System

Abstract

An object of the system according to the embodiment is to verify a correlation between a prompt and an output in a generation AI and propose an optimal prompt.SOLUTION: A system includes a reception part, a verification part, an analysis part, and a proposal part. The reception unit inputs a difference between the Base prompt and the information desired by the user as a result. The verification unit verifies a correlation between the prompt and the output by the generation AI based on the information input by the reception unit. The analyzing unit analyzes the cause of the generated AI based on the result obtained by the verifying unit. In the proposing section, the generating AI proposes a prompt based on the result obtained by the analyzing section.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] Conventional technologies do not efficiently verify the correlation between the generative AI's prompts and output, and do not adequately propose optimal prompts, leaving room for improvement.

[0005] The system of the embodiment aims to verify the correlation between the prompts and output of the generation AI and propose optimal prompts. [Means for solving the problem]

[0006] The system according to the embodiment includes a reception unit, a verification unit, an analysis unit, and a proposal unit. The reception unit allows the user to input the difference between the base prompt and the result, or desired information. The verification unit allows the generation AI to verify the correlation between the prompt and the output based on the information input by the reception unit. The analysis unit allows the generation AI to analyze factors based on the results obtained by the verification unit. The proposal unit allows the generation AI to propose a prompt based on the results obtained by the analysis unit. [Effects of the Invention]

[0007] The system according to the embodiment can verify the correlation between the prompts and output of the generation AI and suggest optimal prompts. [Brief explanation of the drawings]

[0008] [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. DETAILED DESCRIPTION OF THE INVENTION

[0009] 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.

[0010] First, the terms used in the following description will be explained.

[0011] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, the processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), or a TPU (Tensor Processing Unit).

[0012] 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.

[0013] 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.

[0014] 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), and Bluetooth (registered trademark).

[0015] 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."

[0016] [First embodiment] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0017] 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.

[0018] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and 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).

[0019] 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.

[0020] The reception device 38 includes a touch panel 38A and a microphone 38B, and receives user input. The touch panel 38A detects contact with a pointer (for example, a pen or a finger) to receive user input by the touch of the pointer. 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 (see FIG. 2) acquires the data indicating the user input.

[0021] Output device 40 includes a display 40A and a speaker 40B, and presents data to a user by outputting the data in a form of expression that the user can perceive (e.g., audio and / or text). Display 40A displays visible information such as text and images in accordance with instructions from processor 46. Speaker 40B outputs audio in accordance with instructions from processor 46. 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.

[0022] 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.

[0023] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0024] 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.

[0025] 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. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0026] In the smart device 14, the specific processing is performed by the processor 46. The storage 50 stores a specific processing program 60. The specific processing program 60 is used together with the specific processing program 56 by the data processing system 10. The processor 46 reads the specific processing program 60 from the storage 50 and executes the read specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as the control unit 46A in accordance with the specific processing program 60 executed on the RAM 48. Note that the smart device 14 has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can also perform processing similar to that of the specific processing unit 290 using these models.

[0027] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device (e.g., a generation server) may have the data generation model 58. In this case, the data processing device 12 obtains a processing result (prediction result, etc.) using the data generation model 58 by communicating with the server device having the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device owned by a user (e.g., a mobile phone, a robot, a home appliance, etc.). Next, an example of processing by the data processing system 10 according to the first embodiment will be described.

[0028] (Example 1) A prompt optimization system according to an embodiment of the present invention uses a generation AI to verify the correlation between a prompt and its output, analyze factors that lead to an incorrect answer, and propose prompts that bring the user closer to the correct answer. In the prompt optimization system, a user inputs the difference between a base prompt and its result, the desired information. The generation AI then verifies the correlation between the prompt and the output, analyzes factors that lead to an incorrect answer, and proposes prompts that bring the user closer to the correct answer. For example, in the prompt optimization system, a user inputs a specific prompt, its output, and the difference between the desired information in detail. For example, if a user receives an output such as "The price of this product is 1,000 yen" in response to a prompt such as "What is the price of this product?", the user inputs "I would also like to know the product's stock status" as the difference from the desired information. Next, the prompt optimization system uses a generation AI to verify the correlation between the input prompt and the output. The generation AI analyzes the relationship between the prompt and its output and identifies which parts are close to the correct answer and which parts deviate from the correct answer. For example, if a user receives an output such as "The price of this product is 1,000 yen" in response to a prompt such as "What is the price of this product?", the generation AI verifies whether this output is the correct answer for the prompt. Furthermore, the prompt optimization system uses a generation AI to analyze factors that lead to an incorrect answer. Based on the correlation between the prompt and the output, the generation AI identifies which part of the answer is incorrect. For example, if the prompt "Tell me the price of this product" returns the output "The product price is 1,000 yen," but the user enters "I would also like to know the product's stock status" as the information they are looking for, the generation AI analyzes this difference and identifies the factors that caused the answer to be incorrect. Finally, the prompt optimization system uses the generation AI to suggest a prompt that will bring the user closer to the information they are looking for, based on the factors that caused the answer to be incorrect. For example, if the prompt "Tell me the price of this product" returns the output "The product price is 1,000 yen," but the user enters "I would also like to know the product's stock status" as the information they are looking for, the generation AI will suggest a prompt such as "Tell me the product's price and stock status."This allows the prompt optimization system to help users review prompts efficiently and obtain accurate output. This allows the prompt optimization system to help users review prompts efficiently and obtain accurate output. For example, if a user wants to know the price and stock status of a product, they can obtain accurate information by using the prompts suggested by the generation AI. The prompt optimization system also verifies the correlation between the prompt and the output and analyzes the factors that cause the answer to be incorrect, helping users review prompts efficiently.

[0029] A prompt optimization system according to an embodiment includes a reception unit, a verification unit, an analysis unit, and a proposal unit. The reception unit receives a user input of a difference between a base prompt and its result, which represents the desired information. The information input by the user includes, but is not limited to, a specific prompt, its output, and the difference between the desired information. For example, if a user receives an output such as "The price of the product is 1,000 yen" in response to a prompt such as "What is the price of the product?", the reception unit allows the user to input "I would also like to know the product's stock status" as the difference between the desired information. The verification unit uses a generation AI to verify the correlation between the prompt and the output based on the information input by the reception unit. For example, the verification unit analyzes the relationship between the prompt and the output and identifies which parts are close to the correct answer and which parts deviate from the correct answer. For example, if the generation AI receives the output "The price of the product is 1,000 yen" in response to the prompt "What is the price of the product?", the generation AI verifies whether this output is the correct answer to the prompt. The analysis unit uses the generation AI to analyze the reasons for the incorrect answer based on the results obtained by the verification unit. The analysis unit, for example, identifies which part of the generation AI's output is incorrect based on the correlation between the prompt and the output. For example, if the prompt "Tell me the price of this product" returns the output "The price of this product is 1,000 yen," but the user also enters "I would also like to know the product's stock status" as the desired information, the generation AI analyzes this difference and identifies the reason for the incorrect answer. The suggestion unit uses the generation AI to propose a prompt that brings the generation AI closer to the correct answer based on the results obtained by the analysis unit. For example, the suggestion unit proposes a prompt that brings the generation AI closer to the desired information based on the reason for the incorrect answer. For example, if the prompt "Tell me the price of this product" returns the output "The price of this product is 1,000 yen," but the user also enters "I would also like to know the product's stock status" as the desired information, the generation AI proposes a prompt such as "Tell me the price and stock status of this product." This allows the prompt optimization system according to the embodiment to allow the user to efficiently review the prompt and obtain accurate output.

[0030] The verification unit includes a similarity calculation unit that calculates the similarity between the prompt and the output using natural language processing technology. The similarity calculation unit calculates the similarity between the prompt and the output using natural language processing technology. The similarity calculation unit calculates the similarity using methods such as cosine similarity or Jaccard coefficient. For example, the similarity calculation unit calculates the similarity between the prompt and the output using cosine similarity. The similarity calculation unit can also calculate the similarity using the Jaccard coefficient. The similarity calculation unit can also calculate the similarity using TF-IDF. For example, the similarity calculation unit calculates the angle between two vectors in a vector space model as cosine similarity, and determines that the smaller the angle, the higher the similarity. The Jaccard coefficient is the value obtained by dividing the size of the intersection of two sets by the total size, and indicates the degree of overlap between the sets. TF-IDF is an index that evaluates the importance of words in a document and is used to calculate the similarity between the prompt and the output. By calculating the similarity between the prompt and the output in this way, the accuracy of the correlation is improved. Some or all of the above-described processing in the similarity calculation unit may be performed using, for example, AI, or may be performed without using AI. For example, the similarity calculation unit may calculate the similarity using an AI model that takes the prompt and the output as input and outputs the similarity.

[0031] The analysis unit includes an error analysis unit that uses error analysis or a feedback loop. The error analysis unit analyzes factors that result in an incorrect answer using the error analysis or the feedback loop. The error analysis unit performs error analysis, for example, using an error log analysis or an error classification method. For example, the error analysis unit analyzes the error log to identify the cause of the error. The error analysis unit can also identify the type of error using the error classification method. The error analysis unit can also identify the cause of the error using a feedback loop. For example, the error analysis unit uses a feedback loop to identify the cause of the error and correct the cause. In this way, the use of error analysis or a feedback loop improves the accuracy of identifying factors that result in an incorrect answer. Some or all of the above-described processing in the error analysis unit may be performed using, for example, AI, or may be performed without using AI. For example, the error analysis unit can perform error analysis using an AI model that receives an error log as input and outputs the cause of the error.

[0032] The suggestion unit includes an optimization unit that generates a prompt based on user feedback and past data. The optimization unit generates an optimal prompt based on the user feedback and past data. The optimization unit generates the optimal prompt using, for example, a method for evaluating user feedback and a method for using past data. For example, the optimization unit evaluates user feedback and generates an optimal prompt based on the evaluation result. The optimization unit can also generate the optimal prompt using past data. The optimization unit can also generate the optimal prompt by combining user feedback and past data. For example, the optimization unit develops an algorithm that combines user feedback and past data to generate the optimal prompt. This generates the optimal prompt based on the user feedback and past data, thereby improving the accuracy of suggestions. Some or all of the above-described processing in the optimization unit may be performed using, for example, AI, or may be performed without using AI. For example, the optimization unit can generate the optimal prompt using an AI model that receives user feedback and past data as input and outputs the optimal prompt.

[0033] The reception unit can analyze the user's past input history and suggest an input method. For example, the reception unit can automatically display prompts and results frequently entered by the user in the past as candidates. The reception unit can also prioritize suggesting input methods (such as voice and text) that the user has used in the past. The reception unit can also predict and suggest prompts and results to be used in a specific time period based on the user's past input history. For example, the reception unit can automatically display prompts frequently used by the user in the past as candidates, making it easier for the user to select. The reception unit can also prioritize suggesting input methods that the user has used in the past, allowing the user to input efficiently. Furthermore, the reception unit can analyze the user's past input history and predict and suggest prompts and results to be used in a specific time period, allowing the user to input efficiently. In this way, by analyzing the user's past input history, the optimal input method can be suggested, improving input efficiency. Some or all of the above-mentioned processing in the reception unit may be performed using, or without, AI. For example, the reception unit can input the user's past input history data into a generation AI and have the generation AI suggest an optimal input method.

[0034] The reception unit can change the input fields based on the user's current task or area of ​​interest when the user inputs data. For example, the reception unit can prioritize displaying prompts and result input fields related to the task the user is currently performing. The reception unit can also customize the relevant prompts and result input fields based on the user's area of ​​interest. The reception unit can also automatically display input fields related to a specific project the user is working on. For example, the reception unit can prioritize displaying prompts and result input fields related to the task the user is currently performing, allowing the user to input data efficiently. The reception unit can also customize the relevant prompts and result input fields based on the user's area of ​​interest, allowing the user to input data efficiently. Furthermore, the reception unit can automatically display input fields related to a specific project the user is working on, allowing the user to input data efficiently. This improves input efficiency by customizing the input fields based on the user's current task or area of ​​interest. Some or all of the above-described processing in the reception unit can be performed using, or without, AI. For example, the reception unit can input data on the user's current task or area of ​​interest to a generation AI and have the generation AI customize the input fields.

[0035] The reception unit can select the optimal input means depending on the user's input method (voice, text, image, etc.) at the time of input. For example, if the user prefers voice input, the reception unit can prioritize voice input. Furthermore, if the user prefers text input, the reception unit can also prioritize text input. Furthermore, if the user prefers image input, the reception unit can also prioritize image input. For example, if the user prefers voice input, the reception unit can prioritize voice input, allowing the user to input efficiently. Furthermore, if the user prefers text input, the reception unit can prioritize text input, allowing the user to input efficiently. Furthermore, if the user prefers image input, the reception unit can prioritize image input, allowing the user to input efficiently. This improves input efficiency by selecting the optimal input means depending on the user's input method. Some or all of the above-described processing in the reception unit may be performed using, for example, AI, or may be performed without AI. For example, the reception unit can input data on the user's input method to a generation AI and cause the generation AI to select the optimal input means.

[0036] The reception unit can display highly relevant input fields based on the user's geographical location information when the user enters information. For example, if the user is in a specific region, the reception unit can prioritize displaying prompts and result input fields related to the region. Furthermore, if the user is traveling, the reception unit can prioritize displaying prompts and result input fields related to the travel destination. Furthermore, if the user is participating in a specific event, the reception unit can prioritize displaying prompts and result input fields related to the event. For example, if the user is in a specific region, the reception unit can prioritize displaying prompts and result input fields related to the region, allowing the user to enter information efficiently. Furthermore, if the user is traveling, the reception unit can prioritize displaying prompts and result input fields related to the travel destination, allowing the user to enter information efficiently. Furthermore, if the user is participating in a specific event, the reception unit can prioritize displaying prompts and result input fields related to the event, allowing the user to enter information efficiently. Thus, by preferentially displaying highly relevant input fields based on the user's geographical location information, input efficiency is improved. Some or all of the above-described processing in the reception unit may be performed, for example, using AI or without AI. For example, the reception unit can input the user's geographical location information data into the generation AI and cause the generation AI to display highly relevant input fields.

[0037] The reception unit can analyze the user's social media activity at the time of input and suggest related input content. The reception unit can, for example, suggest related prompts and resultant input content based on content frequently posted by the user on social media. The reception unit can also suggest related prompts and resultant input content by referring to the activities of the user's friends on social media. The reception unit can also suggest related prompts and resultant input content based on the user's check-in information on social media. For example, the reception unit can suggest related prompts and resultant input content based on content frequently posted by the user on social media, thereby enabling the user to input efficiently. The reception unit can also suggest related prompts and resultant input content by referring to the activities of the user's friends on social media, thereby enabling the user to input efficiently. The reception unit can also suggest related prompts and resultant input content based on the user's check-in information on social media, thereby enabling the user to input efficiently. In this way, by analyzing the user's social media activity, related input content is suggested and input efficiency is improved. Some or all of the above-mentioned processing by the reception unit may be performed, for example, using AI or without AI. For example, the reception unit can input the user's social media activity data into the generation AI and have the generation AI execute suggestions for related input content.

[0038] The reception unit can change the input method by reflecting the user's past feedback when inputting data. For example, the reception unit can suggest an optimal input method based on feedback provided by the user in the past. The reception unit can also preferentially provide a specific input method based on the user's past feedback. The reception unit can also analyze the user's past feedback and customize the input interface. For example, the reception unit can suggest an optimal input method based on the user's past feedback, allowing the user to input data efficiently. The reception unit can also preferentially provide a specific input method based on the user's past feedback, allowing the user to input data efficiently. The reception unit can also analyze the user's past feedback and customize the input interface, allowing the user to input data efficiently. In this way, the input method is customized by reflecting the user's past feedback, improving input efficiency. Some or all of the above-described processing by the reception unit may be performed using AI, for example, or may be performed without AI. For example, the reception unit can input the user's past feedback data into a generation AI and cause the generation AI to customize the input method.

[0039] During verification, the verification unit can analyze the correlation between the prompt and the output and extract patterns. For example, the verification unit can analyze the correlation between the prompt and the output to identify frequently occurring patterns. The verification unit can also analyze the correlation between the prompt and the output to identify patterns that are close to the correct answer. The verification unit can also analyze the correlation between the prompt and the output to identify patterns that deviate from the correct answer. For example, the verification unit can analyze the correlation between the prompt and the output to identify frequently occurring patterns and help the user efficiently review the prompt. The verification unit can also analyze the correlation between the prompt and the output to identify patterns that are close to the correct answer and help the user obtain accurate output. The verification unit can also analyze the correlation between the prompt and the output to identify patterns that deviate from the correct answer and help the user improve the prompt. In this way, by analyzing the correlation between the prompt and the output in detail, specific patterns can be extracted and the accuracy of the verification can be improved. Some or all of the above-mentioned processing in the verification unit may be performed, for example, using AI or without using AI. For example, the verification unit can input prompt and output data to the generation AI and cause the generation AI to extract patterns.

