative AI.

[1270] 2. The server converts these ideas into tokens using a tokenizer, which are then fed into a generative AI model to generate the following score:

[1271] Idea 1: Creativity 80 points, Feasibility 85 points, Impact 90 points

[1272] Idea 2: Creativity 90 points, Feasibility 80 points, Impact 85 points

[1273] Idea 3: Creativity 85, Feasibility 90, Impact 80

[1274] 3. The server aggregates these scores, sorts them by highest score, and sends the top three ideas to the user as feedback data.

[1275] This series of processes creates a system that uses generative AI to objectively and efficiently evaluate ideas and select the best ones.

[1276] Example 1

[1277] 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."

[1278] Existing systems have had problems in efficiently and objectively evaluating previously submitted concepts and selecting the best ones. In particular, when processing large amounts of concept data, issues often arise in setting evaluation criteria and consistency of evaluation. It has also been difficult to provide users with prompt and accurate feedback.

[1279] 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.

[1280] In this invention, the server includes means for collecting previously submitted concepts, means for loading a generative AI model and setting evaluation criteria, means for inputting the collected concepts into the generative AI model and scoring the concepts, means for tokenizing the text of each concept, means for aggregating the scores and sorting them in descending order of score, and means for feeding back the concepts with high scores to the user. This enables efficient and objective evaluation of concepts and rapid selection and feedback of excellent concepts.

[1281] "Previously Submitted Ideas" means ideas or suggestions previously submitted by you or a third party.

[1282] A "generative artificial intelligence model" refers to an algorithm or system that uses artificial intelligence technology to perform natural language processing and predictive analysis.

[1283] "Evaluation criteria" refers to the specific measures or criteria for evaluating ideas, including creativity, feasibility, and impact.

[1284] "Scoring" refers to the process of assigning a score or rating to an initiative based on established evaluation criteria.

[1285] "Tokenization" refers to the process of breaking down text data into words and phrases for processing.

[1286] "Feedback" refers to the action or process of returning information such as evaluation results or notifications to users.

[1287] "User" refers to any person or entity that uses this system to receive the results of an initiative evaluation.

[1288] MODE FOR CARRYING OUT THE INVENTION

[1289] This invention relates to a system that utilizes a generative artificial intelligence model to collect and evaluate previously submitted concepts and select the best concept. This system operates in cooperation with a server, terminals, and users.

[1290] First, the server accesses the database of the Generative AI Contest and collects previously submitted concepts. The server retrieves the data using queries such as SQL and converts the retrieved concept data into an internal dictionary format. This data conversion allows the server to efficiently manage the data and perform subsequent processing.

[1291] The server then loads the generative AI model (e.g., BERT or GPT-3) to be used for evaluation. This generative AI model is an algorithm that uses generative artificial intelligence technology and includes a tokenizer to process the text data. The tokenizer breaks the text data down into tokens and converts them into a format that can be fed into the generative AI model. The server then sets the criteria for evaluating the ideas: creativity, feasibility, and impact.

[1292] Specifically, the server converts the text of each idea into tokens using a tokenizer, and then inputs these into a generative AI model. The generative AI model generates a score for each idea based on the evaluation criteria of creativity, feasibility, and impact. This scoring process provides a relative evaluation of each idea.

[1293] After the scores are generated, the server aggregates them and sorts them by highest score, clearly identifying the best ideas. The server then provides feedback to the user about the highly scored ideas. Specifically, the server notifies the user of the top few highly rated ideas. This feedback allows the user to view the details of the highly rated ideas and refer to the evaluation results.

[1294] Specific examples

[1295] For example, a user might submit the following idea to the database:

[1296] 1. "AI-based automatic translation system"

[1297] 2. "An app that creates music using generative AI"

[1298] 3. "System for controlling self-driving cars using generative AI"

[1299] The server collects these ideas from a database and uses a generative AI model to score them based on evaluation criteria, for example, the following scores are generated:

[1300] Idea 1: Creativity 80 points, Feasibility 85 points, Impact 90 points

[1301] Idea 2: Creativity 90 points, Feasibility 80 points, Impact 85 points

[1302] Idea 3: Creativity 85, Feasibility 90, Impact 80

[1303] The server aggregates these scores and sorts them by highest score, and the highest scoring ideas are then fed back to the user.

[1304] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1305] Program processing flow

[1306] Step 1: Collect idea data

[1307] The server accesses the database of the Generative AI Contest and queries and retrieves previously submitted idea data. The input is the query result from the database, and the output is the idea data converted into dictionary format.

