System

Abstract

A system is provided.SOLUTION: A system comprising: means for uploading a document summarized by a user and raw data; means for generating a summary from the uploaded raw data according to a generation AI; means for comparing the document summarized by the user and the summary according to the generation AI to detect inconsistencies and information deficiencies; and means for providing feedback to the user based on the comparison.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] The ability to summarize documents and discussions is an important skill in many workplaces and educational settings. However, in reality, many people have not fully mastered this skill, resulting in inaccurate summaries and missing information. Conventional methods struggle to improve summary quality or identify inconsistencies, and lack feedback to help users appropriately improve their summarization skills. This often results in poor quality documents and presentations, making effective communication difficult. There is a need for a system that can solve these problems and enable users to summarize efficiently and accurately. [Means for solving the problem]

[0005] To solve the above problems, the present invention provides the following means. A system is constructed that includes: a means for users to upload summarized documents and original data; a means for generating summaries from the uploaded original data using a generation AI; a means for comparing the user-summarized document with the summary generated by the generation AI to detect inconsistencies and information oversaturation and deficiencies; and a means for providing feedback to the user based on the comparison results. This system checks the format and content of the original data upon receiving it, and performs data formatting and preprocessing if necessary. Furthermore, when evaluating the comparison results, natural language processing technology is used to confirm semantic consistency and context. This allows users to receive specific and useful feedback to improve their summarizing skills, thereby improving the quality of their document creation and presentations.

[0006] A "summary" is a document that extracts the most important parts of the original information and summarizes them concisely.

[0007] "Original data" refers to the initial data that contains all the information or content to be summarized.

[0008] "Generative AI" is a system that uses artificial intelligence technology to automatically generate summaries from input data.

[0009] A "user" is an individual or organization that uses this system to summarize raw data.

[0010] A "contradiction" is a situation in which the summary generated by the AI ​​does not match the summary made by the user and contains conflicting information.

[0011] "Information overload / deficiency" refers to a situation in which important information contained in the original data is missing or unnecessary information is included in the document summarized by the user.

[0012] "Feedback" is information that the system provides to the user based on the comparison results, stating suggestions and areas for improvement.

[0013] "Preprocessing" is a preparatory step carried out to make the format and content of the original data consistent.

[0014] "Natural language processing technology" is a technology that enables computers to understand, analyze, and generate human language. [Brief explanation of the drawings]

[0015] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13]FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0016] 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.

[0017] First, the terms used in the following description will be explained.

[0018] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0019] 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.

[0020] 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.

[0021] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0022] 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."

[0023] [First embodiment]

[0024] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0025] 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.

[0026] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0027] 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.

[0028] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0029] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0030] 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.

[0031] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0032] 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.

[0033] 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.

[0034] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0035] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0036] This invention is a system for improving a user's document summarization ability. It uses a generative AI to compare documents manually summarized by the user with the original data, and identifies inconsistencies and information omissions. The main components of this system and their specific operation are described below.

[0037] System configuration

[0038] 1. Terminal

[0039] It provides an interface for users to input and upload the original data they want to summarize and the summary documents they have manually created. The devices can be computers, smartphones, tablets, etc.

[0040] 2. Server

[0041] It is the main processing unit responsible for generating summaries, comparative analysis, and generating feedback.

[0042] System Operation

[0043] Data reception

[0044] First, users upload original data and summary documents using their devices. This data can include meeting transcripts, reports, articles, etc. The server receives the data, checks its format and content, and performs formatting and preprocessing of the data if necessary.

[0045] Generate a summary

[0046] The server then uses a generative AI to automatically generate a summary using the original data as input. This generative AI works based on natural language processing technology, extracting key points from the original data and creating a shortened summary.

[0047] comparative analysis

[0048] The server compares the generated AI summary with the user's manual summary. This comparison checks the following points:

[0049] Inconsistency detection: Detect inconsistencies or conflicting information between the generated AI summary and the user's summary.

[0050] Information deficiency / excess detection: Check whether the user's summary is missing important information contained in the generated AI summary, or whether it contains unnecessary information.

[0051] Feedback Generation

[0052] The server generates specific feedback based on the comparison results, such as, "This summary does not include the 'progress report of each team,' which is a key point in the original data."

[0053] Providing Feedback

[0054] Finally, the server sends the generated feedback to the user's terminal, where the user can receive it and review their own summaries to improve their summarization skills.

[0055] Specific examples

[0056] Original data (example)

[0057] 1. Discussion on next year's budget

[0058] 2. New project launch plan

[0059] 3. Check the schedule of company events

[0060] 4. Progress reports from each team

[0061] User Summary (Example)

[0062] We discussed the budget for the next fiscal year and new projects, and also confirmed the dates for company events.

[0063] Processing result (example)

[0064] 1. Summary of Generative AI (Generated by the Server)

[0065] Discussions included next year's budget, plans to launch new projects, confirmation of dates for internal events, and progress reports from each team.

[0066] 2. Feedback (Server-Generated)

[0067] "Your summary does not include 'progress reports from each team.'"

[0068] In this way, users can improve the accuracy of their summaries by receiving specific feedback, which will enhance their ability to accurately summarize the main points of materials and discussions, ultimately improving the quality of their documents and presentations.

[0069] The processing flow will be explained below.

[0070] Step 1:

[0071] The user uploads the original data (e.g., transcription data of a meeting) and a summary document created manually via the terminal. The terminal then sends this data to the server.

[0072] Step 2:

[0073] The server receives the original data and the summary sent by the user. At this time, the server checks the consistency of the data format and performs preprocessing if necessary. Preprocessing includes text cleaning and formatting.

[0074] Step 3:

[0075] The server calls the generation AI, passing the original data as input. The generation AI extracts the key points of the original data and generates a short summary, which is later used to compare with the user summary.

[0076] Step 4:

[0077] The server compares the AI-generated summary with the one the user manually created. The comparison checks the following points:

[0078] Inconsistency detection: Detects inconsistencies between the generated AI summary and the user's summary.

[0079] Information deficiency / excess detection: Check whether the user's summary is missing important information contained in the generated AI summary or whether it contains unnecessary information.

[0080] Step 5:

[0081] The server generates feedback based on the comparison, including specific suggestions and areas for improvement, such as identifying important or unnecessary information missing from the user's summary.

[0082] Step 6:

[0083] The server sends the generated feedback to the user's device, allowing the user to review their summary and make corrections if necessary.

[0084] Example 1

[0085] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0086] Conventional document summarization systems have difficulty effectively detecting inconsistencies or excess or deficiencies of information between a user's manually created summary and the original data, and providing appropriate feedback. Furthermore, if data formatting and preprocessing are insufficient, errors are likely to occur in summary generation and comparative analysis. Furthermore, the feedback is not specific enough, so it is not possible to adequately support users in improving their summarization skills.

[0087] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0088] In this invention, the server includes means for uploading a document summarized by a user and the original data, means for checking the format and content of the uploaded original data and reformatting or preprocessing the data as necessary, means for generating a summary from the uploaded original data using a generative AI model, means for generating a summary by providing an appropriate prompt to the generative AI model, means for comparing the document summarized by the user with the summary generated by the generative AI model to detect inconsistencies or excess or deficiencies of information, and means for providing feedback to the user based on the comparison results. This makes it possible to effectively detect inconsistencies or excess or deficiencies of information between the summary created by the user and the original data and provide specific feedback.

[0089] "User" refers to a person who uses the system to create a summary of a document.

[0090] "Original data" refers to the collection of original text or information that a user uses to summarize.

[0091] A "terminal" is a device to which a user uploads original data and manually created summary documents, and includes a computer, a smartphone, a tablet, and the like.

[0092] "Server" refers to a device that performs major processing such as generating summaries, comparative analysis, and generating feedback.

[0093] A "generative AI model" refers to artificial intelligence that automatically generates summaries from original data based on natural language processing technology.

[0094] A "prompt" refers to an input instruction that instructs a generative AI model to perform a specific task.

[0095] "Feedback" refers to the indications and advice that the server provides to the user based on the results of the comparison and analysis.

[0096] "Data formatting" refers to standardizing the format of received data and making it suitable for analysis.

[0097] "Preprocessing" refers to the early stage of data processing carried out to improve the accuracy of data analysis and summary generation.

[0098] This invention is a system for improving a user's document summarization ability. It utilizes a generative AI model to compare documents manually summarized by the user with the original data, and identifies inconsistencies and information omissions. The main components of this system and their specific operation are described below.

[0099] System configuration

[0100] Terminal

[0101] It provides an interface for users to input and upload the original data they want to summarize and the summary documents they have manually created. The devices can be computers, smartphones, tablets, etc.

[0102] server

[0103] This is the main processing unit that generates summaries, performs comparative analysis, and generates feedback. A generative AI model (e.g., GPT-3 (registered trademark)) is installed on the server and used for automatic summary generation.

[0104] System Operation

[0105] Data reception

[0106] First, the user uploads the original data and summary documents to the interface using their terminal. This data includes, for example, minutes of meetings, reports, and articles. The server receives the data and checks its format and content. If any inappropriate content or format is found, the server reformats or preprocesses the data.

[0107] Generate a summary

[0108] The server passes the received raw data as input to the generative AI model, which then automatically generates a summary. In this process, the server provides the generative AI with an appropriate prompt. For example, the following prompt can be used:

[0109] "Please summarize the following sentences: 1. Discussion about next year's budget 2. Plan to launch a new project 3. Confirmation of schedule for company events 4. Progress reports from each team."

[0110] A generative AI model then generates a summary.

[0111] comparative analysis

[0112] The server compares the summaries generated by the generative AI model with the manual summaries created by the user. This comparison checks the following points:

[0113] Inconsistency detection: Detect inconsistencies or conflicting information between the generated AI summary and the user's summary.

[0114] Information deficiency / excess detection: Check whether the user's summary is missing important information contained in the generated AI summary, or whether it contains unnecessary information.

[0115] Feedback Generation

[0116] The server generates specific feedback based on the comparison, such as:

[0117] "Your summary does not include 'progress reports from each team.'"

[0118] Providing Feedback

[0119] Finally, the server sends the generated feedback to the user's terminal, where the user can receive it and review their own summaries to improve their summarization skills.

[0120] Specific examples

[0121] Original data (example)

[0122] 1. Discussion on next year's budget

[0123] 2. New project launch plan

[0124] 3. Check the schedule of company events

[0125] 4. Progress reports from each team

[0126] User Summary (Example)

[0127] We discussed the budget for the next fiscal year and new projects, and also confirmed the dates for company events.

[0128] Processing result (example)

[0129] 1. Summary of Generative AI (Generated by the Server)

[0130] Discussions included next year's budget, plans to launch new projects, confirmation of dates for internal events, and progress reports from each team.

[0131] 2. Feedback (Server-Generated)

[0132] Your summary does not include a "progress report from each team."

[0133] As described above, by receiving specific feedback, users can improve the accuracy of their summaries, thereby enhancing their ability to accurately summarize the main points of materials and discussions, and ultimately improving the quality of their documents and presentations.

[0134] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0135] Step 1:

[0136] Data reception

[0137] The user uses a terminal to upload the original data and a manually created summary document to the interface. The terminal then sends this input data to the server. The server receives the uploaded data and checks its format and content. For example, the data format received by the server is JSON format, and after receiving the input data, it parses it to check for errors. If an error is detected, the server notifies the user with an appropriate error message.

[0138] Step 2:

[0139] Data formatting and preprocessing

[0140] The server formats and preprocesses the received data. This includes standardizing the data format and making it easier to analyze. For example, the server cleans the text data, removes unnecessary whitespace and special characters, and normalizes the document. These operations transform the data into a state suitable for analysis.

[0141] Step 3:

[0142] Generate a summary

[0143] The server passes the formatted original data as input to a generative AI model (e.g., GPT-3) to automatically generate a summary. At this time, the server provides an appropriate prompt to the generative AI model. The following format is used as an example of a prompt:

[0144] "Please summarize the following sentences: 1. Discussion about next year's budget 2. Plan to launch a new project 3. Confirmation of schedule for company events 4. Progress reports from each team."

[0145] The generative AI model receives this prompt and generates a summary based on the original data, which is then received and stored by the server.

[0146] Step 4:

[0147] comparative analysis

[0148] The server compares and analyzes the summaries generated by the AI ​​model with the manual summaries created by the user. This comparison involves text analysis and similarity calculations to detect inconsistencies and information overloads and omissions. The server vectorizes the text of both summaries using, for example, TF-IDF or word embeddings, and calculates the degree of agreement. It also extracts parts where important information is missing or where unnecessary information has been added.

[0149] Step 5:

[0150] Generate feedback

[0151] The server generates specific feedback based on the results of the comparison and analysis. The generated feedback is written in natural language and provided in a format that is easy for users to understand. For example, the following feedback is generated:

[0152] "Your summary does not include 'progress reports from each team.'"

[0153] The server edits the feedback text to include useful suggestions and advice for the user.

[0154] Step 6:

[0155] Providing feedback

[0156] The server sends the generated feedback to the user's device, which receives the feedback and notifies the user. The user can then use the feedback to review their own summaries and create more accurate summaries.

[0157] By clearly indicating the specific operations and the flow of input and output data in each step, the embodiments of the invention can be explained in detail.

[0158] (Application example 1)

[0159] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0160] In traditional advertising production, it is difficult to evaluate the effectiveness of advertising copy and the accuracy of important information, requiring a lot of time and effort. In particular, there is a lack of means to quickly and accurately determine whether the summary created by advertising creators is conveyed to consumers, making it difficult to maximize the effectiveness of advertising.

[0161] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0162] In this invention, the server includes means for comparing the original advertisement data uploaded by the user with the summary and evaluating the accuracy of the information and the effectiveness of the summary, means for generating a summary using a generation AI, and means for comparing the document summarized by the user with the summary generated by the generation AI to detect inconsistencies and excess or deficiency of information. This enables advertisement creators to quickly determine whether their summary matches the generation AI and whether important points are missing, thereby optimizing the message of the advertisement.

[0163] A "document summarized by a user" is a document that has been manually summarized and simplified from original data by a user.

[0164] "Original data" refers to documents or data that contain the information that is the basis for summarizing.

[0165] "Generative AI" is a system that uses artificial intelligence technology to automatically generate summaries from original data.

[0166] A "means for generating a summary" is a mechanism for automatically creating a summary from original data using generative AI.

[0167] The "means for detecting inconsistencies and information omissions" is a mechanism for comparing the generated AI's summary with the user's summary to detect inconsistencies or missing important information.

[0168] The "means for providing feedback" is a mechanism for suggesting specific improvements or corrections to the user based on the comparison results.

[0169] "Advertising source data" refers to the documents and data that are the basis for creating advertisements.

[0170] "Means for assessing the accuracy of information and the effectiveness of summaries" refers to a mechanism for comparing the original advertising data with the summaries to assess whether the information is being conveyed accurately and whether the summaries are effective.

[0171] An embodiment of the present invention is a system and method for evaluating advertising copy created by advertising creators and generating effective advertising summaries. This system is mainly composed of a server and a user terminal, and utilizes a generative AI model to automatically generate summaries of advertising copy and compare them with summaries created by users, thereby providing effective and efficient feedback.

[0172] System configuration:

[0173] server

[0174] The server performs the following main functions:

[0175] Receive the original data (advertising source data) and check the format and content.

[0176] Format and preprocess the data as needed.

[0177] A generative AI model is used to automatically generate summaries from the original data.

[0178] The summary created by the user is compared with the summary generated by the AI ​​to detect inconsistencies and information that is insufficient or excessive.

[0179] Feedback is generated based on the comparison result and transmitted to the user terminal.

[0180] Hardware and software used

[0181] Hardware:

[0182] The server may be a high-performance computer or a cloud computing infrastructure (e.g., AWS (registered trademark), Google (registered trademark) Cloud Platform).

[0183] software:

[0184] For data formatting and preprocessing, we use Python's pandas and numpy libraries.

[0185] The generative AI model used is OpenAI's (registered trademark) GPT-3.

[0186] The Flask framework is used to build the web interface.

[0187] Natural language processing technologies (e.g., spaCy, NLTK) may also be used for comparative analysis and feedback generation.

[0188] Processing flow

[0189] 1. Data reception and preprocessing:

[0190] Users upload the original advertising data and their own summaries from their smartphones or PCs. The server receives the data, checks the format and content, and formats and preprocesses the data as necessary.

[0191] 2. Generate a summary:

[0192] The server uses a generative AI model (e.g., OpenAI's GPT-3) to automatically generate a summary from the original data. This summary is a shorthand representation of the main points of the original data.

[0193] 3. Comparative analysis:

[0194] The summary created by the user is compared with the summary generated by the server to check for consistency and whether there is any excess or deficiency in the information. This is done using the natural language processing technology of the generation AI.

[0195] 4. Feedback Generation:

[0196] Generate specific feedback based on the comparison, such as "Your summary is missing the important information 'full of vitamin C'."

[0197] 5. Providing Feedback:

[0198] The generated feedback is sent to the user's device, and the user can use it to revise and improve the summary.

[0199] Specific examples

[0200] Ad source data:

[0201] Try our new Super Drink this fall! Packed with Vitamin C, it will help relieve fatigue.

[0202] User Summary:

[0203] Try a new vitamin drink.

[0204] Summary of generated AI (server generated):

[0205] This autumn's new product is a super drink packed with vitamin C.

[0206] Feedback (server generated):

[0207] "Your summary doesn't include key points like 'New this fall.'"

[0208] Prompt Sentence Examples

[0209] For the generative AI model, we input:

[0210] Summarize the following ad copy:

[0211] Try our new Super Drink this fall! Packed with Vitamin C, it will help relieve fatigue.

[0212] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0213] Step 1:

[0214] The user uploads the original ad data and a self-created summary to the server via their device. The device then sends the input data to the server. The input at this time is the "original ad data" and the "user summary," and the output to the server is the original data.

[0215] Step 2:

[0216] The server checks the format and content of the received source data and user summary. For example, the server checks whether the data format is correct and the content is entered properly, and performs data formatting (e.g., grammar correction and text format standardization) and preprocessing as necessary. The input to this process is the data received in step 1, and the output is the formatted and preprocessed data.

[0217] Step 3:

[0218] The server uses a generative AI model (OpenAI GPT-3) to generate a summary from the formatted and preprocessed original ad data. The generative AI receives the original data as a prompt in the form of "Please summarize the ad copy below:" The output of the generative AI is an automatically generated summary based on the original ad data.