[0040] During verification, the verification unit can apply a verification algorithm depending on the category of the prompt. For example, the verification unit selects an optimal verification algorithm depending on the category of the prompt. The verification unit can also apply different verification algorithms depending on the category of the prompt to obtain optimal results. The verification unit can also apply a combination of multiple verification algorithms depending on the category of the prompt. For example, the verification unit selects an optimal verification algorithm depending on the category of the prompt to help the user efficiently review the prompt. The verification unit can also apply different verification algorithms depending on the category of the prompt to help the user obtain accurate output. The verification unit can also apply a combination of multiple verification algorithms depending on the category of the prompt to help the user improve the prompt. In this way, applying different verification algorithms depending on the category of the prompt improves the accuracy of the verification. Some or all of the above-mentioned processing in the verification unit may be performed using, for example, AI, or may be performed without using AI. For example, the verification unit can input prompt category data to a generation AI and cause the generation AI to apply a verification algorithm.

[0041] During verification, the verification unit can improve the verification by referring to the user's past verification results. For example, the verification unit adjusts the verification algorithm based on the user's past verification results. The verification unit can also improve the accuracy of the verification by referring to the user's past verification results. The verification unit can also improve the accuracy of the verification by analyzing the user's past verification results. For example, the verification unit adjusts the verification algorithm based on the user's past verification results to help the user efficiently review prompts. The verification unit also improves the accuracy of the verification by referring to the user's past verification results, helping the user obtain accurate output. Furthermore, the verification unit analyzes the user's past verification results to improve the accuracy of the verification and help the user improve the prompts. In this way, the accuracy of the verification is improved by referring to the user's past verification results. Some or all of the above-described processing in the verification unit may be performed using, for example, AI, or may be performed without using AI. For example, the verification unit can input the user's past verification result data into the generation AI and cause the generation AI to improve the verification.

[0042] During verification, the verification unit can determine the verification priority based on the time at which the prompt was submitted. For example, the verification unit prioritizes verification of prompts that were submitted more recently. The verification unit can also prioritize verification of prompts that were submitted earlier. The verification unit can also dynamically adjust the verification priority based on the time at which the prompt was submitted. For example, the verification unit prioritizes verification of prompts that were submitted more recently, helping the user to review the prompts based on the latest information. The verification unit prioritizes verification of prompts that were submitted earlier, helping the user to review the prompts based on past information. Furthermore, the verification unit dynamically adjusts the verification priority based on the time at which the prompt was submitted, helping the user to efficiently review the prompts. In this way, by determining the verification priority based on the time at which the prompt was submitted, the latest information can be prioritized for verification. Some or all of the above-described processing in the verification unit may be performed using, for example, AI, or may be performed without using AI. For example, the verification unit can input prompt submission time data to the generation AI and cause the generation AI to determine the verification priority.

[0043] The verification unit can change the order of verification based on the relevance of the prompts during verification. For example, the verification unit prioritizes verification of prompts with high relevance. The verification unit can also postpone verification of prompts with low relevance. The verification unit can also dynamically adjust the order of verification based on the relevance of the prompts. For example, the verification unit prioritizes verification of prompts with high relevance, helping the user to efficiently review the prompts. The verification unit also postpones prompts with low relevance, helping the user to efficiently review the prompts. Furthermore, the verification unit dynamically adjusts the order of verification based on the relevance of the prompts, helping the user to efficiently review the prompts. In this way, important information can be prioritized for verification by adjusting the order of verification based on the relevance of the prompts. Some or all of the above-described processing in the verification unit may be performed using, for example, AI, or may be performed without using AI. For example, the verification unit can input prompt relevance data to the generation AI and cause the generation AI to adjust the order of verification.

[0044] The verification unit can change the level of verification detail during verification depending on the user's level of expertise. For example, if the user's level of expertise is high, the verification unit provides a detailed verification result. Furthermore, if the user's level of expertise is low, the verification unit can provide a concise verification result. Furthermore, the verification unit can dynamically adjust the level of verification detail depending on the user's level of expertise. For example, if the user's level of expertise is high, the verification unit provides a detailed verification result to help the user efficiently review the prompt. Furthermore, if the user's level of expertise is low, the verification unit provides a concise verification result to help the user efficiently review the prompt. Furthermore, the verification unit dynamically adjusts the level of verification detail depending on the user's level of expertise to help the user efficiently review the prompt. Thus, by adjusting the level of verification detail depending on the user's level of expertise, information appropriate for the user can be provided. Some or all of the above-described processing in the verification unit may be performed using, or without, AI. For example, the verification unit may input the user's level of expertise data to the generation AI and cause the generation AI to adjust the level of verification detail.

[0045] During the analysis, the analysis unit can analyze the frequency and patterns of errors. For example, the analysis unit can analyze the frequency of errors to identify errors that occur frequently. The analysis unit can also analyze error patterns to identify errors that occur under specific conditions. The analysis unit can also combine and analyze the frequency and patterns of errors to identify the root cause of the error. For example, the analysis unit can analyze the frequency of errors to identify frequently occurring errors and help the user efficiently correct the errors. The analysis unit can also analyze error patterns to identify errors that occur under specific conditions and help the user efficiently correct the errors. Furthermore, the analysis unit can combine and analyze the frequency and patterns of errors to identify the root cause of the error and help the user efficiently correct the errors. This makes it easier to identify the root cause of the error by analyzing the frequency and patterns of errors in detail. Some or all of the above-mentioned processing in the analysis unit may be performed using, for example, AI, or may be performed without AI. For example, the analysis unit can input error frequency and pattern data to the generation AI and have the generation AI perform error analysis.

[0046] During analysis, the analysis unit can apply an analysis method according to the category of the prompt. For example, the analysis unit selects an optimal analysis method according to the category of the prompt. The analysis unit can also apply different analysis methods according to the category of the prompt to obtain optimal results. The analysis unit can also apply a combination of multiple analysis methods according to the category of the prompt. For example, the analysis unit selects an optimal analysis method according to the category of the prompt to help the user efficiently correct errors. The analysis unit can also apply different analysis methods according to the category of the prompt to help the user efficiently correct errors. Furthermore, the analysis unit can apply a combination of multiple analysis methods according to the category of the prompt to help the user efficiently correct errors. In this way, applying different analysis methods according to the category of the prompt improves the accuracy of the analysis. Some or all of the above-mentioned processing in the analysis unit may be performed using, for example, AI, or may be performed without using AI. For example, the analysis unit can input prompt category data to a generation AI and cause the generation AI to apply an analysis method.

[0047] During analysis, the analysis unit can improve the analysis by referring to the user's past analysis results. For example, the analysis unit adjusts the analysis method based on the user's past analysis results. The analysis unit can also improve the accuracy of the analysis by referring to the user's past analysis results. The analysis unit can also analyze the user's past analysis results to improve the accuracy of the analysis. For example, the analysis unit adjusts the analysis method based on the user's past analysis results to help the user efficiently correct errors. The analysis unit can also improve the accuracy of the analysis by referring to the user's past analysis results. The analysis unit can also improve the accuracy of the analysis by referring to the user's past analysis results. Furthermore, the analysis unit can analyze the user's past analysis results to improve the accuracy of the analysis and help the user efficiently correct errors. In this way, the accuracy of the analysis is improved by referring to the user's past analysis results. Some or all of the above-described processing in the analysis unit may be performed using, for example, AI, or may be performed without using AI. For example, the analysis unit can input the user's past analysis result data into the generation AI and cause the generation AI to improve the analysis.

[0048] During analysis, the analysis unit can determine the order of analysis based on the time of error occurrence. For example, the analysis unit prioritizes analysis of recent errors. The analysis unit can also prioritize analysis of older errors. The analysis unit can also dynamically adjust the analysis priority based on the time of error occurrence. For example, the analysis unit prioritizes analysis of recent errors to help the user correct errors based on the latest information. The analysis unit prioritizes analysis of older errors to help the user correct errors based on past information. The analysis unit can also dynamically adjust the analysis priority based on the time of error occurrence to help the user efficiently correct errors. In this way, by determining the analysis priority based on the time of error occurrence, the latest information can be prioritized for analysis. Some or all of the above-described processing in the analysis unit may be performed using, for example, AI, or may be performed without AI. For example, the analysis unit can input data on the time of error occurrence to the generation AI and have the generation AI determine the analysis priority.

[0049] During analysis, the analysis unit can change the order of analysis based on the relevance of errors. For example, the analysis unit prioritizes analysis of errors with high relevance. The analysis unit can also postpone analysis of errors with low relevance. The analysis unit can also dynamically adjust the order of analysis based on the relevance of errors. For example, the analysis unit prioritizes analysis of errors with high relevance to help the user efficiently correct errors. The analysis unit also postpones analysis of errors with low relevance to help the user efficiently correct errors. Furthermore, the analysis unit dynamically adjusts the order of analysis based on the relevance of errors to help the user efficiently correct errors. In this way, adjusting the order of analysis based on the relevance of errors allows important information to be analyzed preferentially. Some or all of the above-described processing in the analysis unit may be performed using, for example, AI, or may be performed without using AI. For example, the analysis unit can input error relevance data to the generation AI and cause the generation AI to adjust the order of analysis.

[0050] During analysis, the analysis unit can change the level of detail of the analysis according to the user's level of expertise. For example, if the user's level of expertise is high, the analysis unit can provide detailed analysis results. Furthermore, if the user's level of expertise is low, the analysis unit can provide concise analysis results. Furthermore, the analysis unit can dynamically adjust the level of detail of the analysis according to the user's level of expertise. For example, if the user's level of expertise is high, the analysis unit can provide detailed analysis results to help the user efficiently correct errors. Furthermore, if the user's level of expertise is low, the analysis unit can provide concise analysis results to help the user efficiently correct errors. Furthermore, the analysis unit can dynamically adjust the level of detail of the analysis according to the user's level of expertise to help the user efficiently correct errors. In this way, by adjusting the level of detail of the analysis according to the user's level of expertise, it is possible to provide information appropriate for the user. Some or all of the above-described processing in the analysis unit may be performed using, or without, AI. For example, the analysis unit can input the user's level of expertise data into the generation AI and cause the generation AI to adjust the level of detail of the analysis.

[0051] The suggestion unit can change the level of detail of the suggestion based on the importance of the prompt when making a suggestion. For example, the suggestion unit provides a detailed suggestion when the importance of the prompt is high. The suggestion unit can also provide a concise suggestion when the importance of the prompt is low. The suggestion unit can also dynamically adjust the level of detail of the suggestion based on the importance of the prompt. For example, the suggestion unit provides a detailed suggestion when the importance of the prompt is high, helping the user to efficiently review the prompt. The suggestion unit provides a concise suggestion when the importance of the prompt is low, helping the user to efficiently review the prompt. Furthermore, the suggestion unit dynamically adjusts the level of detail of the suggestion based on the importance of the prompt, helping the user to efficiently review the prompt. In this way, by adjusting the level of detail of the suggestion based on the importance of the prompt, important information can be provided preferentially. Some or all of the above-described processing in the suggestion unit may be performed using AI, for example, or may be performed without using AI. For example, the suggestion unit can input prompt importance data to a generation AI and cause the generation AI to adjust the level of detail of the suggestion.

[0052] The suggestion unit can apply a suggestion algorithm depending on the category of the prompt when making a suggestion. For example, the suggestion unit selects an optimal suggestion algorithm depending on the category of the prompt. The suggestion unit can also apply different suggestion algorithms depending on the category of the prompt to obtain optimal results. The suggestion unit can also apply a combination of multiple suggestion algorithms depending on the category of the prompt. For example, the suggestion unit selects an optimal suggestion algorithm depending on the category of the prompt to help the user efficiently review the prompt. The suggestion unit can also apply different suggestion algorithms depending on the category of the prompt to help the user efficiently review the prompt. Furthermore, the suggestion unit can apply a combination of multiple suggestion algorithms depending on the category of the prompt to help the user efficiently review the prompt. In this way, applying different suggestion algorithms depending on the category of the prompt improves the accuracy of the suggestion. Some or all of the above-described processing in the suggestion unit may be performed using AI, for example, or may be performed without using AI. For example, the suggestion unit can input prompt category data to a generation AI and cause the generation AI to apply a suggestion algorithm.

[0053] When making a suggestion, the suggestion unit can improve the suggestion by referring to the user's past suggestion results. For example, the suggestion unit adjusts the suggestion algorithm based on the user's past suggestion results. The suggestion unit can also improve the accuracy of the suggestion by referring to the user's past suggestion results. The suggestion unit can also improve the accuracy of the suggestion by analyzing the user's past suggestion results. For example, the suggestion unit adjusts the suggestion algorithm based on the user's past suggestion results to help the user efficiently review the prompt. The suggestion unit can also improve the accuracy of the suggestion by referring to the user's past suggestion results, to help the user efficiently review the prompt. Furthermore, the suggestion unit analyzes the user's past suggestion results to improve the accuracy of the suggestion, to help the user efficiently review the prompt. In this way, the accuracy of the suggestion is improved by referring to the user's past suggestion results. Some or all of the above-described processing in the suggestion unit may be performed using AI, for example, or may be performed without using AI. For example, the suggestion unit can input the user's past suggestion result data to the generation AI and cause the generation AI to improve the suggestion.

[0054] When making a suggestion, the suggestion unit can determine the order of the suggestions based on the time at which the prompts were submitted. For example, the suggestion unit can prioritize suggesting prompts that were submitted more recently. The suggestion unit can also prioritize suggesting prompts that were submitted earlier. The suggestion unit can also dynamically adjust the priority of the suggestions based on the time at which the prompts were submitted. For example, the suggestion unit can prioritize suggesting prompts that were submitted more recently, helping the user to review the prompts based on the latest information. The suggestion unit can prioritize suggesting prompts that were submitted earlier, helping the user to review the prompts based on past information. The suggestion unit can also dynamically adjust the priority of the suggestions based on the time at which the prompts were submitted, helping the user to efficiently review the prompts. In this way, by determining the priority of the suggestions based on the time at which the prompts were submitted, the latest information can be prioritized for suggestion. Some or all of the above-described processing by the suggestion unit may be performed using, or without, AI. For example, the suggestion unit can input prompt submission time data to a generation AI and cause the generation AI to determine the priority of the suggestions.

[0055] The suggestion unit can change the order of suggestions based on the relevance of the prompts when making suggestions. For example, the suggestion unit prioritizes suggesting prompts with high relevance. The suggestion unit can also postpone prompts with low relevance. The suggestion unit can also dynamically adjust the order of suggestions based on the relevance of the prompts. For example, the suggestion unit prioritizes suggesting prompts with high relevance, helping the user to efficiently review the prompts. The suggestion unit also postpones prompts with low relevance, helping the user to efficiently review the prompts. Furthermore, the suggestion unit dynamically adjusts the order of suggestions based on the relevance of the prompts, helping the user to efficiently review the prompts. In this way, important information can be prioritized by adjusting the order of suggestions based on the relevance of the prompts. Some or all of the above-described processing in the suggestion unit may be performed using AI, for example, or may be performed without using AI. For example, the suggestion unit can input prompt relevance data to a generation AI and cause the generation AI to adjust the order of suggestions.

[0056] The suggestion unit may change the use of technical terminology in the suggestion depending on the user's level of expertise. For example, if the user's level of expertise is high, the suggestion unit may provide a suggestion that uses a lot of technical terminology. Furthermore, if the user's level of expertise is low, the suggestion unit may provide a concise and easy-to-understand suggestion. Furthermore, the suggestion unit may dynamically adjust the use of technical terminology in the suggestion depending on the user's level of expertise. For example, if the user's level of expertise is high, the suggestion unit may provide a suggestion that uses a lot of technical terminology, thereby helping the user to efficiently review the prompt. Furthermore, if the user's level of expertise is low, the suggestion unit may provide a concise and easy-to-understand suggestion, thereby helping the user to efficiently review the prompt. Furthermore, the suggestion unit may dynamically adjust the use of technical terminology in the suggestion depending on the user's level of expertise, thereby helping the user to efficiently review the prompt. By adjusting the use of technical terminology in the suggestion depending on the user's level of expertise, it is possible to provide information suitable for the user. Some or all of the above-described processing in the suggestion unit may be performed using, for example, AI, or may be performed without using AI. For example, the suggestion unit can input the user's expertise level data into the generation AI and have the generation AI execute the suggested terminology usage.

[0057] The similarity calculation unit can analyze the context of the prompt and the output when calculating the similarity. For example, the similarity calculation unit analyzes the context of the prompt and the output to identify frequently occurring patterns. The similarity calculation unit can also analyze the context of the prompt and the output to identify patterns that are close to the correct answer. The similarity calculation unit can also analyze the context of the prompt and the output to identify patterns that deviate from the correct answer. For example, the similarity calculation unit analyzes the context of the prompt and the output to identify frequently occurring patterns and help the user efficiently review the prompt. The similarity calculation unit also analyzes the context of the prompt and the output to identify patterns that are close to the correct answer and help the user obtain an accurate output. Furthermore, the similarity calculation unit analyzes the context of the prompt and the output to identify patterns that deviate from the correct answer and help the user improve the prompt. In this way, by analyzing the context of the prompt and the output in detail, specific patterns can be extracted and the accuracy of the similarity calculation can be improved. Some or all of the above-mentioned processing in the similarity calculation unit may be performed, for example, using AI or without AI. For example, the similarity calculation unit can input prompt and output data into the generation AI and have the generation AI perform context analysis.