[1308] Specifically, it executes the SQL query "SELECT FROM ideas WHERE submission_date > '2022-01-01'" and converts the results into a Python dictionary, where each idea is managed as a key-value pair. For example, you might get a data structure like this:

[1309] python

[1310] ideas = {

[1311] "idea1": {"title": "AI-based automatic translation system", "submitted_by": "user123", "date": "2023-05-10"},

[1312] "idea2": {"title": "App to create music with generative AI", "submitted_by": "user456", "date": "2023-06-12"},

[1313] }

[1314] Step 2: Prepare the generative AI

[1315] The server loads the generative AI model to be used for evaluation and sets the evaluation criteria. The input is the generative AI model and tokenizer parameters, and the output is the loaded model and the set evaluation criteria. Specifically, it imports libraries such as "from transformers import GPT-3" and creates an instance of the model. It also creates an instance of the tokenizer:

[1316] python

[1317] tokenizer = GPT3Tokenizer.from_pretrained('gpt-3')

[1318] model = GPT3Model.from_pretrained('gpt-3')

[1319] Set the criteria for evaluation: creativity, feasibility, and impact:

[1320] python

[1321] evaluation_criteria = ["creativity", "feasibility", "impact"]

[1322] Step 3: Tokenize your ideas

[1323] The server converts the text of each idea into tokens using a tokenizer. The input is the text data of each idea, and the output is the tokenized text data. Specifically, it converts into tokens as follows:

[1324] python

[1325] tokens = tokenizer.encode(ideas["idea1"]["title"], return_tensors='pt')

[1326] This converts the text of the idea into a format that can be input into a generative AI model.

[1327] Step 4: Evaluate your ideas

[1328] The server inputs the tokenized data into a generative AI model to score ideas. The input is tokenized text data, and the output is a score for each idea. Specifically, the data is input into the model as follows:

[1329] python

[1330] scores = model(tokens)

[1331] You will receive a score based on each criterion:

[1332] python

[1333] scores = {"creativity": 80, "feasibility": 85, "impact": 90}

[1334] Step 5: Counting the scores

[1335] The server aggregates the scores of each idea and sorts them in descending order. The input is the score of each idea, and the output is a list of ideas sorted in descending order. Specifically, the server aggregates and sorts the scores as follows:

[1336] python

[1337] sorted_ideas = sorted(ideas.items(), key=lambda x: x[1]['scores']['total'], reverse=True)

[1338] This allows the best ideas to rise to the top.

[1339] Step 6: Feedback on results

[1340] The server feeds back the highest-scoring ideas to the user. The input is the high-scoring ideas, and the output is a feedback message to the user. Specifically, it sends the following feedback message to the user's device:

[1341] python

[1342] feedback_message = f"The idea with the highest score is "{sorted_ideas[0]['title']}". Learn more here."

[1343] send_feedback_to_user(user_id, feedback_message)

[1344] In this way, the system of the present invention can efficiently and objectively evaluate ideas and quickly present excellent ideas to the user.

[1345] (Application example 1)

[1346] 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."

[1347] In existing factory automation systems, the evaluation and selection of new efficiency improvement measures and automation ideas is often done manually, making objective and efficient evaluation difficult. Furthermore, it is difficult to quickly find the most suitable idea from among many ideas and put it into practice. Given this background, there is a demand for a faster and more reliable idea evaluation system in workplaces seeking innovation and efficiency improvements in factory automation.

[1348] 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.

[1349] In this invention, the server includes means for collecting previously submitted ideas, means for loading a generative AI model and setting evaluation criteria, means for inputting the collected ideas into the generative AI model and scoring the ideas, means for aggregating the scores and sorting them in descending order of score, means for providing feedback on the ideas with the highest scores to the user, and means for applying the idea evaluation results to the factory automation system and providing optimal solutions for improving efficiency. This enables objective and efficient evaluation of new efficiency improvement measures and automation ideas in the factory automation system, and makes it possible to quickly find and implement the most suitable ideas.

[1350] A "server" is an information processing device connected to a computer network, which collects, analyzes, processes, and provides data.

[1351] "Means for collecting idea data" refers to devices or algorithms that retrieve previously submitted ideas from databases, etc., and manage them appropriately.

[1352] A "generative AI model" refers to an artificial intelligence program that uses natural language processing algorithms and other techniques to analyze input data and perform specific tasks.

[1353] "Means for setting evaluation criteria" refers to devices or methods for defining and setting criteria for evaluation based on the creativity, feasibility, impact, etc. of ideas.

[1354] "Idea scoring means" refers to a device or method that uses a generative AI model to evaluate collected ideas and assign scores based on each evaluation criterion.

[1355] "Means for aggregating scores and sorting ideas by highest score" refers to a device or algorithm that aggregates the scores assigned to ideas and orders the ideas based on that aggregated score.

[1356] "Means for providing feedback to the user" refers to a device or method for notifying the user of the ideas with high scores and communicating the results.