[0219] Step 4:

[0220] The server compares the summary generated by the AI ​​with the summary created by the user. This comparison process uses natural language processing technology to check whether the meanings of the two sentences match and whether important information is missing. The inputs to this process are the AI's summary and the user's summary, and the output is information on whether they match or not based on the comparison results.

[0221] Step 5:

[0222] The server generates feedback based on the comparison results. Specifically, if there are any inconsistencies between the AI's summary and the user's summary, or if important information is missing, the server will provide specific feedback. The input for this process is the comparison result from step 4, and the output is a feedback statement to be given to the user.

[0223] Step 6:

[0224] The server sends the generated feedback to the user terminal, which receives it and makes further corrections and improvements. The input of this process is the feedback from step 5, and the output is the feedback sentence displayed on the user terminal.

[0225] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0226] This invention is a system for improving a user's document summarization ability. It utilizes generative AI to compare documents manually summarized by the user with the original data, and points out inconsistencies and excess or deficiency of information. It also combines an emotion engine to provide feedback based on the user's emotions. The main components of this system and their specific operation are described below.

[0227] System configuration

[0228] 1. Terminal

[0229] It provides an interface for users to input and upload the original data they want to summarize and the summary documents they have manually created. The device can be a computer, smartphone, tablet, etc. The device also has an emotion engine to identify the user's emotions.

[0230] 2. Server

[0231] It is the main processing unit for generating summaries, comparative analysis, and generating feedback. The server has the ability to adjust the feedback based on the user's emotions identified by the emotion engine.

[0232] System Operation

[0233] Data reception

[0234] First, the user uploads original data and summary documents using their device. This data can include meeting transcripts, reports, articles, etc. The server receives the data, checks its format and content, and performs data formatting and preprocessing if necessary. The device's on-board emotion engine also identifies the user's current emotion, which is then sent to the server.

[0235] Generate a summary

[0236] The server then uses a generative AI to automatically generate a summary using the original data as input. This generative AI works based on natural language processing technology, extracting key points from the original data and creating a shortened summary.

[0237] comparative analysis

[0238] The server compares the generated AI summary with the user's manual summary. The comparison checks the following points:

[0239] Inconsistency detection: Detect inconsistencies or conflicting information between the generated AI summary and the user's summary.

[0240] Information overload / deficiency detection: Check whether the user's summary is missing important information contained in the generated AI summary, or whether it contains unnecessary information.

[0241] Feedback Generation

[0242] The server generates optimal feedback for the user based on the comparison results and also takes into account information from the emotion engine. For example, if the user is feeling impatient or anxious, the feedback message will be adjusted to be gentle and encouraging.

[0243] Providing Feedback

[0244] Finally, the server sends the generated feedback to the user's device. The user can then review their summary and improve their summarization skills. The server also tracks the user's emotional changes and analyzes long-term patterns to further optimize future feedback.

[0245] Specific examples

[0246] Original data (example)

[0247] 1. Discussion on next year's budget

[0248] 2. New project launch plan

[0249] 3. Check the schedule of company events

[0250] 4. Progress reports from each team

[0251] User Summary (Example)

[0252] We discussed the budget for the next fiscal year and new projects, and also confirmed the dates for company events.

[0253] Processing result (example)

[0254] 1. Summary of Generative AI (Generated by the Server)

[0255] Discussions included next year's budget, plans to launch new projects, confirmation of dates for internal events, and progress reports from each team.

[0256] 2. User's emotions (recognized by the emotion engine)

[0257] "A little impatient"

[0258] 3. Feedback (Server-Generated)

[0259] "Your summary does not include the 'progress report of each team,' but it is a good summary overall. Please take your time and review it again."

[0260] In this way, users can efficiently improve their summarizing skills by receiving feedback that takes into account specific points and emotions, which will improve their ability to accurately summarize the main points of materials and discussions, and ultimately improve the quality of their documents and presentations.

[0261] The processing flow will be explained below.

[0262] Step 1:

[0263] The user uses the device to upload the original data (e.g., a transcript of a meeting) and a manually created summary document. The user uses the device interface to select the file and clicks the send button.

[0264] Step 2:

[0265] The device sends the uploaded original data and summary to the server. At the same time, the device's built-in emotion engine analyzes the user's emotions and sends the results to the server. The emotion engine identifies emotions such as "anxiety" or "calmness" based on the user's facial expressions and input speed.

[0266] Step 3:

[0267] The server receives the original data and summary text sent by the user. At this time, the server checks the consistency of the data format and, if there is any inconsistency, formats and preprocesses the data. Specifically, it removes unnecessary whitespace and standardizes the text.

[0268] Step 4:

[0269] The server passes the original data to the AI ​​generator and requests it to generate a summary. The AI ​​generator extracts the key points of the original data and generates a shortened summary. Natural language processing technology is used in this process.

[0270] Step 5:

[0271] The server compares the AI-generated summary with the summary created manually by the user. The server checks the following points:

[0272] Inconsistency detection: Detects where the generated AI summary and the user's summary do not match.

[0273] Information deficiency / excess detection: Check whether the user's summary is missing important information contained in the generated AI summary or whether it contains unnecessary information.

[0274] Step 6:

[0275] The server generates feedback based on the comparison results, taking into account the emotions expressed by the user when uploading. For example, if the user is feeling "impatient," the server adjusts the feedback message to emphasize a gentle, encouraging tone.

[0276] Step 7:

[0277] The server sends the generated feedback to the user's device, where the user can then revise their summary. Based on the feedback, which takes specific points and emotions into consideration, the user can improve their summarization skills.

[0278] Specific examples

[0279] Let's say the original data looks like this:

[0280] 1. Discussion on next year's budget

[0281] 2. New project launch plan

[0282] 3. Check the schedule of company events

[0283] 4. Progress reports from each team

[0284] Suppose the user's summary was:

[0285] We discussed the budget for the next fiscal year and new projects, and also confirmed the dates for company events.

[0286] The summary generated by the server is:

[0287] Discussions included next year's budget, plans to launch new projects, confirmation of dates for internal events, and progress reports from each team.

[0288] If the user's emotion is recognized as "impatience," the server's feedback will be as follows:

[0289] Your summary does not include the "progress report of each team," but it is a good summary overall. Please take your time and review it again.

[0290] This example allows users to improve their summaries and receive feedback that will help them create their next summaries.

[0291] Example 2

[0292] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0293] Existing document summarization systems have the problem of being unable to accurately identify inconsistencies or information omissions when comparing manually created summaries with automatically generated summaries. Furthermore, feedback is provided without taking into account the user's emotional state, which can reduce the user's motivation and learning effectiveness. Furthermore, the lack of long-term support for user development makes it difficult to achieve continuous improvement.

[0294] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0295] In this invention, the server includes means for uploading a document summarized by a user and its original data, means for generating a summary from the uploaded original data using a generative AI model, means for comparing the document summarized by the user with the summary generated by the generative AI model to detect inconsistencies and excess or deficiency of information, means for providing feedback based on the user's emotions using an emotion engine based on the comparison results, and means for transmitting the feedback to the user's terminal and tracking changes in the user's emotions. This makes it possible to effectively compare the summary created by the user with the automatically generated summary, provide appropriate feedback based on the user's emotional state, and support continuous improvement.

[0296] "User" means an individual or organization that uses the document summarization system and creates and uploads original data and summarized documents.

[0297] "Terminal" means a hardware device through which a user inputs and uploads original data and summary documents, such as a computer, smartphone, or tablet.

[0298] The "server" is the main processing unit responsible for generating summaries, comparative analysis, and feedback generation, and is a computer equipped with the generative AI model and emotion engine.

[0299] "Original data" refers to documents uploaded by users to be summarized, including meeting minutes, articles, reports, and the like.

[0300] A "summary document" is a shortened document that is manually created by a user based on the original data.

[0301] A "generative AI model" is an artificial intelligence algorithm that automatically generates a summary based on original data.

[0302] An "emotion engine" is a computer program that analyzes a user's facial expressions and tone of voice to identify the user's current emotional state.

[0303] "Feedback" refers to comments or suggestions provided to the user's summary document based on the results of the comparative analysis, and is adjusted to take into account the user's emotional state.

[0304] "Data formatting" is the process of checking the format and content of the original data received and converting it into a format that is easy to analyze.

[0305] "Preprocessing" refers to a series of preparatory steps that are performed before analyzing raw data and generating summaries, including text cleaning and preparation for analysis.

[0306] This invention is a system for improving users' document summarization abilities by utilizing a generative AI model to compare documents manually summarized by the user with the original data, and to identify inconsistencies and information omissions. This system also incorporates an emotion engine to provide feedback based on the user's emotions.

[0307] The main components of the system of the present invention are a terminal, a server, a generative AI model, and an emotion engine. First, a user uploads original data and a manually created summary document using a terminal. This terminal can be a computer, smartphone, tablet, or other device. The terminal also has an emotion engine that identifies the user's emotions.

[0308] The original data and manual summaries are sent from the device to the server, which receives them, formats and preprocesses them as needed, and then summarizes the original data using a generative AI model, which works based on natural language processing techniques.

[0309] The server compares the generated AI summary with the user's manual summary, checking for inconsistencies and information oversaturation. Based on the results of this comparison, the server generates optimal feedback, taking into account the user's emotional information received from the device. Finally, the server sends the generated feedback to the user's device, allowing the user to review their own summary based on it and improve their summarization ability.

[0310] Examples of prompt statements

[0311] Specific examples are shown below.

[0312] Original data (example)

[0313] 1. Discussion on next year's budget

[0314] 2. New project launch plan

[0315] 3. Check the schedule of company events

[0316] 4. Progress reports from each team

[0317] User Summary (Example)

[0318] We discussed the budget for the next fiscal year and new projects, and also confirmed the dates for company events.

[0319] Processing result (example)

[0320] 1. Summary of Generative AI (Generated by the Server)

[0321] Discussions included next year's budget, plans to launch new projects, confirmation of dates for internal events, and progress reports from each team.

[0322] 2. User's emotions (recognized by the emotion engine)

[0323] "A little impatient"

[0324] 3. Feedback (Server-Generated)

[0325] "Your summary does not include the 'progress report of each team,' but it is a good summary overall. Please take your time and review it again."

[0326] In this way, users can efficiently improve their summarizing skills by receiving feedback that takes into account specific points and emotions, thereby improving their ability to accurately summarize the main points of materials and discussions, and ultimately improving the quality of their documents and presentations.

[0327] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0328] Step 1:

[0329] A user uses a terminal to upload the original data to be summarized and a manually created summary document.

[0330] Input: Original data file, manual summary document file

[0331] Output: Data transfer request to the server

[0332] Specific operation: The user opens the device's web browser or app, clicks the upload button, selects the original data file and the manual summary document file, and uploads them. The device's built-in emotion engine analyzes the user's facial expressions and tone of voice to identify the user's emotional state.

[0333] Step 2:

[0334] The terminal transmits the received data and the identified emotion information to the server.

[0335] Input: Original data file, manual summary document file, user emotion information

[0336] Output: Server data reception process begins

[0337] Specific operation: The terminal uploads the selected original data file and the manually summarized document file to the server, and simultaneously transmits the user's emotion information.

[0338] Step 3:

[0339] The server analyzes the received original data and manual summary documents, and formats and preprocesses the data.

[0340] Input: Received original data file, manual summary document file

[0341] Output: Formatted and preprocessed data

[0342] Specific operation: The server checks the received raw data files and summary documents based on their format and content, and performs formatting and preprocessing as necessary, including text cleaning and formatting standardization.

[0343] Step 4:

[0344] The server inputs the formatted and preprocessed raw data into a generative AI model to generate an automatic summary.

[0345] Input: Raw data that has been reformatted and preprocessed

[0346] Output: Automatic summary generated by the generative AI model

[0347] How it works: The server inputs the original data into a generative AI model, extracts key points, and automatically generates a summary. This generative AI model uses natural language processing technology to identify important information and output it as a summary.

[0348] Step 5:

[0349] The server compares the generated AI summary with the user's manually summarized document to detect inconsistencies and information omissions.

[0350] Input: Automated summaries by generative AI models, manually summarized documents

[0351] Output: Comparison results (status identifying inconsistencies and information oversaturations)

[0352] How it works: The server displays the AI-generated summary and the user's manual summary side by side and uses a specific algorithm to highlight differences, for example, showing inconsistencies in red and missing information in blue.

[0353] Step 6:

[0354] The server generates feedback based on the comparison results and adjusts it taking into account the user's emotional information provided by the emotion engine.

[0355] Input: Comparison results, user emotion information

[0356] Output: Adjusted feedback message

[0357] Specific behavior: The server generates specific feedback messages based on the differences highlighted, and adjusts the feedback to be gentle and encouraging if the user seems impatient.

[0358] Step 7:

[0359] The server transmits the generated feedback to the user's terminal, and the user receives the feedback.

[0360] Input: Adjusted feedback message

[0361] Output: A feedback message that is displayed on the user's terminal.

[0362] Specific operation: The server sends the generated feedback message to the user's device, and the user checks the feedback in notifications or on the dashboard. The user receives the feedback and reevaluates their own manual summaries to learn from them.

[0363] The above are the specific processing steps of the system. In this way, users can receive feedback based on specific suggestions and emotions, and can efficiently improve their summarization ability.

[0364] (Application example 2)

[0365] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0366] This invention relates to a system for improving the accuracy and efficiency of summarization for users who perform document summarization. In particular, the system aims to improve summarization ability by detecting inconsistencies in summary content and excess or deficiency of information through comparative analysis of automatic summaries generated by generative AI and manual summaries by users, and by providing optimal feedback to the user. Another objective of this invention is to provide a system that reduces the user's mental burden and encourages continuous learning by providing feedback based on the user's emotions in an emotional environment.

[0367] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for uploading a document summarized by a user and original data, means for generating a summary from the uploaded original data using a generation AI, means for comparing the document summarized by the user with the summary generated by the generation AI to detect inconsistencies and excess or deficiency of information, means for providing feedback to the user based on the comparison results, and means including an emotion engine for identifying the user's emotions and adjusting the feedback content. This makes it possible to improve the user's summarizing ability and provide feedback according to emotions.

[0368] "User-summarized documents" refers to information that has been manually summarized by a user based on original data.

[0369] "Original data" refers to a series of information or text data that is the basis for the summary.

[0370] "Generative AI" refers to a system that uses artificial intelligence technology to automatically summarize original data.

[0371] "Inconsistency" refers to inconsistencies in information or content between the generated AI summary and the user's summary.

[0372] "Information overload" refers to a situation in which important information is missing from a summary or unnecessary information is included.

[0373] "Feedback" refers to advice, including instructions for improvement or correction, provided to the user based on the results of comparing the generated AI with the user's summary.

[0374] An "emotion engine" refers to a device or software that identifies a user's emotions and generates feedback appropriate to those emotions.

[0375] "Prompt sentence" refers to the input sentence that the generative AI uses when summarizing.

[0376] The present invention is a system for improving users' summarization skills, and a specific embodiment is described below. This system uses a generative AI model and an emotion engine to perform a comparative analysis of a manually created summary document and the original data, and provides feedback.

[0377] System Configuration

[0378] 1. Terminal

[0379] The device provides an interface for users to input and upload the original data they want to summarize and the summary documents they have manually created. The device is equipped with an emotion engine to identify the user's emotions. This includes smartphones, tablets, and computers.

[0380] 2. Server

[0381] The server is the main processing unit for generating summaries, comparative analysis, and feedback generation. It includes the following means:

[0382] Summary generation method using generative AI

[0383] A comparative method for user-generated summaries and AI-generated summaries

[0384] A means of detecting inconsistencies and information overloads and deficiencies

[0385] Feedback adjustment method by emotion engine

[0386] System Operation

[0387] Data reception

[0388] The user uploads original data and manually summarized documents using a terminal, such as meeting transcripts, reports, articles, etc. The terminal uses an emotion engine to identify the user's emotions and transmits them to the server.

[0389] Generate a summary

[0390] The server uses a generative AI model (e.g., OpenAI GPT-4 (registered trademark)) to automatically generate a summary based on the original data. This generative AI works based on natural language processing technology to extract the main points of the original data and create a short summary.

[0391] comparative analysis

[0392] The server compares the summaries generated by the generative AI model with the manual summaries created by the user. The following points are checked:

[0393] Inconsistency detection: Detects inconsistencies and conflicting information between the generated AI summary and the user's summary.

[0394] Information deficiency / excess detection: Check whether the user's summary is missing important information contained in the generated AI summary or whether it contains unnecessary information.

[0395] Feedback Generation

[0396] Based on the comparison results and taking into account information from the emotion engine, the server generates optimal feedback for the user. For example, if the user is feeling anxious or impatient, the server will adjust the feedback message to be more gentle and encouraging.

[0397] Providing Feedback

[0398] The server generates feedback and sends it to the user's device. The user receives the feedback and reviews their own summarization to improve their summarization skills. The feedback is also optimized based on long-term emotional data.

[0399] Hardware and software used

[0400] Devices: smartphones (compatible with iOS / ANDROID (registered trademark)), tablets, computers

[0401] Emotion engine API: Azure (registered trademark) Emotion API

[0402] Generative AI model: OpenAI GPT-4

[0403] Server: AWS EC2, Node.js, Express

[0404] Prompt Sentence Examples

[0405] Original data: This product uses the latest technology, is highly durable, and remains comfortable even after extended use.

[0406] Generate a summary:

[0407] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0408] Step 1: Enter your data

[0409] The user inputs the original data and the manually summarized text into the application using a device (smartphone, tablet, or computer). The device then uses an emotion engine to identify the user's emotions. The input data includes the original data text, the manually summarized text, and emotion data.

[0410] Step 2: Sending data

[0411] The terminal transmits the input original data, the manual summary document, and the identified emotion data to the server. The transmitted data includes the original data, the manual summary document, and the emotion data.

[0412] Step 3: Data preparation and preprocessing

[0413] The server checks the format and content of the received original data and manual summary document, and formats and preprocesses the data as necessary. Specifically, it deletes unnecessary spaces and symbols and standardizes the text format. This process results in the formatted original data and manual summary document.

[0414] Step 4: Generate a summary

[0415] The server takes the formatted original data as input and generates an automatic summary using a generative AI model (OpenAI GPT-4). The prompt message generated is "Original data: [formatted original data] Please generate a summary:", which is then input into the AI ​​model to output the automatic summary.

[0416] Step 5: Comparative analysis of summaries

[0417] The server compares the automatic summary generated by the generative AI model with the manually summarized document. This comparison uses natural language processing technology to check for semantic agreement and context, and detects inconsistencies and information deficiencies. Specifically, it calculates word agreement rates and differences in contextual meaning, and outputs any inconsistencies or information deficiencies.

