system

Abstract

The system enables efficient document analysis and summary generation, reducing the burden on legal departments and speeding up contract conclusion. [Solution] A system including: a means for a user to upload a document to the system; a means for a server to receive and store the uploaded document; a means for the server to preprocess the document and convert it into a format suitable for analysis; a means for the server to determine the language of the document; a means for the server to analyze each section of the document using natural language processing technology and extract important keywords and phrases; a means for the server to generate a summary of the document based on the analysis results; a means for the server to send the summary generated by the server to the user's terminal; and a means for the user to check the analysis results and the summary via the terminal.

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] It shows the "problem that the invention aims to solve" and the "means for solving the problem."

[0005] Reviewing and summarizing contracts and other important documents is a crucial part of modern business operations, but it can be a time-consuming and labor-intensive task for legal departments. This burden increases especially when dealing with documents written in multiple languages. Therefore, there is a need for a system that can efficiently analyze documents, extract key information, and provide summaries. Such a system would reduce the burden on legal departments and expedite contract signing. [Means for solving the problem]

[0006] To solve the above problems, the present invention provides the following means: a means for a user to upload a document to the system; a means for a server to receive and store the uploaded document; a means for the server to preprocess the document and convert it into a format suitable for analysis; a means for the server to determine the language of the document; a means for the server to analyze each section of the document using natural language processing technology and extract important keywords and phrases; a means for the server to generate a summary of the document based on the analysis results; a means for the server to send the summary generated by the server to a user's terminal; and a means for the user to check the analysis results and the summary through the terminal. Furthermore, if the document is not in a language specified by the user, the server has a means for translating the document or summary into the specified language using an automatic translation function; and a means for the server to automatically detect specific important sections (e.g., contract term, payment terms, liability clauses, etc.) based on the content of the document and include these sections in the summary. This enables efficient document analysis and summary generation, reduces the burden on the legal department, and speeds up contract conclusion.

[0007] A "user" is a person or department that uses the system to upload documents or check analysis results.

[0008] A "terminal" is an electronic device (e.g., computer, tablet, smartphone, etc.) that a user uses to access the system.

[0009] A "server" is a central computing resource that receives, stores, analyzes, summarizes, and transmits the results to the terminals.

[0010] "Documents" are digital files to be analyzed and summarized, such as text or PDF files containing contracts or other important information.

[0011] The "system" is a set of hardware and software in which the server analyzes documents uploaded by users, generates summaries, and provides the results to terminals.

[0012] "Natural language processing technology" is a general term for technologies and methods that enable computers to understand, generate, and analyze human language.

[0013] "Preprocessing" refers to the initial processing of documents to convert them into a format suitable for analysis and to remove noise and unnecessary data.

[0014] A "summary" is a concise summary of the important points or sections of a document.

[0015] "Automatic translation function" means a function in which the system automatically translates documents or summaries in different languages ​​into a specified language.

[0016] "Important sections" are parts of the document that deserve particular attention (e.g., contract terms, payment terms, liability clauses, etc.). [Brief explanation of the drawings]

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

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

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

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

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

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

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

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

[0025] [First embodiment]

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

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

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

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

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

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

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

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

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

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

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

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

[0038] This invention relates to a method for efficiently analyzing contracts and other important documents and generating summaries using a system, in which a user uploads the document to the system using a terminal, a server analyzes the document, generates a summary, and sends it back to the user's terminal.

[0039] Explanation of the specific operation of the program

[0040] 1. Uploading documents

[0041] Users use their devices to upload documents they wish to analyze and summarize to the system, for example by selecting a PDF file of a contract from a PC, smartphone, or other device, and clicking the upload button.

[0042] 2. Document Receipt and Storage

[0043] The server receives the uploaded document and temporarily stores it, at which point it obtains basic information about the document's format (e.g. PDF, DOCX).

[0044] 3. Document Preprocessing

[0045] The server performs pre-processing on the received documents, which includes converting the PDF files to text format and removing noise (e.g., images and redundant text).

[0046] 4. Language Identification

[0047] The server uses natural language processing (NLP) techniques to automatically determine the language of the document, whether it is written in English, Japanese, French, etc.

[0048] 5. Document Analysis

[0049] Based on the language identified, the server uses NLP technology to analyze the document, extracting important keywords and phrases (e.g., contract terms, payment terms, liability clauses, etc.), and analyzes the document structure, dividing it into sections for analysis.

[0050] 6. Generating a Summary

[0051] The server then generates a summary of the document based on the analysis results. The summary is a concise summary of the document's key points, including the most important information extracted. For example, a contract summary might include information such as "Contract period: one year" and "Payment terms: payment by the end of each month."

[0052] 7. Multilingual support

[0053] If the language of the document is different from the language specified by the user, the server will use an automatic translation function to translate the document and summary into the specified language, allowing the user to view the summary in the language they require.

[0054] 8. Submitting the results

[0055] The server sends the generated summary and analysis results to the user's terminal, where the user can check the summary received and perform additional verification or correction as necessary.

[0056] Specific examples

[0057] Example 1: Processing an English contract

[0058] A user uploads an English contract titled "Service Agreement" in PDF format to the system using their PC. The server receives the document, converts it to text format, and removes noise. After determining that the document is in English, it uses NLP technology to extract important information such as the contract period, payment terms, and liability clauses. A summary is generated based on this information and sent to the user's PC. The user then checks the summary on their PC and determines whether the contract contents are reasonable.

[0059] Example 2: Processing multilingual contracts

[0060] A user uploads a contract written in French to the system using their smartphone. The server receives the French document and performs the necessary preprocessing. It determines that the document is in French and uses NLP technology to extract important information. If the user specifies English, the server uses an automatic translation function to translate the French summary into English. The translated summary is sent to the user's smartphone, where the user can review it.

[0061] This allows the system to efficiently analyze the contents of contracts and important documents, generate summaries, and provide them to users, thereby reducing the burden on legal departments and speeding up contract signing processes.

[0062] The processing flow will be explained below.

[0063] Step 1:

[0064] The user uses the terminal to upload the document to be analyzed and summarized to the system. For example, the user selects a PDF file of a contract from the terminal interface and clicks the "Upload" button.

[0065] Step 2:

[0066] The server receives the uploaded document. The received document is temporarily stored on the server and its format information (PDF, DOCX, etc.) is acquired.

[0067] Step 3:

[0068] The server pre-processes the document. If the document is in PDF format, it converts it to text format. It also removes unwanted images, redundant text, and other noise from the document.

[0069] Step 4:

[0070] The server determines the language of the document, automatically detecting the language using natural language processing (NLP) techniques and preparing the next analysis step based on the results.

[0071] Step 5:

[0072] The server analyzes the document. Based on the language identified, it uses NLP techniques to analyze each section of the document and extract important keywords and phrases. For example, in the case of a contract, it identifies important information such as "contract duration," "payment terms," ​​and "liability clauses."

[0073] Step 6:

[0074] The server generates a summary based on the analysis results, organizes the extracted important information, and generates a summary of the document. The summary is concise and easy for users to understand.

[0075] Step 7:

[0076] The server determines whether multilingual support is necessary, and if the document language is different from the language specified by the user, it uses an automatic translation function to translate the document or abstract into the specified language.

[0077] Step 8:

[0078] The server sends the generated summary and analysis results to the user's terminal, where the user receives the results and checks them on the screen.

[0079] Step 9:

[0080] The user checks the analysis results and summary on the terminal, and determines whether the contract contents are valid based on the provided summary, and performs additional verification or corrections as necessary.

[0081] This series of steps results in a system that efficiently analyzes the contents of documents and quickly provides users with summaries, reducing the burden on legal departments and streamlining contract execution.

[0082] Example 1

[0083] 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."

[0084] To efficiently analyze documents and generate summaries, it is necessary to process documents in different formats in a unified manner, accurately extract information using natural language processing technology, and generate summaries that include particularly important sections. Furthermore, when documents are written in different languages, it is necessary to perform appropriate translation. These challenges have been difficult to completely address with conventional systems.

[0085] 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.

[0086] In this invention, the server includes: a means for a user to upload a document to the system; a means for the server to receive and store the uploaded document; a means for the server to preprocess the document and convert it into a format suitable for analysis; a means for the server to determine the language of the document; a means for the server to analyze each section of the document using natural language processing technology and extract important information; a means for the server to generate a summary of the document based on the analysis results; a means for the server to send the summary generated by the server to the user's terminal; a means for the user to check the analysis results and the summary via the terminal; and a means for the server to select an appropriate preprocessing method depending on the document format. This makes it possible to consistently and accurately extract information and generate summaries including important sections for documents of different formats and languages.

[0087] "User" means a person who uses the system to upload documents and check analysis results.

[0088] A "server" is a device or system that receives uploaded documents, stores them, and performs various processes such as preprocessing, analysis, summary generation, and translation.

[0089] "Upload" refers to the act of sending a document that a user possesses to the system.

[0090] "Preprocessing" refers to the process of converting documents into a format suitable for analysis, and includes text data extraction and noise removal.

[0091] "Language determination" is the process of automatically identifying the language in which an uploaded document is written.

[0092] "Natural language processing technology" refers to technology for analyzing documents and extracting information, including technology for extracting noun phrases and identifying important phrases.

[0093] "Summary generation" is the process of creating a summary that succinctly summarizes the important points of a document based on the analysis results.

[0094] The "automatic translation function" is a function that automatically translates documents and summaries in different languages ​​into a language specified by the user.

[0095] "Certain Important Sections" are those parts of the document that contain particularly important information, such as the terms of the contract, payment terms, and liability clauses.

[0096] "Analysis results" refer to important information and keywords extracted using natural language processing technology.

[0097] "Terminal" refers to a device used by a user to upload documents or check analysis results, and includes PCs, smartphones, etc.

[0098] "Noise removal" is the process of eliminating information unnecessary for analysis (such as images and duplicated text).