[1357] "Factory automation system" is a general term for hardware and software that automates manufacturing processes and operations within a factory to improve efficiency.

[1358] "Means for providing optimal solutions" refers to devices and methods that are proposed and introduced based on the evaluation results to improve factory efficiency and optimize automation.

[1359] The system for implementing the present invention involves a process of collecting previously submitted ideas, evaluating them using a generative AI model, and providing optimal solutions to a factory automation system based on the evaluation results. A specific embodiment of the system is shown below.

[1360] First, the server accesses the database and collects previously submitted ideas. This idea data is converted into an internal dictionary format and efficiently managed.

[1361] The server then loads a generative AI model for evaluation. The model includes a natural language processing algorithm and uses a tokenizer to convert the text data of the ideas into a format that can be evaluated. The server also sets evaluation criteria such as creativity, feasibility, and impact.

[1362] Each collected idea text is converted into tokens by a tokenizer and then input into a generative AI model. The generative AI model evaluates the input ideas and generates a score for each idea based on the evaluation criteria of creativity, feasibility, and impact. The server aggregates these scores and sorts them by highest score, allowing it to identify superior ideas.

[1363] The server then provides feedback on the highly scored ideas to users, such as factory managers. Specifically, the top few ideas are selected and notified to the users. This feedback allows users to check the details of the highly rated ideas and refer to the evaluation results.

[1364] Furthermore, the most highly evaluated ideas will be applied to factory automation systems to provide optimal solutions for improving efficiency, thereby optimizing factory automation processes and increasing productivity.

[1365] For example, the following prompts can be fed into a generative AI model to evaluate ideas:

[1366] Prompt statement:

[1367] "Please rate the following ideas:

[1368] Idea: "Inventory management system using unmanned robots"

[1369] Evaluation criteria: Creativity, feasibility, impact.”

[1370] The expected output is, for example, "80, 90, 85", which are scores based on each evaluation criterion.

[1371] Based on the evaluation results, the server provides optimal solutions to the factory automation system, improving factory efficiency.

[1372] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1373] Step 1:

[1374] The server accesses a database to collect previously submitted ideas. The input is a database query, and the output is a list of ideas. This list is converted into an internal dictionary format and efficiently managed.

[1375] Step 2:

[1376] The server loads the generative AI model to be used for evaluation. At this stage, the generative AI model containing a specific natural language processing algorithm is loaded. The input is an instruction to load the AI ​​model, and the output is an AI model ready for evaluation.

[1377] Step 3:

[1378] The server sets the evaluation criteria, specifically defining criteria such as creativity, feasibility, and impact. The input is the evaluation criteria setting, and the output is the set evaluation criteria.

[1379] Step 4:

[1380] Each piece of text in the collected idea data is converted into tokens by a tokenizer. The input to this step is the text data of the idea, and the output is the tokenized data. The tokenizer analyzes the text and divides it into small units that can be analyzed.

[1381] Step 5:

[1382] The server uses a generative AI model to evaluate the tokenized ideas. In this step, the tokenized ideas are input into the generative AI model, and a score based on each evaluation criterion is generated as output. Specifically, the AI ​​model performs semantic analysis of the text and calculates a score for each evaluation criterion.

[1383] Step 6:

[1384] The server aggregates the scores obtained from the evaluation and sorts them in descending order. The input is the score of each idea, and the output is a list of ideas sorted in descending order. The server adds up the scores and sorts the list in descending order.

[1385] Step 7:

[1386] The server feeds back the highly scored ideas to the user. Here, the top few ideas are selected and their details are notified to the user. The input is a list of highly scored ideas, and the output is a notification of the fed back ideas. Specifically, the server sends a notification to the user's device.

[1387] Step 8:

[1388] Based on the feedback, the user applies the optimal solution to the factory automation system. The input of this step is the feedback received by the user, and the output is the applied automation solution. Specifically, the user implements the selected idea to optimize the factory automation process.

[1389] 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.

[1390] MODE FOR CARRYING OUT THE INVENTION

[1391] This invention relates to a system that uses generative AI to collect and evaluate previously submitted ideas and select the best idea, and further combines it with an emotion engine that recognizes the user's emotions. This system is implemented as follows:

[1392] First, the server accesses the database of generative AI contests and collects data on ideas submitted in the past. The collected data is then converted into an internal dictionary format and efficiently managed.

[1393] Next, the server loads the generative AI model to be used for evaluation, prepares the tokenizer, which is used to properly tokenize the text data of the idea, and sets the evaluation criteria of creativity, feasibility, and impact.

[1394] The server then converts the collected text of each idea into tokens using a tokenizer and feeds them into a generative AI model, which scores each idea and generates a score for each idea based on the evaluation criteria set.