[0418] Step 6: Generate feedback

[0419] The server uses a generative AI model to generate optimal feedback for the user based on the results of the comparative analysis and the identified emotional data. The feedback message is generated using the results of the comparative analysis and the emotional data as input, and is output in a format that is easy for the user to understand.

[0420] Step 7: Provide feedback

[0421] The server sends the generated feedback to the user's device, where the user can receive the feedback and use it to improve the summary. The feedback content is adjusted based on the results of the comparative analysis and the user's sentiment.

[0422] The above is the specific processing flow of the system that realizes the application example.

[0423] 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.

[0424] 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> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0425] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0426] [Second embodiment]

[0427] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0428] 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.

[0429] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0430] 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.

[0431] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0432] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0433] 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.

[0434] 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.

[0435] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0436] 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.

[0437] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0438] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0439] This invention is a system for improving a user's document summarization ability. It uses a generative AI to compare documents manually summarized by the user with the original data, and identifies inconsistencies and information omissions. The main components of this system and their specific operation are described below.

[0440] System configuration

[0441] 1. Terminal

[0442] It provides an interface for users to input and upload the original data they want to summarize and the summary documents they have manually created. The devices can be computers, smartphones, tablets, etc.

[0443] 2. Server

[0444] It is the main processing unit responsible for generating summaries, comparative analysis, and generating feedback.

[0445] System Operation

[0446] Data reception

[0447] First, users upload original data and summary documents using their devices. This data can include meeting transcripts, reports, articles, etc. The server receives the data, checks its format and content, and performs formatting and preprocessing of the data if necessary.

[0448] Generate a summary

[0449] The server then uses a generative AI to automatically generate a summary using the original data as input. This generative AI works based on natural language processing technology, extracting key points from the original data and creating a shortened summary.

[0450] comparative analysis

[0451] The server compares the generated AI summary with the user's manual summary. This comparison checks the following points:

[0452] Inconsistency detection: Detect inconsistencies or conflicting information between the generated AI summary and the user's summary.

[0453] Information deficiency / excess detection: Check whether the user's summary is missing important information contained in the generated AI summary, or whether it contains unnecessary information.

[0454] Feedback Generation

[0455] The server generates specific feedback based on the comparison results, such as, "This summary does not include the 'progress report of each team,' which is a key point in the original data."

[0456] Providing Feedback

[0457] Finally, the server sends the generated feedback to the user's terminal, where the user can receive it and review their own summaries to improve their summarization skills.

[0458] Specific examples

[0459] Original data (example)

[0460] 1. Discussion on next year's budget

[0461] 2. New project launch plan

[0462] 3. Check the schedule of company events

[0463] 4. Progress reports from each team

[0464] User Summary (Example)

[0465] We discussed the budget for the next fiscal year and new projects, and also confirmed the dates for company events.

[0466] Processing result (example)

[0467] 1. Summary of Generative AI (Generated by the Server)

[0468] Discussions included next year's budget, plans to launch new projects, confirmation of dates for internal events, and progress reports from each team.

[0469] 2. Feedback (Server-Generated)

[0470] "Your summary does not include 'progress reports from each team.'"

[0471] In this way, users can improve the accuracy of their summaries by receiving specific feedback, which will enhance their ability to accurately summarize the main points of materials and discussions, ultimately improving the quality of their documents and presentations.

[0472] The processing flow will be explained below.

[0473] Step 1:

[0474] The user uploads the original data (e.g., transcription data of a meeting) and a summary document created manually via the terminal. The terminal then sends this data to the server.

[0475] Step 2:

[0476] The server receives the original data and the summary sent by the user. At this time, the server checks the consistency of the data format and performs preprocessing if necessary. Preprocessing includes text cleaning and formatting.

[0477] Step 3:

[0478] The server calls the generation AI, passing the original data as input. The generation AI extracts the key points of the original data and generates a short summary, which is later used to compare with the user summary.

[0479] Step 4:

[0480] The server compares the AI-generated summary with the one the user manually created. The comparison checks the following points:

[0481] Inconsistency detection: Detects inconsistencies between the generated AI summary and the user's summary.

[0482] Information deficiency / excess detection: Check whether the user's summary is missing important information contained in the generated AI summary or whether it contains unnecessary information.

[0483] Step 5:

[0484] The server generates feedback based on the comparison, including specific suggestions and areas for improvement, such as identifying important or unnecessary information missing from the user's summary.

[0485] Step 6:

[0486] The server sends the generated feedback to the user's device, allowing the user to review their summary and make corrections if necessary.

[0487] Example 1

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

[0489] Conventional document summarization systems have difficulty effectively detecting inconsistencies or excess or deficiencies of information between a user's manually created summary and the original data, and providing appropriate feedback. Furthermore, if data formatting and preprocessing are insufficient, errors are likely to occur in summary generation and comparative analysis. Furthermore, the feedback is not specific enough, so it is not possible to adequately support users in improving their summarization skills.

[0490] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0491] In this invention, the server includes means for uploading a document summarized by a user and the original data, means for checking the format and content of the uploaded original data and reformatting or preprocessing the data as necessary, means for generating a summary from the uploaded original data using a generative AI model, means for generating a summary by providing an appropriate prompt to the generative AI model, means for comparing the document summarized by the user with the summary generated by the generative AI model to detect inconsistencies or excess or deficiencies of information, and means for providing feedback to the user based on the comparison results. This makes it possible to effectively detect inconsistencies or excess or deficiencies of information between the summary created by the user and the original data and provide specific feedback.

[0492] "User" refers to a person who uses the system to create a summary of a document.

[0493] "Original data" refers to the collection of original text or information that a user uses to summarize.

[0494] A "terminal" is a device to which a user uploads original data and manually created summary documents, and includes a computer, a smartphone, a tablet, and the like.

[0495] "Server" refers to a device that performs major processing such as generating summaries, comparative analysis, and generating feedback.

[0496] A "generative AI model" refers to artificial intelligence that automatically generates summaries from original data based on natural language processing technology.

[0497] A "prompt" refers to an input instruction that instructs a generative AI model to perform a specific task.

[0498] "Feedback" refers to the indications and advice that the server provides to the user based on the results of the comparison and analysis.

[0499] "Data formatting" refers to standardizing the format of received data and making it suitable for analysis.

[0500] "Preprocessing" refers to the early stage of data processing carried out to improve the accuracy of data analysis and summary generation.

[0501] This invention is a system for improving a user's document summarization ability. It utilizes a generative AI model to compare documents manually summarized by the user with the original data, and identifies inconsistencies and information omissions. The main components of this system and their specific operation are described below.

[0502] System configuration

[0503] Terminal

[0504] It provides an interface for users to input and upload the original data they want to summarize and the summary documents they have manually created. The devices can be computers, smartphones, tablets, etc.

[0505] server

[0506] This is the main processing unit that generates summaries, performs comparative analysis, and generates feedback. A generative AI model (e.g., GPT-3) is installed on the server and used to automatically generate summaries.

[0507] System Operation

[0508] Data reception

[0509] First, the user uploads the original data and summary documents to the interface using their terminal. This data includes, for example, minutes of meetings, reports, and articles. The server receives the data and checks its format and content. If any inappropriate content or format is found, the server reformats or preprocesses the data.

[0510] Generate a summary

[0511] The server passes the received raw data as input to the generative AI model, which then automatically generates a summary. In this process, the server provides the generative AI with an appropriate prompt. For example, the following prompt can be used:

[0512] "Please summarize the following sentences: 1. Discussion about next year's budget 2. Plan to launch a new project 3. Confirmation of schedule for company events 4. Progress reports from each team."

[0513] A generative AI model then generates a summary.

[0514] comparative analysis

[0515] The server compares the summaries generated by the generative AI model with the manual summaries created by the user. This comparison checks the following points:

[0516] Inconsistency detection: Detect inconsistencies or conflicting information between the generated AI summary and the user's summary.

[0517] Information deficiency / excess detection: Check whether the user's summary is missing important information contained in the generated AI summary, or whether it contains unnecessary information.

[0518] Feedback Generation

[0519] The server generates specific feedback based on the comparison, such as:

[0520] "Your summary does not include 'progress reports from each team.'"

[0521] Providing Feedback

[0522] Finally, the server sends the generated feedback to the user's terminal, where the user can receive it and review their own summaries to improve their summarization skills.

[0523] Specific examples

[0524] Original data (example)

[0525] 1. Discussion on next year's budget

[0526] 2. New project launch plan

[0527] 3. Check the schedule of company events

[0528] 4. Progress reports from each team

[0529] User Summary (Example)

[0530] We discussed the budget for the next fiscal year and new projects, and also confirmed the dates for company events.

[0531] Processing result (example)

[0532] 1. Summary of Generative AI (Generated by the Server)

[0533] Discussions included next year's budget, plans to launch new projects, confirmation of dates for internal events, and progress reports from each team.

[0534] 2. Feedback (Server-Generated)

[0535] Your summary does not include a "progress report from each team."

[0536] As described above, by receiving specific feedback, users can improve the accuracy of their summaries, thereby enhancing their ability to accurately summarize the main points of materials and discussions, and ultimately improving the quality of their documents and presentations.

[0537] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0538] Step 1:

[0539] Data reception

[0540] The user uses a terminal to upload the original data and a manually created summary document to the interface. The terminal then sends this input data to the server. The server receives the uploaded data and checks its format and content. For example, the data format received by the server is JSON format, and after receiving the input data, it parses it to check for errors. If an error is detected, the server notifies the user with an appropriate error message.

[0541] Step 2:

[0542] Data formatting and preprocessing

[0543] The server formats and preprocesses the received data. This includes standardizing the data format and making it easier to analyze. For example, the server cleans the text data, removes unnecessary whitespace and special characters, and normalizes the document. These operations transform the data into a state suitable for analysis.

[0544] Step 3:

[0545] Generate a summary

[0546] The server passes the formatted original data as input to a generative AI model (e.g., GPT-3) to automatically generate a summary. At this time, the server provides an appropriate prompt to the generative AI model. The following format is used as an example of a prompt:

[0547] "Please summarize the following sentences: 1. Discussion about next year's budget 2. Plan to launch a new project 3. Confirmation of schedule for company events 4. Progress reports from each team."

[0548] The generative AI model receives this prompt and generates a summary based on the original data, which is then received and stored by the server.

[0549] Step 4:

[0550] comparative analysis

[0551] The server compares and analyzes the summaries generated by the AI ​​model with the manual summaries created by the user. This comparison involves text analysis and similarity calculations to detect inconsistencies and information overloads and omissions. The server vectorizes the text of both summaries using, for example, TF-IDF or word embeddings, and calculates the degree of agreement. It also extracts parts where important information is missing or where unnecessary information has been added.

[0552] Step 5:

[0553] Generate feedback

[0554] The server generates specific feedback based on the results of the comparison and analysis. The generated feedback is written in natural language and provided in a format that is easy for users to understand. For example, the following feedback is generated:

[0555] "Your summary does not include 'progress reports from each team.'"

[0556] The server edits the feedback text to include useful suggestions and advice for the user.

[0557] Step 6:

[0558] Providing feedback

[0559] The server sends the generated feedback to the user's device, which receives the feedback and notifies the user. The user can then use the feedback to review their own summaries and create more accurate summaries.

[0560] By clearly indicating the specific operations and the flow of input and output data in each step, the embodiments of the invention can be explained in detail.

[0561] (Application example 1)

[0562] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0563] In traditional advertising production, it is difficult to evaluate the effectiveness of advertising copy and the accuracy of important information, requiring a lot of time and effort. In particular, there is a lack of means to quickly and accurately determine whether the summary created by advertising creators is conveyed to consumers, making it difficult to maximize the effectiveness of advertising.

[0564] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0565] In this invention, the server includes means for comparing the original advertisement data uploaded by the user with the summary and evaluating the accuracy of the information and the effectiveness of the summary, means for generating a summary using a generation AI, and means for comparing the document summarized by the user with the summary generated by the generation AI to detect inconsistencies and excess or deficiency of information. This enables advertisement creators to quickly determine whether their summary matches the generation AI and whether important points are missing, thereby optimizing the message of the advertisement.

[0566] A "document summarized by a user" is a document that has been manually summarized and simplified from original data by a user.

[0567] "Original data" refers to documents or data that contain the information that is the basis for summarizing.

[0568] "Generative AI" is a system that uses artificial intelligence technology to automatically generate summaries from original data.

[0569] A "means for generating a summary" is a mechanism for automatically creating a summary from original data using generative AI.

[0570] The "means for detecting inconsistencies and information omissions" is a mechanism for comparing the generated AI's summary with the user's summary to detect inconsistencies or missing important information.

[0571] The "means for providing feedback" is a mechanism for suggesting specific improvements or corrections to the user based on the comparison results.

[0572] "Advertising source data" refers to the documents and data that are the basis for creating advertisements.

[0573] "Means for assessing the accuracy of information and the effectiveness of summaries" refers to a mechanism for comparing the original advertising data with the summaries to assess whether the information is being conveyed accurately and whether the summaries are effective.

[0574] An embodiment of the present invention is a system and method for evaluating advertising copy created by advertising creators and generating effective advertising summaries. This system is mainly composed of a server and a user terminal, and utilizes a generative AI model to automatically generate summaries of advertising copy and compare them with summaries created by users, thereby providing effective and efficient feedback.

[0575] System configuration:

[0576] server

[0577] The server performs the following main functions:

[0578] Receive the original data (advertising source data) and check the format and content.

[0579] Format and preprocess the data as needed.

[0580] A generative AI model is used to automatically generate summaries from the original data.

[0581] The summary created by the user is compared with the summary generated by the AI ​​to detect inconsistencies and information that is insufficient or excessive.

[0582] Feedback is generated based on the comparison result and transmitted to the user terminal.

[0583] Hardware and software used

[0584] Hardware:

[0585] As a server, use a high-performance computer or cloud computing infrastructure (e.g., AWS, Google Cloud Platform).

[0586] software:

[0587] For data formatting and preprocessing, we use Python's pandas and numpy libraries.

[0588] The generative AI model used is OpenAI's GPT-3.

[0589] The Flask framework is used to build the web interface.

[0590] Natural language processing technologies (e.g., spaCy, NLTK) may also be used for comparative analysis and feedback generation.

[0591] Processing flow

[0592] 1. Data reception and preprocessing:

[0593] Users upload the original advertising data and their own summaries from their smartphones or PCs. The server receives the data, checks the format and content, and formats and preprocesses the data as necessary.

[0594] 2. Generate a summary:

[0595] The server uses a generative AI model (e.g., OpenAI's GPT-3) to automatically generate a summary from the original data. This summary is a shorthand representation of the main points of the original data.

[0596] 3. Comparative analysis:

[0597] The summary created by the user is compared with the summary generated by the server to check for consistency and whether there is any excess or deficiency in the information. This is done using the natural language processing technology of the generation AI.

[0598] 4. Feedback Generation:

[0599] Generate specific feedback based on the comparison, such as "Your summary is missing the important information 'full of vitamin C'."

[0600] 5. Providing Feedback:

[0601] The generated feedback is sent to the user's device, and the user can use it to revise and improve the summary.

[0602] Specific examples

[0603] Ad source data:

[0604] Try our new Super Drink this fall! Packed with Vitamin C, it will help relieve fatigue.

[0605] User Summary:

[0606] Try a new vitamin drink.

[0607] Summary of generated AI (server generated):

[0608] This autumn's new product is a super drink packed with vitamin C.

[0609] Feedback (server generated):

[0610] "Your summary doesn't include key points like 'New this fall.'"

[0611] Prompt Sentence Examples

[0612] For the generative AI model, we input:

[0613] Summarize the following ad copy:

[0614] Try our new Super Drink this fall! Packed with Vitamin C, it will help relieve fatigue.

[0615] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0616] Step 1:

[0617] The user uploads the original ad data and a self-created summary to the server via their device. The device then sends the input data to the server. The input at this time is the "original ad data" and the "user summary," and the output to the server is the original data.

[0618] Step 2:

[0619] The server checks the format and content of the received source data and user summary. For example, the server checks whether the data format is correct and the content is entered properly, and performs data formatting (e.g., grammar correction and text format standardization) and preprocessing as necessary. The input to this process is the data received in step 1, and the output is the formatted and preprocessed data.

[0620] Step 3:

[0621] The server uses a generative AI model (OpenAI GPT-3) to generate a summary from the formatted and preprocessed original ad data. The generative AI receives the original data as a prompt in the form of "Please summarize the ad copy below:" The output of the generative AI is an automatically generated summary based on the original ad data.

[0622] Step 4:

[0623] The server compares the summary generated by the AI ​​with the summary created by the user. This comparison process uses natural language processing technology to check whether the meanings of the two sentences match and whether important information is missing. The inputs to this process are the AI's summary and the user's summary, and the output is information on whether they match or not based on the comparison results.

[0624] Step 5:

[0625] The server generates feedback based on the comparison results. Specifically, if there are any inconsistencies between the AI's summary and the user's summary, or if important information is missing, the server will provide specific feedback. The input for this process is the comparison result from step 4, and the output is a feedback statement to be given to the user.

[0626] Step 6:

[0627] The server sends the generated feedback to the user terminal, which receives it and makes further corrections and improvements. The input of this process is the feedback from step 5, and the output is the feedback sentence displayed on the user terminal.

[0628] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0629] This invention is a system for improving a user's document summarization ability. It utilizes generative AI to compare documents manually summarized by the user with the original data, and points out inconsistencies and excess or deficiency of information. It also combines an emotion engine to provide feedback based on the user's emotions. The main components of this system and their specific operation are described below.

[0630] System configuration

[0631] 1. Terminal

[0632] It provides an interface for users to input and upload the original data they want to summarize and the summary documents they have manually created. The device can be a computer, smartphone, tablet, etc. The device also has an emotion engine to identify the user's emotions.

[0633] 2. Server

[0634] It is the main processing unit for generating summaries, comparative analysis, and generating feedback. The server has the ability to adjust the feedback based on the user's emotions identified by the emotion engine.

[0635] System Operation

[0636] Data reception

[0637] First, the user uploads original data and summary documents using their device. This data can include meeting transcripts, reports, articles, etc. The server receives the data, checks its format and content, and performs data formatting and preprocessing if necessary. The device's on-board emotion engine also identifies the user's current emotion, which is then sent to the server.

[0638] Generate a summary

[0639] The server then uses a generative AI to automatically generate a summary using the original data as input. This generative AI works based on natural language processing technology, extracting key points from the original data and creating a shortened summary.