[0099] "Text data" refers to character string data extracted from a document.

[0100] "Tokenization" is the process of dividing the text in a document into units (words or phrases) that are easier to analyze.

[0101] "Part-of-speech tagging" is the process of automatically assigning each word its part of speech.

[0102] The above terms are defined so that the technical scope and embodiments of the invention can be clearly understood.

[0103] The present invention relates to a method for efficiently analyzing documents and generating summaries using a system, which is implemented through a series of steps in which a user uploads a document to the system using a terminal, a server analyzes the document, generates a summary, and sends it back to the user's terminal.

[0104] Hardware and software used

[0105] Server: Uses a computer with high-performance computing resources and provides functions for receiving, storing, preprocessing, analyzing, summarizing, translating, and transmitting documents.

[0106] Device: A PC or smartphone used by a user to upload documents and receive and check summaries.

[0107] Software tools:

[0108] PDF text extraction tools (pdfminer, PyMuPDF)

[0109] Natural language processing libraries (spaCy, NLTK)

[0110] Language detection library (langdetect)

[0111] Translation API (Google (registered trademark) Translation API, DeepL API)

[0112] Explanation of program processing

[0113] 1. Uploading documents

[0114] The user uses a device to upload the document they want to analyze and summarize to the system. For example, they select a PDF file of a contract from the file selection screen on their PC or smartphone and click the "Upload" button. The device then sends the selected document to the server.

[0115] 2. Document Receipt and Storage

[0116] The server receives the document sent from the device and temporarily stores it, identifying the document format (PDF, DOCX, etc.) and obtaining the associated metadata.

[0117] 3. Document Preprocessing

[0118] The server preprocesses the documents. Specifically, it uses tools (such as pdfminer or PyMuPDF) to convert PDF files into text format, extracts the text data, and then performs a filtering process to remove noise (such as images or overlapping text) from the documents.

[0119] 4. Language Identification

[0120] The server automatically determines the language of the document, specifically using libraries such as langdetect or spaCy to determine whether the document is written in English, Japanese, French, etc.

[0121] 5. Document Analysis

[0122] Based on the language identified, the server analyzes the document using natural language processing techniques, such as tokenization and part-of-speech tagging (POS tagging), to extract noun phrases and key phrases (such as contract terms, payment terms, and liability clauses), using libraries such as spaCy and NLTK.

[0123] 6. Generating a Summary

[0124] The server generates a summary of the document based on the analyzed data. The summary is generated using a text summarization algorithm (TextRank or an extractive summarization model). The generated summary includes extracted important information (e.g., contract period: one year, payment terms: payment by the end of each month).

[0125] 7. Multilingual support

[0126] If the language specified by the user is different from the language of the document, the server translates the summary into the specified language using an automatic translation function, for example, using the Google Translate API or DeepL API to translate the summary from English to Japanese.

[0127] 8. Submitting the results

[0128] The server sends the generated summary and analysis results to the user's device, where the user can check the received summary and perform additional verification or corrections as necessary.

[0129] Specific examples

[0130] Example 1: Processing an English contract

[0131] Using a PC, a user selects an English contract in PDF format titled "Service Agreement" and uploads it to the system. The device sends the selected PDF file to the server. The server receives the document and uses pdfminer to extract text data and remove noise. The server uses the langdetect library to determine that the document is written in English. The server uses spaCy to extract important keywords (such as contract term, payment terms, and liability clauses). The server uses the TextRank algorithm to summarize the document. The summary includes phrases such as "Contract term: 1 year" and "Payment terms: Payment by the end of each month." The server then sends the generated summary to the user's PC. The user then reviews the summary on their PC and determines whether the contract terms are reasonable.

[0132] Example 2: Processing multilingual contracts

[0133] A user uploads a French contract titled "Contrat de Service" to the system using a smartphone. The device sends the selected DOCX file to the server. The server receives the document and uses PyMuPDF to extract text data and remove noise. The server uses spaCy's multilingual model to determine that the document is in French. The server uses a French NLP model to extract important keywords. The server generates a summary based on the extracted data. The server uses the DeepL API to translate the French summary into English. The server sends the translated summary to the user's smartphone. The user reviews the summary and understands the contract.

[0134] Prompt Sentence Examples

[0135] Below is an example of a prompt sentence.

[0136] 1. "Generate a summary of the contract written in English."

[0137] 2. "Analyze the uploaded PDF document and provide key keywords and a brief summary."

[0138] 3. "Translate this French document into English and summarize the key points."

[0139] In this way, the present invention improves the efficiency of document analysis and summary generation, thereby improving the efficiency of legal departments and business operations.

[0140] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0141] Step 1:

[0142] A user uses a terminal to upload a document to the system. Specifically, the user selects the document to be processed (e.g., PDF or DOCX format) from the terminal's file selection screen and clicks the "Upload" button. The input of the terminal is the document file selected by the user, and the output is the transmission of the document file to the server.

[0143] Step 2:

[0144] The server receives documents sent from the device and temporarily stores them. The input is the document file sent from the device, and the output is a file stored in the server's storage area. At this time, the server identifies the document format (PDF, DOCX, etc.) and obtains related metadata (file format, file size, etc.).

[0145] Step 3:

[0146] The server preprocesses the documents. Specifically, it uses tools such as pdfminer and PyMuPDF to convert PDF files into text format and extracts the text data. It also performs a filtering process to remove noise (images, overlapping text, etc.) from the documents. The input is the saved document file, and the output is the text data converted into text format and with noise removed.

[0147] Step 4:

[0148] The server automatically determines the language of a document using natural language processing technology. Specifically, it uses libraries such as langdetect and spaCy to determine the language in which the document is written. The input is preprocessed text data, and the output is the document's language information.

[0149] Step 5:

[0150] Based on the determined language, the server uses natural language processing techniques (such as spaCy or NLTK) to analyze the document. Specifically, it performs tokenization and part-of-speech tagging (POS tagging) to extract noun phrases and important phrases (such as contract terms, payment terms, and liability clauses). The input is the document's linguistic information and preprocessed text data, and the output is the extracted important information and keywords.

[0151] Step 6:

[0152] The server generates a document summary based on the extracted data. It uses a text summarization algorithm such as TextRank or an extractive summarization model to generate the summary. The generated summary includes extracted key information (e.g., contract period: one year, payment terms: payment by the end of each month). The input is the extracted key information and keywords, and the output is the generated summary.

[0153] Step 7:

[0154] If the language specified by the user is different from the language of the document, the server translates the summary into the specified language using an automatic translation function (such as Google Translate API or DeepL API). The input is the generated summary and the language information specified by the user, and the output is the translated summary.

[0155] Step 8:

[0156] The server sends the generated summary and analysis results to the user's device. The user checks the received summary on their own device and performs additional verification or correction as necessary. The input is the generated (translated) summary and analysis results, and the output is transmission to the user's device.

[0157] (Application example 1)

[0158] 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."

[0159] In recent years, logistics centers have become required to process large volumes of documents, and there is a demand for the rapid and accurate understanding of the contents of contracts and other important documents. However, manual document analysis and summary generation is time-consuming and places a burden on human resources. Furthermore, when dealing with documents in different languages, translation costs and time increase, resulting in reduced work efficiency. Even with the introduction of systems using smart glasses, the above issues remain unresolved because existing technologies do not automate document content analysis.

[0160] 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.

[0161] In this invention, the server includes: a means for a user to upload a document to the system; a means for the server to receive and store the uploaded document; a means for the server to preprocess the document and convert it into a format suitable for analysis; a means for the server to determine the language of the document; a means for the server to analyze each section of the document using natural language processing technology and extract important keywords and phrases; a means for the server to generate a summary of the document based on the analysis results; a means for the server to send the summary generated by the server to a user's terminal; a means for the user to check the analysis results and the summary through the terminal; and a means for logistics center staff to wear smart glasses and scan contracts or important documents in packages with a camera, causing the system to automatically analyze the documents and display the summaries. This makes it possible to automatically analyze the contents of contracts and important documents and provide summaries quickly and accurately at logistics centers by using smart glasses. The summaries can automatically determine and translate language differences, allowing documents in different languages ​​to be processed efficiently.

[0162] A "user" is an individual or company employee who accesses the system to upload documents and review analysis results.

[0163] "Document" refers to text data containing contracts and other important information, and may be in formats such as PDF or DOCX.

[0164] "System" refers to the set of servers and associated software that receives documents uploaded by users, processes, analyzes, summarizes, and in some cases translates them, and provides the results to users.

[0165] The "server" is the device that forms the core of the system, and is a computer that processes documents received from users, analyzes them, generates summaries, and finally transmits the results to the users.

[0166] "Preprocessing" refers to the initial stage of processing to convert uploaded documents into a state suitable for analysis, and specifically includes text extraction and noise removal.

[0167] "Natural language processing technology" is a technology that enables computers to understand and use human language, specifically, technology that performs text analysis and extracts important keywords.

[0168] A "summary" is a concise text summary of the important information of a document analyzed by the server, presented in a user-friendly format.

[0169] "Translation" refers to the process of converting a document or abstract into a specified language, and is achieved by using an automatic translation function.

[0170] "Smart glasses" are visual devices that have built-in cameras and displays and can present augmented reality (AR) and information when worn by the user.

[0171] A "logistics center" is a facility where goods and packages are collected and prepared for shipment and delivery.

[0172] "Scanning" is the act of taking an image of a document using the camera in the smart glasses and importing it into the system as digital data.

[0173] The present invention is a system in which staff working at a logistics center wear smart glasses and scan contracts and important documents attached to packages with a camera, allowing the system to automatically analyze the documents and display summaries.

[0174] Explanation of the specific operation of the program

[0175] 1. Scanning documents

[0176] A user wears the smart glasses and uses the camera to take an image of a document, allowing work at the logistics center to proceed quickly and efficiently.