[1395] Furthermore, this system uses an emotion engine to recognize the user's emotions in real time. When the user provides feedback on an idea, the device inputs the feedback into the emotion engine for analysis. Emotion data based on the analysis results is reflected in the evaluation of the idea.

[1396] The server then recalculates a final score for each idea, taking into account the emotional data obtained by the emotion engine. The scores obtained in this way are aggregated and sorted by highest score.

[1397] Finally, the server provides feedback on the top ideas to the user. Feedback data containing details of the top ideas is generated and sent to the user. This allows the user to check the details of the highly rated ideas and to identify areas for future improvement based on the emotional data obtained from the emotion engine.

[1398] Natural language explanation of the process

[1399] First, the server accesses a database to retrieve previously submitted ideas, which are then converted into an internal dictionary format.

[1400] The server then loads the generative AI model and tokenizer, ready to evaluate each idea based on the set criteria, which include creativity, feasibility, and impact.

[1401] The server then converts the text of each idea into tokens using a tokenizer, feeds them into a generative AI model, and calculates a score, which indicates how good each idea is.

[1402] Next, when receiving feedback from users, they input the feedback into the emotion engine using their terminal. The emotion engine analyzes the user's emotions in real time and obtains the results. This emotion data is incorporated into the score so that it is also reflected in the evaluation of the idea.

[1403] The final scores are then tallied and sorted by highest score, thereby identifying the best ideas.

[1404] Finally, the server provides feedback on the top ideas to the user, who can then check the details of the highly rated ideas and refer to the evaluation results and the analysis results of the emotion engine.

[1405] Specific examples

[1406] For example, the following idea is submitted:

[1407] 1. "AI-based automatic translation system"

[1408] 2. "An app that creates music using generative AI"

[1409] 3. "System for controlling self-driving cars using generative AI"

[1410] The server collects these ideas from the database, converts them into tokens using a tokenizer, and feeds them into a generative AI model, which generates a score for each idea, such as:

[1411] Idea 1: Creativity 80 points, Feasibility 85 points, Impact 90 points

[1412] Idea 2: Creativity 90 points, Feasibility 80 points, Impact 85 points

[1413] Idea 3: Creativity 85, Feasibility 90, Impact 80

[1414] When users provide feedback on these ideas, the device uses an emotion engine to analyze their emotions in real time and reflects the analysis results in the score. For example, if a user shows strong interest in idea 2, the emotion analysis may increase the evaluation score.

[1415] Finally, the server recalculates the score taking into account the emotion data and sends the top three ideas as feedback to the user, realizing an advanced evaluation system that combines generative AI and emotion recognition.

[1416] The processing flow will be explained below.

[1417] MODE FOR CARRYING OUT THE INVENTION

[1418] Idea data collection

[1419] Step 1: Query the database

[1420] The server queries the generative AI contest database to retrieve all previously submitted idea data.

[1421] Step 2: Transform the data

[1422] The server converts the acquired idea data into an internal dictionary format and manages it efficiently.

[1423] Preparing for generative AI

[1424] Step 1: Loading the Generative AI Model

[1425] The server loads generative AI models (natural language processing algorithms) for evaluation.

[1426] Step 2: Preparing the tokenizer

[1427] The server loads the generative AI model and the corresponding tokenizer, preparing it to properly tokenize the idea text data.

[1428] Step 3: Set evaluation criteria

[1429] The server sets the criteria for evaluation: creativity, feasibility, and impact.

[1430] Idea Evaluation

[1431] Step 1: Tokenize your ideas

[1432] The server converts the text of each collected idea into tokens using a tokenizer.

[1433] Step 2: Score with the model

[1434] The server inputs each tokenized idea into a generative AI model and calculates a score for each.

[1435] Step 3: Create a score list

[1436] The server stores the scores generated for each idea in the form of a list.

[1437] Using the Emotion Engine

[1438] Step 1: Accepting user feedback

[1439] The device receives feedback from the user, including detailed thoughts on the ideas that the user has rated.

[1440] Step 2: Analysis by the emotion engine

[1441] The device inputs the received feedback into an emotion engine and analyzes the user's emotions in real time.

[1442] Step 3: Generate emotion data

[1443] The terminal generates emotion data of the user based on the analysis results.

[1444] Step 4: Reflecting emotional data

[1445] The server reflects the generated emotion data in the score of the idea and updates the evaluation result.

[1446] Score tally

[1447] Step 1: Sort the scores

[1448] The server sorts the final scores from highest to lowest.

[1449] Step 2: Extracting top ideas

[1450] The server extracts the top few ideas (e.g., the top three).

[1451] Feedback of results

[1452] Step 1: Generate feedback data

[1453] The server generates feedback data including details of the top ideas.