[0640] comparative analysis

[0641] The server compares the generated AI summary with the user's manual summary. The comparison checks the following points:

[0642] Inconsistency detection: Detect inconsistencies or conflicting information between the generated AI summary and the user's summary.

[0643] Information overload / deficiency detection: Check whether the user's summary is missing important information contained in the generated AI summary, or whether it contains unnecessary information.

[0644] Feedback Generation

[0645] The server generates optimal feedback for the user based on the comparison results and also takes into account information from the emotion engine. For example, if the user is feeling impatient or anxious, the feedback message will be adjusted to be gentle and encouraging.

[0646] Providing Feedback

[0647] Finally, the server sends the generated feedback to the user's device. The user can then review their summary and improve their summarization skills. The server also tracks the user's emotional changes and analyzes long-term patterns to further optimize future feedback.

[0648] Specific examples

[0649] Original data (example)

[0650] 1. Discussion on next year's budget

[0651] 2. New project launch plan

[0652] 3. Check the schedule of company events

[0653] 4. Progress reports from each team

[0654] User Summary (Example)

[0655] We discussed the budget for the next fiscal year and new projects, and also confirmed the dates for company events.

[0656] Processing result (example)

[0657] 1. Summary of Generative AI (Generated by the Server)

[0658] Discussions included next year's budget, plans to launch new projects, confirmation of dates for internal events, and progress reports from each team.

[0659] 2. User's emotions (recognized by the emotion engine)

[0660] "A little impatient"

[0661] 3. Feedback (Server-Generated)

[0662] "Your summary does not include the 'progress report of each team,' but it is a good summary overall. Please take your time and review it again."

[0663] In this way, users can efficiently improve their summarizing skills by receiving feedback that takes into account specific points and emotions, which will improve their ability to accurately summarize the main points of materials and discussions, and ultimately improve the quality of their documents and presentations.

[0664] The processing flow will be explained below.

[0665] Step 1:

[0666] The user uses the device to upload the original data (e.g., a transcript of a meeting) and a manually created summary document. The user uses the device interface to select the file and clicks the send button.

[0667] Step 2:

[0668] The device sends the uploaded original data and summary to the server. At the same time, the device's built-in emotion engine analyzes the user's emotions and sends the results to the server. The emotion engine identifies emotions such as "anxiety" or "calmness" based on the user's facial expressions and input speed.

[0669] Step 3:

[0670] The server receives the original data and summary text sent by the user. At this time, the server checks the consistency of the data format and, if there is any inconsistency, formats and preprocesses the data. Specifically, it removes unnecessary whitespace and standardizes the text.

[0671] Step 4:

[0672] The server passes the original data to the AI ​​generator and requests it to generate a summary. The AI ​​generator extracts the key points of the original data and generates a shortened summary. Natural language processing technology is used in this process.

[0673] Step 5:

[0674] The server compares the AI-generated summary with the summary created manually by the user. The server checks the following points:

[0675] Inconsistency detection: Detects where the generated AI summary and the user's summary do not match.

[0676] Information deficiency / excess detection: Check whether the user's summary is missing important information contained in the generated AI summary or whether it contains unnecessary information.

[0677] Step 6:

[0678] The server generates feedback based on the comparison results, taking into account the emotions expressed by the user when uploading. For example, if the user is feeling "impatient," the server adjusts the feedback message to emphasize a gentle, encouraging tone.

[0679] Step 7:

[0680] The server sends the generated feedback to the user's device, where the user can then revise their summary. Based on the feedback, which takes specific points and emotions into consideration, the user can improve their summarization skills.

[0681] Specific examples

[0682] Let's say the original data looks like this:

[0683] 1. Discussion on next year's budget

[0684] 2. New project launch plan

[0685] 3. Check the schedule of company events

[0686] 4. Progress reports from each team

[0687] Suppose the user's summary was:

[0688] We discussed the budget for the next fiscal year and new projects, and also confirmed the dates for company events.

[0689] The summary generated by the server is:

[0690] Discussions included next year's budget, plans to launch new projects, confirmation of dates for internal events, and progress reports from each team.

[0691] If the user's emotion is recognized as "impatience," the server's feedback will be as follows:

[0692] Your summary does not include the "progress report of each team," but it is a good summary overall. Please take your time and review it again.

[0693] This example allows users to improve their summaries and receive feedback that will help them create their next summaries.

[0694] Example 2

[0695] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0696] Existing document summarization systems have the problem of being unable to accurately identify inconsistencies or information omissions when comparing manually created summaries with automatically generated summaries. Furthermore, feedback is provided without taking into account the user's emotional state, which can reduce the user's motivation and learning effectiveness. Furthermore, the lack of long-term support for user development makes it difficult to achieve continuous improvement.

[0697] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0698] In this invention, the server includes means for uploading a document summarized by a user and its original data, means for generating a summary from the uploaded original data using a generative AI model, means for comparing the document summarized by the user with the summary generated by the generative AI model to detect inconsistencies and excess or deficiency of information, means for providing feedback based on the user's emotions using an emotion engine based on the comparison results, and means for transmitting the feedback to the user's terminal and tracking changes in the user's emotions. This makes it possible to effectively compare the summary created by the user with the automatically generated summary, provide appropriate feedback based on the user's emotional state, and support continuous improvement.

[0699] "User" means an individual or organization that uses the document summarization system and creates and uploads original data and summarized documents.

[0700] "Terminal" means a hardware device through which a user inputs and uploads original data and summary documents, such as a computer, smartphone, or tablet.

[0701] The "server" is the main processing unit responsible for generating summaries, comparative analysis, and feedback generation, and is a computer equipped with the generative AI model and emotion engine.

[0702] "Original data" refers to documents uploaded by users to be summarized, including meeting minutes, articles, reports, and the like.

[0703] A "summary document" is a shortened document that is manually created by a user based on the original data.

[0704] A "generative AI model" is an artificial intelligence algorithm that automatically generates a summary based on original data.

[0705] An "emotion engine" is a computer program that analyzes a user's facial expressions and tone of voice to identify the user's current emotional state.

[0706] "Feedback" refers to comments or suggestions provided to the user's summary document based on the results of the comparative analysis, and is adjusted to take into account the user's emotional state.

[0707] "Data formatting" is the process of checking the format and content of the original data received and converting it into a format that is easy to analyze.

[0708] "Preprocessing" refers to a series of preparatory steps that are performed before analyzing raw data and generating summaries, including text cleaning and preparation for analysis.

[0709] This invention is a system for improving users' document summarization abilities by utilizing a generative AI model to compare documents manually summarized by the user with the original data, and to identify inconsistencies and information omissions. This system also incorporates an emotion engine to provide feedback based on the user's emotions.

[0710] The main components of the system of the present invention are a terminal, a server, a generative AI model, and an emotion engine. First, a user uploads original data and a manually created summary document using a terminal. This terminal can be a computer, smartphone, tablet, or other device. The terminal also has an emotion engine that identifies the user's emotions.

[0711] The original data and manual summaries are sent from the device to the server, which receives them, formats and preprocesses them as needed, and then summarizes the original data using a generative AI model, which works based on natural language processing techniques.

[0712] The server compares the generated AI summary with the user's manual summary, checking for inconsistencies and information oversaturation. Based on the results of this comparison, the server generates optimal feedback, taking into account the user's emotional information received from the device. Finally, the server sends the generated feedback to the user's device, allowing the user to review their own summary based on it and improve their summarization ability.

[0713] Examples of prompt statements

[0714] Specific examples are shown below.

[0715] Original data (example)

[0716] 1. Discussion on next year's budget

[0717] 2. New project launch plan

[0718] 3. Check the schedule of company events

[0719] 4. Progress reports from each team

[0720] User Summary (Example)

[0721] We discussed the budget for the next fiscal year and new projects, and also confirmed the dates for company events.

[0722] Processing result (example)

[0723] 1. Summary of Generative AI (Generated by the Server)

[0724] Discussions included next year's budget, plans to launch new projects, confirmation of dates for internal events, and progress reports from each team.

[0725] 2. User's emotions (recognized by the emotion engine)

[0726] "A little impatient"

[0727] 3. Feedback (Server-Generated)

[0728] "Your summary does not include the 'progress report of each team,' but it is a good summary overall. Please take your time and review it again."

[0729] In this way, users can efficiently improve their summarizing skills by receiving feedback that takes into account specific points and emotions, thereby improving their ability to accurately summarize the main points of materials and discussions, and ultimately improving the quality of their documents and presentations.

[0730] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0731] Step 1:

[0732] A user uses a terminal to upload the original data to be summarized and a manually created summary document.

[0733] Input: Original data file, manual summary document file

[0734] Output: Data transfer request to the server

[0735] Specific operation: The user opens the device's web browser or app, clicks the upload button, selects the original data file and the manual summary document file, and uploads them. The device's built-in emotion engine analyzes the user's facial expressions and tone of voice to identify the user's emotional state.

[0736] Step 2:

[0737] The terminal transmits the received data and the identified emotion information to the server.

[0738] Input: Original data file, manual summary document file, user emotion information

[0739] Output: Server data reception process begins

[0740] Specific operation: The terminal uploads the selected original data file and the manually summarized document file to the server, and simultaneously transmits the user's emotion information.

[0741] Step 3:

[0742] The server analyzes the received original data and manual summary documents, and formats and preprocesses the data.

[0743] Input: Received original data file, manual summary document file

[0744] Output: Formatted and preprocessed data

[0745] Specific operation: The server checks the received raw data files and summary documents based on their format and content, and performs formatting and preprocessing as necessary, including text cleaning and formatting standardization.

[0746] Step 4:

[0747] The server inputs the formatted and preprocessed raw data into a generative AI model to generate an automatic summary.

[0748] Input: Raw data that has been reformatted and preprocessed

[0749] Output: Automatic summary generated by the generative AI model

[0750] How it works: The server inputs the original data into a generative AI model, extracts key points, and automatically generates a summary. This generative AI model uses natural language processing technology to identify important information and output it as a summary.

[0751] Step 5:

[0752] The server compares the generated AI summary with the user's manually summarized document to detect inconsistencies and information omissions.

[0753] Input: Automated summaries by generative AI models, manually summarized documents

[0754] Output: Comparison results (status identifying inconsistencies and information oversaturations)

[0755] How it works: The server displays the AI-generated summary and the user's manual summary side by side and uses a specific algorithm to highlight differences, for example, showing inconsistencies in red and missing information in blue.

[0756] Step 6:

[0757] The server generates feedback based on the comparison results and adjusts it taking into account the user's emotional information provided by the emotion engine.

[0758] Input: Comparison results, user emotion information

[0759] Output: Adjusted feedback message

[0760] Specific behavior: The server generates specific feedback messages based on the differences highlighted, and adjusts the feedback to be gentle and encouraging if the user seems impatient.

[0761] Step 7:

[0762] The server transmits the generated feedback to the user's terminal, and the user receives the feedback.

[0763] Input: Adjusted feedback message

[0764] Output: A feedback message that is displayed on the user's terminal.

[0765] Specific operation: The server sends the generated feedback message to the user's device, and the user checks the feedback in notifications or on the dashboard. The user receives the feedback and reevaluates their own manual summaries to learn from them.

[0766] The above are the specific processing steps of the system. In this way, users can receive feedback based on specific suggestions and emotions, and can efficiently improve their summarization ability.

[0767] (Application example 2)

[0768] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0769] This invention relates to a system for improving the accuracy and efficiency of summarization for users who perform document summarization. In particular, the system aims to improve summarization ability by detecting inconsistencies in summary content and excess or deficiency of information through comparative analysis of automatic summaries generated by generative AI and manual summaries by users, and by providing optimal feedback to the user. Another objective of this invention is to provide a system that reduces the user's mental burden and encourages continuous learning by providing feedback based on the user's emotions in an emotional environment.

[0770] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for uploading a document summarized by a user and original data, means for generating a summary from the uploaded original data using a generation AI, means for comparing the document summarized by the user with the summary generated by the generation AI to detect inconsistencies and excess or deficiency of information, means for providing feedback to the user based on the comparison results, and means including an emotion engine for identifying the user's emotions and adjusting the feedback content. This makes it possible to improve the user's summarizing ability and provide feedback according to emotions.

[0771] "User-summarized documents" refers to information that has been manually summarized by a user based on original data.

[0772] "Original data" refers to a series of information or text data that is the basis for the summary.

[0773] "Generative AI" refers to a system that uses artificial intelligence technology to automatically summarize original data.

[0774] "Inconsistency" refers to inconsistencies in information or content between the generated AI summary and the user's summary.

[0775] "Information overload" refers to a situation in which important information is missing from a summary or unnecessary information is included.

[0776] "Feedback" refers to advice, including instructions for improvement or correction, provided to the user based on the results of comparing the generated AI with the user's summary.

[0777] An "emotion engine" refers to a device or software that identifies a user's emotions and generates feedback appropriate to those emotions.

[0778] "Prompt sentence" refers to the input sentence that the generative AI uses when summarizing.

[0779] The present invention is a system for improving users' summarization skills, and a specific embodiment is described below. This system uses a generative AI model and an emotion engine to perform a comparative analysis of a manually created summary document and the original data, and provides feedback.

[0780] System Configuration

[0781] 1. Terminal

[0782] The device provides an interface for users to input and upload the original data they want to summarize and the summary documents they have manually created. The device is equipped with an emotion engine to identify the user's emotions. This includes smartphones, tablets, and computers.

[0783] 2. Server

[0784] The server is the main processing unit for generating summaries, comparative analysis, and feedback generation. It includes the following means:

[0785] Summary generation method using generative AI

[0786] A comparative method for user-generated summaries and AI-generated summaries

[0787] A means of detecting inconsistencies and information overloads and deficiencies

[0788] Feedback adjustment method by emotion engine

[0789] System Operation

[0790] Data reception

[0791] The user uploads original data and manually summarized documents using a terminal, such as meeting transcripts, reports, articles, etc. The terminal uses an emotion engine to identify the user's emotions and transmits them to the server.

[0792] Generate a summary

[0793] The server uses a generative AI model (e.g., OpenAI GPT-4) to automatically generate a summary based on the original data. This generative AI works based on natural language processing technology, extracting the main points of the original data and creating a short summary.

[0794] comparative analysis

[0795] The server compares the summaries generated by the generative AI model with the manual summaries created by the user. The following points are checked:

[0796] Inconsistency detection: Detects inconsistencies and conflicting information between the generated AI summary and the user's summary.

[0797] Information deficiency / excess detection: Check whether the user's summary is missing important information contained in the generated AI summary or whether it contains unnecessary information.

[0798] Feedback Generation

[0799] Based on the comparison results and taking into account information from the emotion engine, the server generates optimal feedback for the user. For example, if the user is feeling anxious or impatient, the server will adjust the feedback message to be more gentle and encouraging.

[0800] Providing Feedback

[0801] The server generates feedback and sends it to the user's device. The user receives the feedback and reviews their own summarization to improve their summarization skills. The feedback is also optimized based on long-term emotional data.

[0802] Hardware and software used

[0803] Devices: smartphones (iOS / Android compatible), tablets, computers

[0804] Emotion engine API: Azure Emotion API

[0805] Generative AI model: OpenAI GPT-4

[0806] Server: AWS EC2, Node.js, Express

[0807] Prompt Sentence Examples

[0808] Original data: This product uses the latest technology, is highly durable, and remains comfortable even after extended use.

[0809] Generate a summary:

[0810] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0811] Step 1: Enter your data

[0812] The user inputs the original data and the manually summarized text into the application using a device (smartphone, tablet, or computer). The device then uses an emotion engine to identify the user's emotions. The input data includes the original data text, the manually summarized text, and emotion data.

[0813] Step 2: Sending data

[0814] The terminal transmits the input original data, the manual summary document, and the identified emotion data to the server. The transmitted data includes the original data, the manual summary document, and the emotion data.

[0815] Step 3: Data preparation and preprocessing

[0816] The server checks the format and content of the received original data and manual summary document, and formats and preprocesses the data as necessary. Specifically, it deletes unnecessary spaces and symbols and standardizes the text format. This process results in the formatted original data and manual summary document.

[0817] Step 4: Generate a summary

[0818] The server takes the formatted original data as input and generates an automatic summary using a generative AI model (OpenAI GPT-4). The prompt message generated is "Original data: [formatted original data] Please generate a summary:", which is then input into the AI ​​model to output the automatic summary.

[0819] Step 5: Comparative analysis of summaries

[0820] The server compares the automatic summary generated by the generative AI model with the manually summarized document. This comparison uses natural language processing technology to check for semantic agreement and context, and detects inconsistencies and information deficiencies. Specifically, it calculates word agreement rates and differences in contextual meaning, and outputs any inconsistencies or information deficiencies.

[0821] Step 6: Generate feedback

[0822] The server uses a generative AI model to generate optimal feedback for the user based on the results of the comparative analysis and the identified emotional data. The feedback message is generated using the results of the comparative analysis and the emotional data as input, and is output in a format that is easy for the user to understand.

[0823] Step 7: Provide feedback

[0824] The server sends the generated feedback to the user's device, where the user can receive the feedback and use it to improve the summary. The feedback content is adjusted based on the results of the comparative analysis and the user's sentiment.

[0825] The above is the specific processing flow of the system that realizes the application example.

[0826] 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.

[0827] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0828] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0829] [Third embodiment]

[0830] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0831] 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.

[0832] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0833] 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.

[0834] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0835] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0836] 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.

[0837] 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.

[0838] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0839] 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.

[0840] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0841] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0842] This invention is a system for improving a user's document summarization ability. It uses a generative AI to compare documents manually summarized by the user with the original data, and identifies inconsistencies and information omissions. The main components of this system and their specific operation are described below.

[0843] System configuration

[0844] 1. Terminal

[0845] It provides an interface for users to input and upload the original data they want to summarize and the summary documents they have manually created. The devices can be computers, smartphones, tablets, etc.

[0846] 2. Server

[0847] It is the main processing unit responsible for generating summaries, comparative analysis, and generating feedback.

[0848] System Operation

[0849] Data reception

[0850] First, users upload original data and summary documents using their devices. This data can include meeting transcripts, reports, articles, etc. The server receives the data, checks its format and content, and performs formatting and preprocessing of the data if necessary.

[0851] Generate a summary

[0852] The server then uses a generative AI to automatically generate a summary using the original data as input. This generative AI works based on natural language processing technology, extracting key points from the original data and creating a shortened summary.