[0058] The similarity calculation unit can apply a calculation algorithm depending on the category of the prompt when calculating the similarity. The similarity calculation unit, for example, selects an optimal calculation algorithm depending on the category of the prompt. The similarity calculation unit can also apply different calculation algorithms depending on the category of the prompt to obtain optimal results. The similarity calculation unit can also apply a combination of multiple calculation algorithms depending on the category of the prompt. For example, the similarity calculation unit selects an optimal calculation algorithm depending on the category of the prompt to help the user efficiently review the prompt. The similarity calculation unit can also apply different calculation algorithms depending on the category of the prompt to help the user efficiently review the prompt. Furthermore, the similarity calculation unit can apply a combination of multiple calculation algorithms depending on the category of the prompt to help the user efficiently review the prompt. In this way, applying different calculation algorithms depending on the category of the prompt improves the accuracy of the similarity calculation. Some or all of the above-mentioned processing in the similarity calculation unit may be performed using, for example, AI, or may be performed without using AI. For example, the similarity calculation unit can input prompt category data to a generation AI and cause the generation AI to apply a calculation algorithm.

[0059] The similarity calculation unit can improve the calculation by referring to the user's past calculation results when calculating the similarity. For example, the similarity calculation unit adjusts the calculation algorithm based on the user's past calculation results. The similarity calculation unit can also improve the accuracy of the calculation by referring to the user's past calculation results. The similarity calculation unit can also improve the accuracy of the calculation by analyzing the user's past calculation results. For example, the similarity calculation unit adjusts the calculation algorithm based on the user's past calculation results to help the user efficiently review the prompt. The similarity calculation unit can also improve the accuracy of the calculation by referring to the user's past calculation results to help the user efficiently review the prompt. The similarity calculation unit can also analyze the user's past calculation results to improve the accuracy of the calculation to help the user efficiently review the prompt. In addition, the similarity calculation unit can analyze the user's past calculation results to improve the accuracy of the calculation to help the user efficiently review the prompt. In this way, the accuracy of the calculation is improved by referring to the user's past calculation results. Some or all of the above-mentioned processing in the similarity calculation unit may be performed using, for example, AI, or may be performed without using AI. For example, the similarity calculation unit can input the user's past calculation result data into the generation AI and cause the generation AI to improve the calculation.

[0060] When calculating the similarity, the similarity calculation unit can determine the calculation order based on the time at which the prompt was submitted. For example, the similarity calculation unit prioritizes the most recent prompt in the calculation. The similarity calculation unit can also prioritize the oldest prompt in the calculation. The similarity calculation unit can also dynamically adjust the calculation priority based on the time at which the prompt was submitted. For example, the similarity calculation unit prioritizes the most recent prompt in the calculation, helping the user to review the prompt based on the latest information. The similarity calculation unit prioritizes the oldest prompt in the calculation, helping the user to review the prompt based on past information. The similarity calculation unit also dynamically adjusts the calculation priority based on the time at which the prompt was submitted, helping the user to efficiently review the prompt. In this way, by determining the calculation priority based on the time at which the prompt was submitted, the most recent information can be prioritized in the calculation. Some or all of the above-described processing in the similarity calculation unit may be performed, for example, using AI or without AI. For example, the similarity calculation unit can input prompt submission timing data to the generation AI and have the generation AI determine the priority of the calculations.

[0061] The similarity calculation unit can change the order of calculations based on the relevance of the prompts when calculating the similarity. For example, the similarity calculation unit prioritizes calculations of prompts with high relevance. The similarity calculation unit can also postpone calculations of prompts with low relevance. The similarity calculation unit can also dynamically adjust the order of calculations based on the relevance of the prompts. For example, the similarity calculation unit prioritizes calculations of prompts with high relevance, helping the user to efficiently review the prompts. The similarity calculation unit also postpones prompts with low relevance, helping the user to efficiently review the prompts. Furthermore, the similarity calculation unit dynamically adjusts the order of calculations based on the relevance of the prompts, helping the user to efficiently review the prompts. In this way, by adjusting the order of calculations based on the relevance of the prompts, important information can be calculated with priority. Some or all of the above-described processing in the similarity calculation unit may be performed using, for example, AI, or may be performed without using AI. For example, the similarity calculation unit can input prompt relevance data to the generation AI and cause the generation AI to adjust the order of calculations.

[0062] The similarity calculation unit can change the level of detail of the calculation according to the user's level of expertise when calculating the similarity. For example, if the user's level of expertise is high, the similarity calculation unit provides a detailed calculation result. Furthermore, if the user's level of expertise is low, the similarity calculation unit can provide a concise calculation result. Furthermore, the similarity calculation unit can dynamically adjust the level of detail of the calculation according to the user's level of expertise. For example, if the user's level of expertise is high, the similarity calculation unit provides a detailed calculation result to help the user efficiently review the prompt. Furthermore, if the user's level of expertise is low, the similarity calculation unit provides a concise calculation result to help the user efficiently review the prompt. Furthermore, the similarity calculation unit dynamically adjusts the level of detail of the calculation according to the user's level of expertise to help the user efficiently review the prompt. By adjusting the level of detail of the calculation according to the user's level of expertise, information suitable for the user can be provided. Some or all of the above-described processing in the similarity calculation unit may be performed, for example, using AI or without AI. For example, the similarity calculation unit can input the user's expertise level data into the generation AI and cause the generation AI to adjust the level of detail of the calculation.

[0063] The error analysis unit can analyze the frequency and pattern of errors during error analysis. For example, the error analysis unit analyzes the frequency of error occurrence and identifies errors that occur frequently. The error analysis unit can also analyze error patterns and identify errors that occur under specific conditions. The error analysis unit can also combine and analyze the frequency and pattern of errors to identify the root cause of the error. For example, the error analysis unit analyzes the frequency of error occurrence and identifies frequently occurring errors, helping the user to efficiently correct the errors. The error analysis unit can also analyze the pattern of errors and identify errors that occur under specific conditions, helping the user to efficiently correct the errors. Furthermore, the error analysis unit combines and analyzes the frequency and pattern of errors to identify the root cause of the error, helping the user to efficiently correct the errors. This makes it easier to identify the root cause of the error by analyzing the frequency and pattern of the error in detail. Some or all of the above-described processing in the error analysis unit may be performed using, or without, AI. For example, the error analysis unit can input error frequency and pattern data to the generation AI and have the generation AI perform error analysis.

[0064] When analyzing an error, the error analysis unit can apply an analysis method depending on the category of the prompt. For example, the error analysis unit selects an optimal analysis method depending on the category of the prompt. The error analysis unit can also apply different analysis methods depending on the category of the prompt to obtain optimal results. The error analysis unit can also apply a combination of multiple analysis methods depending on the category of the prompt. For example, the error analysis unit selects an optimal analysis method depending on the category of the prompt to help the user efficiently correct the error. The error analysis unit can also apply different analysis methods depending on the category of the prompt to help the user efficiently correct the error. Furthermore, the error analysis unit can apply a combination of multiple analysis methods depending on the category of the prompt to help the user efficiently correct the error. In this way, applying different analysis methods depending on the category of the prompt improves the accuracy of the analysis. Some or all of the above-mentioned processing in the error analysis unit may be performed using, for example, AI, or may be performed without using AI. For example, the error analysis unit can input prompt category data to the generation AI and cause the generation AI to apply the analysis method.

[0065] During error analysis, the error analysis unit can improve the analysis by referring to the user's past analysis results. For example, the error analysis unit adjusts the analysis method based on the user's past analysis results. The error analysis unit can also improve the accuracy of the analysis by referring to the user's past analysis results. The error analysis unit can also analyze the user's past analysis results to improve the accuracy of the analysis. For example, the error analysis unit adjusts the analysis method based on the user's past analysis results to help the user efficiently correct errors. The error analysis unit also improves the accuracy of the analysis by referring to the user's past analysis results, helping the user efficiently correct errors. Furthermore, the error analysis unit analyzes the user's past analysis results to improve the accuracy of the analysis, helping the user efficiently correct errors. In this way, the accuracy of the analysis is improved by referring to the user's past analysis results. Some or all of the above-described processing in the error analysis unit may be performed, for example, using AI, or may be performed without using AI. For example, the error analysis unit can input the user's past analysis result data into the generation AI and cause the generation AI to improve the analysis.

[0066] During error analysis, the error analysis unit can determine the analysis priority based on the time of error occurrence. For example, the error analysis unit prioritizes analysis of recent errors. The error analysis unit can also prioritize analysis of older errors. The error analysis unit can also dynamically adjust the analysis priority based on the time of error occurrence. For example, the error analysis unit prioritizes analysis of recent errors to help the user correct errors based on the latest information. The error analysis unit also prioritizes analysis of older errors to help the user correct errors based on past information. The error analysis unit can also dynamically adjust the analysis priority based on the time of error occurrence to help the user efficiently correct errors. By determining the analysis priority based on the time of error occurrence, the latest information can be prioritized for analysis. Some or all of the above-described processing in the error analysis unit may be performed using, or without, AI. For example, the error analysis unit can input error occurrence time data to the generation AI and have the generation AI determine the analysis priority.

[0067] During error analysis, the error analysis unit can change the analysis order based on the relevance of the errors. For example, the error analysis unit prioritizes analysis of errors with high relevance. The error analysis unit can also postpone analysis of errors with low relevance. The error analysis unit can also dynamically adjust the analysis order based on the relevance of the errors. For example, the error analysis unit prioritizes analysis of errors with high relevance, helping the user to efficiently correct errors. The error analysis unit also postpones analysis of errors with low relevance, helping the user to efficiently correct errors. Furthermore, the error analysis unit dynamically adjusts the analysis order based on the relevance of the errors, helping the user to efficiently correct errors. In this way, adjusting the analysis order based on the relevance of the errors allows important information to be analyzed preferentially. Some or all of the above-described processing in the error analysis unit may be performed using, for example, AI, or may be performed without using AI. For example, the error analysis unit can input error relevance data to the generation AI and cause the generation AI to adjust the analysis order.

[0068] During error analysis, the error analysis unit can change the level of detail of the analysis according to the user's level of expertise. For example, if the user's level of expertise is high, the error analysis unit can provide detailed analysis results. Furthermore, if the user's level of expertise is low, the error analysis unit can provide concise analysis results. Furthermore, the error analysis unit can dynamically adjust the level of detail of the analysis according to the user's level of expertise. For example, if the user's level of expertise is high, the error analysis unit can provide detailed analysis results to help the user efficiently correct errors. Furthermore, if the user's level of expertise is low, the error analysis unit can provide concise analysis results to help the user efficiently correct errors. Furthermore, the error analysis unit can dynamically adjust the level of detail of the analysis according to the user's level of expertise to help the user efficiently correct errors. In this way, by adjusting the level of detail of the analysis according to the user's level of expertise, it is possible to provide information appropriate for the user. Some or all of the above-described processing in the error analysis unit may be performed using, or without, AI. For example, the error analysis unit can input the user's level of expertise data into the generation AI and cause the generation AI to adjust the level of detail of the analysis.

[0069] The optimization unit can change the level of detail of the optimization based on the importance of the prompt during optimization. For example, when the importance of the prompt is high, the optimization unit provides a detailed optimization result. Furthermore, when the importance of the prompt is low, the optimization unit can provide a concise optimization result. Furthermore, the optimization unit can dynamically adjust the level of detail of the optimization based on the importance of the prompt. For example, when the importance of the prompt is high, the optimization unit provides a detailed optimization result to help the user efficiently review the prompt. Furthermore, when the importance of the prompt is low, the optimization unit provides a concise optimization result to help the user efficiently review the prompt. Furthermore, the optimization unit dynamically adjusts the level of detail of the optimization based on the importance of the prompt to help the user efficiently review the prompt. Thus, by adjusting the level of detail of the optimization based on the importance of the prompt, important information can be provided preferentially. Some or all of the above-described processing in the optimization unit may be performed using, or without, AI. For example, the optimization unit can input prompt importance data to the generation AI and cause the generation AI to adjust the level of detail of the optimization.

[0070] During optimization, the optimization unit can apply an optimization algorithm depending on the category of the prompt. For example, the optimization unit selects an optimal optimization algorithm depending on the category of the prompt. The optimization unit can also apply different optimization algorithms depending on the category of the prompt to obtain optimal results. The optimization unit can also apply a combination of multiple optimization algorithms depending on the category of the prompt. For example, the optimization unit selects an optimal optimization algorithm depending on the category of the prompt to help the user efficiently review the prompt. The optimization unit can also apply different optimization algorithms depending on the category of the prompt to help the user efficiently review the prompt. Furthermore, the optimization unit can apply a combination of multiple optimization algorithms depending on the category of the prompt to help the user efficiently review the prompt. In this way, applying different optimization algorithms depending on the category of the prompt improves the accuracy of optimization. Some or all of the above-mentioned processing in the optimization unit may be performed using AI, for example, or may be performed without using AI. For example, the optimization unit can input prompt category data to the generation AI and cause the generation AI to apply the optimization algorithm.

[0071] During optimization, the optimization unit can improve optimization by referring to the user's past optimization results. For example, the optimization unit adjusts the optimization algorithm based on the user's past optimization results. The optimization unit can also improve the accuracy of optimization by referring to the user's past optimization results. The optimization unit can also analyze the user's past optimization results to improve the accuracy of optimization. For example, the optimization unit adjusts the optimization algorithm based on the user's past optimization results to help the user efficiently review prompts. The optimization unit can also improve the accuracy of optimization by referring to the user's past optimization results to help the user efficiently review prompts. Furthermore, the optimization unit analyzes the user's past optimization results to improve the accuracy of optimization to help the user efficiently review prompts. In this way, the accuracy of optimization is improved by referring to the user's past optimization results. Some or all of the above-described processing in the optimization unit may be performed using, for example, AI, or may be performed without using AI. For example, the optimization unit can input the user's past optimization result data into the generation AI and cause the generation AI to improve the optimization.

[0072] During optimization, the optimization unit can determine the order of optimization based on the submission time of the prompt. For example, the optimization unit prioritizes optimizing a prompt that was submitted more recently. The optimization unit can also prioritize optimizing a prompt that was submitted earlier. The optimization unit can also dynamically adjust the optimization priority based on the submission time of the prompt. For example, the optimization unit prioritizes optimizing a prompt that was submitted more recently, helping the user to review the prompt based on the latest information. The optimization unit prioritizes optimizing a prompt that was submitted earlier, helping the user to review the prompt based on past information. Furthermore, the optimization unit dynamically adjusts the optimization priority based on the submission time of the prompt, helping the user to efficiently review the prompt. In this way, by determining the optimization priority based on the submission time of the prompt, the latest information can be prioritized for optimization. Some or all of the above-described processing in the optimization unit may be performed using, for example, AI, or may be performed without using AI. For example, the optimization unit can input prompt submission time data to the generation AI and cause the generation AI to determine the optimization priority.

[0073] The optimization unit can change the order of optimization based on the relevance of the prompts during optimization. For example, the optimization unit prioritizes optimizing prompts with high relevance. The optimization unit can also postpone prompts with low relevance. The optimization unit can also dynamically adjust the order of optimization based on the relevance of the prompts. For example, the optimization unit prioritizes optimizing prompts with high relevance, helping the user to efficiently review the prompts. The optimization unit also postpones prompts with low relevance, helping the user to efficiently review the prompts. Furthermore, the optimization unit dynamically adjusts the order of optimization based on the relevance of the prompts, helping the user to efficiently review the prompts. In this way, by adjusting the order of optimization based on the relevance of the prompts, important information can be prioritized for optimization. Some or all of the above-described processing in the optimization unit may be performed using AI, for example, or may be performed without using AI. For example, the optimization unit can input prompt relevance data to the generation AI and cause the generation AI to adjust the order of optimization.

[0074] During optimization, the optimization unit can change the level of detail of the optimization according to the user's level of expertise. For example, if the user's level of expertise is high, the optimization unit provides detailed optimization results. Furthermore, if the user's level of expertise is low, the optimization unit can provide concise optimization results. Furthermore, the optimization unit can dynamically adjust the level of detail of the optimization according to the user's level of expertise. For example, if the user's level of expertise is high, the optimization unit provides detailed optimization results to help the user efficiently review prompts. Furthermore, if the user's level of expertise is low, the optimization unit provides concise optimization results to help the user efficiently review prompts. Furthermore, the optimization unit dynamically adjusts the level of detail of the optimization according to the user's level of expertise to help the user efficiently review prompts. Thus, by adjusting the level of detail of the optimization according to the user's level of expertise, information suitable for the user can be provided. Some or all of the above-described processing in the optimization unit may be performed using, or without, AI. For example, the optimization unit may input the user's expertise level data to the generation AI and cause the generation AI to adjust the level of detail of the optimization.