[1454] Step 2: Submit your feedback

[1455] The server transmits the generated feedback data to the user.

[1456] Specific examples

[1457] 1. The server retrieves ideas from the Generative AI Utilization Contest Database, such as "an automatic translation system using AI," "an app that creates music using generative AI," and "a system that controls self-driving cars using generative AI."

[1458] 2. The server converts these ideas into tokens using a tokenizer, which are then fed into a generative AI model to generate a score:

[1459] Idea 1: Creativity 80 points, Feasibility 85 points, Impact 90 points

[1460] Idea 2: Creativity 90 points, Feasibility 80 points, Impact 85 points

[1461] Idea 3: Creativity 85, Feasibility 90, Impact 80

[1462] 3. When users provide feedback on these ideas, the device uses an emotion engine to analyze the feedback in real time and generate emotion data based on the analysis results. For example, if the user shows strong interest in idea 2, the emotion analysis may increase the evaluation score.

[1463] 4. The server recalculates the score taking into account the emotional data and sends the top three ideas to the user as feedback data.

[1464] This series of processes realizes an advanced evaluation system that combines generative AI and emotion recognition.

[1465] Example 2

[1466] 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."

[1467] Conventional idea evaluation systems have limited evaluation criteria, making it difficult to provide flexible evaluations that reflect user sentiment. Furthermore, the accuracy and reliability of the evaluations are lacking, making it difficult to accurately identify the best ideas. This has led to the problem that many excellent ideas are not properly evaluated and end up being buried.

[1468] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1469] In this invention, the server includes means for collecting previously submitted information, means for loading an AI model and setting evaluation criteria, means for inputting the collected information into the AI ​​model and scoring the information, means for analyzing the user's emotions, means for recalculating the score based on the emotion data, means for aggregating the scores and sorting them in descending order, and means for feeding back information on the high scores to the user. This enables flexible and accurate evaluation that reflects the user's emotions.

[1470] "Previously Submitted Information" means data, such as ideas or suggestions, previously submitted by users or other interested parties.

[1471] An "artificial intelligence model" is a system that analyzes and scores data using machine learning algorithms trained for a specific purpose.

[1472] "Metrics" are the indicators or standards used to evaluate the quality or usefulness of information or ideas.

[1473] "Means for analyzing user emotions" refers to software or devices that analyze emotions and impressions based on feedback and reactions from users.

[1474] "Emotional data" refers to information about the analyzed user's emotions and reactions.

[1475] "Means for recalculating scores" means software or computational algorithms for reevaluating existing scores based on emotion data and generating updated scores.

[1476] "Means for aggregating scores and sorting by highest score" refers to software or algorithms that calculate the score for each piece of information and sort it in descending order.

[1477] The "means for feeding back information with high scores to the user" refers to a method or means for reporting information with high scores from among the sorted information to the user.

[1478] MODE FOR CARRYING OUT THE INVENTION

[1479] This invention relates to a system that uses generative AI to collect and evaluate previously submitted information and select the best information, and also combines it with an emotion engine that recognizes the user's emotions. This system can realize consistent evaluations that also reflect the user's emotions.

[1480] First, the server accesses the AI-powered contest database and collects previously submitted information. The software used here can be a database management system (e.g., MySQL or PostgreSQL). The collected information is internally converted into a dictionary format (e.g., Python dictionary data structure) for efficient management.

[1481] Next, the server loads the generative AI model and tokenizer to be used for evaluation. Specifically, it uses a natural language processing algorithm (e.g., GPT-3 or BERT). The tokenizer (e.g., Hugging Face's tokenizer) is used to properly tokenize the information text data. At this stage, the evaluation criteria are set: creativity, feasibility, and impact.

[1482] The server then converts the text of each piece of collected information into tokens using a tokenizer and feeds them into a generative AI model. The generative AI model scores each piece of information and generates a score based on the set evaluation criteria. These scores are stored in a temporary data store (e.g., Redis).

[1483] Furthermore, this system uses an emotion engine to recognize the user's emotions in real time. When the user provides feedback on information, the terminal inputs the feedback into the emotion engine (e.g., emotion analysis model Sentiment 140) for analysis. Emotion data based on the analysis results is sent back to the server and reflected in the evaluation of the information.

[1484] The server recalculates the scores based on the emotion data, aggregates these scores, and sorts them in order of highest score. Finally, feedback data is generated to provide the top ranking information to the user. This data is sent to the user's device via API or email. This process allows the user to check the details of the highly rated information and to identify areas for future improvement based on the emotion data obtained from the emotion engine.