[0177] 2. Image to text conversion

[0178] The server receives the document image sent from the smart glasses and uses Tesseract OCR to extract text from the image, which then becomes the base data for analysis.

[0179] 3. Preprocessing and Language Identification

[0180] The server preprocesses the extracted text, removing noise and converting it into a format suitable for analysis, and then automatically determines the language of the document using natural language processing techniques.

[0181] 4. Text Analysis and Summary Generation

[0182] The server then analyzes the text using natural language processing technology such as SpaCy to extract important keywords and phrases based on the language identified, and generates a summary of the document based on the analysis results.

[0183] 5. Translation and result display

[0184] If the document is not in the language specified by the user, the server uses its automatic translation function to translate the summary into the specified language, and the translated summary is sent back to the smart glasses, where the user can view it on the glasses' display.

[0185] Hardware and software used

[0186] Hardware: Smart glasses (camera, display, CPU, memory) and server

[0187] Software: Tesseract OCR, SpaCy, natural language processing technology, translation API

[0188] Specific examples

[0189] Example 1: Processing an English contract

[0190] A staff member at the logistics center wears smart glasses and uses the camera to scan a PDF English contract titled "Service Agreement." The server receives the document, converts it to text format, and removes noise. After determining that the document is written in English, it uses natural language processing technology to extract important information such as the contract period, payment terms, and liability clauses. A summary is generated based on this information and sent to the user's smart glasses. The user then reviews the summary on the smart glasses and determines whether the contract terms are reasonable.

[0191] Example 2: Processing multilingual contracts

[0192] A staff member at a logistics center uses smart glasses to scan a contract written in French with a camera. The server receives the French document and performs the necessary preprocessing. It determines that the document is in French and uses natural language processing technology to extract key information. If the user specifies English, the server uses an automatic translation function to translate the French summary into English. The translated summary is sent to the user's smart glasses, where the user can review it.

[0193] This means that by using smart glasses, it will be possible to automatically analyze the contents of contracts and important documents quickly and accurately on-site at logistics centers and provide summaries.

[0194] Example prompts to input to the generative AI model

[0195] "When a logistics center staff member scans a contract with smart glasses, the text is automatically extracted and displayed."

[0196] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0197] Step 1:

[0198] A user puts on the smart glasses and uses the camera to take an image of a document.

[0199] (Input) An image of a document captured by smart glasses

[0200] (Operation) A user points the camera at a document and presses the shutter button to capture an image of the document.

[0201] (Output) Document image data

[0202] Step 2:

[0203] The terminal transmits the captured image data to the server.

[0204] (Input) Image data of the photographed document

[0205] (Operation) Image data is transferred to the server using a transmission module in the terminal.

[0206] (Output) Image data sent to the server

[0207] Step 3:

[0208] The server converts the received image data into text data using Tesseract OCR.

[0209] (Input) Image data sent to the server

[0210] (Operation) The server calls Tesseract OCR and performs the process of extracting text from the image.

[0211] (Output) Extracted text data

[0212] Step 4:

[0213] The server preprocesses the extracted text data and converts it into a format suitable for analysis.

[0214] (Input) Extracted text data

[0215] (Operation) The server performs preprocessing to remove noise and unnecessary information from the text data.

[0216] (Output) Preprocessed text data

[0217] Step 5:

[0218] The server determines the language of the preprocessed text data using natural language processing techniques.

[0219] (Input) Preprocessed text data

[0220] (Operation) The server uses a natural language processing algorithm to identify the language of the text data.

[0221] (Output) Language information of text data

[0222] Step 6:

[0223] The server analyzes the text data and extracts important keywords and phrases.

[0224] (Input) Text data and its linguistic information

[0225] (Operation) The server uses natural language processing tools such as SpaCy to analyze the text and extract important keywords and phrases.

[0226] (Output) Extracted keywords and phrases

[0227] Step 7:

[0228] The server generates a summary of the document based on important keywords and phrases.

[0229] (Input) Extracted keywords and phrases

[0230] (Operation) The server constructs a summary based on keywords and phrases, concisely summarizing important information.

[0231] (Output) Generated summary

[0232] Step 8:

[0233] If the document is not in the language specified by the user, the server uses an automatic translation function to translate the summary into the specified language.

[0234] (Input) Generated summary and user-specified language information

[0235] (Operation) The server calls the translation API and converts the summary into the specified language.

[0236] (Output) Translated summary

[0237] Step 9:

[0238] The server transmits the generated summary or the translated summary back to the terminal.

[0239] (Input) (Translated) Generated Summary

[0240] (Operation) The server executes a process to transmit the summary data to the terminal.

[0241] (Output) Summary sent to terminal

[0242] Step 10:

[0243] The terminal displays the received summary on the display of the smart glasses.

[0244] (Input) Summary data sent from the server

[0245] (Operation) The terminal formats the summary data for display and displays it on the smart glasses display.

[0246] (Output) Summary displayed on the smart glasses display

[0247] 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.

[0248] This invention relates to a method for efficiently analyzing contracts and other important documents and generating summaries using a system. It also aims to combine an emotion engine that recognizes the user's emotions to analyze and adjust summaries according to the user's emotional state. Specifically, the system involves a series of steps: a user uploads a document to the system using a terminal; a server analyzes the document, generates a summary, and then transmits the summary, adjusted based on the user's emotion, back to the user's terminal.

[0249] Explanation of the specific operation of the program

[0250] 1. Uploading documents

[0251] Users use a device to upload documents to the system for analysis and summarization. The device is equipped with a camera and microphone to collect emotional data through the user's facial expressions and voice.

[0252] 2. Document Receipt and Storage

[0253] The server receives the uploaded document and temporarily stores it. At the same time, the user's emotional data is also recorded. Basic information such as the document format (PDF, DOCX, etc.) and the emotional data is acquired.

[0254] 3. Document Preprocessing

[0255] The server performs pre-processing on the received documents, which includes converting the PDF files to text format and removing noise (e.g., images and redundant text).

[0256] 4. Language Identification

[0257] The server uses natural language processing (NLP) techniques to determine the language of the document, whether it is written in English, Japanese, French, etc.

[0258] 5. Document Analysis

[0259] Based on the language identified, the server uses NLP techniques to analyze the document, specifically extracting important keywords and phrases (e.g., contract terms, payment terms, liability clauses, etc.).

[0260] 6. Generating a Summary

[0261] Based on the analysis results, the server generates a summary of the document, which is a concise summary of the document's key points, including the extracted key points.

[0262] 7. Emotional Engine Adjustment

[0263] The server uses an emotion engine to adapt the analysis results to the emotional data collected from the user, for example, adjusting the results to produce a more concise and clear summary if the user is feeling stressed or confused.

[0264] 8. Multilingual support

[0265] If the language of the document is different from the language specified by the user, the server will use an automatic translation function to translate the document or abstract into the specified language, allowing the user to view the abstract in the language they require.

[0266] 9. Submitting the results

[0267] The server then sends the generated summary and analysis results to the user's device, including summaries adjusted by the emotion engine. The user receives the results and checks them on the screen.

[0268] Specific examples

[0269] Example 1: Processing an English contract

[0270] When a user uploads a PDF English contract titled "Service Agreement" to the system using their PC, a camera and microphone monitor the user's facial expressions and voice to collect emotional data. The server receives the document, converts it to text format, and removes noise. After determining that the document is written in English, it uses NLP technology to extract important information. A summary is generated based on this information and adjusted according to the user's emotional state. The adjusted summary is then sent to the user's PC. The user then reviews the summary on their PC and determines whether the contract terms are appropriate.

[0271] Example 2: Processing multilingual contracts

[0272] When a user uploads a contract written in French to the system using their smartphone, the smartphone's camera and microphone collect the user's emotions. The server receives the French document and performs the necessary preprocessing. It determines the document's language as French and uses NLP technology to extract key information. If the user specifies English, the French summary is translated into English using an automatic translation function. The translated summary is adjusted by the emotion engine and sent to the user's smartphone. The user then reviews the summary and reevaluates the contract content if necessary.

[0273] In this way, a system is realized that efficiently analyzes the content of documents and provides summaries that correspond to the user's emotional state, thereby reducing the burden on legal departments and speeding up contract signing.

[0274] The processing flow will be explained below.

[0275] Step 1:

[0276] The user uploads a document to the system using a device. From the device interface, the user selects the PDF file of the contract they wish to analyze and summarize, and clicks the "Upload" button. At the same time, the device's camera and microphone collect the user's facial expressions and voice data, which are then recorded as emotion data.

[0277] Step 2:

[0278] The server receives the uploaded document and temporarily stores it. The format information of the saved document (e.g., PDF, DOCX) is acquired. The user's emotion data sent from the device is also stored at the same time.

[0279] Step 3:

[0280] The server pre-processes the document: if the document is in PDF format, it performs operations to convert it to text format, and also removes noise from the document, such as unnecessary images and redundant text.

[0281] Step 4:

[0282] The server determines the language of the document. It uses natural language processing (NLP) techniques to automatically detect the language the document is written in. Once the language is determined, it launches an analysis process for that language.

[0283] Step 5:

[0284] The server analyzes the document. Based on the language identified, it uses NLP techniques to analyze each section of the document and extract important keywords and phrases. Specifically, it identifies important information such as contract terms, payment terms, and liability clauses, and extracts the necessary data.

[0285] Step 6:

[0286] The server generates a summary based on the analysis results, organizes the extracted important information, and generates a summary of the document. The summary is concisely compiled in a format that is easy for users to understand.

[0287] Step 7:

[0288] The server uses an emotion engine to analyze the user's emotion data, specifically through facial expression recognition and voice analysis, to determine the user's current emotional state (e.g., stress, excitement, calm).

[0289] Step 8:

[0290] The server then uses the emotion engine to adjust the summary of the document based on the analyzed emotion data. For example, if the user is feeling confused or stressed, the summary will be adjusted to be more concise and clear.