[0075] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0076] The reception unit can analyze the user's input content in real time and provide appropriate suggestions based on the input content. For example, if the user inputs "Tell me the price of the product," the reception unit can display a suggestion such as "Would you also like to know the product's stock status?". Also, if the user inputs "Tell me the product reviews," the reception unit can display a suggestion such as "Would you also like to know the product's rating score?". Furthermore, if the user inputs "Tell me the product's delivery status," the reception unit can display a suggestion such as "Would you also like to know the estimated delivery date?". In this way, appropriate suggestions can be provided according to the user's input content, allowing the user to input information efficiently.

[0077] The similarity calculation unit can search for related past data and calculate the similarity based on the user's input. For example, if the user inputs "Tell me the price of this product," the similarity calculation unit searches for data in which a similar prompt was previously input and calculates the similarity. Also, if the user inputs "Tell me product reviews," the similarity calculation unit can search for data related to past reviews and calculate the similarity. Furthermore, if the user inputs "Tell me the delivery status of this product," the similarity calculation unit can search for data related to past delivery status and calculate the similarity. In this way, by searching for related past data based on the user's input and calculating the similarity, more accurate information can be provided.

[0078] The error analysis unit can identify the cause of the error based on the user's input and suggest a correction method. For example, if a user inputs "Tell me the price of this product" but the correct price is not displayed, the error analysis unit can identify the cause and suggest a correction method such as "The price information of this product may not have been updated." Also, if a user inputs "Tell me reviews of this product" but no reviews are displayed, the error analysis unit can identify the cause and suggest a correction method such as "The review information may not be sufficient." Furthermore, if a user inputs "Tell me the delivery status of this product" but the delivery status is not displayed, the error analysis unit can identify the cause and suggest a correction method such as "The delivery information may not have been updated." In this way, by identifying the cause of the error based on the user's input and suggesting a correction method, the user can efficiently correct errors.

[0079] The optimization unit can generate an optimal prompt based on the user's input. For example, if the user inputs "Tell me the price of the product," the optimization unit can generate an optimal prompt such as "Tell me the price and stock status of the product." Also, if the user inputs "Tell me the reviews of the product," the optimization unit can generate an optimal prompt such as "Tell me the reviews and rating score of the product." Furthermore, if the user inputs "Tell me the delivery status of the product," the optimization unit can generate an optimal prompt such as "Tell me the delivery status and expected delivery date of the product." In this way, generating an optimal prompt based on the user's input allows the user to input information efficiently.

[0080] The reception unit can analyze the user's past input history and suggest an input method. For example, it can automatically display prompts that the user has frequently used in the past and the results thereof as candidates. The reception unit can also preferentially suggest input methods (such as voice and text) that the user has used in the past. The reception unit can also predict and suggest prompts and results to be used in a specific time period based on the user's past input history. For example, the reception unit can automatically display prompts that the user has frequently used in the past as candidates, making it easier for the user to select. The reception unit can also preferentially suggest input methods that the user has used in the past, allowing the user to input efficiently. The reception unit can also analyze the user's past input history and predict and suggest prompts and results to be used in a specific time period, allowing the user to input efficiently. In this way, the optimal input method can be suggested by analyzing the user's past input history, improving input efficiency.

[0081] The reception unit can change the input fields based on the user's current task or area of ​​interest when the user is entering data. For example, the reception unit can preferentially display prompts and result input fields related to the task the user is currently performing. The reception unit can also customize related prompts and result input fields based on the user's area of ​​interest. The reception unit can also automatically display input fields related to a specific project when the user is working on that project. For example, the reception unit preferentially displays prompts and result input fields related to the task the user is currently performing, allowing the user to enter data efficiently. The reception unit can also customize related prompts and result input fields based on the user's area of ​​interest, allowing the user to enter data efficiently. The reception unit can also automatically display input fields related to the project when the user is working on a specific project, allowing the user to enter data efficiently. This improves input efficiency by customizing input fields based on the user's current task or area of ​​interest.

[0082] The reception unit can select the optimal input means depending on the user's input method (voice, text, image, etc.) at the time of input. For example, if the user prefers voice input, the reception unit can provide voice input preferentially. Furthermore, if the user prefers text input, the reception unit can also provide text input preferentially. Furthermore, if the user prefers image input, the reception unit can also provide image input preferentially. For example, if the user prefers voice input, the reception unit can provide voice input preferentially, allowing the user to input efficiently. Furthermore, if the user prefers text input, the reception unit can provide text input preferentially, allowing the user to input efficiently. Furthermore, if the user prefers image input, the reception unit can provide image input preferentially, allowing the user to input efficiently. In this way, input efficiency is improved by selecting the optimal input means depending on the user's input method.

[0083] The reception unit can display highly relevant input fields based on the user's geographical location information when the user enters information. For example, if the user is in a specific region, the reception unit can prioritize displaying prompts and result input fields related to that region. Furthermore, if the user is traveling, the reception unit can prioritize displaying prompts and result input fields related to the travel destination. Furthermore, if the user is participating in a specific event, the reception unit can prioritize displaying prompts and result input fields related to the event. For example, if the user is in a specific region, the reception unit can prioritize displaying prompts and result input fields related to the region, allowing the user to enter information efficiently. Furthermore, if the user is traveling, the reception unit can prioritize displaying prompts and result input fields related to the travel destination, allowing the user to enter information efficiently. Furthermore, if the user is participating in a specific event, the reception unit can prioritize displaying prompts and result input fields related to the event, allowing the user to enter information efficiently. In this way, input efficiency is improved by preferentially displaying highly relevant input fields based on the user's geographical location information.

[0084] The processing flow of the first embodiment will be briefly explained below.

[0085] Step 1: The reception unit receives input from the user of the Base prompt, and the difference between the result and the desired information. The information input by the user includes the specific prompt, its output, and the difference between the desired information. For example, if the user receives an output such as "The price of the product is 1,000 yen" in response to a prompt such as "Please tell me the price of the product," the user can input "I would also like to know the product's stock status" as the difference between the desired information and the actual information. Step 2: The verification unit uses the generation AI to verify the correlation between the prompt and the output based on the information entered in the reception unit. For example, the generation AI analyzes the relationship between the prompt and its output, and identifies which parts are close to the correct answer and which parts deviate from the correct answer. For example, if the output obtained in response to the prompt "Tell me the price of this product" is "The price of this product is 1,000 yen," the verification unit verifies whether this output is the correct answer to the prompt. Step 3: The analysis section uses the generation AI to analyze the factors that cause the answer to be incorrect based on the results obtained by the verification section. For example, the generation AI can identify which part is causing the incorrect answer based on the correlation between the prompt and the output. For example, if the prompt "Tell me the price of this product" returns the output "The price of this product is 1,000 yen," but the user enters the desired information as "I also want to know the product's stock status," the generation AI will analyze this difference and identify the factors that cause the incorrect answer. Step 4: The suggestion unit uses the generation AI to propose prompts that bring the user closer to the correct answer based on the results obtained by the analysis unit. For example, based on the factors that caused the generation AI to give an incorrect answer, the suggestion unit proposes prompts that bring the user closer to the information they want. For example, if the output obtained in response to the prompt "Tell me the price of this product" is "The price of this product is 1,000 yen," but the user enters the desired information as "I would also like to know the product's stock status," the generation AI will propose a prompt such as "Tell me the price and stock status of this product."

[0086] (Example 2) A prompt optimization system according to an embodiment of the present invention uses a generation AI to verify the correlation between a prompt and its output, analyze factors that lead to an incorrect answer, and propose prompts that bring the user closer to the correct answer. In the prompt optimization system, a user inputs the difference between a base prompt and its result, the desired information. The generation AI then verifies the correlation between the prompt and the output, analyzes factors that lead to an incorrect answer, and proposes prompts that bring the user closer to the correct answer. For example, in the prompt optimization system, a user inputs a specific prompt, its output, and the difference between the desired information in detail. For example, if a user receives an output such as "The price of this product is 1,000 yen" in response to a prompt such as "What is the price of this product?", the user inputs "I would also like to know the product's stock status" as the difference from the desired information. Next, the prompt optimization system uses a generation AI to verify the correlation between the input prompt and the output. The generation AI analyzes the relationship between the prompt and its output and identifies which parts are close to the correct answer and which parts deviate from the correct answer. For example, if a user receives an output such as "The price of this product is 1,000 yen" in response to a prompt such as "What is the price of this product?", the generation AI verifies whether this output is the correct answer for the prompt. Furthermore, the prompt optimization system uses a generation AI to analyze factors that lead to an incorrect answer. Based on the correlation between the prompt and the output, the generation AI identifies which part of the answer is incorrect. For example, if the prompt "Tell me the price of this product" returns the output "The product price is 1,000 yen," but the user enters "I would also like to know the product's stock status" as the information they are looking for, the generation AI analyzes this difference and identifies the factors that caused the answer to be incorrect. Finally, the prompt optimization system uses the generation AI to suggest a prompt that will bring the user closer to the information they are looking for, based on the factors that caused the answer to be incorrect. For example, if the prompt "Tell me the price of this product" returns the output "The product price is 1,000 yen," but the user enters "I would also like to know the product's stock status" as the information they are looking for, the generation AI will suggest a prompt such as "Tell me the product's price and stock status."This allows the prompt optimization system to help users review prompts efficiently and obtain accurate output. This allows the prompt optimization system to help users review prompts efficiently and obtain accurate output. For example, if a user wants to know the price and stock status of a product, they can obtain accurate information by using the prompts suggested by the generation AI. The prompt optimization system also verifies the correlation between the prompt and the output and analyzes the factors that cause the answer to be incorrect, helping users review prompts efficiently.

[0087] A prompt optimization system according to an embodiment includes a reception unit, a verification unit, an analysis unit, and a proposal unit. The reception unit receives a user input of a difference between a base prompt and its result, which represents the desired information. The information input by the user includes, but is not limited to, a specific prompt, its output, and the difference between the desired information. For example, if a user receives an output such as "The price of the product is 1,000 yen" in response to a prompt such as "What is the price of the product?", the reception unit allows the user to input "I would also like to know the product's stock status" as the difference between the desired information. The verification unit uses a generation AI to verify the correlation between the prompt and the output based on the information input by the reception unit. For example, the verification unit analyzes the relationship between the prompt and the output and identifies which parts are close to the correct answer and which parts deviate from the correct answer. For example, if the generation AI receives the output "The price of the product is 1,000 yen" in response to the prompt "What is the price of the product?", the generation AI verifies whether this output is the correct answer to the prompt. The analysis unit uses the generation AI to analyze the reasons for the incorrect answer based on the results obtained by the verification unit. The analysis unit, for example, identifies which part of the generation AI's output is incorrect based on the correlation between the prompt and the output. For example, if the prompt "Tell me the price of this product" returns the output "The price of this product is 1,000 yen," but the user also enters "I would also like to know the product's stock status" as the desired information, the generation AI analyzes this difference and identifies the reason for the incorrect answer. The suggestion unit uses the generation AI to propose a prompt that brings the generation AI closer to the correct answer based on the results obtained by the analysis unit. For example, the suggestion unit proposes a prompt that brings the generation AI closer to the desired information based on the reason for the incorrect answer. For example, if the prompt "Tell me the price of this product" returns the output "The price of this product is 1,000 yen," but the user also enters "I would also like to know the product's stock status" as the desired information, the generation AI proposes a prompt such as "Tell me the price and stock status of this product." This allows the prompt optimization system according to the embodiment to allow the user to efficiently review the prompt and obtain accurate output.

[0088] The verification unit includes a similarity calculation unit that calculates the similarity between the prompt and the output using natural language processing technology. The similarity calculation unit calculates the similarity between the prompt and the output using natural language processing technology. The similarity calculation unit calculates the similarity using methods such as cosine similarity or Jaccard coefficient. For example, the similarity calculation unit calculates the similarity between the prompt and the output using cosine similarity. The similarity calculation unit can also calculate the similarity using the Jaccard coefficient. The similarity calculation unit can also calculate the similarity using TF-IDF. For example, the similarity calculation unit calculates the angle between two vectors in a vector space model as cosine similarity, and determines that the smaller the angle, the higher the similarity. The Jaccard coefficient is the value obtained by dividing the size of the intersection of two sets by the total size, and indicates the degree of overlap between the sets. TF-IDF is an index that evaluates the importance of words in a document and is used to calculate the similarity between the prompt and the output. By calculating the similarity between the prompt and the output in this way, the accuracy of the correlation is improved. Some or all of the above-described processing in the similarity calculation unit may be performed using, for example, AI, or may be performed without using AI. For example, the similarity calculation unit may calculate the similarity using an AI model that takes the prompt and the output as input and outputs the similarity.

[0089] The analysis unit includes an error analysis unit that uses error analysis or a feedback loop. The error analysis unit analyzes factors that result in an incorrect answer using the error analysis or the feedback loop. The error analysis unit performs error analysis, for example, using an error log analysis or an error classification method. For example, the error analysis unit analyzes the error log to identify the cause of the error. The error analysis unit can also identify the type of error using the error classification method. The error analysis unit can also identify the cause of the error using a feedback loop. For example, the error analysis unit uses a feedback loop to identify the cause of the error and correct the cause. In this way, the use of error analysis or a feedback loop improves the accuracy of identifying factors that result in an incorrect answer. Some or all of the above-described processing in the error analysis unit may be performed using, for example, AI, or may be performed without using AI. For example, the error analysis unit can perform error analysis using an AI model that receives an error log as input and outputs the cause of the error.

[0090] The suggestion unit includes an optimization unit that generates a prompt based on user feedback and past data. The optimization unit generates an optimal prompt based on the user feedback and past data. The optimization unit generates the optimal prompt using, for example, a method for evaluating user feedback and a method for using past data. For example, the optimization unit evaluates user feedback and generates an optimal prompt based on the evaluation result. The optimization unit can also generate the optimal prompt using past data. The optimization unit can also generate the optimal prompt by combining user feedback and past data. For example, the optimization unit develops an algorithm that combines user feedback and past data to generate the optimal prompt. This generates the optimal prompt based on the user feedback and past data, thereby improving the accuracy of suggestions. Some or all of the above-described processing in the optimization unit may be performed using, for example, AI, or may be performed without using AI. For example, the optimization unit can generate the optimal prompt using an AI model that receives user feedback and past data as input and outputs the optimal prompt.

[0091] The reception unit can estimate the user's emotion and change the display method of the input interface based on the estimated user emotion. For example, if the user is feeling stressed, the reception unit can provide a simple interface and minimize input steps. Furthermore, if the user is relaxed, the reception unit can provide detailed input options and suggest customizable input methods. Furthermore, if the user is in a hurry, the reception unit can prioritize voice input and allow the user to quickly input the difference between the base prompt and the result, or the desired information. This reduces user stress and enables efficient input by adjusting the display method of the input interface according to the user's emotion. Emotion estimation is realized using an emotion estimation function, such as an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the reception unit can be performed using, for example, an AI. For example, the reception unit can input the user's emotion data into the generation AI and have the generation AI perform emotion estimation.

[0092] The reception unit can analyze the user's past input history and suggest an input method. For example, the reception unit can automatically display prompts and results frequently entered by the user in the past as candidates. The reception unit can also prioritize suggesting input methods (such as voice and text) that the user has used in the past. The reception unit can also predict and suggest prompts and results to be used in a specific time period based on the user's past input history. For example, the reception unit can automatically display prompts frequently used by the user in the past as candidates, making it easier for the user to select. The reception unit can also prioritize suggesting input methods that the user has used in the past, allowing the user to input efficiently. Furthermore, the reception unit can analyze the user's past input history and predict and suggest prompts and results to be used in a specific time period, allowing the user to input efficiently. In this way, by analyzing the user's past input history, the optimal input method can be suggested, improving input efficiency. Some or all of the above-mentioned processing in the reception unit may be performed using, or without, AI. For example, the reception unit can input the user's past input history data into a generation AI and have the generation AI suggest an optimal input method.

[0093] The reception unit can change the input fields based on the user's current task or area of ​​interest when the user inputs data. For example, the reception unit can prioritize displaying prompts and result input fields related to the task the user is currently performing. The reception unit can also customize the relevant prompts and result input fields based on the user's area of ​​interest. The reception unit can also automatically display input fields related to a specific project the user is working on. For example, the reception unit can prioritize displaying prompts and result input fields related to the task the user is currently performing, allowing the user to input data efficiently. The reception unit can also customize the relevant prompts and result input fields based on the user's area of ​​interest, allowing the user to input data efficiently. Furthermore, the reception unit can automatically display input fields related to a specific project the user is working on, allowing the user to input data efficiently. This improves input efficiency by customizing the input fields based on the user's current task or area of ​​interest. Some or all of the above-described processing in the reception unit can be performed using, or without, AI. For example, the reception unit can input data on the user's current task or area of ​​interest to a generation AI and have the generation AI customize the input fields.