[1485] Specific examples

[1486] For example, if the following information is submitted:

[1487] 1. "AI-based automatic translation system"

[1488] 2. "An app that creates music using generative AI"

[1489] 3. "System for controlling self-driving cars using generative AI"

[1490] The server collects this information from the database, converts it into tokens using a tokenizer, and inputs them into the generative AI model, which generates the following scores for each piece of information:

[1491] Information 1: Creativity 80 points, Feasibility 85 points, Impact 90 points

[1492] Information 2: Creativity 90 points, Feasibility 80 points, Impact 85 points

[1493] Information 3: Creativity 85 points, Feasibility 90 points, Impact 80 points

[1494] When the user provides feedback on this information, the device uses an emotion engine to analyze the user's emotions in real time. For example, if the user shows strong interest in information 2, the evaluation score for information 2 may increase based on the emotion analysis results.

[1495] Finally, the server recalculates the score taking into account the emotional data and sends the top three scores to the user as feedback data, realizing an advanced evaluation system that combines generative AI and emotional recognition.

[1496] Prompt Sentence Examples

[1497] To facilitate information evaluation in an AI-based machine translation system, the following prompts could be input to the generative AI model:

[1498] "AI-based automatic translation system" is a system that enables automatic translation between multiple languages. Please rate this idea in terms of creativity, feasibility, and impact.

[1499] For the emotion engine, the following prompt is an example:

[1500] Analyze user feedback on your AI-powered machine translation system to identify emotions such as interest, delight, and surprise.

[1501] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1502] Step 1: Obtaining collected information

[1503] The server collects previously submitted information. The input for this process is the database connection information and query. Specifically, the server executes an SQL query to extract past information, parses the retrieved information into JSON format, and then converts it into an internal dictionary format. The output is the information data converted into dictionary format.

[1504] Step 2: Load the generative AI model and tokenizer

[1505] The server loads the generative AI model and tokenizer. The input to this process is the path to the model and tokenizer. Specifically, the server imports a machine learning library (e.g., TensorFlow, PyTorch) and loads the model file and tokenizer. The output is the loaded generative AI model and tokenizer.

[1506] Step 3: Set the evaluation criteria

[1507] The server sets the criteria for evaluation. The input of this process is the setting information of the evaluation criteria (creativity, feasibility, impact). In concrete terms, the server reads the evaluation criteria information from a setting file or database and sets these criteria for the model. The output is the set evaluation criteria.

[1508] Step 4: Tokenize your idea data

[1509] The server converts the collected text data into tokens using a tokenizer. The inputs to this process are the information data converted into dictionary format and the loaded tokenizer. Specifically, each piece of text data is passed to the tokenizer and converted into tokens. The output is the tokenized data.

[1510] Step 5: Score your idea data

[1511] The server inputs the tokenized data into the generative AI model and calculates the score. The inputs for this process are the tokenized data and the loaded generative AI model. Specifically, it inputs a prompt sentence into the generative AI model and performs scoring. The output is the calculated score.

[1512] Step 6: User feedback analysis

[1513] The terminal analyzes the feedback provided by the user using an emotion engine. The input of this process is the user's feedback text. Specifically, the feedback text is input to the emotion engine (e.g., Sentiment 140) to obtain emotion data. The output is the analyzed emotion data.

[1514] Step 7: Recalculate the score using emotion data

[1515] The server recalculates the score based on the obtained emotion data. The inputs to this process are the original score and the analyzed emotion data. Specifically, the server updates the score using recalculation logic to generate a final score. The output is the recalculated final score.

[1516] Step 8: Count and sort the scores

[1517] The server tally the final scores of each piece of information and sorts them in descending order. The input to this process is the recalculated final scores. Specific operations include using an algorithm to sort the scores in descending order and listing the top pieces of information. The output is a sorted list of information.

[1518] Step 9: User Feedback

[1519] The server feeds back the top-level information to the user. The input to this process is a sorted information list. Specific operations include generating feedback data and sending it to the user's device via a communication method such as an API or email. The output is the feedback data sent to the user.

[1520] (Application example 2)

[1521] 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."

[1522] Conventional idea evaluation systems using generative AI have the problem of being unable to fully reflect user opinions and emotions, making it difficult to select truly valuable ideas. Particularly for ideas for advertising campaigns, where user emotions are an important evaluation factor, improvement in this area was required.

[1523] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting previously submitted ideas, means for loading a generative AI model and setting evaluation criteria, means for inputting the collected ideas into the generative AI model and scoring the ideas, means for receiving feedback from users and analyzing them with a sentiment analysis engine, means for re-evaluating ideas based on the sentiment analysis results and calculating a final score, and means for feeding back high-scoring ideas to the user. This enables evaluation that takes into account the user's emotions, making it possible to select more appropriate ideas.

[1524] "Means of collection" refers to the means of acquiring ideas that have been submitted in the past and storing them in a database.