[0291] Step 9:

[0292] If the language of the document is different from the language specified by the user, the server uses an automatic translation function to translate the document or abstract into the specified language.

[0293] Step 10:

[0294] The server sends the generated summary and analysis results to the user's device, including summaries adjusted by the emotion engine. The user receives the results and checks them on the screen.

[0295] Step 11:

[0296] The user checks the analysis results and summary on the terminal, and determines whether the contract contents are valid based on the provided summary, and performs additional verification or corrections as necessary.

[0297] This series of steps results in a system that efficiently analyzes the content of documents and provides summaries tailored to the user's emotional state, reducing the burden on legal departments and speeding up contract execution.

[0298] Example 2

[0299] 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."

[0300] Many systems exist for quickly and accurately analyzing contracts and other important documents and generating summaries. However, these systems lack the ability to adjust their analysis results to take into account the user's emotional state. As a result, if a user is stressed or confused, the summary may be difficult to understand, making it difficult for them to make effective decisions. Another issue is that multilingual support is often limited, meaning that the summary cannot be viewed in the user's preferred language.

[0301] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: a means for a user to upload data to the system; a means for the server to receive and store the uploaded data; a means for the server to preprocess the data and convert it into a format suitable for analysis; a means for the server to determine the language of the data; a means for the server to analyze each portion of the data using natural language processing technology and extract important information; a means for the server to generate a summary of the data based on the analysis results; a means for the server to send the summary generated by the server to the user's terminal; a means for the user to check the analysis results and the summary through the terminal; an emotion data collection means including a sensor attached to the terminal for collecting user emotion data; and a means for the server to analyze the user's emotion data and adjust the generated summary. This allows the server to provide a summary adjusted according to the user's emotional state, support multiple languages, and more quickly and accurately understand the content of a document.

[0302] "User" means any person or entity that uses the System to upload documents or receive summaries.

[0303] "Data" refers to information in any form, including contracts and other important documents.

[0304] "System" refers to the collection of devices and software that provides a set of processes for uploading data, analyzing, generating summaries, translating, and adjusting emotion data.

[0305] "Server" refers to a computing device within the system that receives, stores, pre-processes, analyzes, language-determines, summarizes, adjusts, translates, and transmits results.

[0306] "Terminal" refers to a computer device (PC, smartphone, tablet, etc.) used by a user to upload data and check analysis results.

[0307] "Preprocessing" refers to a series of operations that convert data into a format suitable for analysis, including conversion to text format and noise removal.

[0308] "Language determination" refers to the process of identifying the language in which data is written.

[0309] "Natural language processing technology" refers to computer technology used to analyze data and extract important information.

[0310] "Key information" refers to specific keywords or phrases extracted from each piece of data.

[0311] A "summary" is a document that briefly summarizes the key points of the data.

[0312] "Emotion data" refers to data that indicates the emotional state of a user, collected from facial expressions and voice.

[0313] "Emotion data collection means" refers to a device that collects emotional data using sensors, cameras, and microphones attached to a terminal.

[0314] "Tuning" refers to the process of optimizing summaries and analysis results based on the user's emotional state.

[0315] "Automatic translation" refers to technology for translating data or summaries into another language.

[0316] This invention relates to a method for efficiently analyzing contracts and other important documents and generating summaries using a system. Furthermore, it aims to combine this system with an emotion engine that recognizes the user's emotions to adjust the analysis and summaries according to the user's emotional state. Specific embodiments for implementing this invention are described below.

[0317] System configuration

[0318] Document upload

[0319] Users upload documents such as contracts to the system using devices (PCs, smartphones, tablets, etc.). The devices are equipped with cameras and microphones, allowing for the collection of emotional data through the user's facial expressions and voice.

[0320] Document Receipt and Storage

[0321] The server receives the uploaded document and temporarily stores it in storage. At this time, basic information such as the document format (PDF, DOCX, etc.) and emotion data is acquired and recorded as metadata.

[0322] Document Preprocessing

[0323] The server performs pre-processing on the received documents, which includes converting the PDF files to text format and removing noise (e.g., images and overlapping text) using specific document analysis software (e.g., OCR technology).

[0324] Language Identification

[0325] The server uses natural language processing (NLP) techniques to determine the language of the document, whether it is written in English, Japanese, French, etc. The techniques used include, for example, language identification algorithms.

[0326] Document analysis and summary generation

[0327] The server analyzes the document using NLP technology based on the determined language. It extracts important keywords and phrases (e.g., contract terms, payment terms, liability clauses, etc.) and generates a document summary based on the analysis results. The summary is a concise summary of the document's key points. Specific technologies used include keyword extraction algorithms and summary generation models.

[0328] Emotional Engine Adjustment

[0329] The server uses an emotion engine to analyze the emotional data collected from the user and incorporate it into the analysis results. For example, if the user is feeling stressed or confused, the server will adjust the summary to generate a more concise and clear summary. Facial recognition software and voice analysis technology are used to analyze the emotional data.

[0330] Multilingual support and results transmission

[0331] If the document language is different from the language specified by the user, the server uses an automatic translation function to translate the document or summary into the specified language. This function allows the user to view the summary in the language they require. This is achieved using a translation API. The generated summary and analysis results are then sent to the user's device, where the user can view the results.

[0332] Specific examples

[0333] Example 1: Processing an English contract

[0334] When a user uploads a PDF English contract titled "Service Agreement" to the system using their PC, a camera and microphone monitor the user's facial expressions and voice to collect emotional data. The server receives the document, converts it to text format, and removes noise. After determining that the document is written in English, it uses NLP technology to extract key information. A summary is generated based on this information and adjusted according to the user's emotional state. The server then sends the adjusted summary to the user's PC. The user then reviews the summary on their PC and determines whether the contract terms are appropriate.

[0335] Example 2: Processing multilingual contracts

[0336] When a user uploads a contract written in French to the system using their smartphone, the smartphone's camera and microphone collect the user's emotions. The server receives the French document and performs the necessary preprocessing. It determines the document's language as French and uses NLP technology to extract key information. If the user specifies English, the French summary is translated into English using an automatic translation function. The translated summary is then adjusted by the emotion engine and sent to the user's smartphone. The user then reviews the summary and reevaluates the contract content if necessary.

[0337] Prompt Sentence Examples

[0338] "Please upload the English contract and summarize the key points. Also, please adjust it to take into account sentiment data."

[0339] "I have uploaded a document in French. Please translate it into English and provide a summary."

[0340] In this way, the system can efficiently analyze contracts and important documents and provide summaries tailored to the user's emotional state, reducing the burden on legal departments and speeding up contract closures.

[0341] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0342] Step 1: Upload your documents

[0343] The user uses a terminal to upload a document to the system. The input is a contract file selected by the user (e.g., a PDF file), and the output is the transfer of the file to the system. The terminal is equipped with a camera and microphone, which also collects emotional data from the user's facial expressions and voice. Specifically, the user clicks the "Upload" button and selects a document file.

[0344] Step 2: Receiving and storing documents

[0345] The server receives the uploaded document and temporarily stores it in storage. The input is the uploaded contract file and emotion data, and the output is the file stored on the server and the collected emotion data meta information. Specifically, the server receives the file and stores it in storage with a file name such as "contract.pdf." At this time, basic information such as the document format and emotion data is also saved.

[0346] Step 3: Preprocessing the Documents

[0347] The server performs preprocessing on the received document. The input is the saved contract file, and the output is text data suitable for analysis. Specifically, it converts the PDF file to text and removes noise such as images and overlapping text. Specifically, it uses OCR technology to extract text from the PDF and runs a script to remove unnecessary noise.

[0348] Step 4: Language Determination

[0349] The server uses natural language processing (NLP) techniques to determine the language of a document. The input is preprocessed text data, and the output is the document's language information (e.g., English, Japanese, French, etc.). Specifically, it analyzes the frequency of words and phrases in the document and runs a model to identify the language based on that.

[0350] Step 5: Parsing the Document

[0351] The server then uses NLP technology to analyze the document based on the language it has identified. The input is linguistic information and preprocessed text data, and the output is key keywords and phrases extracted through the analysis. Specifically, it runs an algorithm to extract keywords such as contract terms, payment terms, and liability clauses.

[0352] Step 6: Generate a summary

[0353] The server generates a document summary based on the analysis results. The input is the extracted keywords and phrases, and the output is the document summary. Specifically, it runs a script that generates a document summary based on the extracted keywords and an existing format.

[0354] Step 7: Emotional Engine Alignment

[0355] The server uses an emotion engine to adjust the summary based on the user's emotion data. The input is the generated summary and emotion data, and the output is a summary adjusted to reflect the emotion. Specifically, if the user's facial expression data indicates "stress," the server runs an algorithm to reconstruct the summary into simpler language.

[0356] Step 8: Multilingual

[0357] If the document language is different from the language specified by the user, the server uses an automatic translation function to translate the summary into the specified language. The input is the adjusted summary, and the output is the translated summary. Specifically, the server calls an API to translate the summary generated in French into English.

[0358] Step 9: Sending the results

[0359] The server sends the generated summary and analysis results to the user's device. The input is the adjusted and translated summary, and the output is the summary sent to the user's device. Specifically, the generated summary and analysis results are sent to the user's device in a data format such as JSON, and the device displays them.

[0360] (Application example 2)

[0361] 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."

[0362] Conventional document analysis systems can analyze documents and generate summaries, but they do not take into account the user's emotional state, resulting in the inability to provide optimal information to users. Furthermore, they are unable to provide appropriate support when users feel stressed or confused, and are particularly inadequate in situations requiring quick and accurate judgments, particularly in the security service field. There is a need for a system that can improve this situation by adjusting document analysis results and summaries based on the user's emotional state.