[0094] The reception unit can select the optimal input means depending on the user's input method (voice, text, image, etc.) at the time of input. For example, if the user prefers voice input, the reception unit can prioritize voice input. Furthermore, if the user prefers text input, the reception unit can also prioritize text input. Furthermore, if the user prefers image input, the reception unit can also prioritize image input. For example, if the user prefers voice input, the reception unit can prioritize voice input, allowing the user to input efficiently. Furthermore, if the user prefers text input, the reception unit can prioritize text input, allowing the user to input efficiently. Furthermore, if the user prefers image input, the reception unit can prioritize image input, allowing the user to input efficiently. This improves input efficiency by selecting the optimal input means depending on the user's input method. Some or all of the above-described processing in the reception unit may be performed using, for example, AI, or may be performed without AI. For example, the reception unit can input data on the user's input method to a generation AI and cause the generation AI to select the optimal input means.

[0095] The reception unit can estimate the user's emotions and change the design of the input interface based on the estimated user emotions. For example, if the user is nervous, the reception unit can provide an interface with subdued colors to reduce visual stress. Alternatively, if the user is having fun, the reception unit can provide an interface with bright colors to make input work more enjoyable. Alternatively, if the user is tired, the reception unit can provide a simple, highly visible interface to make input work easier. This allows the input interface design to be adjusted according to the user's emotions, reducing user stress and enabling efficient input. Emotion estimation is achieved using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the reception unit can be performed using AI, for example, or without AI. For example, the reception unit can input the user's emotion data into the generation AI and have the generation AI perform emotion estimation.

[0096] The reception unit can display highly relevant input fields based on the user's geographical location information when the user enters information. For example, if the user is in a specific region, the reception unit can prioritize displaying prompts and result input fields related to the region. Furthermore, if the user is traveling, the reception unit can prioritize displaying prompts and result input fields related to the travel destination. Furthermore, if the user is participating in a specific event, the reception unit can prioritize displaying prompts and result input fields related to the event. For example, if the user is in a specific region, the reception unit can prioritize displaying prompts and result input fields related to the region, allowing the user to enter information efficiently. Furthermore, if the user is traveling, the reception unit can prioritize displaying prompts and result input fields related to the travel destination, allowing the user to enter information efficiently. Furthermore, if the user is participating in a specific event, the reception unit can prioritize displaying prompts and result input fields related to the event, allowing the user to enter information efficiently. Thus, by preferentially displaying highly relevant input fields based on the user's geographical location information, input efficiency is improved. Some or all of the above-described processing in the reception unit may be performed, for example, using AI or without AI. For example, the reception unit can input the user's geographical location information data into the generation AI and cause the generation AI to display highly relevant input fields.

[0097] The reception unit can analyze the user's social media activity at the time of input and suggest related input content. The reception unit can, for example, suggest related prompts and resultant input content based on content frequently posted by the user on social media. The reception unit can also suggest related prompts and resultant input content by referring to the activities of the user's friends on social media. The reception unit can also suggest related prompts and resultant input content based on the user's check-in information on social media. For example, the reception unit can suggest related prompts and resultant input content based on content frequently posted by the user on social media, thereby enabling the user to input efficiently. The reception unit can also suggest related prompts and resultant input content by referring to the activities of the user's friends on social media, thereby enabling the user to input efficiently. The reception unit can also suggest related prompts and resultant input content based on the user's check-in information on social media, thereby enabling the user to input efficiently. In this way, by analyzing the user's social media activity, related input content is suggested and input efficiency is improved. Some or all of the above-mentioned processing by the reception unit may be performed, for example, using AI or without AI. For example, the reception unit can input the user's social media activity data into the generation AI and have the generation AI execute suggestions for related input content.

[0098] The reception unit can change the input method by reflecting the user's past feedback when inputting data. For example, the reception unit can suggest an optimal input method based on feedback provided by the user in the past. The reception unit can also preferentially provide a specific input method based on the user's past feedback. The reception unit can also analyze the user's past feedback and customize the input interface. For example, the reception unit can suggest an optimal input method based on the user's past feedback, allowing the user to input data efficiently. The reception unit can also preferentially provide a specific input method based on the user's past feedback, allowing the user to input data efficiently. The reception unit can also analyze the user's past feedback and customize the input interface, allowing the user to input data efficiently. In this way, the input method is customized by reflecting the user's past feedback, improving input efficiency. Some or all of the above-described processing by the reception unit may be performed using AI, for example, or may be performed without AI. For example, the reception unit can input the user's past feedback data into a generation AI and cause the generation AI to customize the input method.

[0099] The verification unit can estimate the user's emotions and change the display method of the verification results based on the estimated user emotions. For example, if the user is nervous, the verification unit can provide a simple, highly visible display method. Furthermore, if the user is relaxed, the verification unit can provide a display method that includes detailed information. Furthermore, if the user is in a hurry, the verification unit can provide a display method that focuses on the main points. By adjusting the display method of the verification results according to the user's emotions, the user's stress can be reduced and verification can be performed efficiently. The emotion estimation is realized using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be, for example, a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples. Some or all of the above-described processing in the verification unit can be performed using, for example, an AI, or without an AI. For example, the verification unit can input the user's emotion data into the generation AI and have the generation AI perform emotion estimation.

[0100] During verification, the verification unit can analyze the correlation between the prompt and the output and extract patterns. For example, the verification unit can analyze the correlation between the prompt and the output to identify frequently occurring patterns. The verification unit can also analyze the correlation between the prompt and the output to identify patterns that are close to the correct answer. The verification unit can also analyze the correlation between the prompt and the output to identify patterns that deviate from the correct answer. For example, the verification unit can analyze the correlation between the prompt and the output to identify frequently occurring patterns and help the user efficiently review the prompt. The verification unit can also analyze the correlation between the prompt and the output to identify patterns that are close to the correct answer and help the user obtain accurate output. The verification unit can also analyze the correlation between the prompt and the output to identify patterns that deviate from the correct answer and help the user improve the prompt. In this way, by analyzing the correlation between the prompt and the output in detail, specific patterns can be extracted and the accuracy of the verification can be improved. Some or all of the above-mentioned processing in the verification unit may be performed, for example, using AI or without using AI. For example, the verification unit can input prompt and output data to the generation AI and cause the generation AI to extract patterns.

[0101] During verification, the verification unit can apply a verification algorithm depending on the category of the prompt. For example, the verification unit selects an optimal verification algorithm depending on the category of the prompt. The verification unit can also apply different verification algorithms depending on the category of the prompt to obtain optimal results. The verification unit can also apply a combination of multiple verification algorithms depending on the category of the prompt. For example, the verification unit selects an optimal verification algorithm depending on the category of the prompt to help the user efficiently review the prompt. The verification unit can also apply different verification algorithms depending on the category of the prompt to help the user obtain accurate output. The verification unit can also apply a combination of multiple verification algorithms depending on the category of the prompt to help the user improve the prompt. In this way, applying different verification algorithms depending on the category of the prompt improves the accuracy of the verification. Some or all of the above-mentioned processing in the verification unit may be performed using, for example, AI, or may be performed without using AI. For example, the verification unit can input prompt category data to a generation AI and cause the generation AI to apply a verification algorithm.

[0102] During verification, the verification unit can improve the verification by referring to the user's past verification results. For example, the verification unit adjusts the verification algorithm based on the user's past verification results. The verification unit can also improve the accuracy of the verification by referring to the user's past verification results. The verification unit can also improve the accuracy of the verification by analyzing the user's past verification results. For example, the verification unit adjusts the verification algorithm based on the user's past verification results to help the user efficiently review prompts. The verification unit also improves the accuracy of the verification by referring to the user's past verification results, helping the user obtain accurate output. Furthermore, the verification unit analyzes the user's past verification results to improve the accuracy of the verification and help the user improve the prompts. In this way, the accuracy of the verification is improved by referring to the user's past verification results. Some or all of the above-described processing in the verification unit may be performed using, for example, AI, or may be performed without using AI. For example, the verification unit can input the user's past verification result data into the generation AI and cause the generation AI to improve the verification.

[0103] The verification unit can estimate the user's emotions and rank the verification results based on the estimated user emotions. For example, if the user is nervous, the verification unit can prioritize displaying important verification results. Furthermore, if the user is relaxed, the verification unit can prioritize displaying detailed verification results. Furthermore, if the user is in a hurry, the verification unit can prioritize displaying verification results that highlight the key points. By prioritizing the verification results according to the user's emotions, important information can be displayed preferentially. Emotion estimation is achieved using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the verification unit can be performed using, for example, an AI, or without an AI. For example, the verification unit can input the user's emotion data into the generation AI and cause the generation AI to estimate the emotion.

[0104] During verification, the verification unit can determine the verification priority based on the time at which the prompt was submitted. For example, the verification unit prioritizes verification of prompts that were submitted more recently. The verification unit can also prioritize verification of prompts that were submitted earlier. The verification unit can also dynamically adjust the verification priority based on the time at which the prompt was submitted. For example, the verification unit prioritizes verification of prompts that were submitted more recently, helping the user to review the prompts based on the latest information. The verification unit prioritizes verification of prompts that were submitted earlier, helping the user to review the prompts based on past information. Furthermore, the verification unit dynamically adjusts the verification priority based on the time at which the prompt was submitted, helping the user to efficiently review the prompts. In this way, by determining the verification priority based on the time at which the prompt was submitted, the latest information can be prioritized for verification. Some or all of the above-described processing in the verification unit may be performed using, for example, AI, or may be performed without using AI. For example, the verification unit can input prompt submission time data to the generation AI and cause the generation AI to determine the verification priority.

[0105] The verification unit can change the order of verification based on the relevance of the prompts during verification. For example, the verification unit prioritizes verification of prompts with high relevance. The verification unit can also postpone verification of prompts with low relevance. The verification unit can also dynamically adjust the order of verification based on the relevance of the prompts. For example, the verification unit prioritizes verification of prompts with high relevance, helping the user to efficiently review the prompts. The verification unit also postpones prompts with low relevance, helping the user to efficiently review the prompts. Furthermore, the verification unit dynamically adjusts the order of verification based on the relevance of the prompts, helping the user to efficiently review the prompts. In this way, important information can be prioritized for verification by adjusting the order of verification based on the relevance of the prompts. Some or all of the above-described processing in the verification unit may be performed using, for example, AI, or may be performed without using AI. For example, the verification unit can input prompt relevance data to the generation AI and cause the generation AI to adjust the order of verification.

[0106] The verification unit can change the level of verification detail during verification depending on the user's level of expertise. For example, if the user's level of expertise is high, the verification unit provides a detailed verification result. Furthermore, if the user's level of expertise is low, the verification unit can provide a concise verification result. Furthermore, the verification unit can dynamically adjust the level of verification detail depending on the user's level of expertise. For example, if the user's level of expertise is high, the verification unit provides a detailed verification result to help the user efficiently review the prompt. Furthermore, if the user's level of expertise is low, the verification unit provides a concise verification result to help the user efficiently review the prompt. Furthermore, the verification unit dynamically adjusts the level of verification detail depending on the user's level of expertise to help the user efficiently review the prompt. Thus, by adjusting the level of verification detail depending on the user's level of expertise, information appropriate for the user can be provided. Some or all of the above-described processing in the verification unit may be performed using, or without, AI. For example, the verification unit may input the user's level of expertise data to the generation AI and cause the generation AI to adjust the level of verification detail.

[0107] The analysis unit can estimate the user's emotions and change the display method of the analysis results based on the estimated user emotions. For example, if the user is nervous, the analysis unit can provide a simple, highly visible display method. Furthermore, if the user is relaxed, the analysis unit can provide a display method that includes detailed information. Furthermore, if the user is in a hurry, the analysis unit can provide a display method that focuses on the main points. By adjusting the display method of the analysis results according to the user's emotions, the user's stress can be reduced and analysis can be performed efficiently. The emotion estimation is realized using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be, for example, a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples. Some or all of the above-described processing in the analysis unit can be performed using, for example, an AI. For example, the analysis unit can input the user's emotion data into the generation AI and have the generation AI perform emotion estimation.

[0108] During the analysis, the analysis unit can analyze the frequency and patterns of errors. For example, the analysis unit can analyze the frequency of errors to identify errors that occur frequently. The analysis unit can also analyze error patterns to identify errors that occur under specific conditions. The analysis unit can also combine and analyze the frequency and patterns of errors to identify the root cause of the error. For example, the analysis unit can analyze the frequency of errors to identify frequently occurring errors and help the user efficiently correct the errors. The analysis unit can also analyze error patterns to identify errors that occur under specific conditions and help the user efficiently correct the errors. Furthermore, the analysis unit can combine and analyze the frequency and patterns of errors to identify the root cause of the error and help the user efficiently correct the errors. This makes it easier to identify the root cause of the error by analyzing the frequency and patterns of errors in detail. Some or all of the above-mentioned processing in the analysis unit may be performed using, for example, AI, or may be performed without AI. For example, the analysis unit can input error frequency and pattern data to the generation AI and have the generation AI perform error analysis.

[0109] During analysis, the analysis unit can apply an analysis method according to the category of the prompt. For example, the analysis unit selects an optimal analysis method according to the category of the prompt. The analysis unit can also apply different analysis methods according to the category of the prompt to obtain optimal results. The analysis unit can also apply a combination of multiple analysis methods according to the category of the prompt. For example, the analysis unit selects an optimal analysis method according to the category of the prompt to help the user efficiently correct errors. The analysis unit can also apply different analysis methods according to the category of the prompt to help the user efficiently correct errors. Furthermore, the analysis unit can apply a combination of multiple analysis methods according to the category of the prompt to help the user efficiently correct errors. In this way, applying different analysis methods according to the category of the prompt improves the accuracy of the analysis. Some or all of the above-mentioned processing in the analysis unit may be performed using, for example, AI, or may be performed without using AI. For example, the analysis unit can input prompt category data to a generation AI and cause the generation AI to apply an analysis method.

[0110] During analysis, the analysis unit can improve the analysis by referring to the user's past analysis results. For example, the analysis unit adjusts the analysis method based on the user's past analysis results. The analysis unit can also improve the accuracy of the analysis by referring to the user's past analysis results. The analysis unit can also analyze the user's past analysis results to improve the accuracy of the analysis. For example, the analysis unit adjusts the analysis method based on the user's past analysis results to help the user efficiently correct errors. The analysis unit can also improve the accuracy of the analysis by referring to the user's past analysis results. The analysis unit can also improve the accuracy of the analysis by referring to the user's past analysis results. Furthermore, the analysis unit can analyze the user's past analysis results to improve the accuracy of the analysis and help the user efficiently correct errors. In this way, the accuracy of the analysis is improved by referring to the user's past analysis results. Some or all of the above-described processing in the analysis unit may be performed using, for example, AI, or may be performed without using AI. For example, the analysis unit can input the user's past analysis result data into the generation AI and cause the generation AI to improve the analysis.

[0111] The analysis unit can estimate the user's emotions and rank the analysis results based on the estimated user emotions. For example, if the user is nervous, the analysis unit can prioritize displaying important analysis results. Furthermore, if the user is relaxed, the analysis unit can prioritize displaying detailed analysis results. Furthermore, if the user is in a hurry, the analysis unit can prioritize displaying key analysis results. Thus, by prioritizing the analysis results according to the user's emotions, important information can be displayed preferentially. The emotion estimation is realized using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the analysis unit can be performed using, for example, an AI, or without an AI. For example, the analysis unit can input the user's emotion data into the generation AI and have the generation AI perform emotion estimation.

[0112] During analysis, the analysis unit can determine the order of analysis based on the time of error occurrence. For example, the analysis unit prioritizes analysis of recent errors. The analysis unit can also prioritize analysis of older errors. The analysis unit can also dynamically adjust the analysis priority based on the time of error occurrence. For example, the analysis unit prioritizes analysis of recent errors to help the user correct errors based on the latest information. The analysis unit prioritizes analysis of older errors to help the user correct errors based on past information. The analysis unit can also dynamically adjust the analysis priority based on the time of error occurrence to help the user efficiently correct errors. In this way, by determining the analysis priority based on the time of error occurrence, the latest information can be prioritized for analysis. Some or all of the above-described processing in the analysis unit may be performed using, for example, AI, or may be performed without AI. For example, the analysis unit can input data on the time of error occurrence to the generation AI and have the generation AI determine the analysis priority.