[1525] A "generative AI model" is an AI model that uses natural language processing algorithms to analyze and evaluate text data.

[1526] "Evaluation criteria" are the standards set to evaluate ideas for their creativity, feasibility, and impact.

[1527] A "means for scoring" is a means for analyzing ideas input into a generative AI model and assigning a score based on set evaluation criteria.

[1528] The "emotion analysis engine" is an engine that analyzes feedback from users and generates emotion data based on that feedback.

[1529] The "re-evaluation means" is a means for adjusting the initial evaluation score and calculating the final score in consideration of the sentiment analysis results.

[1530] The "means of providing feedback" is a means of reporting high-scoring ideas and their details to the user.

[1531] A system for implementing this invention includes a server, a sentiment analysis engine, and a terminal, and uses a generative AI model to evaluate and select ideas for an advertising campaign.

[1532] System program generation

[1533] System Overview

[1534] The server collects previously submitted ideas and sets evaluation criteria. Next, the collected ideas are input into a generative AI model and scored. It also receives feedback from users, analyzes it in real time using a sentiment analysis engine, and generates emotional data. The ideas are then re-evaluated based on the emotional data and a final score is calculated. Finally, the ideas with the highest scores are fed back to the user.

[1535] Hardware and Software

[1536] Hardware: Smartphones, servers

[1537] software:

[1538] Hugging Face's Transformers library: uses a pre-trained BERT model and tokenizer

[1539] VADER (Valence Aware Dictionary and sEntiment Reasoner): for sentiment analysis

[1540] OpenAI API: Text analysis and evaluation using generative AI models

[1541] Data processing and calculation

[1542] 1. Tokenization of ideas

[1543] The server converts the text data of the idea into tokens using Hugging Face's tokenizer.

[1544] 2. Idea scoring

[1545] The server inputs the tokenized idea data into the BERT model and scores it based on the criteria of creativity, feasibility, and impact.

[1546] 3. Sentiment Analysis of Feedback

[1547] The device collects feedback from users in real time and performs sentiment analysis using VADER to generate a sentiment score.

[1548] 4. Calculation of Final Score

[1549] The server calculates a weighted average of the idea score and the emotion score to arrive at a final score.

[1550] 5. Providing Feedback

[1551] The server reports the ideas to the user sorted by highest score.

[1552] Specific examples

[1553] Below are some examples of specific ideas and feedback prompts:

[1554] Example ideas:

[1555] 1. "A video advertising campaign highlighting the features of a new smartphone"

[1556] 2. "Personalized advertising using generative AI"

[1557] 3. "Advertisements that recommend products that match the user's preferences"

[1558] Feedback example:

[1559] 1. "I think this is a really great idea!"

[1560] 2. "I'm not interested."

[1561] 3. "It's an interesting idea, but it seems difficult to implement."

[1562] Example prompt:

[1563] Prompt: Generate an advertising campaign based on the idea.

[1564] Example: A video ad campaign highlighting the features of a new smartphone

[1565] Feedback: I think this is a really great idea!

[1566] The above system makes it possible to evaluate ideas by incorporating the user's emotions, allowing for the selection of more appropriate ideas.

[1567] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1568] Step 1:

[1569] The server collects previously submitted ideas from a database. Specifically, the server accesses the database and retrieves idea data. The input at this stage is the idea data in the database, and the output is a list of collected ideas.

[1570] Step 2:

[1571] The server converts the collected ideas into tokens using Hugging Face's tokenizer. The tokenization process converts text data into a set of tokens. The input of this step is the collected idea list, and the output is the tokenized idea data.

[1572] Step 3:

[1573] The server inputs the tokenized idea data into the BERT model and scores them based on the criteria of creativity, feasibility, and impact. Here, the input is the tokenized idea data, and the output is a score for each idea.

[1574] Step 4:

[1575] The server collects feedback from users through their terminals. Users provide feedback on ideas, and the data is sent to the server. The input of this step is the feedback from users, and the output is the collected feedback data.

[1576] Step 5:

[1577] The device uses VADER to perform real-time sentiment analysis of the collected feedback data and generate a sentiment score. Specifically, the feedback text is input into the sentiment analysis engine, and the analysis results are output as a positive, negative, or neutral score. The input of this step is the user's feedback data, and the output is a sentiment score.

[1578] Step 6:

[1579] The server calculates the final score by weighting the initial evaluation score and the sentiment score. The inputs at this stage are the idea score and the sentiment score, and the output is the re-evaluated final score.

[1580] Step 7:

[1581] The server sorts the ideas by final score and feeds back the ideas with the highest scores to the user. Specifically, the server selects the top sorted ideas and notifies the user of their details. The input of this step is the re-evaluated final score, and the output is the feedback data of the highly rated ideas.