[0853] comparative analysis

[0854] The server compares the generated AI summary with the user's manual summary. This comparison checks the following points:

[0855] Inconsistency detection: Detect inconsistencies or conflicting information between the generated AI summary and the user's summary.

[0856] Information deficiency / excess detection: Check whether the user's summary is missing important information contained in the generated AI summary, or whether it contains unnecessary information.

[0857] Feedback Generation

[0858] The server generates specific feedback based on the comparison results, such as, "This summary does not include the 'progress report of each team,' which is a key point in the original data."

[0859] Providing Feedback

[0860] Finally, the server sends the generated feedback to the user's terminal, where the user can receive it and review their own summaries to improve their summarization skills.

[0861] Specific examples

[0862] Original data (example)

[0863] 1. Discussion on next year's budget

[0864] 2. New project launch plan

[0865] 3. Check the schedule of company events

[0866] 4. Progress reports from each team

[0867] User Summary (Example)

[0868] We discussed the budget for the next fiscal year and new projects, and also confirmed the dates for company events.

[0869] Processing result (example)

[0870] 1. Summary of Generative AI (Generated by the Server)

[0871] Discussions included next year's budget, plans to launch new projects, confirmation of dates for internal events, and progress reports from each team.

[0872] 2. Feedback (Server-Generated)

[0873] "Your summary does not include 'progress reports from each team.'"

[0874] In this way, users can improve the accuracy of their summaries by receiving specific feedback, which will enhance their ability to accurately summarize the main points of materials and discussions, ultimately improving the quality of their documents and presentations.

[0875] The processing flow will be explained below.

[0876] Step 1:

[0877] The user uploads the original data (e.g., transcription data of a meeting) and a summary document created manually via the terminal. The terminal then sends this data to the server.

[0878] Step 2:

[0879] The server receives the original data and the summary sent by the user. At this time, the server checks the consistency of the data format and performs preprocessing if necessary. Preprocessing includes text cleaning and formatting.

[0880] Step 3:

[0881] The server calls the generation AI, passing the original data as input. The generation AI extracts the key points of the original data and generates a short summary, which is later used to compare with the user summary.

[0882] Step 4:

[0883] The server compares the AI-generated summary with the one the user manually created. The comparison checks the following points:

[0884] Inconsistency detection: Detects inconsistencies between the generated AI summary and the user's summary.

[0885] Information deficiency / excess detection: Check whether the user's summary is missing important information contained in the generated AI summary or whether it contains unnecessary information.

[0886] Step 5:

[0887] The server generates feedback based on the comparison, including specific suggestions and areas for improvement, such as identifying important or unnecessary information missing from the user's summary.

[0888] Step 6:

[0889] The server sends the generated feedback to the user's device, allowing the user to review their summary and make corrections if necessary.

[0890] Example 1

[0891] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0892] Conventional document summarization systems have difficulty effectively detecting inconsistencies or excess or deficiencies of information between a user's manually created summary and the original data, and providing appropriate feedback. Furthermore, if data formatting and preprocessing are insufficient, errors are likely to occur in summary generation and comparative analysis. Furthermore, the feedback is not specific enough, so it is not possible to adequately support users in improving their summarization skills.

[0893] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0894] In this invention, the server includes means for uploading a document summarized by a user and the original data, means for checking the format and content of the uploaded original data and reformatting or preprocessing the data as necessary, means for generating a summary from the uploaded original data using a generative AI model, means for generating a summary by providing an appropriate prompt to the generative AI model, means for comparing the document summarized by the user with the summary generated by the generative AI model to detect inconsistencies or excess or deficiencies of information, and means for providing feedback to the user based on the comparison results. This makes it possible to effectively detect inconsistencies or excess or deficiencies of information between the summary created by the user and the original data and provide specific feedback.

[0895] "User" refers to a person who uses the system to create a summary of a document.

[0896] "Original data" refers to the collection of original text or information that a user uses to summarize.

[0897] A "terminal" is a device to which a user uploads original data and manually created summary documents, and includes a computer, a smartphone, a tablet, and the like.

[0898] "Server" refers to a device that performs major processing such as generating summaries, comparative analysis, and generating feedback.

[0899] A "generative AI model" refers to artificial intelligence that automatically generates summaries from original data based on natural language processing technology.

[0900] A "prompt" refers to an input instruction that instructs a generative AI model to perform a specific task.

[0901] "Feedback" refers to the indications and advice that the server provides to the user based on the results of the comparison and analysis.

[0902] "Data formatting" refers to standardizing the format of received data and making it suitable for analysis.

[0903] "Preprocessing" refers to the early stage of data processing carried out to improve the accuracy of data analysis and summary generation.

[0904] This invention is a system for improving a user's document summarization ability. It utilizes a generative AI model to compare documents manually summarized by the user with the original data, and identifies inconsistencies and information omissions. The main components of this system and their specific operation are described below.

[0905] System configuration

[0906] Terminal

[0907] It provides an interface for users to input and upload the original data they want to summarize and the summary documents they have manually created. The devices can be computers, smartphones, tablets, etc.

[0908] server

[0909] This is the main processing unit that generates summaries, performs comparative analysis, and generates feedback. A generative AI model (e.g., GPT-3) is installed on the server and used to automatically generate summaries.

[0910] System Operation

[0911] Data reception

[0912] First, the user uploads the original data and summary documents to the interface using their terminal. This data includes, for example, minutes of meetings, reports, and articles. The server receives the data and checks its format and content. If any inappropriate content or format is found, the server reformats or preprocesses the data.

[0913] Generate a summary

[0914] The server passes the received raw data as input to the generative AI model, which then automatically generates a summary. In this process, the server provides the generative AI with an appropriate prompt. For example, the following prompt can be used:

[0915] "Please summarize the following sentences: 1. Discussion about next year's budget 2. Plan to launch a new project 3. Confirmation of schedule for company events 4. Progress reports from each team."

[0916] A generative AI model then generates a summary.

[0917] comparative analysis

[0918] The server compares the summaries generated by the generative AI model with the manual summaries created by the user. This comparison checks the following points:

[0919] Inconsistency detection: Detect inconsistencies or conflicting information between the generated AI summary and the user's summary.

[0920] Information deficiency / excess detection: Check whether the user's summary is missing important information contained in the generated AI summary, or whether it contains unnecessary information.

[0921] Feedback Generation

[0922] The server generates specific feedback based on the comparison, such as:

[0923] "Your summary does not include 'progress reports from each team.'"

[0924] Providing Feedback

[0925] Finally, the server sends the generated feedback to the user's terminal, where the user can receive it and review their own summaries to improve their summarization skills.

[0926] Specific examples

[0927] Original data (example)

[0928] 1. Discussion on next year's budget

[0929] 2. New project launch plan

[0930] 3. Check the schedule of company events

[0931] 4. Progress reports from each team

[0932] User Summary (Example)

[0933] We discussed the budget for the next fiscal year and new projects, and also confirmed the dates for company events.

[0934] Processing result (example)

[0935] 1. Summary of Generative AI (Generated by the Server)

[0936] Discussions included next year's budget, plans to launch new projects, confirmation of dates for internal events, and progress reports from each team.

[0937] 2. Feedback (Server-Generated)

[0938] Your summary does not include a "progress report from each team."

[0939] As described above, by receiving specific feedback, users can improve the accuracy of their summaries, thereby enhancing their ability to accurately summarize the main points of materials and discussions, and ultimately improving the quality of their documents and presentations.

[0940] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0941] Step 1:

[0942] Data reception

[0943] The user uses a terminal to upload the original data and a manually created summary document to the interface. The terminal then sends this input data to the server. The server receives the uploaded data and checks its format and content. For example, the data format received by the server is JSON format, and after receiving the input data, it parses it to check for errors. If an error is detected, the server notifies the user with an appropriate error message.

[0944] Step 2:

[0945] Data formatting and preprocessing

[0946] The server formats and preprocesses the received data. This includes standardizing the data format and making it easier to analyze. For example, the server cleans the text data, removes unnecessary whitespace and special characters, and normalizes the document. These operations transform the data into a state suitable for analysis.

[0947] Step 3:

[0948] Generate a summary

[0949] The server passes the formatted original data as input to a generative AI model (e.g., GPT-3) to automatically generate a summary. At this time, the server provides an appropriate prompt to the generative AI model. The following format is used as an example of a prompt:

[0950] "Please summarize the following sentences: 1. Discussion about next year's budget 2. Plan to launch a new project 3. Confirmation of schedule for company events 4. Progress reports from each team."

[0951] The generative AI model receives this prompt and generates a summary based on the original data, which is then received and stored by the server.

[0952] Step 4:

[0953] comparative analysis

[0954] The server compares and analyzes the summaries generated by the AI ​​model with the manual summaries created by the user. This comparison involves text analysis and similarity calculations to detect inconsistencies and information overloads and omissions. The server vectorizes the text of both summaries using, for example, TF-IDF or word embeddings, and calculates the degree of agreement. It also extracts parts where important information is missing or where unnecessary information has been added.

[0955] Step 5:

[0956] Generate feedback

[0957] The server generates specific feedback based on the results of the comparison and analysis. The generated feedback is written in natural language and provided in a format that is easy for users to understand. For example, the following feedback is generated:

[0958] "Your summary does not include 'progress reports from each team.'"

[0959] The server edits the feedback text to include useful suggestions and advice for the user.

[0960] Step 6:

[0961] Providing feedback

[0962] The server sends the generated feedback to the user's device, which receives the feedback and notifies the user. The user can then use the feedback to review their own summaries and create more accurate summaries.

[0963] By clearly indicating the specific operations and the flow of input and output data in each step, the embodiments of the invention can be explained in detail.

[0964] (Application example 1)

[0965] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0966] In traditional advertising production, it is difficult to evaluate the effectiveness of advertising copy and the accuracy of important information, requiring a lot of time and effort. In particular, there is a lack of means to quickly and accurately determine whether the summary created by advertising creators is conveyed to consumers, making it difficult to maximize the effectiveness of advertising.

[0967] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0968] In this invention, the server includes means for comparing the original advertisement data uploaded by the user with the summary and evaluating the accuracy of the information and the effectiveness of the summary, means for generating a summary using a generation AI, and means for comparing the document summarized by the user with the summary generated by the generation AI to detect inconsistencies and excess or deficiency of information. This enables advertisement creators to quickly determine whether their summary matches the generation AI and whether important points are missing, thereby optimizing the message of the advertisement.

[0969] A "document summarized by a user" is a document that has been manually summarized and simplified from original data by a user.

[0970] "Original data" refers to documents or data that contain the information that is the basis for summarizing.

[0971] "Generative AI" is a system that uses artificial intelligence technology to automatically generate summaries from original data.

[0972] A "means for generating a summary" is a mechanism for automatically creating a summary from original data using generative AI.

[0973] The "means for detecting inconsistencies and information omissions" is a mechanism for comparing the generated AI's summary with the user's summary to detect inconsistencies or missing important information.

[0974] The "means for providing feedback" is a mechanism for suggesting specific improvements or corrections to the user based on the comparison results.

[0975] "Advertising source data" refers to the documents and data that are the basis for creating advertisements.

[0976] "Means for assessing the accuracy of information and the effectiveness of summaries" refers to a mechanism for comparing the original advertising data with the summaries to assess whether the information is being conveyed accurately and whether the summaries are effective.

[0977] An embodiment of the present invention is a system and method for evaluating advertising copy created by advertising creators and generating effective advertising summaries. This system is mainly composed of a server and a user terminal, and utilizes a generative AI model to automatically generate summaries of advertising copy and compare them with summaries created by users, thereby providing effective and efficient feedback.

[0978] System configuration:

[0979] server

[0980] The server performs the following main functions:

[0981] Receive the original data (advertising source data) and check the format and content.

[0982] Format and preprocess the data as needed.

[0983] A generative AI model is used to automatically generate summaries from the original data.

[0984] The summary created by the user is compared with the summary generated by the AI ​​to detect inconsistencies and information that is insufficient or excessive.

[0985] Feedback is generated based on the comparison result and transmitted to the user terminal.

[0986] Hardware and software used

[0987] Hardware:

[0988] As a server, use a high-performance computer or cloud computing infrastructure (e.g., AWS, Google Cloud Platform).

[0989] software:

[0990] For data formatting and preprocessing, we use Python's pandas and numpy libraries.

[0991] The generative AI model used is OpenAI's GPT-3.

[0992] The Flask framework is used to build the web interface.

[0993] Natural language processing technologies (e.g., spaCy, NLTK) may also be used for comparative analysis and feedback generation.

[0994] Processing flow

[0995] 1. Data reception and preprocessing:

[0996] Users upload the original advertising data and their own summaries from their smartphones or PCs. The server receives the data, checks the format and content, and formats and preprocesses the data as necessary.

[0997] 2. Generate a summary:

[0998] The server uses a generative AI model (e.g., OpenAI's GPT-3) to automatically generate a summary from the original data. This summary is a shorthand representation of the main points of the original data.

[0999] 3. Comparative analysis:

[1000] The summary created by the user is compared with the summary generated by the server to check for consistency and whether there is any excess or deficiency in the information. This is done using the natural language processing technology of the generation AI.

[1001] 4. Feedback Generation:

[1002] Generate specific feedback based on the comparison, such as "Your summary is missing the important information 'full of vitamin C'."

[1003] 5. Providing Feedback:

[1004] The generated feedback is sent to the user's device, and the user can use it to revise and improve the summary.

[1005] Specific examples

[1006] Ad source data:

[1007] Try our new Super Drink this fall! Packed with Vitamin C, it will help relieve fatigue.

[1008] User Summary:

[1009] Try a new vitamin drink.

[1010] Summary of generated AI (server generated):

[1011] This autumn's new product is a super drink packed with vitamin C.

[1012] Feedback (server generated):

[1013] "Your summary doesn't include key points like 'New this fall.'"

[1014] Prompt Sentence Examples

[1015] For the generative AI model, we input:

[1016] Summarize the following ad copy:

[1017] Try our new Super Drink this fall! Packed with Vitamin C, it will help relieve fatigue.

[1018] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1019] Step 1:

[1020] The user uploads the original ad data and a self-created summary to the server via their device. The device then sends the input data to the server. The input at this time is the "original ad data" and the "user summary," and the output to the server is the original data.

[1021] Step 2:

[1022] The server checks the format and content of the received source data and user summary. For example, the server checks whether the data format is correct and the content is entered properly, and performs data formatting (e.g., grammar correction and text format standardization) and preprocessing as necessary. The input to this process is the data received in step 1, and the output is the formatted and preprocessed data.

[1023] Step 3:

[1024] The server uses a generative AI model (OpenAI GPT-3) to generate a summary from the formatted and preprocessed original ad data. The generative AI receives the original data as a prompt in the form of "Please summarize the ad copy below:" The output of the generative AI is an automatically generated summary based on the original ad data.

[1025] Step 4:

[1026] The server compares the summary generated by the AI ​​with the summary created by the user. This comparison process uses natural language processing technology to check whether the meanings of the two sentences match and whether important information is missing. The inputs to this process are the AI's summary and the user's summary, and the output is information on whether they match or not based on the comparison results.

[1027] Step 5:

[1028] The server generates feedback based on the comparison results. Specifically, if there are any inconsistencies between the AI's summary and the user's summary, or if important information is missing, the server will provide specific feedback. The input for this process is the comparison result from step 4, and the output is a feedback statement to be given to the user.

[1029] Step 6:

[1030] The server sends the generated feedback to the user terminal, which receives it and makes further corrections and improvements. The input of this process is the feedback from step 5, and the output is the feedback sentence displayed on the user terminal.

[1031] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1032] This invention is a system for improving a user's document summarization ability. It utilizes generative AI to compare documents manually summarized by the user with the original data, and points out inconsistencies and excess or deficiency of information. It also combines an emotion engine to provide feedback based on the user's emotions. The main components of this system and their specific operation are described below.

[1033] System configuration

[1034] 1. Terminal

[1035] It provides an interface for users to input and upload the original data they want to summarize and the summary documents they have manually created. The device can be a computer, smartphone, tablet, etc. The device also has an emotion engine to identify the user's emotions.

[1036] 2. Server

[1037] It is the main processing unit for generating summaries, comparative analysis, and generating feedback. The server has the ability to adjust the feedback based on the user's emotions identified by the emotion engine.

[1038] System Operation

[1039] Data reception

[1040] First, the user uploads original data and summary documents using their device. This data can include meeting transcripts, reports, articles, etc. The server receives the data, checks its format and content, and performs data formatting and preprocessing if necessary. The device's on-board emotion engine also identifies the user's current emotion, which is then sent to the server.

[1041] Generate a summary

[1042] The server then uses a generative AI to automatically generate a summary using the original data as input. This generative AI works based on natural language processing technology, extracting key points from the original data and creating a shortened summary.

[1043] comparative analysis

[1044] The server compares the generated AI summary with the user's manual summary. The comparison checks the following points:

[1045] Inconsistency detection: Detect inconsistencies or conflicting information between the generated AI summary and the user's summary.

[1046] Information overload / deficiency detection: Check whether the user's summary is missing important information contained in the generated AI summary, or whether it contains unnecessary information.

[1047] Feedback Generation

[1048] The server generates optimal feedback for the user based on the comparison results and also takes into account information from the emotion engine. For example, if the user is feeling impatient or anxious, the feedback message will be adjusted to be gentle and encouraging.

[1049] Providing Feedback

[1050] Finally, the server sends the generated feedback to the user's device. The user can then review their summary and improve their summarization skills. The server also tracks the user's emotional changes and analyzes long-term patterns to further optimize future feedback.

[1051] Specific examples

[1052] Original data (example)

[1053] 1. Discussion on next year's budget

[1054] 2. New project launch plan

[1055] 3. Check the schedule of company events

[1056] 4. Progress reports from each team

[1057] User Summary (Example)

[1058] We discussed the budget for the next fiscal year and new projects, and also confirmed the dates for company events.

[1059] Processing result (example)

[1060] 1. Summary of Generative AI (Generated by the Server)

[1061] Discussions included next year's budget, plans to launch new projects, confirmation of dates for internal events, and progress reports from each team.

[1062] 2. User's emotions (recognized by the emotion engine)

[1063] "A little impatient"

[1064] 3. Feedback (Server-Generated)

[1065] "Your summary does not include the 'progress report of each team,' but it is a good summary overall. Please take your time and review it again."

[1066] In this way, users can efficiently improve their summarizing skills by receiving feedback that takes into account specific points and emotions, which will improve their ability to accurately summarize the main points of materials and discussions, and ultimately improve the quality of their documents and presentations.

[1067] The processing flow will be explained below.