[0113] During analysis, the analysis unit can change the order of analysis based on the relevance of errors. For example, the analysis unit prioritizes analysis of errors with high relevance. The analysis unit can also postpone analysis of errors with low relevance. The analysis unit can also dynamically adjust the order of analysis based on the relevance of errors. For example, the analysis unit prioritizes analysis of errors with high relevance to help the user efficiently correct errors. The analysis unit also postpones analysis of errors with low relevance to help the user efficiently correct errors. Furthermore, the analysis unit dynamically adjusts the order of analysis based on the relevance of errors to help the user efficiently correct errors. In this way, adjusting the order of analysis based on the relevance of errors allows important information to be analyzed preferentially. Some or all of the above-described processing in the analysis unit may be performed using, for example, AI, or may be performed without using AI. For example, the analysis unit can input error relevance data to the generation AI and cause the generation AI to adjust the order of analysis.

[0114] During analysis, the analysis unit can change the level of detail of the analysis according to the user's level of expertise. For example, if the user's level of expertise is high, the analysis unit can provide detailed analysis results. Furthermore, if the user's level of expertise is low, the analysis unit can provide concise analysis results. Furthermore, the analysis unit can dynamically adjust the level of detail of the analysis according to the user's level of expertise. For example, if the user's level of expertise is high, the analysis unit can provide detailed analysis results to help the user efficiently correct errors. Furthermore, if the user's level of expertise is low, the analysis unit can provide concise analysis results to help the user efficiently correct errors. Furthermore, the analysis unit can dynamically adjust the level of detail of the analysis according to the user's level of expertise to help the user efficiently correct errors. In this way, by adjusting the level of detail of the analysis according to the user's level of expertise, it is possible to provide information appropriate for the user. Some or all of the above-described processing in the analysis unit may be performed using, or without, AI. For example, the analysis unit can input the user's level of expertise data into the generation AI and cause the generation AI to adjust the level of detail of the analysis.

[0115] The suggestion unit can estimate the user's emotions and change the way the suggestions are expressed based on the estimated user emotions. For example, if the user is nervous, the suggestion unit can provide a simple, highly visible expression. If the user is relaxed, the suggestion unit can also provide an expression that includes detailed information. If the user is in a hurry, the suggestion unit can also provide an expression that focuses on the main points. This allows the suggestion to be adjusted according to the user's emotions, reducing the user's stress and enabling efficient suggestions. The emotion estimation is achieved using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples. Some or all of the above-described processing in the suggestion unit can be performed using, for example, an AI, or without an AI. For example, the suggestion unit can input the user's emotion data into the generation AI and cause the generation AI to estimate the emotion.

[0116] The suggestion unit can change the level of detail of the suggestion based on the importance of the prompt when making a suggestion. For example, the suggestion unit provides a detailed suggestion when the importance of the prompt is high. The suggestion unit can also provide a concise suggestion when the importance of the prompt is low. The suggestion unit can also dynamically adjust the level of detail of the suggestion based on the importance of the prompt. For example, the suggestion unit provides a detailed suggestion when the importance of the prompt is high, helping the user to efficiently review the prompt. The suggestion unit provides a concise suggestion when the importance of the prompt is low, helping the user to efficiently review the prompt. Furthermore, the suggestion unit dynamically adjusts the level of detail of the suggestion based on the importance of the prompt, helping the user to efficiently review the prompt. In this way, by adjusting the level of detail of the suggestion based on the importance of the prompt, important information can be provided preferentially. Some or all of the above-described processing in the suggestion unit may be performed using AI, for example, or may be performed without using AI. For example, the suggestion unit can input prompt importance data to a generation AI and cause the generation AI to adjust the level of detail of the suggestion.

[0117] The suggestion unit can apply a suggestion algorithm depending on the category of the prompt when making a suggestion. For example, the suggestion unit selects an optimal suggestion algorithm depending on the category of the prompt. The suggestion unit can also apply different suggestion algorithms depending on the category of the prompt to obtain optimal results. The suggestion unit can also apply a combination of multiple suggestion algorithms depending on the category of the prompt. For example, the suggestion unit selects an optimal suggestion algorithm depending on the category of the prompt to help the user efficiently review the prompt. The suggestion unit can also apply different suggestion algorithms depending on the category of the prompt to help the user efficiently review the prompt. Furthermore, the suggestion unit can apply a combination of multiple suggestion algorithms depending on the category of the prompt to help the user efficiently review the prompt. In this way, applying different suggestion algorithms depending on the category of the prompt improves the accuracy of the suggestion. Some or all of the above-described processing in the suggestion unit may be performed using AI, for example, or may be performed without using AI. For example, the suggestion unit can input prompt category data to a generation AI and cause the generation AI to apply a suggestion algorithm.

[0118] When making a suggestion, the suggestion unit can improve the suggestion by referring to the user's past suggestion results. For example, the suggestion unit adjusts the suggestion algorithm based on the user's past suggestion results. The suggestion unit can also improve the accuracy of the suggestion by referring to the user's past suggestion results. The suggestion unit can also improve the accuracy of the suggestion by analyzing the user's past suggestion results. For example, the suggestion unit adjusts the suggestion algorithm based on the user's past suggestion results to help the user efficiently review the prompt. The suggestion unit can also improve the accuracy of the suggestion by referring to the user's past suggestion results, to help the user efficiently review the prompt. Furthermore, the suggestion unit analyzes the user's past suggestion results to improve the accuracy of the suggestion, to help the user efficiently review the prompt. In this way, the accuracy of the suggestion is improved by referring to the user's past suggestion results. Some or all of the above-described processing in the suggestion unit may be performed using AI, for example, or may be performed without using AI. For example, the suggestion unit can input the user's past suggestion result data to the generation AI and cause the generation AI to improve the suggestion.

[0119] The suggestion unit can estimate the user's emotions and change the length of the suggestions based on the estimated user emotions. For example, if the user is nervous, the suggestion unit can provide short, to-the-point suggestions. If the user is relaxed, the suggestion unit can provide longer suggestions with detailed explanations. If the user is in a hurry, the suggestion unit can provide quick, concise suggestions. This allows the length of suggestions to be adjusted according to the user's emotions, reducing the user's stress and enabling efficient suggestions. The emotion estimation is achieved using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to these examples. Some or all of the above-described processing in the suggestion unit can be performed using, for example, an AI. For example, the suggestion unit can input the user's emotion data into the generation AI and cause the generation AI to estimate the emotion.

[0120] When making a suggestion, the suggestion unit can determine the order of the suggestions based on the time at which the prompts were submitted. For example, the suggestion unit can prioritize suggesting prompts that were submitted more recently. The suggestion unit can also prioritize suggesting prompts that were submitted earlier. The suggestion unit can also dynamically adjust the priority of the suggestions based on the time at which the prompts were submitted. For example, the suggestion unit can prioritize suggesting prompts that were submitted more recently, helping the user to review the prompts based on the latest information. The suggestion unit can prioritize suggesting prompts that were submitted earlier, helping the user to review the prompts based on past information. The suggestion unit can also dynamically adjust the priority of the suggestions based on the time at which the prompts were submitted, helping the user to efficiently review the prompts. In this way, by determining the priority of the suggestions based on the time at which the prompts were submitted, the latest information can be prioritized for suggestion. Some or all of the above-described processing by the suggestion unit may be performed using, or without, AI. For example, the suggestion unit can input prompt submission time data to a generation AI and cause the generation AI to determine the priority of the suggestions.

[0121] The suggestion unit can change the order of suggestions based on the relevance of the prompts when making suggestions. For example, the suggestion unit prioritizes suggesting prompts with high relevance. The suggestion unit can also postpone prompts with low relevance. The suggestion unit can also dynamically adjust the order of suggestions based on the relevance of the prompts. For example, the suggestion unit prioritizes suggesting prompts with high relevance, helping the user to efficiently review the prompts. The suggestion unit also postpones prompts with low relevance, helping the user to efficiently review the prompts. Furthermore, the suggestion unit dynamically adjusts the order of suggestions based on the relevance of the prompts, helping the user to efficiently review the prompts. In this way, important information can be prioritized by adjusting the order of suggestions based on the relevance of the prompts. Some or all of the above-described processing in the suggestion unit may be performed using AI, for example, or may be performed without using AI. For example, the suggestion unit can input prompt relevance data to a generation AI and cause the generation AI to adjust the order of suggestions.

[0122] The suggestion unit may change the use of technical terminology in the suggestion depending on the user's level of expertise. For example, if the user's level of expertise is high, the suggestion unit may provide a suggestion that uses a lot of technical terminology. Furthermore, if the user's level of expertise is low, the suggestion unit may provide a concise and easy-to-understand suggestion. Furthermore, the suggestion unit may dynamically adjust the use of technical terminology in the suggestion depending on the user's level of expertise. For example, if the user's level of expertise is high, the suggestion unit may provide a suggestion that uses a lot of technical terminology, thereby helping the user to efficiently review the prompt. Furthermore, if the user's level of expertise is low, the suggestion unit may provide a concise and easy-to-understand suggestion, thereby helping the user to efficiently review the prompt. Furthermore, the suggestion unit may dynamically adjust the use of technical terminology in the suggestion depending on the user's level of expertise, thereby helping the user to efficiently review the prompt. By adjusting the use of technical terminology in the suggestion depending on the user's level of expertise, it is possible to provide information suitable for the user. Some or all of the above-described processing in the suggestion unit may be performed using, for example, AI, or may be performed without using AI. For example, the suggestion unit can input the user's expertise level data into the generation AI and have the generation AI execute the suggested terminology usage.

[0123] The similarity calculation unit can estimate the user's emotion and change the display method of the similarity calculation results based on the estimated user emotion. For example, if the user is nervous, the similarity calculation unit can provide a simple, highly visible display method. Furthermore, if the user is relaxed, the similarity calculation unit can provide a display method that includes detailed information. Furthermore, if the user is in a hurry, the similarity calculation unit can provide a display method that focuses on the main points. By adjusting the display method of the similarity calculation results according to the user's emotion, the user's stress can be reduced and efficient display is possible. Emotion estimation is realized using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples. Some or all of the above-described processing in the similarity calculation unit can be performed using, for example, an AI, or without an AI. For example, the similarity calculation unit can input the user's emotion data into the generation AI and have the generation AI perform emotion estimation.

[0124] The similarity calculation unit can analyze the context of the prompt and the output when calculating the similarity. For example, the similarity calculation unit analyzes the context of the prompt and the output to identify frequently occurring patterns. The similarity calculation unit can also analyze the context of the prompt and the output to identify patterns that are close to the correct answer. The similarity calculation unit can also analyze the context of the prompt and the output to identify patterns that deviate from the correct answer. For example, the similarity calculation unit analyzes the context of the prompt and the output to identify frequently occurring patterns and help the user efficiently review the prompt. The similarity calculation unit also analyzes the context of the prompt and the output to identify patterns that are close to the correct answer and help the user obtain an accurate output. Furthermore, the similarity calculation unit analyzes the context of the prompt and the output to identify patterns that deviate from the correct answer and help the user improve the prompt. In this way, by analyzing the context of the prompt and the output in detail, specific patterns can be extracted and the accuracy of the similarity calculation can be improved. Some or all of the above-mentioned processing in the similarity calculation unit may be performed, for example, using AI or without AI. For example, the similarity calculation unit can input prompt and output data into the generation AI and have the generation AI perform context analysis.

[0125] The similarity calculation unit can apply a calculation algorithm depending on the category of the prompt when calculating the similarity. The similarity calculation unit, for example, selects an optimal calculation algorithm depending on the category of the prompt. The similarity calculation unit can also apply different calculation algorithms depending on the category of the prompt to obtain optimal results. The similarity calculation unit can also apply a combination of multiple calculation algorithms depending on the category of the prompt. For example, the similarity calculation unit selects an optimal calculation algorithm depending on the category of the prompt to help the user efficiently review the prompt. The similarity calculation unit can also apply different calculation algorithms depending on the category of the prompt to help the user efficiently review the prompt. Furthermore, the similarity calculation unit can apply a combination of multiple calculation algorithms depending on the category of the prompt to help the user efficiently review the prompt. In this way, applying different calculation algorithms depending on the category of the prompt improves the accuracy of the similarity calculation. Some or all of the above-mentioned processing in the similarity calculation unit may be performed using, for example, AI, or may be performed without using AI. For example, the similarity calculation unit can input prompt category data to a generation AI and cause the generation AI to apply a calculation algorithm.

[0126] The similarity calculation unit can improve the calculation by referring to the user's past calculation results when calculating the similarity. For example, the similarity calculation unit adjusts the calculation algorithm based on the user's past calculation results. The similarity calculation unit can also improve the accuracy of the calculation by referring to the user's past calculation results. The similarity calculation unit can also improve the accuracy of the calculation by analyzing the user's past calculation results. For example, the similarity calculation unit adjusts the calculation algorithm based on the user's past calculation results to help the user efficiently review the prompt. The similarity calculation unit can also improve the accuracy of the calculation by referring to the user's past calculation results to help the user efficiently review the prompt. The similarity calculation unit can also analyze the user's past calculation results to improve the accuracy of the calculation to help the user efficiently review the prompt. In addition, the similarity calculation unit can analyze the user's past calculation results to improve the accuracy of the calculation to help the user efficiently review the prompt. In this way, the accuracy of the calculation is improved by referring to the user's past calculation results. Some or all of the above-mentioned processing in the similarity calculation unit may be performed using, for example, AI, or may be performed without using AI. For example, the similarity calculation unit can input the user's past calculation result data into the generation AI and cause the generation AI to improve the calculation.

[0127] The similarity calculation unit can estimate the user's emotion and determine the ranking of similarity calculations based on the estimated user's emotion. For example, when the user is nervous, the similarity calculation unit can prioritize displaying important similarity calculation results. Furthermore, when the user is relaxed, the similarity calculation unit can prioritize displaying detailed similarity calculation results. Furthermore, when the user is in a hurry, the similarity calculation unit can prioritize displaying similarity calculation results that highlight the main points. By determining the priority of similarity calculations according to the user's emotion, important information can be displayed preferentially. Emotion estimation is realized using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the similarity calculation unit can be performed using, for example, an AI. For example, the similarity calculation unit can input the user's emotion data into the generation AI and have the generation AI perform emotion estimation.

[0128] When calculating the similarity, the similarity calculation unit can determine the calculation order based on the time at which the prompt was submitted. For example, the similarity calculation unit prioritizes the most recent prompt in the calculation. The similarity calculation unit can also prioritize the oldest prompt in the calculation. The similarity calculation unit can also dynamically adjust the calculation priority based on the time at which the prompt was submitted. For example, the similarity calculation unit prioritizes the most recent prompt in the calculation, helping the user to review the prompt based on the latest information. The similarity calculation unit prioritizes the oldest prompt in the calculation, helping the user to review the prompt based on past information. The similarity calculation unit also dynamically adjusts the calculation priority based on the time at which the prompt was submitted, helping the user to efficiently review the prompt. In this way, by determining the calculation priority based on the time at which the prompt was submitted, the most recent information can be prioritized in the calculation. Some or all of the above-described processing in the similarity calculation unit may be performed, for example, using AI or without AI. For example, the similarity calculation unit can input prompt submission timing data to the generation AI and have the generation AI determine the priority of the calculations.

[0129] The similarity calculation unit can change the order of calculations based on the relevance of the prompts when calculating the similarity. For example, the similarity calculation unit prioritizes calculations of prompts with high relevance. The similarity calculation unit can also postpone calculations of prompts with low relevance. The similarity calculation unit can also dynamically adjust the order of calculations based on the relevance of the prompts. For example, the similarity calculation unit prioritizes calculations of prompts with high relevance, helping the user to efficiently review the prompts. The similarity calculation unit also postpones prompts with low relevance, helping the user to efficiently review the prompts. Furthermore, the similarity calculation unit dynamically adjusts the order of calculations based on the relevance of the prompts, helping the user to efficiently review the prompts. In this way, by adjusting the order of calculations based on the relevance of the prompts, important information can be calculated with priority. Some or all of the above-described processing in the similarity calculation unit may be performed using, for example, AI, or may be performed without using AI. For example, the similarity calculation unit can input prompt relevance data to the generation AI and cause the generation AI to adjust the order of calculations.

[0130] The similarity calculation unit can change the level of detail of the calculation according to the user's level of expertise when calculating the similarity. For example, if the user's level of expertise is high, the similarity calculation unit provides a detailed calculation result. Furthermore, if the user's level of expertise is low, the similarity calculation unit can provide a concise calculation result. Furthermore, the similarity calculation unit can dynamically adjust the level of detail of the calculation according to the user's level of expertise. For example, if the user's level of expertise is high, the similarity calculation unit provides a detailed calculation result to help the user efficiently review the prompt. Furthermore, if the user's level of expertise is low, the similarity calculation unit provides a concise calculation result to help the user efficiently review the prompt. Furthermore, the similarity calculation unit dynamically adjusts the level of detail of the calculation according to the user's level of expertise to help the user efficiently review the prompt. By adjusting the level of detail of the calculation according to the user's level of expertise, information suitable for the user can be provided. Some or all of the above-described processing in the similarity calculation unit may be performed, for example, using AI or without AI. For example, the similarity calculation unit can input the user's expertise level data into the generation AI and cause the generation AI to adjust the level of detail of the calculation.