[0363] 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 a user to upload a document to the system; means for the server to receive and store the uploaded document; means for the server to preprocess the document and convert it into a format suitable for analysis; means for the server to determine the language of the document; means for the server to analyze each section of the document using natural language processing technology and extract important keywords and phrases; means for the server to generate a summary of the document based on the analysis results; means for the server to recognize the user's emotional state and adjust the summary based on that state; means for the server to send the summary generated by the server to the user's terminal; and means for the user to check the analysis results and the summary through the terminal. This makes it possible to provide a summary adjusted according to the user's emotional state.

[0364] "User" refers to the person who uploads documents to the analysis system and reviews the analysis results and summaries.

[0365] "Document" refers to the documents or information to be analyzed and summarized, and may exist in formats such as PDF or DOCX.

[0366] "System" refers to the entire device and software configuration that performs a series of processes in which a user uploads a document and the server analyzes and summarizes it.

[0367] "Server" refers to a computer and associated software for receiving, storing, and processing documents uploaded by users.

[0368] "Preprocessing" refers to a series of operations to convert a document into a format suitable for analysis, such as converting PDF files to text or removing noise.

[0369] "Natural language processing technology" is a technology in the field of computer science that refers to methods and algorithms that allow machines to understand, analyze, and generate human language.

[0370] "Emotional state" refers to the psychological or emotional state recognized from the user's facial expression or voice, and includes, for example, stress, happiness, surprise, etc.

[0371] A "summary" is a concise summary of the important points from an analyzed document that can be quickly understood by a user.

[0372] "Upload" refers to the operation of a user sending a document from their own terminal to the system.

[0373] "Terminal" refers to a device used by a user, such as a smartphone or PC, that interfaces with the system.

[0374] "Adjustment" refers to making changes or modifications to the content of an already generated summary based on the user's emotional state.

[0375] "Automatic translation function" refers to the function by which the server mechanically translates documents and summaries into a language specified by the user.

[0376] "Important sections" are those parts of the document that require special attention, such as specific matters like contract terms, payment terms, and liability clauses.

[0377] This invention is a system for efficiently analyzing documents and generating summaries in the field of security services, and can further adjust the summaries according to the emotional state of the user. The system is implemented through the following steps:

[0378] composition

[0379] 1. Hardware configuration:

[0380] The user's device is a smartphone or a PC, which is equipped with a camera and a microphone.

[0381] The server is equipped with a high-performance CPU and sufficient memory.

[0382] Data storage requires large amounts of storage for analysis.

[0383] 2. Software configuration:

[0384] For natural language processing technology, Google Cloud Natural Language API is used.

[0385] Amazon Rekognition is used for sentiment analysis.

[0386] The automatic translation function uses the Google Translate API.

[0387] The server software uses Apache (registered trademark) HTTP Server and Node.js.

[0388] MySQL (registered trademark) is used as the database.

[0389] Explanation of program processing

[0390] The server receives documents uploaded by users, analyzes them, and summarizes them. This process involves the following major steps:

[0391] 1. Upload and save documents:

[0392] Users upload documents to the system via their smartphones or PCs. The server receives and stores them, and the documents are used for analysis and summary generation.

[0393] 2. Document preprocessing:

[0394] The server pre-processes the uploaded documents, which includes converting PDF files to text format and removing unwanted noise.

[0395] 3. Document Language Determination:

[0396] The server uses the Google Cloud Natural Language API to determine the language of the document, which tells it what language it is written in.

[0397] 4. Document Analysis:

[0398] The server uses natural language processing techniques to analyze the content of the document and extract important keywords and phrases.

[0399] 5. Generate a summary:

[0400] The server generates a summary of the document based on the extracted keywords and phrases, which is concise and easy for users to understand.

[0401] 6. Emotional engine regulation:

[0402] The server uses Amazon Rekognition to analyze the user's emotional data and adjusts the generated summary according to the user's emotional state—for example, if the user is feeling stressed, the summary will be more concise and clear.

[0403] 7. Submit and review your abstract:

[0404] The server then sends the adjusted summary to the user's device, where the user can check the summary on their smartphone or computer and take any necessary action.

[0405] Specific examples

[0406] For example, if a lawyer uploads a security services contract via smartphone, the system analyzes the document and summarizes key sections. Furthermore, if the lawyer is under stress, the summary can be adjusted to be concise and clear. This process ensures fast and accurate information delivery.

[0407] Prompt Sentence Examples

[0408] Analyze and summarize the contracts uploaded to the system. Adjust the summary and reply based on the user's sentiment.

[0409] In this way, the present invention can efficiently analyze the content of documents and provide summaries according to the emotional state of the user, thereby improving reliability and efficiency in the field of security services.

[0410] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0411] Step 1:

[0412] The user uploads a document to the system. The user uses a smartphone or PC to upload the document to be analyzed to the system. The input is a document file (PDF, DOCX, etc.), and the output is the data uploaded to the server.

[0413] Step 2:

[0414] The server receives and stores the uploaded document. The server receives the document file sent by the user and stores it in a database. The input is the document file sent by the user, and the output is the document data stored in the database.

[0415] Step 3:

[0416] The server preprocesses the document and converts it into a format suitable for analysis. The server converts the document into a text format and removes unnecessary noise (images, duplicated text, etc.). The input is the document data stored in the database, and the output is the preprocessed text data.

[0417] Step 4:

[0418] The server determines the language of the document. The server determines the language of the document using the Google Cloud Natural Language API. The input is preprocessed text data, and the output is the language information of the document.

[0419] Step 5:

[0420] The server uses natural language processing technology to analyze each section of the document and extract important keywords and phrases. The server uses natural language processing technology (Google Cloud Natural Language API) to analyze the document and extract important keywords and phrases. The input is the text data of the document, and the output is the extracted keywords and phrases.

[0421] Step 6:

[0422] The server generates a document summary based on the analysis results. The server generates a document summary based on the extracted keywords and phrases. The input is the extracted keywords and phrases, and the output is the generated summary.

[0423] Step 7:

[0424] The server recognizes the user's emotional state and adjusts the summary based on that state. The server uses Amazon Rekognition to analyze the user's emotional state and adjust the generated summary. The input is the user's emotional data (facial expressions and voice) and the generated summary, and the output is the adjusted summary.

[0425] Step 8:

[0426] The server sends the generated summary to the user's terminal. The server sends the adjusted summary to the user's terminal. The input is the adjusted summary, and the output is the summary sent to the user's terminal.

[0427] Step 9:

[0428] The user checks the analysis results and summary via their device. The user checks the analysis results and summary sent from the server via their smartphone or PC. The input is the summary sent from the server, and the output is the user's understanding and any necessary actions.

[0429] 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.

[0430] 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.

[0431] 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.

[0432] [Second embodiment]

[0433] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0434] 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.

[0435] 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).

[0436] 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.

[0437] 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.

[0438] 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).

[0439] 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. 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.

[0440] 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.

[0441] 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.

[0442] 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.

[0443] In the smart glasses 214, 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.

[0444] 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."

[0445] This invention relates to a method for efficiently analyzing contracts and other important documents and generating summaries using a system, in which a user uploads the document to the system using a terminal, a server analyzes the document, generates a summary, and sends it back to the user's terminal.

[0446] Explanation of the specific operation of the program

[0447] 1. Uploading documents

[0448] Users use their devices to upload documents they wish to analyze and summarize to the system, for example by selecting a PDF file of a contract from a PC, smartphone, or other device, and clicking the upload button.

[0449] 2. Document Receipt and Storage

[0450] The server receives the uploaded document and temporarily stores it, at which point it obtains basic information about the document's format (e.g. PDF, DOCX).

[0451] 3. Document Preprocessing

[0452] The server performs pre-processing on the received documents, which includes converting the PDF files to text format and removing noise (e.g., images and redundant text).

[0453] 4. Language Identification

[0454] The server uses natural language processing (NLP) techniques to automatically determine the language of the document, whether it is written in English, Japanese, French, etc.

[0455] 5. Document Analysis

[0456] Based on the language identified, the server uses NLP technology to analyze the document, extracting important keywords and phrases (e.g., contract terms, payment terms, liability clauses, etc.), and analyzes the document structure, dividing it into sections for analysis.

[0457] 6. Generating a Summary

[0458] The server then generates a summary of the document based on the analysis results. The summary is a concise summary of the document's key points, including the most important information extracted. For example, a contract summary might include information such as "Contract period: one year" and "Payment terms: payment by the end of each month."

[0459] 7. Multilingual support

[0460] If the language of the document is different from the language specified by the user, the server will use an automatic translation function to translate the document and summary into the specified language, allowing the user to view the summary in the language they require.

[0461] 8. Submitting the results

[0462] The server sends the generated summary and analysis results to the user's terminal, where the user can check the summary received and perform additional verification or correction as necessary.

[0463] Specific examples

[0464] Example 1: Processing an English contract

[0465] A user uploads an English contract titled "Service Agreement" in PDF format to the system using their PC. The server receives the document, converts it to text format, and removes noise. After determining that the document is in English, it uses NLP technology to extract important information such as the contract period, payment terms, and liability clauses. A summary is generated based on this information and sent to the user's PC. The user then checks the summary on their PC and determines whether the contract contents are reasonable.

[0466] Example 2: Processing multilingual contracts

[0467] A user uploads a contract written in French to the system using their smartphone. The server receives the French document and performs the necessary preprocessing. It determines that the document is in French and uses NLP technology to extract important information. If the user specifies English, the server uses an automatic translation function to translate the French summary into English. The translated summary is sent to the user's smartphone, where the user can review it.

[0468] This allows the system to efficiently analyze the contents of contracts and important documents, generate summaries, and provide them to users, thereby reducing the burden on legal departments and speeding up contract signing processes.

[0469] The processing flow will be explained below.

[0470] Step 1:

[0471] The user uses the terminal to upload the document to be analyzed and summarized to the system. For example, the user selects a PDF file of a contract from the terminal interface and clicks the "Upload" button.