[1068] Step 1:

[1069] The user uses the device to upload the original data (e.g., a transcript of a meeting) and a manually created summary document. The user uses the device interface to select the file and clicks the send button.

[1070] Step 2:

[1071] The device sends the uploaded original data and summary to the server. At the same time, the device's built-in emotion engine analyzes the user's emotions and sends the results to the server. The emotion engine identifies emotions such as "anxiety" or "calmness" based on the user's facial expressions and input speed.

[1072] Step 3:

[1073] The server receives the original data and summary text sent by the user. At this time, the server checks the consistency of the data format and, if there is any inconsistency, formats and preprocesses the data. Specifically, it removes unnecessary whitespace and standardizes the text.

[1074] Step 4:

[1075] The server passes the original data to the AI ​​generator and requests it to generate a summary. The AI ​​generator extracts the key points of the original data and generates a shortened summary. Natural language processing technology is used in this process.

[1076] Step 5:

[1077] The server compares the AI-generated summary with the summary created manually by the user. The server checks the following points:

[1078] Inconsistency detection: Detects where the generated AI summary and the user's summary do not match.

[1079] Information deficiency / excess detection: Check whether the user's summary is missing important information contained in the generated AI summary or whether it contains unnecessary information.

[1080] Step 6:

[1081] The server generates feedback based on the comparison results, taking into account the emotions expressed by the user when uploading. For example, if the user is feeling "impatient," the server adjusts the feedback message to emphasize a gentle, encouraging tone.

[1082] Step 7:

[1083] The server sends the generated feedback to the user's device, where the user can then revise their summary. Based on the feedback, which takes specific points and emotions into consideration, the user can improve their summarization skills.

[1084] Specific examples

[1085] Let's say the original data looks like this:

[1086] 1. Discussion on next year's budget

[1087] 2. New project launch plan

[1088] 3. Check the schedule of company events

[1089] 4. Progress reports from each team

[1090] Suppose the user's summary was:

[1091] We discussed the budget for the next fiscal year and new projects, and also confirmed the dates for company events.

[1092] The summary generated by the server is:

[1093] Discussions included next year's budget, plans to launch new projects, confirmation of dates for internal events, and progress reports from each team.

[1094] If the user's emotion is recognized as "impatience," the server's feedback will be as follows:

[1095] Your summary does not include the "progress report of each team," but it is a good summary overall. Please take your time and review it again.

[1096] This example allows users to improve their summaries and receive feedback that will help them create their next summaries.

[1097] Example 2

[1098] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1099] Existing document summarization systems have the problem of being unable to accurately identify inconsistencies or information omissions when comparing manually created summaries with automatically generated summaries. Furthermore, feedback is provided without taking into account the user's emotional state, which can reduce the user's motivation and learning effectiveness. Furthermore, the lack of long-term support for user development makes it difficult to achieve continuous improvement.

[1100] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1101] In this invention, the server includes means for uploading a document summarized by a user and its original data, means for generating a summary from the uploaded original data using a generative AI model, means for comparing the document summarized by the user with the summary generated by the generative AI model to detect inconsistencies and excess or deficiency of information, means for providing feedback based on the user's emotions using an emotion engine based on the comparison results, and means for transmitting the feedback to the user's terminal and tracking changes in the user's emotions. This makes it possible to effectively compare the summary created by the user with the automatically generated summary, provide appropriate feedback based on the user's emotional state, and support continuous improvement.

[1102] "User" means an individual or organization that uses the document summarization system and creates and uploads original data and summarized documents.

[1103] "Terminal" means a hardware device through which a user inputs and uploads original data and summary documents, such as a computer, smartphone, or tablet.

[1104] The "server" is the main processing unit responsible for generating summaries, comparative analysis, and feedback generation, and is a computer equipped with the generative AI model and emotion engine.

[1105] "Original data" refers to documents uploaded by users to be summarized, including meeting minutes, articles, reports, and the like.

[1106] A "summary document" is a shortened document that is manually created by a user based on the original data.

[1107] A "generative AI model" is an artificial intelligence algorithm that automatically generates a summary based on original data.

[1108] An "emotion engine" is a computer program that analyzes a user's facial expressions and tone of voice to identify the user's current emotional state.

[1109] "Feedback" refers to comments or suggestions provided to the user's summary document based on the results of the comparative analysis, and is adjusted to take into account the user's emotional state.

[1110] "Data formatting" is the process of checking the format and content of the original data received and converting it into a format that is easy to analyze.

[1111] "Preprocessing" refers to a series of preparatory steps that are performed before analyzing raw data and generating summaries, including text cleaning and preparation for analysis.

[1112] This invention is a system for improving users' document summarization abilities by utilizing a generative AI model to compare documents manually summarized by the user with the original data, and to identify inconsistencies and information omissions. This system also incorporates an emotion engine to provide feedback based on the user's emotions.

[1113] The main components of the system of the present invention are a terminal, a server, a generative AI model, and an emotion engine. First, a user uploads original data and a manually created summary document using a terminal. This terminal can be a computer, smartphone, tablet, or other device. The terminal also has an emotion engine that identifies the user's emotions.

[1114] The original data and manual summaries are sent from the device to the server, which receives them, formats and preprocesses them as needed, and then summarizes the original data using a generative AI model, which works based on natural language processing techniques.

[1115] The server compares the generated AI summary with the user's manual summary, checking for inconsistencies and information oversaturation. Based on the results of this comparison, the server generates optimal feedback, taking into account the user's emotional information received from the device. Finally, the server sends the generated feedback to the user's device, allowing the user to review their own summary based on it and improve their summarization ability.

[1116] Examples of prompt statements

[1117] Specific examples are shown below.

[1118] Original data (example)

[1119] 1. Discussion on next year's budget

[1120] 2. New project launch plan

[1121] 3. Check the schedule of company events

[1122] 4. Progress reports from each team

[1123] User Summary (Example)

[1124] We discussed the budget for the next fiscal year and new projects, and also confirmed the dates for company events.

[1125] Processing result (example)

[1126] 1. Summary of Generative AI (Generated by the Server)

[1127] Discussions included next year's budget, plans to launch new projects, confirmation of dates for internal events, and progress reports from each team.

[1128] 2. User's emotions (recognized by the emotion engine)

[1129] "A little impatient"

[1130] 3. Feedback (Server-Generated)

[1131] "Your summary does not include the 'progress report of each team,' but it is a good summary overall. Please take your time and review it again."

[1132] In this way, users can efficiently improve their summarizing skills by receiving feedback that takes into account specific points and emotions, thereby improving their ability to accurately summarize the main points of materials and discussions, and ultimately improving the quality of their documents and presentations.

[1133] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1134] Step 1:

[1135] A user uses a terminal to upload the original data to be summarized and a manually created summary document.

[1136] Input: Original data file, manual summary document file

[1137] Output: Data transfer request to the server

[1138] Specific operation: The user opens the device's web browser or app, clicks the upload button, selects the original data file and the manual summary document file, and uploads them. The device's built-in emotion engine analyzes the user's facial expressions and tone of voice to identify the user's emotional state.

[1139] Step 2:

[1140] The terminal transmits the received data and the identified emotion information to the server.

[1141] Input: Original data file, manual summary document file, user emotion information

[1142] Output: Server data reception process begins

[1143] Specific operation: The terminal uploads the selected original data file and the manually summarized document file to the server, and simultaneously transmits the user's emotion information.

[1144] Step 3:

[1145] The server analyzes the received original data and manual summary documents, and formats and preprocesses the data.

[1146] Input: Received original data file, manual summary document file

[1147] Output: Formatted and preprocessed data

[1148] Specific operation: The server checks the received raw data files and summary documents based on their format and content, and performs formatting and preprocessing as necessary, including text cleaning and formatting standardization.

[1149] Step 4:

[1150] The server inputs the formatted and preprocessed raw data into a generative AI model to generate an automatic summary.

[1151] Input: Raw data that has been reformatted and preprocessed

[1152] Output: Automatic summary generated by the generative AI model

[1153] How it works: The server inputs the original data into a generative AI model, extracts key points, and automatically generates a summary. This generative AI model uses natural language processing technology to identify important information and output it as a summary.

[1154] Step 5:

[1155] The server compares the generated AI summary with the user's manually summarized document to detect inconsistencies and information omissions.

[1156] Input: Automated summaries by generative AI models, manually summarized documents

[1157] Output: Comparison results (status identifying inconsistencies and information oversaturations)

[1158] How it works: The server displays the AI-generated summary and the user's manual summary side by side and uses a specific algorithm to highlight differences, for example, showing inconsistencies in red and missing information in blue.

[1159] Step 6:

[1160] The server generates feedback based on the comparison results and adjusts it taking into account the user's emotional information provided by the emotion engine.

[1161] Input: Comparison results, user emotion information

[1162] Output: Adjusted feedback message

[1163] Specific behavior: The server generates specific feedback messages based on the differences highlighted, and adjusts the feedback to be gentle and encouraging if the user seems impatient.

[1164] Step 7:

[1165] The server transmits the generated feedback to the user's terminal, and the user receives the feedback.

[1166] Input: Adjusted feedback message

[1167] Output: A feedback message that is displayed on the user's terminal.

[1168] Specific operation: The server sends the generated feedback message to the user's device, and the user checks the feedback in notifications or on the dashboard. The user receives the feedback and reevaluates their own manual summaries to learn from them.

[1169] The above are the specific processing steps of the system. In this way, users can receive feedback based on specific suggestions and emotions, and can efficiently improve their summarization ability.

[1170] (Application example 2)

[1171] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1172] This invention relates to a system for improving the accuracy and efficiency of summarization for users who perform document summarization. In particular, the system aims to improve summarization ability by detecting inconsistencies in summary content and excess or deficiency of information through comparative analysis of automatic summaries generated by generative AI and manual summaries by users, and by providing optimal feedback to the user. Another objective of this invention is to provide a system that reduces the user's mental burden and encourages continuous learning by providing feedback based on the user's emotions in an emotional environment.

[1173] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for uploading a document summarized by a user and original data, means for generating a summary from the uploaded original data using a generation AI, means for comparing the document summarized by the user with the summary generated by the generation AI to detect inconsistencies and excess or deficiency of information, means for providing feedback to the user based on the comparison results, and means including an emotion engine for identifying the user's emotions and adjusting the feedback content. This makes it possible to improve the user's summarizing ability and provide feedback according to emotions.

[1174] "User-summarized documents" refers to information that has been manually summarized by a user based on original data.

[1175] "Original data" refers to a series of information or text data that is the basis for the summary.

[1176] "Generative AI" refers to a system that uses artificial intelligence technology to automatically summarize original data.

[1177] "Inconsistency" refers to inconsistencies in information or content between the generated AI summary and the user's summary.

[1178] "Information overload" refers to a situation in which important information is missing from a summary or unnecessary information is included.

[1179] "Feedback" refers to advice, including instructions for improvement or correction, provided to the user based on the results of comparing the generated AI with the user's summary.

[1180] An "emotion engine" refers to a device or software that identifies a user's emotions and generates feedback appropriate to those emotions.

[1181] "Prompt sentence" refers to the input sentence that the generative AI uses when summarizing.

[1182] The present invention is a system for improving users' summarization skills, and a specific embodiment is described below. This system uses a generative AI model and an emotion engine to perform a comparative analysis of a manually created summary document and the original data, and provides feedback.

[1183] System Configuration

[1184] 1. Terminal

[1185] The device provides an interface for users to input and upload the original data they want to summarize and the summary documents they have manually created. The device is equipped with an emotion engine to identify the user's emotions. This includes smartphones, tablets, and computers.

[1186] 2. Server

[1187] The server is the main processing unit for generating summaries, comparative analysis, and feedback generation. It includes the following means:

[1188] Summary generation method using generative AI

[1189] A comparative method for user-generated summaries and AI-generated summaries

[1190] A means of detecting inconsistencies and information overloads and deficiencies

[1191] Feedback adjustment method by emotion engine

[1192] System Operation

[1193] Data reception

[1194] The user uploads original data and manually summarized documents using a terminal, such as meeting transcripts, reports, articles, etc. The terminal uses an emotion engine to identify the user's emotions and transmits them to the server.

[1195] Generate a summary

[1196] The server uses a generative AI model (e.g., OpenAI GPT-4) to automatically generate a summary based on the original data. This generative AI works based on natural language processing technology, extracting the main points of the original data and creating a short summary.

[1197] comparative analysis

[1198] The server compares the summaries generated by the generative AI model with the manual summaries created by the user. The following points are checked:

[1199] Inconsistency detection: Detects inconsistencies and conflicting information between the generated AI summary and the user's summary.

[1200] Information deficiency / excess detection: Check whether the user's summary is missing important information contained in the generated AI summary or whether it contains unnecessary information.

[1201] Feedback Generation

[1202] Based on the comparison results and taking into account information from the emotion engine, the server generates optimal feedback for the user. For example, if the user is feeling anxious or impatient, the server will adjust the feedback message to be more gentle and encouraging.

[1203] Providing Feedback

[1204] The server generates feedback and sends it to the user's device. The user receives the feedback and reviews their own summarization to improve their summarization skills. The feedback is also optimized based on long-term emotional data.

[1205] Hardware and software used

[1206] Devices: smartphones (iOS / Android compatible), tablets, computers

[1207] Emotion engine API: Azure Emotion API

[1208] Generative AI model: OpenAI GPT-4

[1209] Server: AWS EC2, Node.js, Express

[1210] Prompt Sentence Examples

[1211] Original data: This product uses the latest technology, is highly durable, and remains comfortable even after extended use.

[1212] Generate a summary:

[1213] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1214] Step 1: Enter your data

[1215] The user inputs the original data and the manually summarized text into the application using a device (smartphone, tablet, or computer). The device then uses an emotion engine to identify the user's emotions. The input data includes the original data text, the manually summarized text, and emotion data.

[1216] Step 2: Sending data

[1217] The terminal transmits the input original data, the manual summary document, and the identified emotion data to the server. The transmitted data includes the original data, the manual summary document, and the emotion data.

[1218] Step 3: Data preparation and preprocessing

[1219] The server checks the format and content of the received original data and manual summary document, and formats and preprocesses the data as necessary. Specifically, it deletes unnecessary spaces and symbols and standardizes the text format. This process results in the formatted original data and manual summary document.

[1220] Step 4: Generate a summary

[1221] The server takes the formatted original data as input and generates an automatic summary using a generative AI model (OpenAI GPT-4). The prompt message generated is "Original data: [formatted original data] Please generate a summary:", which is then input into the AI ​​model to output the automatic summary.

[1222] Step 5: Comparative analysis of summaries

[1223] The server compares the automatic summary generated by the generative AI model with the manually summarized document. This comparison uses natural language processing technology to check for semantic agreement and context, and detects inconsistencies and information deficiencies. Specifically, it calculates word agreement rates and differences in contextual meaning, and outputs any inconsistencies or information deficiencies.

[1224] Step 6: Generate feedback

[1225] The server uses a generative AI model to generate optimal feedback for the user based on the results of the comparative analysis and the identified emotional data. The feedback message is generated using the results of the comparative analysis and the emotional data as input, and is output in a format that is easy for the user to understand.

[1226] Step 7: Provide feedback

[1227] The server sends the generated feedback to the user's device, where the user can receive the feedback and use it to improve the summary. The feedback content is adjusted based on the results of the comparative analysis and the user's sentiment.

[1228] The above is the specific processing flow of the system that realizes the application example.

[1229] 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.

[1230] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1231] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1232] [Fourth embodiment]

[1233] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1234] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1235] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1236] 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.

[1237] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1238] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1239] 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.

[1240] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1241] 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.

[1242] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1243] 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.

[1244] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1245] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1246] This invention is a system for improving a user's document summarization ability. It uses a generative AI to compare documents manually summarized by the user with the original data, and identifies inconsistencies and information omissions. The main components of this system and their specific operation are described below.

[1247] System configuration

[1248] 1. Terminal

[1249] It provides an interface for users to input and upload the original data they want to summarize and the summary documents they have manually created. The devices can be computers, smartphones, tablets, etc.

[1250] 2. Server

[1251] It is the main processing unit responsible for generating summaries, comparative analysis, and generating feedback.

[1252] System Operation

[1253] Data reception

[1254] First, users upload original data and summary documents using their devices. This data can include meeting transcripts, reports, articles, etc. The server receives the data, checks its format and content, and performs formatting and preprocessing of the data if necessary.

[1255] Generate a summary

[1256] The server then uses a generative AI to automatically generate a summary using the original data as input. This generative AI works based on natural language processing technology, extracting key points from the original data and creating a shortened summary.

[1257] comparative analysis

[1258] The server compares the generated AI summary with the user's manual summary. This comparison checks the following points:

[1259] Inconsistency detection: Detect inconsistencies or conflicting information between the generated AI summary and the user's summary.

[1260] Information deficiency / excess detection: Check whether the user's summary is missing important information contained in the generated AI summary, or whether it contains unnecessary information.

[1261] Feedback Generation

[1262] The server generates specific feedback based on the comparison results, such as, "This summary does not include the 'progress report of each team,' which is a key point in the original data."

[1263] Providing Feedback

[1264] Finally, the server sends the generated feedback to the user's terminal, where the user can receive it and review their own summaries to improve their summarization skills.

[1265] Specific examples

[1266] Original data (example)

[1267] 1. Discussion on next year's budget

[1268] 2. New project launch plan

[1269] 3. Check the schedule of company events

[1270] 4. Progress reports from each team

[1271] User Summary (Example)

[1272] We discussed the budget for the next fiscal year and new projects, and also confirmed the dates for company events.

[1273] Processing result (example)

[1274] 1. Summary of Generative AI (Generated by the Server)

[1275] Discussions included next year's budget, plans to launch new projects, confirmation of dates for internal events, and progress reports from each team.

[1276] 2. Feedback (Server-Generated)

[1277] "Your summary does not include 'progress reports from each team.'"

[1278] In this way, users can improve the accuracy of their summaries by receiving specific feedback, which will enhance their ability to accurately summarize the main points of materials and discussions, ultimately improving the quality of their documents and presentations.

[1279] The processing flow will be explained below.

[1280] Step 1:

[1281] The user uploads the original data (e.g., transcription data of a meeting) and a summary document created manually via the terminal. The terminal then sends this data to the server.

[1282] Step 2:

[1283] The server receives the original data and the summary sent by the user. At this time, the server checks the consistency of the data format and performs preprocessing if necessary. Preprocessing includes text cleaning and formatting.

[1284] Step 3:

[1285] The server calls the generation AI, passing the original data as input. The generation AI extracts the key points of the original data and generates a short summary, which is later used to compare with the user summary.