[0131] The error analysis unit can estimate the user's emotions and change the display method of the error analysis results based on the estimated user emotions. For example, if the user is nervous, the error analysis unit can provide a simple, highly visible display method. Furthermore, if the user is relaxed, the error analysis unit can provide a display method that includes detailed information. Furthermore, if the user is in a hurry, the error analysis unit can provide a display method that focuses on the main points. By adjusting the display method of the error analysis results according to the user's emotions, the user's stress can be reduced and an efficient display can be achieved. The emotion estimation is realized using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the error analysis unit can be performed using, for example, an AI, or without an AI. For example, the error analysis unit can input the user's emotion data into the generation AI and have the generation AI perform emotion estimation.

[0132] The error analysis unit can analyze the frequency and pattern of errors during error analysis. For example, the error analysis unit analyzes the frequency of error occurrence and identifies errors that occur frequently. The error analysis unit can also analyze error patterns and identify errors that occur under specific conditions. The error analysis unit can also combine and analyze the frequency and pattern of errors to identify the root cause of the error. For example, the error analysis unit analyzes the frequency of error occurrence and identifies frequently occurring errors, helping the user to efficiently correct the errors. The error analysis unit can also analyze the pattern of errors and identify errors that occur under specific conditions, helping the user to efficiently correct the errors. Furthermore, the error analysis unit combines and analyzes the frequency and pattern of errors to identify the root cause of the error, helping the user to efficiently correct the errors. This makes it easier to identify the root cause of the error by analyzing the frequency and pattern of the error in detail. Some or all of the above-described processing in the error analysis unit may be performed using, or without, AI. For example, the error analysis unit can input error frequency and pattern data to the generation AI and have the generation AI perform error analysis.

[0133] When analyzing an error, the error analysis unit can apply an analysis method depending on the category of the prompt. For example, the error analysis unit selects an optimal analysis method depending on the category of the prompt. The error analysis unit can also apply different analysis methods depending on the category of the prompt to obtain optimal results. The error analysis unit can also apply a combination of multiple analysis methods depending on the category of the prompt. For example, the error analysis unit selects an optimal analysis method depending on the category of the prompt to help the user efficiently correct the error. The error analysis unit can also apply different analysis methods depending on the category of the prompt to help the user efficiently correct the error. Furthermore, the error analysis unit can apply a combination of multiple analysis methods depending on the category of the prompt to help the user efficiently correct the error. In this way, applying different analysis methods depending on the category of the prompt improves the accuracy of the analysis. Some or all of the above-mentioned processing in the error analysis unit may be performed using, for example, AI, or may be performed without using AI. For example, the error analysis unit can input prompt category data to the generation AI and cause the generation AI to apply the analysis method.

[0134] During error analysis, the error analysis unit can improve the analysis by referring to the user's past analysis results. For example, the error analysis unit adjusts the analysis method based on the user's past analysis results. The error analysis unit can also improve the accuracy of the analysis by referring to the user's past analysis results. The error analysis unit can also analyze the user's past analysis results to improve the accuracy of the analysis. For example, the error analysis unit adjusts the analysis method based on the user's past analysis results to help the user efficiently correct errors. The error analysis unit also improves the accuracy of the analysis by referring to the user's past analysis results, helping the user efficiently correct errors. Furthermore, the error analysis unit analyzes the user's past analysis results to improve the accuracy of the analysis, helping the user efficiently correct errors. In this way, the accuracy of the analysis is improved by referring to the user's past analysis results. Some or all of the above-described processing in the error analysis unit may be performed, for example, using AI, or may be performed without using AI. For example, the error analysis unit can input the user's past analysis result data into the generation AI and cause the generation AI to improve the analysis.

[0135] The error analysis unit can estimate the user's emotions and determine the order of error analysis based on the estimated user emotions. For example, if the user is nervous, the error analysis unit can prioritize displaying important error analysis results. Furthermore, if the user is relaxed, the error analysis unit can prioritize displaying detailed error analysis results. Furthermore, if the user is in a hurry, the error analysis unit can prioritize displaying error analysis results that highlight the main points. By determining the priority of error analysis according to the user's emotions, important information can be displayed preferentially. The emotion estimation is realized using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the error analysis unit can be performed using, for example, an AI, or without an AI. For example, the error analysis unit can input the user's emotion data into the generation AI and have the generation AI perform emotion estimation.

[0136] During error analysis, the error analysis unit can determine the analysis priority based on the time of error occurrence. For example, the error analysis unit prioritizes analysis of recent errors. The error analysis unit can also prioritize analysis of older errors. The error analysis unit can also dynamically adjust the analysis priority based on the time of error occurrence. For example, the error analysis unit prioritizes analysis of recent errors to help the user correct errors based on the latest information. The error analysis unit also prioritizes analysis of older errors to help the user correct errors based on past information. The error analysis unit can also dynamically adjust the analysis priority based on the time of error occurrence to help the user efficiently correct errors. By determining the analysis priority based on the time of error occurrence, the latest information can be prioritized for analysis. Some or all of the above-described processing in the error analysis unit may be performed using, or without, AI. For example, the error analysis unit can input error occurrence time data to the generation AI and have the generation AI determine the analysis priority.

[0137] During error analysis, the error analysis unit can change the analysis order based on the relevance of the errors. For example, the error analysis unit prioritizes analysis of errors with high relevance. The error analysis unit can also postpone analysis of errors with low relevance. The error analysis unit can also dynamically adjust the analysis order based on the relevance of the errors. For example, the error analysis unit prioritizes analysis of errors with high relevance, helping the user to efficiently correct errors. The error analysis unit also postpones analysis of errors with low relevance, helping the user to efficiently correct errors. Furthermore, the error analysis unit dynamically adjusts the analysis order based on the relevance of the errors, helping the user to efficiently correct errors. In this way, adjusting the analysis order based on the relevance of the errors allows important information to be analyzed preferentially. Some or all of the above-described processing in the error analysis unit may be performed using, for example, AI, or may be performed without using AI. For example, the error analysis unit can input error relevance data to the generation AI and cause the generation AI to adjust the analysis order.

[0138] During error analysis, the error analysis unit can change the level of detail of the analysis according to the user's level of expertise. For example, if the user's level of expertise is high, the error analysis unit can provide detailed analysis results. Furthermore, if the user's level of expertise is low, the error analysis unit can provide concise analysis results. Furthermore, the error analysis unit can dynamically adjust the level of detail of the analysis according to the user's level of expertise. For example, if the user's level of expertise is high, the error analysis unit can provide detailed analysis results to help the user efficiently correct errors. Furthermore, if the user's level of expertise is low, the error analysis unit can provide concise analysis results to help the user efficiently correct errors. Furthermore, the error analysis unit can dynamically adjust the level of detail of the analysis according to the user's level of expertise to help the user efficiently correct errors. In this way, by adjusting the level of detail of the analysis according to the user's level of expertise, it is possible to provide information appropriate for the user. Some or all of the above-described processing in the error analysis unit may be performed using, or without, AI. For example, the error analysis unit can input the user's level of expertise data into the generation AI and cause the generation AI to adjust the level of detail of the analysis.

[0139] The optimization unit can estimate the user's emotions and change the display method of the optimization results based on the estimated user emotions. For example, if the user is nervous, the optimization unit can provide a simple, highly visible display method. Furthermore, if the user is relaxed, the optimization unit can provide a display method that includes detailed information. Furthermore, if the user is in a hurry, the optimization unit can provide a display method that focuses on the main points. This allows the optimization result display method to be adjusted according to the user's emotions, thereby reducing the user's stress and enabling efficient display. Emotion estimation is achieved using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be a text generation AI (e.g., LLM) or a multimodal generation AI, but is not limited to such examples. Some or all of the above-described processing in the optimization unit can be performed using, for example, an AI, or without an AI. For example, the optimization unit can input user emotion data into the generation AI and cause the generation AI to estimate the emotion.

[0140] The optimization unit can change the level of detail of the optimization based on the importance of the prompt during optimization. For example, when the importance of the prompt is high, the optimization unit provides a detailed optimization result. Furthermore, when the importance of the prompt is low, the optimization unit can provide a concise optimization result. Furthermore, the optimization unit can dynamically adjust the level of detail of the optimization based on the importance of the prompt. For example, when the importance of the prompt is high, the optimization unit provides a detailed optimization result to help the user efficiently review the prompt. Furthermore, when the importance of the prompt is low, the optimization unit provides a concise optimization result to help the user efficiently review the prompt. Furthermore, the optimization unit dynamically adjusts the level of detail of the optimization based on the importance of the prompt to help the user efficiently review the prompt. Thus, by adjusting the level of detail of the optimization based on the importance of the prompt, important information can be provided preferentially. Some or all of the above-described processing in the optimization unit may be performed using, or without, AI. For example, the optimization unit can input prompt importance data to the generation AI and cause the generation AI to adjust the level of detail of the optimization.

[0141] During optimization, the optimization unit can apply an optimization algorithm depending on the category of the prompt. For example, the optimization unit selects an optimal optimization algorithm depending on the category of the prompt. The optimization unit can also apply different optimization algorithms depending on the category of the prompt to obtain optimal results. The optimization unit can also apply a combination of multiple optimization algorithms depending on the category of the prompt. For example, the optimization unit selects an optimal optimization algorithm depending on the category of the prompt to help the user efficiently review the prompt. The optimization unit can also apply different optimization algorithms depending on the category of the prompt to help the user efficiently review the prompt. Furthermore, the optimization unit can apply a combination of multiple optimization algorithms depending on the category of the prompt to help the user efficiently review the prompt. In this way, applying different optimization algorithms depending on the category of the prompt improves the accuracy of optimization. Some or all of the above-mentioned processing in the optimization unit may be performed using AI, for example, or may be performed without using AI. For example, the optimization unit can input prompt category data to the generation AI and cause the generation AI to apply the optimization algorithm.

[0142] During optimization, the optimization unit can improve optimization by referring to the user's past optimization results. For example, the optimization unit adjusts the optimization algorithm based on the user's past optimization results. The optimization unit can also improve the accuracy of optimization by referring to the user's past optimization results. The optimization unit can also analyze the user's past optimization results to improve the accuracy of optimization. For example, the optimization unit adjusts the optimization algorithm based on the user's past optimization results to help the user efficiently review prompts. The optimization unit can also improve the accuracy of optimization by referring to the user's past optimization results to help the user efficiently review prompts. Furthermore, the optimization unit analyzes the user's past optimization results to improve the accuracy of optimization to help the user efficiently review prompts. In this way, the accuracy of optimization is improved by referring to the user's past optimization results. Some or all of the above-described processing in the optimization unit may be performed using, for example, AI, or may be performed without using AI. For example, the optimization unit can input the user's past optimization result data into the generation AI and cause the generation AI to improve the optimization.

[0143] The optimization unit can estimate the user's emotions and determine the order of optimization based on the estimated user emotions. For example, if the user is nervous, the optimization unit can prioritize displaying important optimization results. Furthermore, if the user is relaxed, the optimization unit can prioritize displaying detailed optimization results. Furthermore, if the user is in a hurry, the optimization unit can prioritize displaying optimization results that focus on the main points. Thus, by determining the priority of optimization according to the user's emotions, important information can be displayed preferentially. The emotion estimation is realized using an emotion estimation function, for example, an emotion engine or a generation AI. The generation AI can be, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the optimization unit can be performed using, for example, an AI, or without an AI. For example, the optimization unit can input the user's emotion data into the generation AI and cause the generation AI to estimate the emotion.

[0144] During optimization, the optimization unit can determine the order of optimization based on the submission time of the prompt. For example, the optimization unit prioritizes optimizing a prompt that was submitted more recently. The optimization unit can also prioritize optimizing a prompt that was submitted earlier. The optimization unit can also dynamically adjust the optimization priority based on the submission time of the prompt. For example, the optimization unit prioritizes optimizing a prompt that was submitted more recently, helping the user to review the prompt based on the latest information. The optimization unit prioritizes optimizing a prompt that was submitted earlier, helping the user to review the prompt based on past information. Furthermore, the optimization unit dynamically adjusts the optimization priority based on the submission time of the prompt, helping the user to efficiently review the prompt. In this way, by determining the optimization priority based on the submission time of the prompt, the latest information can be prioritized for optimization. Some or all of the above-described processing in the optimization unit may be performed using, for example, AI, or may be performed without using AI. For example, the optimization unit can input prompt submission time data to the generation AI and cause the generation AI to determine the optimization priority.

[0145] The optimization unit can change the order of optimization based on the relevance of the prompts during optimization. For example, the optimization unit prioritizes optimizing prompts with high relevance. The optimization unit can also postpone prompts with low relevance. The optimization unit can also dynamically adjust the order of optimization based on the relevance of the prompts. For example, the optimization unit prioritizes optimizing prompts with high relevance, helping the user to efficiently review the prompts. The optimization unit also postpones prompts with low relevance, helping the user to efficiently review the prompts. Furthermore, the optimization unit dynamically adjusts the order of optimization based on the relevance of the prompts, helping the user to efficiently review the prompts. In this way, by adjusting the order of optimization based on the relevance of the prompts, important information can be prioritized for optimization. Some or all of the above-described processing in the optimization unit may be performed using AI, for example, or may be performed without using AI. For example, the optimization unit can input prompt relevance data to the generation AI and cause the generation AI to adjust the order of optimization.

[0146] During optimization, the optimization unit can change the level of detail of the optimization according to the user's level of expertise. For example, if the user's level of expertise is high, the optimization unit provides detailed optimization results. Furthermore, if the user's level of expertise is low, the optimization unit can provide concise optimization results. Furthermore, the optimization unit can dynamically adjust the level of detail of the optimization according to the user's level of expertise. For example, if the user's level of expertise is high, the optimization unit provides detailed optimization results to help the user efficiently review prompts. Furthermore, if the user's level of expertise is low, the optimization unit provides concise optimization results to help the user efficiently review prompts. Furthermore, the optimization unit dynamically adjusts the level of detail of the optimization according to the user's level of expertise to help the user efficiently review prompts. Thus, by adjusting the level of detail of the optimization according to the user's level of expertise, information suitable for the user can be provided. Some or all of the above-described processing in the optimization unit may be performed using, or without, AI. For example, the optimization unit may input the user's expertise level data to the generation AI and cause the generation AI to adjust the level of detail of the optimization. === Hard Collateral 1-1 === Each of the multiple elements, including the above-mentioned reception unit, verification unit, analysis unit, and suggestion unit, is realized, for example, by at least one of the smart device 14 and the data processing device 12. For example, the reception unit is realized by the control unit 46A of the smart device 14 and allows the user to input the difference between a base prompt and the resulting desired information. The verification unit is realized, for example, by the specific processing unit 290 of the data processing device 12 and verifies the correlation between the prompt and the output using a generation AI. The analysis unit is realized, for example, by the specific processing unit 290 of the data processing device 12 and analyzes factors that result in an incorrect answer. The suggestion unit is realized, for example, by the control unit 46A of the smart device 14 and suggests a prompt to bring the user closer to the desired information. === Hard Collateral 1-2 === Each of the multiple elements, including the above-mentioned reception unit, verification unit, analysis unit, and suggestion unit, is realized, for example, by at least one of the smart glasses 214 and the data processing device 12. For example, the reception unit is realized by the control unit 46A of the smart glasses 214 and allows the user to input the difference between the base prompt and the resulting desired information. The verification unit is realized, for example, by the specific processing unit 290 of the data processing device 12 and verifies the correlation between the prompt and the output using a generation AI. The analysis unit is realized, for example, by the specific processing unit 290 of the data processing device 12 and analyzes factors that result in an incorrect answer. The suggestion unit is realized, for example, by the control unit 46A of the smart glasses 214 and suggests a prompt to bring the user closer to the desired information. === Hard Collateral 1-3 === Each of the multiple elements, including the above-mentioned reception unit, verification unit, analysis unit, and suggestion unit, is realized, for example, by at least one of the headset-type terminal 314 and the data processing device 12. For example, the reception unit is realized by the control unit 46A of the headset-type terminal 314 and allows the user to input the difference between a base prompt and the resulting desired information. The verification unit is realized, for example, by the specific processing unit 290 of the data processing device 12 and verifies the correlation between the prompt and the output using a generation AI. The analysis unit is realized, for example, by the specific processing unit 290 of the data processing device 12 and analyzes factors that result in an incorrect answer. The suggestion unit is realized, for example, by the control unit 46A of the headset-type terminal 314 and suggests a prompt to bring the user closer to the desired information. === Hard Collateral 1-4 === Each of the multiple elements, including the above-mentioned reception unit, verification unit, analysis unit, and suggestion unit, is realized, for example, by at least one of the robot 414 and the data processing device 12. For example, the reception unit is realized by the control unit 46A of the robot 414 and allows the user to input the difference between the base prompt and the resulting desired information. The verification unit is realized, for example, by the specific processing unit 290 of the data processing device 12 and verifies the correlation between the prompt and the output using a generation AI. The analysis unit is realized, for example, by the specific processing unit 290 of the data processing device 12 and analyzes factors that result in an incorrect answer. The suggestion unit is realized, for example, by the control unit 46A of the robot 414 and suggests a prompt to bring the user closer to the desired information.