[0472] Step 2:

[0473] The server receives the uploaded document. The received document is temporarily stored on the server and its format information (PDF, DOCX, etc.) is acquired.

[0474] Step 3:

[0475] The server pre-processes the document. If the document is in PDF format, it converts it to text format. It also removes unwanted images, redundant text, and other noise from the document.

[0476] Step 4:

[0477] The server determines the language of the document, automatically detecting the language using natural language processing (NLP) techniques and preparing the next analysis step based on the results.

[0478] Step 5:

[0479] The server analyzes the document. Based on the language identified, it uses NLP techniques to analyze each section of the document and extract important keywords and phrases. For example, in the case of a contract, it identifies important information such as "contract duration," "payment terms," ​​and "liability clauses."

[0480] Step 6:

[0481] The server generates a summary based on the analysis results, organizes the extracted important information, and generates a summary of the document. The summary is concise and easy for users to understand.

[0482] Step 7:

[0483] The server determines whether multilingual support is necessary, and if the document language is different from the language specified by the user, it uses an automatic translation function to translate the document or abstract into the specified language.

[0484] Step 8:

[0485] The server sends the generated summary and analysis results to the user's terminal, where the user receives the results and checks them on the screen.

[0486] Step 9:

[0487] The user checks the analysis results and summary on the terminal, and determines whether the contract contents are valid based on the provided summary, and performs additional verification or corrections as necessary.

[0488] This series of steps results in a system that efficiently analyzes the contents of documents and quickly provides users with summaries, reducing the burden on legal departments and streamlining contract execution.

[0489] Example 1

[0490] 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."

[0491] To efficiently analyze documents and generate summaries, it is necessary to process documents in different formats in a unified manner, accurately extract information using natural language processing technology, and generate summaries that include particularly important sections. Furthermore, when documents are written in different languages, it is necessary to perform appropriate translation. These challenges have been difficult to completely address with conventional systems.

[0492] 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.

[0493] In this invention, the server includes: a means for a user to upload a document to the system; a means for the server to receive and store the uploaded document; a means for the server to preprocess the document and convert it into a format suitable for analysis; a means for the server to determine the language of the document; a means for the server to analyze each section of the document using natural language processing technology and extract important information; a means for the server to generate a summary of the document based on the analysis results; a means for the server to send the summary generated by the server to the user's terminal; a means for the user to check the analysis results and the summary via the terminal; and a means for the server to select an appropriate preprocessing method depending on the document format. This makes it possible to consistently and accurately extract information and generate summaries including important sections for documents of different formats and languages.

[0494] "User" means a person who uses the system to upload documents and check analysis results.

[0495] A "server" is a device or system that receives uploaded documents, stores them, and performs various processes such as preprocessing, analysis, summary generation, and translation.

[0496] "Upload" refers to the act of sending a document that a user possesses to the system.

[0497] "Preprocessing" refers to the process of converting documents into a format suitable for analysis, and includes text data extraction and noise removal.

[0498] "Language determination" is the process of automatically identifying the language in which an uploaded document is written.

[0499] "Natural language processing technology" refers to technology for analyzing documents and extracting information, including technology for extracting noun phrases and identifying important phrases.

[0500] "Summary generation" is the process of creating a summary that succinctly summarizes the important points of a document based on the analysis results.

[0501] The "automatic translation function" is a function that automatically translates documents and summaries in different languages ​​into a language specified by the user.

[0502] "Certain Important Sections" are those parts of the document that contain particularly important information, such as the terms of the contract, payment terms, and liability clauses.

[0503] "Analysis results" refer to important information and keywords extracted using natural language processing technology.

[0504] "Terminal" refers to a device used by a user to upload documents or check analysis results, and includes PCs, smartphones, etc.

[0505] "Noise removal" is the process of eliminating information unnecessary for analysis (such as images and duplicated text).

[0506] "Text data" refers to character string data extracted from a document.

[0507] "Tokenization" is the process of dividing the text in a document into units (words or phrases) that are easier to analyze.

[0508] "Part-of-speech tagging" is the process of automatically assigning each word its part of speech.

[0509] The above terms are defined so that the technical scope and embodiments of the invention can be clearly understood.

[0510] The present invention relates to a method for efficiently analyzing documents and generating summaries using a system, which is implemented through a series of steps in which a user uploads a document to the system using a terminal, a server analyzes the document, generates a summary, and sends it back to the user's terminal.

[0511] Hardware and software used

[0512] Server: Uses a computer with high-performance computing resources and provides functions for receiving, storing, preprocessing, analyzing, summarizing, translating, and transmitting documents.

[0513] Device: A PC or smartphone used by a user to upload documents and receive and check summaries.

[0514] Software tools:

[0515] PDF text extraction tools (pdfminer, PyMuPDF)

[0516] Natural language processing libraries (spaCy, NLTK)

[0517] Language detection library (langdetect)

[0518] Translation API (Google Translate API, DeepL API)

[0519] Explanation of program processing

[0520] 1. Uploading documents

[0521] The user uses a device to upload the document they want to analyze and summarize to the system. For example, they select a PDF file of a contract from the file selection screen on their PC or smartphone and click the "Upload" button. The device then sends the selected document to the server.

[0522] 2. Document Receipt and Storage

[0523] The server receives the document sent from the device and temporarily stores it, identifying the document format (PDF, DOCX, etc.) and obtaining the associated metadata.

[0524] 3. Document Preprocessing

[0525] The server preprocesses the documents. Specifically, it uses tools (such as pdfminer or PyMuPDF) to convert PDF files into text format, extracts the text data, and then performs a filtering process to remove noise (such as images or overlapping text) from the documents.

[0526] 4. Language Identification

[0527] The server automatically determines the language of the document, specifically using libraries such as langdetect or spaCy to determine whether the document is written in English, Japanese, French, etc.

[0528] 5. Document Analysis

[0529] Based on the language identified, the server analyzes the document using natural language processing techniques, such as tokenization and part-of-speech tagging (POS tagging), to extract noun phrases and key phrases (such as contract terms, payment terms, and liability clauses), using libraries such as spaCy and NLTK.

[0530] 6. Generating a Summary

[0531] The server generates a summary of the document based on the analyzed data. The summary is generated using a text summarization algorithm (TextRank or an extractive summarization model). The generated summary includes extracted important information (e.g., contract period: one year, payment terms: payment by the end of each month).

[0532] 7. Multilingual support

[0533] If the language specified by the user is different from the language of the document, the server translates the summary into the specified language using an automatic translation function, for example, using the Google Translate API or DeepL API to translate the summary from English to Japanese.

[0534] 8. Submitting the results

[0535] The server sends the generated summary and analysis results to the user's device, where the user can check the received summary and perform additional verification or corrections as necessary.

[0536] Specific examples

[0537] Example 1: Processing an English contract

[0538] Using a PC, a user selects an English contract in PDF format titled "Service Agreement" and uploads it to the system. The device sends the selected PDF file to the server. The server receives the document and uses pdfminer to extract text data and remove noise. The server uses the langdetect library to determine that the document is written in English. The server uses spaCy to extract important keywords (such as contract term, payment terms, and liability clauses). The server uses the TextRank algorithm to summarize the document. The summary includes phrases such as "Contract term: 1 year" and "Payment terms: Payment by the end of each month." The server then sends the generated summary to the user's PC. The user then reviews the summary on their PC and determines whether the contract terms are reasonable.

[0539] Example 2: Processing multilingual contracts

[0540] A user uploads a French contract titled "Contrat de Service" to the system using a smartphone. The device sends the selected DOCX file to the server. The server receives the document and uses PyMuPDF to extract text data and remove noise. The server uses spaCy's multilingual model to determine that the document is in French. The server uses a French NLP model to extract important keywords. The server generates a summary based on the extracted data. The server uses the DeepL API to translate the French summary into English. The server sends the translated summary to the user's smartphone. The user reviews the summary and understands the contract.

[0541] Prompt Sentence Examples

[0542] Below is an example of a prompt sentence.

[0543] 1. "Generate a summary of the contract written in English."

[0544] 2. "Analyze the uploaded PDF document and provide key keywords and a brief summary."

[0545] 3. "Translate this French document into English and summarize the key points."

[0546] In this way, the present invention improves the efficiency of document analysis and summary generation, thereby improving the efficiency of legal departments and business operations.

[0547] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0548] Step 1:

[0549] A user uses a terminal to upload a document to the system. Specifically, the user selects the document to be processed (e.g., PDF or DOCX format) from the terminal's file selection screen and clicks the "Upload" button. The input of the terminal is the document file selected by the user, and the output is the transmission of the document file to the server.

[0550] Step 2:

[0551] The server receives documents sent from the device and temporarily stores them. The input is the document file sent from the device, and the output is a file stored in the server's storage area. At this time, the server identifies the document format (PDF, DOCX, etc.) and obtains related metadata (file format, file size, etc.).

[0552] Step 3:

[0553] The server preprocesses the documents. Specifically, it uses tools such as pdfminer and PyMuPDF to convert PDF files into text format and extracts the text data. It also performs a filtering process to remove noise (images, overlapping text, etc.) from the documents. The input is the saved document file, and the output is the text data converted into text format and with noise removed.

[0554] Step 4:

[0555] The server automatically determines the language of a document using natural language processing technology. Specifically, it uses libraries such as langdetect and spaCy to determine the language in which the document is written. The input is preprocessed text data, and the output is the document's language information.

[0556] Step 5:

[0557] Based on the determined language, the server uses natural language processing techniques (such as spaCy or NLTK) to analyze the document. Specifically, it performs tokenization and part-of-speech tagging (POS tagging) to extract noun phrases and important phrases (such as contract terms, payment terms, and liability clauses). The input is the document's linguistic information and preprocessed text data, and the output is the extracted important information and keywords.