[1286] Step 4:

[1287] The server compares the AI-generated summary with the one the user manually created. The comparison checks the following points:

[1288] Inconsistency detection: Detects inconsistencies between the generated AI summary and the user's summary.

[1289] Information deficiency / excess detection: Check whether the user's summary is missing important information contained in the generated AI summary or whether it contains unnecessary information.

[1290] Step 5:

[1291] The server generates feedback based on the comparison, including specific suggestions and areas for improvement, such as identifying important or unnecessary information missing from the user's summary.

[1292] Step 6:

[1293] The server sends the generated feedback to the user's device, allowing the user to review their summary and make corrections if necessary.

[1294] Example 1

[1295] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1296] Conventional document summarization systems have difficulty effectively detecting inconsistencies or excess or deficiencies of information between a user's manually created summary and the original data, and providing appropriate feedback. Furthermore, if data formatting and preprocessing are insufficient, errors are likely to occur in summary generation and comparative analysis. Furthermore, the feedback is not specific enough, so it is not possible to adequately support users in improving their summarization skills.

[1297] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1298] In this invention, the server includes means for uploading a document summarized by a user and the original data, means for checking the format and content of the uploaded original data and reformatting or preprocessing the data as necessary, means for generating a summary from the uploaded original data using a generative AI model, means for generating a summary by providing an appropriate prompt to the generative AI model, means for comparing the document summarized by the user with the summary generated by the generative AI model to detect inconsistencies or excess or deficiencies of information, and means for providing feedback to the user based on the comparison results. This makes it possible to effectively detect inconsistencies or excess or deficiencies of information between the summary created by the user and the original data and provide specific feedback.

[1299] "User" refers to a person who uses the system to create a summary of a document.

[1300] "Original data" refers to the collection of original text or information that a user uses to summarize.

[1301] A "terminal" is a device to which a user uploads original data and manually created summary documents, and includes a computer, a smartphone, a tablet, and the like.

[1302] "Server" refers to a device that performs major processing such as generating summaries, comparative analysis, and generating feedback.

[1303] A "generative AI model" refers to artificial intelligence that automatically generates summaries from original data based on natural language processing technology.

[1304] A "prompt" refers to an input instruction that instructs a generative AI model to perform a specific task.

[1305] "Feedback" refers to the indications and advice that the server provides to the user based on the results of the comparison and analysis.

[1306] "Data formatting" refers to standardizing the format of received data and making it suitable for analysis.

[1307] "Preprocessing" refers to the early stage of data processing carried out to improve the accuracy of data analysis and summary generation.

[1308] This invention is a system for improving a user's document summarization ability. It utilizes a generative AI model to compare documents manually summarized by the user with the original data, and identifies inconsistencies and information omissions. The main components of this system and their specific operation are described below.

[1309] System configuration

[1310] Terminal

[1311] It provides an interface for users to input and upload the original data they want to summarize and the summary documents they have manually created. The devices can be computers, smartphones, tablets, etc.

[1312] server

[1313] This is the main processing unit that generates summaries, performs comparative analysis, and generates feedback. A generative AI model (e.g., GPT-3) is installed on the server and used to automatically generate summaries.

[1314] System Operation

[1315] Data reception

[1316] First, the user uploads the original data and summary documents to the interface using their terminal. This data includes, for example, minutes of meetings, reports, and articles. The server receives the data and checks its format and content. If any inappropriate content or format is found, the server reformats or preprocesses the data.

[1317] Generate a summary

[1318] The server passes the received raw data as input to the generative AI model, which then automatically generates a summary. In this process, the server provides the generative AI with an appropriate prompt. For example, the following prompt can be used:

[1319] "Please summarize the following sentences: 1. Discussion about next year's budget 2. Plan to launch a new project 3. Confirmation of schedule for company events 4. Progress reports from each team."

[1320] A generative AI model then generates a summary.

[1321] comparative analysis

[1322] The server compares the summaries generated by the generative AI model with the manual summaries created by the user. This comparison checks the following points:

[1323] Inconsistency detection: Detect inconsistencies or conflicting information between the generated AI summary and the user's summary.

[1324] Information deficiency / excess detection: Check whether the user's summary is missing important information contained in the generated AI summary, or whether it contains unnecessary information.

[1325] Feedback Generation

[1326] The server generates specific feedback based on the comparison, such as:

[1327] "Your summary does not include 'progress reports from each team.'"

[1328] Providing Feedback

[1329] Finally, the server sends the generated feedback to the user's terminal, where the user can receive it and review their own summaries to improve their summarization skills.

[1330] Specific examples

[1331] Original data (example)

[1332] 1. Discussion on next year's budget

[1333] 2. New project launch plan

[1334] 3. Check the schedule of company events

[1335] 4. Progress reports from each team

[1336] User Summary (Example)

[1337] We discussed the budget for the next fiscal year and new projects, and also confirmed the dates for company events.

[1338] Processing result (example)

[1339] 1. Summary of Generative AI (Generated by the Server)

[1340] Discussions included next year's budget, plans to launch new projects, confirmation of dates for internal events, and progress reports from each team.

[1341] 2. Feedback (Server-Generated)

[1342] Your summary does not include a "progress report from each team."

[1343] As described above, by receiving specific feedback, users can improve the accuracy of their summaries, thereby enhancing their ability to accurately summarize the main points of materials and discussions, and ultimately improving the quality of their documents and presentations.

[1344] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1345] Step 1:

[1346] Data reception

[1347] The user uses a terminal to upload the original data and a manually created summary document to the interface. The terminal then sends this input data to the server. The server receives the uploaded data and checks its format and content. For example, the data format received by the server is JSON format, and after receiving the input data, it parses it to check for errors. If an error is detected, the server notifies the user with an appropriate error message.

[1348] Step 2:

[1349] Data formatting and preprocessing

[1350] The server formats and preprocesses the received data. This includes standardizing the data format and making it easier to analyze. For example, the server cleans the text data, removes unnecessary whitespace and special characters, and normalizes the document. These operations transform the data into a state suitable for analysis.

[1351] Step 3:

[1352] Generate a summary

[1353] The server passes the formatted original data as input to a generative AI model (e.g., GPT-3) to automatically generate a summary. At this time, the server provides an appropriate prompt to the generative AI model. The following format is used as an example of a prompt:

[1354] "Please summarize the following sentences: 1. Discussion about next year's budget 2. Plan to launch a new project 3. Confirmation of schedule for company events 4. Progress reports from each team."

[1355] The generative AI model receives this prompt and generates a summary based on the original data, which is then received and stored by the server.

[1356] Step 4:

[1357] comparative analysis

[1358] The server compares and analyzes the summaries generated by the AI ​​model with the manual summaries created by the user. This comparison involves text analysis and similarity calculations to detect inconsistencies and information overloads and omissions. The server vectorizes the text of both summaries using, for example, TF-IDF or word embeddings, and calculates the degree of agreement. It also extracts parts where important information is missing or where unnecessary information has been added.

[1359] Step 5:

[1360] Generate feedback

[1361] The server generates specific feedback based on the results of the comparison and analysis. The generated feedback is written in natural language and provided in a format that is easy for users to understand. For example, the following feedback is generated:

[1362] "Your summary does not include 'progress reports from each team.'"

[1363] The server edits the feedback text to include useful suggestions and advice for the user.

[1364] Step 6:

[1365] Providing feedback

[1366] The server sends the generated feedback to the user's device, which receives the feedback and notifies the user. The user can then use the feedback to review their own summaries and create more accurate summaries.

[1367] By clearly indicating the specific operations and the flow of input and output data in each step, the embodiments of the invention can be explained in detail.

[1368] (Application example 1)

[1369] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1370] In traditional advertising production, it is difficult to evaluate the effectiveness of advertising copy and the accuracy of important information, requiring a lot of time and effort. In particular, there is a lack of means to quickly and accurately determine whether the summary created by advertising creators is conveyed to consumers, making it difficult to maximize the effectiveness of advertising.

[1371] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1372] In this invention, the server includes means for comparing the original advertisement data uploaded by the user with the summary and evaluating the accuracy of the information and the effectiveness of the summary, means for generating a summary using a generation AI, and means for comparing the document summarized by the user with the summary generated by the generation AI to detect inconsistencies and excess or deficiency of information. This enables advertisement creators to quickly determine whether their summary matches the generation AI and whether important points are missing, thereby optimizing the message of the advertisement.

[1373] A "document summarized by a user" is a document that has been manually summarized and simplified from original data by a user.

[1374] "Original data" refers to documents or data that contain the information that is the basis for summarizing.

[1375] "Generative AI" is a system that uses artificial intelligence technology to automatically generate summaries from original data.

[1376] A "means for generating a summary" is a mechanism for automatically creating a summary from original data using generative AI.

[1377] The "means for detecting inconsistencies and information omissions" is a mechanism for comparing the generated AI's summary with the user's summary to detect inconsistencies or missing important information.

[1378] The "means for providing feedback" is a mechanism for suggesting specific improvements or corrections to the user based on the comparison results.

[1379] "Advertising source data" refers to the documents and data that are the basis for creating advertisements.

[1380] "Means for assessing the accuracy of information and the effectiveness of summaries" refers to a mechanism for comparing the original advertising data with the summaries to assess whether the information is being conveyed accurately and whether the summaries are effective.

[1381] An embodiment of the present invention is a system and method for evaluating advertising copy created by advertising creators and generating effective advertising summaries. This system is mainly composed of a server and a user terminal, and utilizes a generative AI model to automatically generate summaries of advertising copy and compare them with summaries created by users, thereby providing effective and efficient feedback.

[1382] System configuration:

[1383] server

[1384] The server performs the following main functions:

[1385] Receive the original data (advertising source data) and check the format and content.

[1386] Format and preprocess the data as needed.

[1387] A generative AI model is used to automatically generate summaries from the original data.

[1388] The summary created by the user is compared with the summary generated by the AI ​​to detect inconsistencies and information that is insufficient or excessive.

[1389] Feedback is generated based on the comparison result and transmitted to the user terminal.

[1390] Hardware and software used

[1391] Hardware:

[1392] As a server, use a high-performance computer or cloud computing infrastructure (e.g., AWS, Google Cloud Platform).

[1393] software:

[1394] For data formatting and preprocessing, we use Python's pandas and numpy libraries.

[1395] The generative AI model used is OpenAI's GPT-3.

[1396] The Flask framework is used to build the web interface.

[1397] Natural language processing technologies (e.g., spaCy, NLTK) may also be used for comparative analysis and feedback generation.

[1398] Processing flow

[1399] 1. Data reception and preprocessing:

[1400] Users upload the original advertising data and their own summaries from their smartphones or PCs. The server receives the data, checks the format and content, and formats and preprocesses the data as necessary.

[1401] 2. Generate a summary:

[1402] The server uses a generative AI model (e.g., OpenAI's GPT-3) to automatically generate a summary from the original data. This summary is a shorthand representation of the main points of the original data.

[1403] 3. Comparative analysis:

[1404] The summary created by the user is compared with the summary generated by the server to check for consistency and whether there is any excess or deficiency in the information. This is done using the natural language processing technology of the generation AI.

[1405] 4. Feedback Generation:

[1406] Generate specific feedback based on the comparison, such as "Your summary is missing the important information 'full of vitamin C'."

[1407] 5. Providing Feedback:

[1408] The generated feedback is sent to the user's device, and the user can use it to revise and improve the summary.

[1409] Specific examples

[1410] Ad source data:

[1411] Try our new Super Drink this fall! Packed with Vitamin C, it will help relieve fatigue.

[1412] User Summary:

[1413] Try a new vitamin drink.

[1414] Summary of generated AI (server generated):

[1415] This autumn's new product is a super drink packed with vitamin C.

[1416] Feedback (server generated):

[1417] "Your summary doesn't include key points like 'New this fall.'"

[1418] Prompt Sentence Examples

[1419] For the generative AI model, we input:

[1420] Summarize the following ad copy:

[1421] Try our new Super Drink this fall! Packed with Vitamin C, it will help relieve fatigue.

[1422] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1423] Step 1:

[1424] The user uploads the original ad data and a self-created summary to the server via their device. The device then sends the input data to the server. The input at this time is the "original ad data" and the "user summary," and the output to the server is the original data.

[1425] Step 2:

[1426] The server checks the format and content of the received source data and user summary. For example, the server checks whether the data format is correct and the content is entered properly, and performs data formatting (e.g., grammar correction and text format standardization) and preprocessing as necessary. The input to this process is the data received in step 1, and the output is the formatted and preprocessed data.

[1427] Step 3:

[1428] The server uses a generative AI model (OpenAI GPT-3) to generate a summary from the formatted and preprocessed original ad data. The generative AI receives the original data as a prompt in the form of "Please summarize the ad copy below:" The output of the generative AI is an automatically generated summary based on the original ad data.

[1429] Step 4:

[1430] The server compares the summary generated by the AI ​​with the summary created by the user. This comparison process uses natural language processing technology to check whether the meanings of the two sentences match and whether important information is missing. The inputs to this process are the AI's summary and the user's summary, and the output is information on whether they match or not based on the comparison results.

[1431] Step 5:

[1432] The server generates feedback based on the comparison results. Specifically, if there are any inconsistencies between the AI's summary and the user's summary, or if important information is missing, the server will provide specific feedback. The input for this process is the comparison result from step 4, and the output is a feedback statement to be given to the user.

[1433] Step 6:

[1434] The server sends the generated feedback to the user terminal, which receives it and makes further corrections and improvements. The input of this process is the feedback from step 5, and the output is the feedback sentence displayed on the user terminal.

[1435] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1436] This invention is a system for improving a user's document summarization ability. It utilizes generative AI to compare documents manually summarized by the user with the original data, and points out inconsistencies and excess or deficiency of information. It also combines an emotion engine to provide feedback based on the user's emotions. The main components of this system and their specific operation are described below.

[1437] System configuration

[1438] 1. Terminal

[1439] It provides an interface for users to input and upload the original data they want to summarize and the summary documents they have manually created. The device can be a computer, smartphone, tablet, etc. The device also has an emotion engine to identify the user's emotions.

[1440] 2. Server

[1441] It is the main processing unit for generating summaries, comparative analysis, and generating feedback. The server has the ability to adjust the feedback based on the user's emotions identified by the emotion engine.

[1442] System Operation

[1443] Data reception

[1444] First, the user uploads original data and summary documents using their device. This data can include meeting transcripts, reports, articles, etc. The server receives the data, checks its format and content, and performs data formatting and preprocessing if necessary. The device's on-board emotion engine also identifies the user's current emotion, which is then sent to the server.

[1445] Generate a summary

[1446] The server then uses a generative AI to automatically generate a summary using the original data as input. This generative AI works based on natural language processing technology, extracting key points from the original data and creating a shortened summary.

[1447] comparative analysis

[1448] The server compares the generated AI summary with the user's manual summary. The comparison checks the following points:

[1449] Inconsistency detection: Detect inconsistencies or conflicting information between the generated AI summary and the user's summary.

[1450] Information overload / deficiency detection: Check whether the user's summary is missing important information contained in the generated AI summary, or whether it contains unnecessary information.

[1451] Feedback Generation

[1452] The server generates optimal feedback for the user based on the comparison results and also takes into account information from the emotion engine. For example, if the user is feeling impatient or anxious, the feedback message will be adjusted to be gentle and encouraging.

[1453] Providing Feedback

[1454] Finally, the server sends the generated feedback to the user's device. The user can then review their summary and improve their summarization skills. The server also tracks the user's emotional changes and analyzes long-term patterns to further optimize future feedback.

[1455] Specific examples

[1456] Original data (example)

[1457] 1. Discussion on next year's budget

[1458] 2. New project launch plan

[1459] 3. Check the schedule of company events

[1460] 4. Progress reports from each team

[1461] User Summary (Example)

[1462] We discussed the budget for the next fiscal year and new projects, and also confirmed the dates for company events.

[1463] Processing result (example)

[1464] 1. Summary of Generative AI (Generated by the Server)

[1465] Discussions included next year's budget, plans to launch new projects, confirmation of dates for internal events, and progress reports from each team.

[1466] 2. User's emotions (recognized by the emotion engine)

[1467] "A little impatient"

[1468] 3. Feedback (Server-Generated)

[1469] "Your summary does not include the 'progress report of each team,' but it is a good summary overall. Please take your time and review it again."

[1470] In this way, users can efficiently improve their summarizing skills by receiving feedback that takes into account specific points and emotions, which will improve their ability to accurately summarize the main points of materials and discussions, and ultimately improve the quality of their documents and presentations.

[1471] The processing flow will be explained below.

[1472] Step 1:

[1473] The user uses the device to upload the original data (e.g., a transcript of a meeting) and a manually created summary document. The user uses the device interface to select the file and clicks the send button.

[1474] Step 2:

[1475] The device sends the uploaded original data and summary to the server. At the same time, the device's built-in emotion engine analyzes the user's emotions and sends the results to the server. The emotion engine identifies emotions such as "anxiety" or "calmness" based on the user's facial expressions and input speed.

[1476] Step 3:

[1477] The server receives the original data and summary text sent by the user. At this time, the server checks the consistency of the data format and, if there is any inconsistency, formats and preprocesses the data. Specifically, it removes unnecessary whitespace and standardizes the text.

[1478] Step 4:

[1479] The server passes the original data to the AI ​​generator and requests it to generate a summary. The AI ​​generator extracts the key points of the original data and generates a shortened summary. Natural language processing technology is used in this process.

[1480] Step 5:

[1481] The server compares the AI-generated summary with the summary created manually by the user. The server checks the following points:

[1482] Inconsistency detection: Detects where the generated AI summary and the user's summary do not match.

[1483] Information deficiency / excess detection: Check whether the user's summary is missing important information contained in the generated AI summary or whether it contains unnecessary information.

[1484] Step 6:

[1485] The server generates feedback based on the comparison results, taking into account the emotions expressed by the user when uploading. For example, if the user is feeling "impatient," the server adjusts the feedback message to emphasize a gentle, encouraging tone.

[1486] Step 7:

[1487] The server sends the generated feedback to the user's device, where the user can then revise their summary. Based on the feedback, which takes specific points and emotions into consideration, the user can improve their summarization skills.

[1488] Specific examples

[1489] Let's say the original data looks like this:

[1490] 1. Discussion on next year's budget

[1491] 2. New project launch plan

[1492] 3. Check the schedule of company events

[1493] 4. Progress reports from each team

[1494] Suppose the user's summary was:

[1495] We discussed the budget for the next fiscal year and new projects, and also confirmed the dates for company events.