[0147] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0148] The reception unit can analyze the user's input content in real time and provide appropriate suggestions based on the input content. For example, if the user inputs "Tell me the price of the product," the reception unit can display a suggestion such as "Would you also like to know the product's stock status?". Also, if the user inputs "Tell me the product reviews," the reception unit can display a suggestion such as "Would you also like to know the product's rating score?". Furthermore, if the user inputs "Tell me the product's delivery status," the reception unit can display a suggestion such as "Would you also like to know the estimated delivery date?". In this way, appropriate suggestions can be provided according to the user's input content, allowing the user to input information efficiently.

[0149] The similarity calculation unit can search for related past data and calculate the similarity based on the user's input. For example, if the user inputs "Tell me the price of this product," the similarity calculation unit searches for data in which a similar prompt was previously input and calculates the similarity. Also, if the user inputs "Tell me product reviews," the similarity calculation unit can search for data related to past reviews and calculate the similarity. Furthermore, if the user inputs "Tell me the delivery status of this product," the similarity calculation unit can search for data related to past delivery status and calculate the similarity. In this way, by searching for related past data based on the user's input and calculating the similarity, more accurate information can be provided.

[0150] The error analysis unit can identify the cause of the error based on the user's input and suggest a correction method. For example, if a user inputs "Tell me the price of this product" but the correct price is not displayed, the error analysis unit can identify the cause and suggest a correction method such as "The price information of this product may not have been updated." Also, if a user inputs "Tell me reviews of this product" but no reviews are displayed, the error analysis unit can identify the cause and suggest a correction method such as "The review information may not be sufficient." Furthermore, if a user inputs "Tell me the delivery status of this product" but the delivery status is not displayed, the error analysis unit can identify the cause and suggest a correction method such as "The delivery information may not have been updated." In this way, by identifying the cause of the error based on the user's input and suggesting a correction method, the user can efficiently correct errors.

[0151] The optimization unit can generate an optimal prompt based on the user's input. For example, if the user inputs "Tell me the price of the product," the optimization unit can generate an optimal prompt such as "Tell me the price and stock status of the product." Also, if the user inputs "Tell me the reviews of the product," the optimization unit can generate an optimal prompt such as "Tell me the reviews and rating score of the product." Furthermore, if the user inputs "Tell me the delivery status of the product," the optimization unit can generate an optimal prompt such as "Tell me the delivery status and expected delivery date of the product." In this way, generating an optimal prompt based on the user's input allows the user to input information efficiently.

[0152] The reception unit can estimate the user's emotions and change the display method of the input interface based on the estimated user emotions. For example, if the user is feeling stressed, a simple interface can be provided to minimize the input steps. Alternatively, if the user is relaxed, detailed input options can be provided and a customizable input method can be suggested. Alternatively, if the user is in a hurry, voice input can be prioritized to allow the user to quickly input the difference between the base prompt and the result, or the desired information. In this way, adjusting the display method of the input interface according to the user's emotions reduces user stress and enables efficient input.

[0153] The reception unit can analyze the user's past input history and suggest an input method. For example, it can automatically display prompts that the user has frequently used in the past and the results thereof as candidates. The reception unit can also preferentially suggest input methods (such as voice and text) that the user has used in the past. The reception unit can also predict and suggest prompts and results to be used in a specific time period based on the user's past input history. For example, the reception unit can automatically display prompts that the user has frequently used in the past as candidates, making it easier for the user to select. The reception unit can also preferentially suggest input methods that the user has used in the past, allowing the user to input efficiently. The reception unit can also analyze the user's past input history and predict and suggest prompts and results to be used in a specific time period, allowing the user to input efficiently. In this way, the optimal input method can be suggested by analyzing the user's past input history, improving input efficiency.

[0154] The reception unit can change the input fields based on the user's current task or area of ​​interest when the user is entering data. For example, the reception unit can preferentially display prompts and result input fields related to the task the user is currently performing. The reception unit can also customize related prompts and result input fields based on the user's area of ​​interest. The reception unit can also automatically display input fields related to a specific project when the user is working on that project. For example, the reception unit preferentially displays prompts and result input fields related to the task the user is currently performing, allowing the user to enter data efficiently. The reception unit can also customize related prompts and result input fields based on the user's area of ​​interest, allowing the user to enter data efficiently. The reception unit can also automatically display input fields related to the project when the user is working on a specific project, allowing the user to enter data efficiently. This improves input efficiency by customizing input fields based on the user's current task or area of ​​interest.

[0155] The reception unit can select the optimal input means depending on the user's input method (voice, text, image, etc.) at the time of input. For example, if the user prefers voice input, the reception unit can provide voice input preferentially. Furthermore, if the user prefers text input, the reception unit can also provide text input preferentially. Furthermore, if the user prefers image input, the reception unit can also provide image input preferentially. For example, if the user prefers voice input, the reception unit can provide voice input preferentially, allowing the user to input efficiently. Furthermore, if the user prefers text input, the reception unit can provide text input preferentially, allowing the user to input efficiently. Furthermore, if the user prefers image input, the reception unit can provide image input preferentially, allowing the user to input efficiently. In this way, input efficiency is improved by selecting the optimal input means depending on the user's input method.

[0156] The reception unit can estimate the user's emotions and change the design of the input interface based on the estimated user's emotions. For example, if the user is nervous, the reception unit can provide an interface with subdued colors to reduce visual stress. If the user is having fun, the reception unit can provide an interface with bright colors to make input work more enjoyable. If the user is tired, the reception unit can provide a simple, highly visible interface to make input work easier. In this way, adjusting the design of the input interface according to the user's emotions reduces user stress and enables efficient input.

[0157] The reception unit can display highly relevant input fields based on the user's geographical location information when the user enters information. For example, if the user is in a specific region, the reception unit can prioritize displaying prompts and result input fields related to that region. Furthermore, if the user is traveling, the reception unit can prioritize displaying prompts and result input fields related to the travel destination. Furthermore, if the user is participating in a specific event, the reception unit can prioritize displaying prompts and result input fields related to the event. For example, if the user is in a specific region, the reception unit can prioritize displaying prompts and result input fields related to the region, allowing the user to enter information efficiently. Furthermore, if the user is traveling, the reception unit can prioritize displaying prompts and result input fields related to the travel destination, allowing the user to enter information efficiently. Furthermore, if the user is participating in a specific event, the reception unit can prioritize displaying prompts and result input fields related to the event, allowing the user to enter information efficiently. In this way, input efficiency is improved by preferentially displaying highly relevant input fields based on the user's geographical location information.

[0158] The processing flow of the second embodiment will be briefly explained below.

[0159] Step 1: The reception unit receives input from the user of the Base prompt, and the difference between the result and the desired information. The information input by the user includes the specific prompt, its output, and the difference between the desired information. For example, if the user receives an output such as "The price of the product is 1,000 yen" in response to a prompt such as "Please tell me the price of the product," the user can input "I would also like to know the product's stock status" as the difference between the desired information and the actual information. Step 2: The verification unit uses the generation AI to verify the correlation between the prompt and the output based on the information entered in the reception unit. For example, the generation AI analyzes the relationship between the prompt and its output, and identifies which parts are close to the correct answer and which parts deviate from the correct answer. For example, if the output obtained in response to the prompt "Tell me the price of this product" is "The price of this product is 1,000 yen," the verification unit verifies whether this output is the correct answer to the prompt. Step 3: The analysis section uses the generation AI to analyze the factors that cause the answer to be incorrect based on the results obtained by the verification section. For example, the generation AI can identify which part is causing the incorrect answer based on the correlation between the prompt and the output. For example, if the prompt "Tell me the price of this product" returns the output "The price of this product is 1,000 yen," but the user enters the desired information as "I also want to know the product's stock status," the generation AI will analyze this difference and identify the factors that cause the incorrect answer. Step 4: The suggestion unit uses the generation AI to propose prompts that bring the user closer to the correct answer based on the results obtained by the analysis unit. For example, based on the factors that caused the generation AI to give an incorrect answer, the suggestion unit proposes prompts that bring the user closer to the information they want. For example, if the output obtained in response to the prompt "Tell me the price of this product" is "The price of this product is 1,000 yen," but the user enters the desired information as "I would also like to know the product's stock status," the generation AI will propose a prompt such as "Tell me the price and stock status of this product."

[0160] 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.

[0161] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> Examples of generative AIs include the data generation model 58, such as a neural network model (e.g., a neural network model), and a neural network model (e.g., a neural network model). 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 speech, text data indicating text, and image data indicating an image is also input to the data generation model 58. 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. The specification processing unit 290 performs the above-mentioned specification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AIs other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processes of each of the above-mentioned parts are performed by AI, the processes may be performed in part or entirely by AI, but are not limited to these examples. The processes performed by AI, including the generative AI, may be replaced with rule-based processes.

[0162] Furthermore, the processing by the data processing system 10 described above is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart device 14, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart device 14. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the smart device 14 or an external device, and the smart device 14 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0163] The correspondence between each part and the device or control part is not limited to the example described above, and various modifications are possible.

[0164] [Second embodiment] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0165] 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.

[0166] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and 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 and / or a LAN.

[0167] 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.

[0168] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to instructions from the processor 46.

[0169] 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 user's surroundings (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0170] 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.

[0171] 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.

[0172] 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.

[0173] 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. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0174] In the smart glasses 214, the specific processing is performed by the processor 46. A specific processing program 60 is stored in the storage 50. The processor 46 reads the specific processing program 60 from the storage 50 and executes the read specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as the control unit 46A in accordance with the specific processing program 60 executed on the RAM 48. The smart glasses 214 also have a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models.

[0175] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 communicates with the server device having the data generation model 58 to obtain a processing result (such as a prediction result) using the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device (for example, a mobile phone, a robot, a home appliance, etc.) owned by a user.

[0176] 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.

[0177] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt containing an instruction, as well as inference data such as voice data representing speech, text data representing text, and image data representing an image. 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. The identification processing unit 290 performs the above-mentioned identification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AIs other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processes of each of the above-mentioned parts are performed by AI, the processes may be performed in part or entirely by AI, but are not limited to these examples. The processes performed by AI, including the generative AI, may be replaced with rule-based processes.

[0178] The data processing system 210 according to the second embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 210 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart glasses 214, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart glasses 214. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the smart glasses 214 or an external device, etc., and the smart glasses 214 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0179] The correspondence between each part and the device or control part is not limited to the example described above, and various modifications are possible.

[0180] [Third embodiment] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0181] 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.

[0182] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and 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 and / or a LAN.

[0183] 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.

[0184] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to instructions from the processor 46.

[0185] 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 user's surroundings (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0186] 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.

[0187] 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.

[0188] 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.

[0189] 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. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0190] In the headset type terminal 314, the identification process is performed by the processor 46. A identification program 60 is stored in the storage 50. The processor 46 reads the identification program 60 from the storage 50 and executes the read identification program 60 on the RAM 48. The identification process is realized by the processor 46 operating as a control unit 46A in accordance with the identification program 60 executed on the RAM 48. Note that the headset type terminal 314 has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can also perform processing similar to that of the identification processing unit 290 using these models.

[0191] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 communicates with the server device having the data generation model 58 to obtain a processing result (such as a prediction result) using the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device (for example, a mobile phone, a robot, a home appliance, etc.) owned by a user.

[0192] 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.

[0193] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt containing an instruction, as well as inference data such as voice data representing speech, text data representing text, and image data representing an image. 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. The identification processing unit 290 performs the above-mentioned identification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AIs other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processes of each of the above-mentioned parts are performed by AI, the processes may be performed in part or entirely by AI, but are not limited to these examples. The processes performed by AI, including the generative AI, may be replaced with rule-based processes.

[0194] The data processing system 310 according to the third embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 310 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the headset type terminal 314, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the headset type terminal 314. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the headset type terminal 314 or an external device, etc., and the headset type terminal 314 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0195] The correspondence between each part and the device or control part is not limited to the example described above, and various modifications are possible.

[0196] [Fourth embodiment] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0197] 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0198] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, RAM 30, and 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 and / or a LAN.

[0199] 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.

[0200] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to instructions from the processor 46.

[0201] 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 image sensor or a CCD image sensor, and captures images of the user's surroundings (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0202] 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.

[0203] The control object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the 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.

[0204] 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.

[0205] 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.

[0206] 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. The identification processing unit 290 can estimate a user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, but is not limited to these examples. Furthermore, the estimation and prediction of emotion also includes, for example, emotion analysis.

[0207] In the robot 414, the processor 46 performs the identification process. The storage 50 stores the identification program 60. The processor 46 reads the identification program 60 from the storage 50 and executes the read identification program 60 on the RAM 48. The identification process is realized by the processor 46 operating as the control unit 46A in accordance with the identification program 60 executed on the RAM 48. The robot 414 also has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can perform the same process as the identification processing unit 290 using these models.

[0208] Note that a device other than the data processing device 12 may have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 communicates with the server device having the data generation model 58 to obtain a processing result (such as a prediction result) using the data generation model 58. Furthermore, the data processing device 12 may be a server device, or may be a terminal device (for example, a mobile phone, a robot, a home appliance, etc.) owned by a user.

[0209] 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.

[0210] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 receives a prompt containing an instruction, as well as inference data such as voice data representing speech, text data representing text, and image data representing an image. 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. The identification processing unit 290 performs the above-mentioned identification processing using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction. In this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. The data processing device 12 and the like include multiple types of data generation models 58, and the data generation model 58 includes AIs other than the generative AI. The AI ​​other than the generative AI may be, for example, linear regression, logistic regression, decision tree, random forest, support vector machine (SVM), k-means clustering, convolutional neural network (CNN), recurrent neural network (RNN), generative adversarial network (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. The AI ​​may also be an AI agent. When the processes of each of the above-mentioned parts are performed by AI, the processes may be performed in part or entirely by AI, but are not limited to these examples. The processes performed by AI, including the generative AI, may be replaced with rule-based processes.

[0211] The data processing system 410 according to the fourth embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 410 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the robot 414, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the robot 414. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the robot 414 or an external device, etc., and the robot 414 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0212] The correspondence between each part and the device or control part is not limited to the example described above, and various modifications are possible.

[0213] 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.

[0214] 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 encompasses both emotions 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.

[0215] 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.

[0216] 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).

[0217] 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 expressed, and when they approach the ideal, a state of pleasure is expressed. Emotions can also be created for robots, cars, 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 expressed, and when they approach the ideal, a state of pleasure is expressed. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on speech emotion recognition and brain physiological signal analysis systems for emotions, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the area called "reaction," where sensation is dominant. The right half of the emotion map lists emotions belonging to the area called "situation," where situational awareness is dominant.

[0218] 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."

[0219] 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.

[0220] In the above embodiment, an example was given in which a specific process is performed by one computer 22, but the technology disclosed herein is not limited to this, and distributed processing of the specific process may be performed by multiple computers including computer 22.

[0221] 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.

[0222] 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.

[0223] 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.

[0224] The hardware resource for executing a specific process can be any of the following types of processors: A processor, for example, 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. A processor also includes 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.

[0225] The hardware resource that executes the specific process 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 process may be a single processor.

[0226] 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.

[0227] 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.

[0228] In the above example, the first to fourth embodiments have been described separately, but some or all of these embodiments may be combined. The smart device 14, smart glasses 214, headset terminal 314, and robot 414 are merely examples, and they may be combined, or other devices may be used. In the above example, the first and second embodiments have been described separately, but they may be combined.

[0229] 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.

[0230] 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.

[0231] [Explanation of symbols]

[0232] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot

Claims

1. A reception section where a user inputs the difference between the Base prompt and the resultant desired information; A verification unit in which the generation AI verifies the correlation between the prompt and the output based on the information input by the reception unit; an analysis unit that analyzes factors based on the results obtained by the verification unit; a suggestion unit that proposes a prompt based on the results obtained by the analysis unit; Equipped with A system characterized by:

2. The verification unit Equipped with a similarity calculation unit that uses natural language processing technology to calculate the similarity between prompts and outputs 2. The system of claim 1.

3. The analysis unit An error analysis unit that uses error analysis or a feedback loop 2. The system of claim 1.

4. The proposal unit It has an optimization section that generates prompts based on user feedback and past data.

2. The system of claim 1.

5. The reception unit Estimates the user's emotions and changes the display method of the input interface based on the estimated user emotions.

2. The system of claim 1.

6. The reception unit Analyzes the user's past input history and suggests input methods 2. The system of claim 1.

7. The reception unit Change input fields as you type based on the user's current task or area of ​​interest 2. The system of claim 1.

8. The reception unit When inputting, select the input method according to the user's input method 2. The system of claim 1.

9. The reception unit Estimate the user's emotions and change the design of the input interface based on the estimated user emotions.

2. The system of claim 1.

Images