[0558] Step 6:

[0559] The server generates a document summary based on the extracted data. It uses a text summarization algorithm such as TextRank or an extractive summarization model to generate the summary. The generated summary includes extracted key information (e.g., contract period: one year, payment terms: payment by the end of each month). The input is the extracted key information and keywords, and the output is the generated summary.

[0560] Step 7:

[0561] If the language specified by the user is different from the language of the document, the server translates the summary into the specified language using an automatic translation function (such as Google Translate API or DeepL API). The input is the generated summary and the language information specified by the user, and the output is the translated summary.

[0562] Step 8:

[0563] The server sends the generated summary and analysis results to the user's device. The user checks the received summary on their own device and performs additional verification or correction as necessary. The input is the generated (translated) summary and analysis results, and the output is transmission to the user's device.

[0564] (Application example 1)

[0565] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0566] In recent years, logistics centers have become required to process large volumes of documents, and there is a demand for the rapid and accurate understanding of the contents of contracts and other important documents. However, manual document analysis and summary generation is time-consuming and places a burden on human resources. Furthermore, when dealing with documents in different languages, translation costs and time increase, resulting in reduced work efficiency. Even with the introduction of systems using smart glasses, the above issues remain unresolved because existing technologies do not automate document content analysis.

[0567] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0568] In this invention, the server includes: a means for a user to upload a document to the system; a means for the server to receive and store the uploaded document; a means for the server to preprocess the document and convert it into a format suitable for analysis; a means for the server to determine the language of the document; a means for the server to analyze each section of the document using natural language processing technology and extract important keywords and phrases; a means for the server to generate a summary of the document based on the analysis results; a means for the server to send the summary generated by the server to a user's terminal; a means for the user to check the analysis results and the summary through the terminal; and a means for logistics center staff to wear smart glasses and scan contracts or important documents in packages with a camera, causing the system to automatically analyze the documents and display the summaries. This makes it possible to automatically analyze the contents of contracts and important documents and provide summaries quickly and accurately at logistics centers by using smart glasses. The summaries can automatically determine and translate language differences, allowing documents in different languages ​​to be processed efficiently.

[0569] A "user" is an individual or company employee who accesses the system to upload documents and review analysis results.

[0570] "Document" refers to text data containing contracts and other important information, and may be in formats such as PDF or DOCX.

[0571] "System" refers to the set of servers and associated software that receives documents uploaded by users, processes, analyzes, summarizes, and in some cases translates them, and provides the results to users.

[0572] The "server" is the device that forms the core of the system, and is a computer that processes documents received from users, analyzes them, generates summaries, and finally transmits the results to the users.

[0573] "Preprocessing" refers to the initial stage of processing to convert uploaded documents into a state suitable for analysis, and specifically includes text extraction and noise removal.

[0574] "Natural language processing technology" is a technology that enables computers to understand and use human language, specifically, technology that performs text analysis and extracts important keywords.

[0575] A "summary" is a concise text summary of the important information of a document analyzed by the server, presented in a user-friendly format.

[0576] "Translation" refers to the process of converting a document or abstract into a specified language, and is achieved by using an automatic translation function.

[0577] "Smart glasses" are visual devices that have built-in cameras and displays and can present augmented reality (AR) and information when worn by the user.

[0578] A "logistics center" is a facility where goods and packages are collected and prepared for shipment and delivery.

[0579] "Scanning" is the act of taking an image of a document using the camera in the smart glasses and importing it into the system as digital data.

[0580] The present invention is a system in which staff working at a logistics center wear smart glasses and scan contracts and important documents attached to packages with a camera, allowing the system to automatically analyze the documents and display summaries.

[0581] Explanation of the specific operation of the program

[0582] 1. Scanning documents

[0583] A user wears the smart glasses and uses the camera to take an image of a document, allowing work at the logistics center to proceed quickly and efficiently.

[0584] 2. Image to text conversion

[0585] The server receives the document image sent from the smart glasses and uses Tesseract OCR to extract text from the image, which then becomes the base data for analysis.

[0586] 3. Preprocessing and Language Identification

[0587] The server preprocesses the extracted text, removing noise and converting it into a format suitable for analysis, and then automatically determines the language of the document using natural language processing techniques.

[0588] 4. Text Analysis and Summary Generation

[0589] The server then analyzes the text using natural language processing technology such as SpaCy to extract important keywords and phrases based on the language identified, and generates a summary of the document based on the analysis results.

[0590] 5. Translation and result display

[0591] If the document is not in the language specified by the user, the server uses its automatic translation function to translate the summary into the specified language, and the translated summary is sent back to the smart glasses, where the user can view it on the glasses' display.

[0592] Hardware and software used

[0593] Hardware: Smart glasses (camera, display, CPU, memory) and server

[0594] Software: Tesseract OCR, SpaCy, natural language processing technology, translation API

[0595] Specific examples

[0596] Example 1: Processing an English contract

[0597] A staff member at the logistics center wears smart glasses and uses the camera to scan a PDF English contract titled "Service Agreement." The server receives the document, converts it to text format, and removes noise. After determining that the document is written in English, it uses natural language processing technology to extract important information such as the contract period, payment terms, and liability clauses. A summary is generated based on this information and sent to the user's smart glasses. The user then reviews the summary on the smart glasses and determines whether the contract terms are reasonable.

[0598] Example 2: Processing multilingual contracts

[0599] A staff member at a logistics center uses smart glasses to scan a contract written in French with a camera. The server receives the French document and performs the necessary preprocessing. It determines that the document is in French and uses natural language processing technology to extract key information. If the user specifies English, the server uses an automatic translation function to translate the French summary into English. The translated summary is sent to the user's smart glasses, where the user can review it.

[0600] This means that by using smart glasses, it will be possible to automatically analyze the contents of contracts and important documents quickly and accurately on-site at logistics centers and provide summaries.

[0601] Example prompts to input to the generative AI model

[0602] "When a logistics center staff member scans a contract with smart glasses, the text is automatically extracted and displayed."

[0603] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0604] Step 1:

[0605] A user puts on the smart glasses and uses the camera to take an image of a document.

[0606] (Input) An image of a document captured by smart glasses

[0607] (Operation) A user points the camera at a document and presses the shutter button to capture an image of the document.

[0608] (Output) Document image data

[0609] Step 2:

[0610] The terminal transmits the captured image data to the server.

[0611] (Input) Image data of the photographed document

[0612] (Operation) Image data is transferred to the server using a transmission module in the terminal.

[0613] (Output) Image data sent to the server

[0614] Step 3:

[0615] The server converts the received image data into text data using Tesseract OCR.

[0616] (Input) Image data sent to the server

[0617] (Operation) The server calls Tesseract OCR and performs the process of extracting text from the image.

[0618] (Output) Extracted text data

[0619] Step 4:

[0620] The server preprocesses the extracted text data and converts it into a format suitable for analysis.

[0621] (Input) Extracted text data

[0622] (Operation) The server performs preprocessing to remove noise and unnecessary information from the text data.

[0623] (Output) Preprocessed text data

[0624] Step 5:

[0625] The server determines the language of the preprocessed text data using natural language processing techniques.

[0626] (Input) Preprocessed text data

[0627] (Operation) The server uses a natural language processing algorithm to identify the language of the text data.

[0628] (Output) Language information of text data

[0629] Step 6:

[0630] The server analyzes the text data and extracts important keywords and phrases.

[0631] (Input) Text data and its linguistic information

[0632] (Operation) The server uses natural language processing tools such as SpaCy to analyze the text and extract important keywords and phrases.

[0633] (Output) Extracted keywords and phrases

[0634] Step 7:

[0635] The server generates a summary of the document based on important keywords and phrases.

[0636] (Input) Extracted keywords and phrases

[0637] (Operation) The server constructs a summary based on keywords and phrases, concisely summarizing important information.

[0638] (Output) Generated summary

[0639] Step 8:

[0640] If the document is not in the language specified by the user, the server uses an automatic translation function to translate the summary into the specified language.

[0641] (Input) Generated summary and user-specified language information

[0642] (Operation) The server calls the translation API and converts the summary into the specified language.

[0643] (Output) Translated summary

[0644] Step 9:

[0645] The server transmits the generated summary or the translated summary back to the terminal.

[0646] (Input) (Translated) Generated Summary

[0647] (Operation) The server executes a process to transmit the summary data to the terminal.

[0648] (Output) Summary sent to terminal

[0649] Step 10:

[0650] The terminal displays the received summary on the display of the smart glasses.

[0651] (Input) Summary data sent from the server

[0652] (Operation) The terminal formats the summary data for display and displays it on the smart glasses display.

[0653] (Output) Summary displayed on the smart glasses display

[0654] 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.

[0655] This invention relates to a method for efficiently analyzing contracts and other important documents and generating summaries using a system. It also aims to combine an emotion engine that recognizes the user's emotions to analyze and adjust summaries according to the user's emotional state. Specifically, the system involves a series of steps: a user uploads a document to the system using a terminal; a server analyzes the document, generates a summary, and then transmits the summary, adjusted based on the user's emotion, back to the user's terminal.

[0656] Explanation of the specific operation of the program

[0657] 1. Uploading documents

[0658] Users use a device to upload documents to the system for analysis and summarization. The device is equipped with a camera and microphone to collect emotional data through the user's facial expressions and voice.

[0659] 2. Document Receipt and Storage

[0660] The server receives the uploaded document and temporarily stores it. At the same time, the user's emotional data is also recorded. Basic information such as the document format (PDF, DOCX, etc.) and the emotional data is acquired.

[0661] 3. Document Preprocessing

[0662] The server performs pre-processing on the received documents, which includes converting the PDF files to text format and removing noise (e.g., images and redundant text).

[0663] 4. Language Identification

[0664] The server uses natural language processing (NLP) techniques to determine the language of the document, whether it is written in English, Japanese, French, etc.

[0665] 5. Document Analysis

[0666] Based on the language identified, the server uses NLP techniques to analyze the document, specifically extracting important keywords and phrases (e.g., contract terms, payment terms, liability clauses, etc.).