[1496] The summary generated by the server is:

[1497] Discussions included next year's budget, plans to launch new projects, confirmation of dates for internal events, and progress reports from each team.

[1498] If the user's emotion is recognized as "impatience," the server's feedback will be as follows:

[1499] Your summary does not include the "progress report of each team," but it is a good summary overall. Please take your time and review it again.

[1500] This example allows users to improve their summaries and receive feedback that will help them create their next summaries.

[1501] Example 2

[1502] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1503] Existing document summarization systems have the problem of being unable to accurately identify inconsistencies or information omissions when comparing manually created summaries with automatically generated summaries. Furthermore, feedback is provided without taking into account the user's emotional state, which can reduce the user's motivation and learning effectiveness. Furthermore, the lack of long-term support for user development makes it difficult to achieve continuous improvement.

[1504] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1505] In this invention, the server includes means for uploading a document summarized by a user and its original data, means for generating a summary from the uploaded original data using a generative AI model, means for comparing the document summarized by the user with the summary generated by the generative AI model to detect inconsistencies and excess or deficiency of information, means for providing feedback based on the user's emotions using an emotion engine based on the comparison results, and means for transmitting the feedback to the user's terminal and tracking changes in the user's emotions. This makes it possible to effectively compare the summary created by the user with the automatically generated summary, provide appropriate feedback based on the user's emotional state, and support continuous improvement.

[1506] "User" means an individual or organization that uses the document summarization system and creates and uploads original data and summarized documents.

[1507] "Terminal" means a hardware device through which a user inputs and uploads original data and summary documents, such as a computer, smartphone, or tablet.

[1508] The "server" is the main processing unit responsible for generating summaries, comparative analysis, and feedback generation, and is a computer equipped with the generative AI model and emotion engine.

[1509] "Original data" refers to documents uploaded by users to be summarized, including meeting minutes, articles, reports, and the like.

[1510] A "summary document" is a shortened document that is manually created by a user based on the original data.

[1511] A "generative AI model" is an artificial intelligence algorithm that automatically generates a summary based on original data.

[1512] An "emotion engine" is a computer program that analyzes a user's facial expressions and tone of voice to identify the user's current emotional state.

[1513] "Feedback" refers to comments or suggestions provided to the user's summary document based on the results of the comparative analysis, and is adjusted to take into account the user's emotional state.

[1514] "Data formatting" is the process of checking the format and content of the original data received and converting it into a format that is easy to analyze.

[1515] "Preprocessing" refers to a series of preparatory steps that are performed before analyzing raw data and generating summaries, including text cleaning and preparation for analysis.

[1516] This invention is a system for improving users' document summarization abilities by utilizing a generative AI model to compare documents manually summarized by the user with the original data, and to identify inconsistencies and information omissions. This system also incorporates an emotion engine to provide feedback based on the user's emotions.

[1517] The main components of the system of the present invention are a terminal, a server, a generative AI model, and an emotion engine. First, a user uploads original data and a manually created summary document using a terminal. This terminal can be a computer, smartphone, tablet, or other device. The terminal also has an emotion engine that identifies the user's emotions.

[1518] The original data and manual summaries are sent from the device to the server, which receives them, formats and preprocesses them as needed, and then summarizes the original data using a generative AI model, which works based on natural language processing techniques.

[1519] The server compares the generated AI summary with the user's manual summary, checking for inconsistencies and information oversaturation. Based on the results of this comparison, the server generates optimal feedback, taking into account the user's emotional information received from the device. Finally, the server sends the generated feedback to the user's device, allowing the user to review their own summary based on it and improve their summarization ability.

[1520] Examples of prompt statements

[1521] Specific examples are shown below.

[1522] Original data (example)

[1523] 1. Discussion on next year's budget

[1524] 2. New project launch plan

[1525] 3. Check the schedule of company events

[1526] 4. Progress reports from each team

[1527] User Summary (Example)

[1528] We discussed the budget for the next fiscal year and new projects, and also confirmed the dates for company events.

[1529] Processing result (example)

[1530] 1. Summary of Generative AI (Generated by the Server)

[1531] Discussions included next year's budget, plans to launch new projects, confirmation of dates for internal events, and progress reports from each team.

[1532] 2. User's emotions (recognized by the emotion engine)

[1533] "A little impatient"

[1534] 3. Feedback (Server-Generated)

[1535] "Your summary does not include the 'progress report of each team,' but it is a good summary overall. Please take your time and review it again."

[1536] In this way, users can efficiently improve their summarizing skills by receiving feedback that takes into account specific points and emotions, thereby improving their ability to accurately summarize the main points of materials and discussions, and ultimately improving the quality of their documents and presentations.

[1537] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1538] Step 1:

[1539] A user uses a terminal to upload the original data to be summarized and a manually created summary document.

[1540] Input: Original data file, manual summary document file

[1541] Output: Data transfer request to the server

[1542] Specific operation: The user opens the device's web browser or app, clicks the upload button, selects the original data file and the manual summary document file, and uploads them. The device's built-in emotion engine analyzes the user's facial expressions and tone of voice to identify the user's emotional state.

[1543] Step 2:

[1544] The terminal transmits the received data and the identified emotion information to the server.

[1545] Input: Original data file, manual summary document file, user emotion information

[1546] Output: Server data reception process begins

[1547] Specific operation: The terminal uploads the selected original data file and the manually summarized document file to the server, and simultaneously transmits the user's emotion information.

[1548] Step 3:

[1549] The server analyzes the received original data and manual summary documents, and formats and preprocesses the data.

[1550] Input: Received original data file, manual summary document file

[1551] Output: Formatted and preprocessed data

[1552] Specific operation: The server checks the received raw data files and summary documents based on their format and content, and performs formatting and preprocessing as necessary, including text cleaning and formatting standardization.

[1553] Step 4:

[1554] The server inputs the formatted and preprocessed raw data into a generative AI model to generate an automatic summary.

[1555] Input: Raw data that has been reformatted and preprocessed

[1556] Output: Automatic summary generated by the generative AI model

[1557] How it works: The server inputs the original data into a generative AI model, extracts key points, and automatically generates a summary. This generative AI model uses natural language processing technology to identify important information and output it as a summary.

[1558] Step 5:

[1559] The server compares the generated AI summary with the user's manually summarized document to detect inconsistencies and information omissions.

[1560] Input: Automated summaries by generative AI models, manually summarized documents

[1561] Output: Comparison results (status identifying inconsistencies and information oversaturations)

[1562] How it works: The server displays the AI-generated summary and the user's manual summary side by side and uses a specific algorithm to highlight differences, for example, showing inconsistencies in red and missing information in blue.

[1563] Step 6:

[1564] The server generates feedback based on the comparison results and adjusts it taking into account the user's emotional information provided by the emotion engine.

[1565] Input: Comparison results, user emotion information

[1566] Output: Adjusted feedback message

[1567] Specific behavior: The server generates specific feedback messages based on the differences highlighted, and adjusts the feedback to be gentle and encouraging if the user seems impatient.

[1568] Step 7:

[1569] The server transmits the generated feedback to the user's terminal, and the user receives the feedback.

[1570] Input: Adjusted feedback message

[1571] Output: A feedback message that is displayed on the user's terminal.

[1572] Specific operation: The server sends the generated feedback message to the user's device, and the user checks the feedback in notifications or on the dashboard. The user receives the feedback and reevaluates their own manual summaries to learn from them.

[1573] The above are the specific processing steps of the system. In this way, users can receive feedback based on specific suggestions and emotions, and can efficiently improve their summarization ability.

[1574] (Application example 2)

[1575] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1576] This invention relates to a system for improving the accuracy and efficiency of summarization for users who perform document summarization. In particular, the system aims to improve summarization ability by detecting inconsistencies in summary content and excess or deficiency of information through comparative analysis of automatic summaries generated by generative AI and manual summaries by users, and by providing optimal feedback to the user. Another objective of this invention is to provide a system that reduces the user's mental burden and encourages continuous learning by providing feedback based on the user's emotions in an emotional environment.

[1577] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for uploading a document summarized by a user and original data, means for generating a summary from the uploaded original data using a generation AI, means for comparing the document summarized by the user with the summary generated by the generation AI to detect inconsistencies and excess or deficiency of information, means for providing feedback to the user based on the comparison results, and means including an emotion engine for identifying the user's emotions and adjusting the feedback content. This makes it possible to improve the user's summarizing ability and provide feedback according to emotions.

[1578] "User-summarized documents" refers to information that has been manually summarized by a user based on original data.

[1579] "Original data" refers to a series of information or text data that is the basis for the summary.

[1580] "Generative AI" refers to a system that uses artificial intelligence technology to automatically summarize original data.

[1581] "Inconsistency" refers to inconsistencies in information or content between the generated AI summary and the user's summary.

[1582] "Information overload" refers to a situation in which important information is missing from a summary or unnecessary information is included.

[1583] "Feedback" refers to advice, including instructions for improvement or correction, provided to the user based on the results of comparing the generated AI with the user's summary.

[1584] An "emotion engine" refers to a device or software that identifies a user's emotions and generates feedback appropriate to those emotions.

[1585] "Prompt sentence" refers to the input sentence that the generative AI uses when summarizing.

[1586] The present invention is a system for improving users' summarization skills, and a specific embodiment is described below. This system uses a generative AI model and an emotion engine to perform a comparative analysis of a manually created summary document and the original data, and provides feedback.

[1587] System Configuration

[1588] 1. Terminal

[1589] The device provides an interface for users to input and upload the original data they want to summarize and the summary documents they have manually created. The device is equipped with an emotion engine to identify the user's emotions. This includes smartphones, tablets, and computers.

[1590] 2. Server

[1591] The server is the main processing unit for generating summaries, comparative analysis, and feedback generation. It includes the following means:

[1592] Summary generation method using generative AI

[1593] A comparative method for user-generated summaries and AI-generated summaries

[1594] A means of detecting inconsistencies and information overloads and deficiencies

[1595] Feedback adjustment method by emotion engine

[1596] System Operation

[1597] Data reception

[1598] The user uploads original data and manually summarized documents using a terminal, such as meeting transcripts, reports, articles, etc. The terminal uses an emotion engine to identify the user's emotions and transmits them to the server.

[1599] Generate a summary

[1600] The server uses a generative AI model (e.g., OpenAI GPT-4) to automatically generate a summary based on the original data. This generative AI works based on natural language processing technology, extracting the main points of the original data and creating a short summary.

[1601] comparative analysis

[1602] The server compares the summaries generated by the generative AI model with the manual summaries created by the user. The following points are checked:

[1603] Inconsistency detection: Detects inconsistencies and conflicting information between the generated AI summary and the user's summary.

[1604] Information deficiency / excess detection: Check whether the user's summary is missing important information contained in the generated AI summary or whether it contains unnecessary information.

[1605] Feedback Generation

[1606] Based on the comparison results and taking into account information from the emotion engine, the server generates optimal feedback for the user. For example, if the user is feeling anxious or impatient, the server will adjust the feedback message to be more gentle and encouraging.

[1607] Providing Feedback

[1608] The server generates feedback and sends it to the user's device. The user receives the feedback and reviews their own summarization to improve their summarization skills. The feedback is also optimized based on long-term emotional data.

[1609] Hardware and software used

[1610] Devices: smartphones (iOS / Android compatible), tablets, computers

[1611] Emotion engine API: Azure Emotion API

[1612] Generative AI model: OpenAI GPT-4

[1613] Server: AWS EC2, Node.js, Express

[1614] Prompt Sentence Examples

[1615] Original data: This product uses the latest technology, is highly durable, and remains comfortable even after extended use.

[1616] Generate a summary:

[1617] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1618] Step 1: Enter your data

[1619] The user inputs the original data and the manually summarized text into the application using a device (smartphone, tablet, or computer). The device then uses an emotion engine to identify the user's emotions. The input data includes the original data text, the manually summarized text, and emotion data.

[1620] Step 2: Sending data

[1621] The terminal transmits the input original data, the manual summary document, and the identified emotion data to the server. The transmitted data includes the original data, the manual summary document, and the emotion data.

[1622] Step 3: Data preparation and preprocessing

[1623] The server checks the format and content of the received original data and manual summary document, and formats and preprocesses the data as necessary. Specifically, it deletes unnecessary spaces and symbols and standardizes the text format. This process results in the formatted original data and manual summary document.

[1624] Step 4: Generate a summary

[1625] The server takes the formatted original data as input and generates an automatic summary using a generative AI model (OpenAI GPT-4). The prompt message generated is "Original data: [formatted original data] Please generate a summary:", which is then input into the AI ​​model to output the automatic summary.

[1626] Step 5: Comparative analysis of summaries

[1627] The server compares the automatic summary generated by the generative AI model with the manually summarized document. This comparison uses natural language processing technology to check for semantic agreement and context, and detects inconsistencies and information deficiencies. Specifically, it calculates word agreement rates and differences in contextual meaning, and outputs any inconsistencies or information deficiencies.

[1628] Step 6: Generate feedback

[1629] The server uses a generative AI model to generate optimal feedback for the user based on the results of the comparative analysis and the identified emotional data. The feedback message is generated using the results of the comparative analysis and the emotional data as input, and is output in a format that is easy for the user to understand.

[1630] Step 7: Provide feedback

[1631] The server sends the generated feedback to the user's device, where the user can receive the feedback and use it to improve the summary. The feedback content is adjusted based on the results of the comparative analysis and the user's sentiment.

[1632] The above is the specific processing flow of the system that realizes the application example.

[1633] 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.

[1634] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1635] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1636] 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.

[1637] FIG. 9 illustrates an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and behaviors arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1638] 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.

[1639] 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).

[1640] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1641] 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."

[1642] 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.

[1643] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1644] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1645] 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.

[1646] 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.

[1647] 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.

[1648] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1649] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1650] 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.

[1651] 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.

[1652] 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.

[1653] 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.

[1654] The following is further disclosed regarding the above embodiment.

[1655] (Claim 1)

[1656] a means for users to upload summarized documents and raw data;

[1657] A means to generate a summary from the uploaded original data using AI;

[1658] A means to compare the document summarized by the user with the summary generated by the AI ​​to detect inconsistencies and information omissions.

[1659] means for providing feedback to the user based on the comparison;

[1660] A system including:

[1661] (Claim 2)

[1662] 10. The system of claim 1, further comprising means for checking the format and content of the original data upon receipt and, if necessary, formatting or pre-processing the data.

[1663] (Claim 3)

[1664] 10. The system of claim 1, further comprising means for using natural language processing techniques to verify semantic agreement and context when evaluating the comparison results.

[1665] "Example 1"

[1666] (Claim 1)

[1667] a means for users to upload summarized documents and raw data;

[1668] A means to check the format and content of the uploaded original data, and to format and preprocess the data as necessary;

[1669] A means for generating summaries from uploaded raw data using a generative AI model; and

[1670] A means for providing an appropriate prompt to the generative AI model to generate a summary;

[1671] A means to compare a user-summarized document with a summary generated by an AI model to detect inconsistencies and information oversaturations.

[1672] means for providing feedback to the user based on the comparison;

[1673] A system including:

[1674] (Claim 2)

[1675] 10. The system of claim 1, further comprising means for evaluating the comparison results using natural language processing techniques to verify semantic agreement and context.

[1676] (Claim 3)

[1677] 10. The system of claim 1, further comprising means for providing specific instructions in natural language when generating feedback.

[1678] "Application Example 1"

[1679] (Claim 1)

[1680] a means for users to upload summarized documents and raw data;

[1681] A means to generate a summary from the uploaded original data using AI;

[1682] A means to compare the document summarized by the user with the summary generated by the AI ​​to detect inconsistencies and information omissions.

[1683] means for comparing the summary with the advertising source data uploaded by the user to evaluate the accuracy of the information and the effectiveness of the summary;

[1684] means for providing feedback to the user based on the comparison;

[1685] A system including:

[1686] (Claim 2)

[1687] 10. The system of claim 1, further comprising means for checking the format and content of the original data upon receipt and, if necessary, formatting or pre-processing the data.

[1688] (Claim 3)

[1689] 10. The system of claim 1, further comprising means for using natural language processing techniques to verify semantic agreement and context when evaluating the comparison results.

[1690] "Example 2: Combining Emotion Engines"

[1691] (Claim 1)

[1692] a means for users to upload summarized documents and raw data;

[1693] A means for generating summaries from uploaded raw data using a generative AI model; and

[1694] A means to compare a user-summarized document with a summary generated by an AI model to detect inconsistencies and information oversaturations.

[1695] means for providing feedback based on the user's emotions according to the emotion engine based on the comparison result;

[1696] means for transmitting feedback to a user's device and tracking changes in the user's emotions;

[1697] A system including:

[1698] (Claim 2)

[1699] 10. The system of claim 1, further comprising means for checking the format and content of the original data upon receipt and, if necessary, formatting or pre-processing the data.

[1700] (Claim 3)

[1701] 10. The system of claim 1, further comprising means for using natural language processing techniques to verify semantic agreement and context when evaluating the comparison results.

[1702] "Application example 2 when combining emotion engines"

[1703] (Claim 1)

[1704] a means for users to upload summarized documents and raw data;

[1705] A means to generate a summary from the uploaded original data using AI;

[1706] A means to compare the document summarized by the user with the summary generated by the AI ​​to detect inconsistencies and information omissions.

[1707] means for providing feedback to the user based on the comparison;

[1708] means for identifying a user's emotion and adjusting the feedback content, the means including an emotion engine;

[1709] A system including:

[1710] (Claim 2)

[1711] 10. The system of claim 1, further comprising means for checking the format and content of the original data upon receipt and, if necessary, formatting or pre-processing the data.

[1712] (Claim 3)

[1713] 10. The system of claim 1, further comprising means for using natural language processing techniques to verify semantic agreement and context when evaluating the comparison results.

[1714] (Claim 4)

[1715] 10. The system of claim 1, further comprising: means for generating prompt sentences using a generative AI model based on source data and summary documents uploaded by a user. [Explanation of symbols]

[1716] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a means for users to upload summarized documents and raw data; A means to generate a summary from the uploaded original data using AI; A means to compare the document summarized by the user with the summary generated by the AI ​​to detect inconsistencies and information omissions. means for providing feedback to the user based on the comparison; A system including:

2. 2. The system of claim 1, further comprising means for checking the format and content of the original data upon receipt and, if necessary, for formatting or pre-processing the data.

3. 10. The system of claim 1, further comprising means for using natural language processing techniques to verify semantic agreement and context when evaluating the comparison results.

Images