[1582] 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.

[1583] 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.

[1584] 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.

[1585] 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.

[1586] FIG. 9 illustrates an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and behaviors arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1587] 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.

[1588] 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).

[1589] 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.

[1590] 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."

[1591] 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.

[1592] 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).

[1593] 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.

[1594] 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.

[1595] 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.

[1596] 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.

[1597] 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.

[1598] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1599] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1600] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1601] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1602] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1603] The following is further disclosed regarding the above embodiment.

[1604] (Claim 1)

[1605] A means of collecting previously submitted ideas;

[1606] A means for loading generative AI models and setting evaluation criteria;

[1607] A means to input collected ideas into a generative AI model and score the ideas;

[1608] A means to tally the scores and sort them by highest score,

[1609] A means of providing feedback to users on ideas with high scores;

[1610] A system including:

[1611] (Claim 2)

[1612] 10. The system of claim 1, which uses a natural language processing algorithm as the generative AI model.

[1613] (Claim 3)

[1614] 10. The system of claim 1, wherein the evaluation criteria for an idea include creativity, feasibility, and impact.

[1615] "Example 1"

[1616] (Claim 1)

[1617] A means of collecting previously submitted ideas;

[1618] means for loading a generative artificial intelligence model and setting evaluation criteria;

[1619] means for inputting the collected ideas into a generative artificial intelligence model and scoring the ideas;

[1620] A means of tokenizing the text of each concept,

[1621] A means to tally the scores and sort them by highest score,

[1622] A means of providing feedback to users on high-scoring concepts,

[1623] A system including:

[1624] (Claim 2)

[1625] 10. The system of claim 1, wherein the generative artificial intelligence model uses a natural language processing algorithm.

[1626] (Claim 3)

[1627] 10. The system of claim 1, wherein the evaluation criteria for an idea include creativity, feasibility, and impact.

[1628] "Application Example 1"

[1629] (Claim 1)

[1630] A means of collecting previously submitted ideas;

[1631] A means for loading generative AI models and setting evaluation criteria;

[1632] A means to input collected ideas into a generative AI model and score the ideas;

[1633] A means to tally the scores and sort them by highest score,

[1634] A means of providing feedback to users on ideas with high scores;

[1635] A means to apply the idea evaluation results to factory automation systems and provide optimal solutions for improving efficiency;

[1636] A system including:

[1637] (Claim 2)

[1638] 10. The system of claim 1, which uses a natural language processing algorithm as the generative AI model.

[1639] (Claim 3)

[1640] 10. The system of claim 1, wherein the evaluation criteria for an idea include creativity, feasibility, and impact.

[1641] "Example 2: Combining Emotion Engines"

[1642] (Claim 1)

[1643] a means of collecting previously submitted information;

[1644] means for loading an artificial intelligence model and setting evaluation criteria;

[1645] A means for inputting the collected information into an artificial intelligence model and scoring the information;

[1646] means for analyzing user emotions;

[1647] A means to recalculate the score based on the emotion data,

[1648] A means to tally the scores and sort them by highest score,

[1649] A means for providing feedback on high score information to the user;

[1650] A system including:

[1651] (Claim 2)

[1652] 10. The system of claim 1, wherein the artificial intelligence model uses a natural language processing algorithm.

[1653] (Claim 3)

[1654] 10. The system of claim 1, wherein the criteria for evaluating information include creativity, feasibility, and impact.

[1655] "Application example 2 when combining emotion engines"

[1656] (Claim 1)

[1657] A means of collecting previously submitted ideas;

[1658] A means for loading generative AI models and setting evaluation criteria;

[1659] A means to input collected ideas into a generative AI model and score the ideas;

[1660] A means to tally the scores and sort them by highest score,

[1661] A means of receiving user feedback and analyzing it with a sentiment analysis engine;

[1662] A means to re-evaluate ideas based on the sentiment analysis results and calculate a final score;

[1663] A means of providing feedback to users on ideas with high scores;

[1664] A system including:

[1665] (Claim 2)

[1666] 10. The system of claim 1, which uses a natural language processing algorithm as the generative AI model.

[1667] (Claim 3)

[1668] 10. The system of claim 1, wherein the evaluation criteria for an idea include creativity, feasibility, and impact. [Explanation of symbols]

[1669] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of collecting previously submitted ideas; A means for loading generative AI models and setting evaluation criteria; A means to input collected ideas into a generative AI model and score the ideas; A means to tally the scores and sort them by highest score, A means of providing feedback to users on ideas with high scores; A system including:

2. The system of claim 1 , which uses a natural language processing algorithm as the generative AI model.

3. 2. The system of claim 1, wherein the idea evaluation criteria include creativity, feasibility, and impact.

Images