[0667] 6. Generating a Summary

[0668] Based on the analysis results, the server generates a summary of the document, which is a concise summary of the document's key points, including the extracted key points.

[0669] 7. Emotional Engine Adjustment

[0670] The server uses an emotion engine to adapt the analysis results to the emotional data collected from the user, for example, adjusting the results to produce a more concise and clear summary if the user is feeling stressed or confused.

[0671] 8. Multilingual support

[0672] If the language of the document is different from the language specified by the user, the server will use an automatic translation function to translate the document or abstract into the specified language, allowing the user to view the abstract in the language they require.

[0673] 9. Submitting the results

[0674] The server then sends the generated summary and analysis results to the user's device, including summaries adjusted by the emotion engine. The user receives the results and checks them on the screen.

[0675] Specific examples

[0676] Example 1: Processing an English contract

[0677] When a user uploads a PDF English contract titled "Service Agreement" to the system using their PC, a camera and microphone monitor the user's facial expressions and voice to collect emotional data. The server receives the document, converts it to text format, and removes noise. After determining that the document is written in English, it uses NLP technology to extract important information. A summary is generated based on this information and adjusted according to the user's emotional state. The adjusted summary is then sent to the user's PC. The user then reviews the summary on their PC and determines whether the contract terms are appropriate.

[0678] Example 2: Processing multilingual contracts

[0679] When a user uploads a contract written in French to the system using their smartphone, the smartphone's camera and microphone collect the user's emotions. The server receives the French document and performs the necessary preprocessing. It determines the document's language as French and uses NLP technology to extract key information. If the user specifies English, the French summary is translated into English using an automatic translation function. The translated summary is adjusted by the emotion engine and sent to the user's smartphone. The user then reviews the summary and reevaluates the contract content if necessary.

[0680] In this way, a system is realized that efficiently analyzes the content of documents and provides summaries that correspond to the user's emotional state, thereby reducing the burden on legal departments and speeding up contract signing.

[0681] The processing flow will be explained below.

[0682] Step 1:

[0683] The user uploads a document to the system using a device. From the device interface, the user selects the PDF file of the contract they wish to analyze and summarize, and clicks the "Upload" button. At the same time, the device's camera and microphone collect the user's facial expressions and voice data, which are then recorded as emotion data.

[0684] Step 2:

[0685] The server receives the uploaded document and temporarily stores it. The format information of the saved document (e.g., PDF, DOCX) is acquired. The user's emotion data sent from the device is also stored at the same time.

[0686] Step 3:

[0687] The server pre-processes the document: if the document is in PDF format, it performs operations to convert it to text format, and also removes noise from the document, such as unnecessary images and redundant text.

[0688] Step 4:

[0689] The server determines the language of the document. It uses natural language processing (NLP) techniques to automatically detect the language the document is written in. Once the language is determined, it launches an analysis process for that language.

[0690] Step 5:

[0691] The server analyzes the document. Based on the language identified, it uses NLP techniques to analyze each section of the document and extract important keywords and phrases. Specifically, it identifies important information such as contract terms, payment terms, and liability clauses, and extracts the necessary data.

[0692] Step 6:

[0693] The server generates a summary based on the analysis results, organizes the extracted important information, and generates a summary of the document. The summary is concisely compiled in a format that is easy for users to understand.

[0694] Step 7:

[0695] The server uses an emotion engine to analyze the user's emotion data, specifically through facial expression recognition and voice analysis, to determine the user's current emotional state (e.g., stress, excitement, calm).

[0696] Step 8:

[0697] The server then uses the emotion engine to adjust the summary of the document based on the analyzed emotion data. For example, if the user is feeling confused or stressed, the summary will be adjusted to be more concise and clear.

[0698] Step 9:

[0699] If the language of the document is different from the language specified by the user, the server uses an automatic translation function to translate the document or abstract into the specified language.

[0700] Step 10:

[0701] The server sends the generated summary and analysis results to the user's device, including summaries adjusted by the emotion engine. The user receives the results and checks them on the screen.

[0702] Step 11:

[0703] The user checks the analysis results and summary on the terminal, and determines whether the contract contents are valid based on the provided summary, and performs additional verification or corrections as necessary.

[0704] This series of steps results in a system that efficiently analyzes the content of documents and provides summaries tailored to the user's emotional state, reducing the burden on legal departments and speeding up contract execution.

[0705] Example 2

[0706] 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."

[0707] Many systems exist for quickly and accurately analyzing contracts and other important documents and generating summaries. However, these systems lack the ability to adjust their analysis results to take into account the user's emotional state. As a result, if a user is stressed or confused, the summary may be difficult to understand, making it difficult for them to make effective decisions. Another issue is that multilingual support is often limited, meaning that the summary cannot be viewed in the user's preferred language.

[0708] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: a means for a user to upload data to the system; a means for the server to receive and store the uploaded data; a means for the server to preprocess the data and convert it into a format suitable for analysis; a means for the server to determine the language of the data; a means for the server to analyze each portion of the data using natural language processing technology and extract important information; a means for the server to generate a summary of the data based on the analysis results; a means for the server to send the summary generated by the server to the user's terminal; a means for the user to check the analysis results and the summary through the terminal; an emotion data collection means including a sensor attached to the terminal for collecting user emotion data; and a means for the server to analyze the user's emotion data and adjust the generated summary. This allows the server to provide a summary adjusted according to the user's emotional state, support multiple languages, and more quickly and accurately understand the content of a document.

[0709] "User" means any person or entity that uses the System to upload documents or receive summaries.

[0710] "Data" refers to information in any form, including contracts and other important documents.

[0711] "System" refers to the collection of devices and software that provides a set of processes for uploading data, analyzing, generating summaries, translating, and adjusting emotion data.

[0712] "Server" refers to a computing device within the system that receives, stores, pre-processes, analyzes, language-determines, summarizes, adjusts, translates, and transmits results.

[0713] "Terminal" refers to a computer device (PC, smartphone, tablet, etc.) used by a user to upload data and check analysis results.

[0714] "Preprocessing" refers to a series of operations that convert data into a format suitable for analysis, including conversion to text format and noise removal.

[0715] "Language determination" refers to the process of identifying the language in which data is written.

[0716] "Natural language processing technology" refers to computer technology used to analyze data and extract important information.

[0717] "Key information" refers to specific keywords or phrases extracted from each piece of data.

[0718] A "summary" is a document that briefly summarizes the key points of the data.

[0719] "Emotion data" refers to data that indicates the emotional state of a user, collected from facial expressions and voice.

[0720] "Emotion data collection means" refers to a device that collects emotional data using sensors, cameras, and microphones attached to a terminal.

[0721] "Tuning" refers to the process of optimizing summaries and analysis results based on the user's emotional state.

[0722] "Automatic translation" refers to technology for translating data or summaries into another language.

[0723] This invention relates to a method for efficiently analyzing contracts and other important documents and generating summaries using a system. Furthermore, it aims to combine this system with an emotion engine that recognizes the user's emotions to adjust the analysis and summaries according to the user's emotional state. Specific embodiments for implementing this invention are described below.

[0724] System configuration

[0725] Document upload

[0726] Users upload documents such as contracts to the system using devices (PCs, smartphones, tablets, etc.). The devices are equipped with cameras and microphones, allowing for the collection of emotional data through the user's facial expressions and voice.

[0727] Document Receipt and Storage

[0728] The server receives the uploaded document and temporarily stores it in storage. At this time, basic information such as the document format (PDF, DOCX, etc.) and emotion data is acquired and recorded as metadata.

[0729] Document Preprocessing

[0730] The server performs pre-processing on the received documents, which includes converting the PDF files to text format and removing noise (e.g., images and overlapping text) using specific document analysis software (e.g., OCR technology).

[0731] Language Identification

[0732] The server uses natural language processing (NLP) techniques to determine the language of the document, whether it is written in English, Japanese, French, etc. The techniques used include, for example, language identification algorithms.

[0733] Document analysis and summary generation

[0734] The server analyzes the document using NLP technology based on the determined language. It extracts important keywords and phrases (e.g., contract terms, payment terms, liability clauses, etc.) and generates a document summary based on the analysis results. The summary is a concise summary of the document's key points. Specific technologies used include keyword extraction algorithms and summary generation models.

[0735] Emotional Engine Adjustment

[0736] The server uses an emotion engine to analyze the emotional data collected from the user and incorporate it into the analysis results. For example, if the user is feeling stressed or confused, the server will adjust the summary to generate a more concise and clear summary. Facial recognition software and voice analysis technology are used to analyze the emotional data.

[0737] Multilingual support and results transmission

[0738] If the document language is different from the language specified by the user, the server uses an automatic translation function to translate the document or summary into the specified language. This function allows the user to view the summary in the language they require. This is achieved using a translation API. The generated summary and analysis results are then sent to the user's device, where the user can view the results.

[0739] Specific examples

[0740] Example 1: Processing an English contract

[0741] When a user uploads a PDF English contract titled "Service Agreement" to the system using their PC, a camera and microphone monitor the user's facial expressions and voice to collect emotional data. The server receives the document, converts it to text format, and removes noise. After determining that the document is written in English, it uses NLP technology to extract key information. A summary is generated based on this information and adjusted according to the user's emotional state. The server then sends the adjusted summary to the user's PC. The user then reviews the summary on their PC and determines whether the contract terms are appropriate.

[0742] Example 2: Processing multilingual contracts

[0743] When a user uploads a contract written in French to the system using their smartphone, the smartphone's camera and microphone collect the user's emotions. The server receives the French document and performs the necessary preprocessing. It determines the document's language as French and uses NLP technology to extract key information. If the user specifies English, the French summary is translated into English using an automatic translation function. The translated summary is then adjusted by the emotion engine and sent to the user's smartphone. The user then reviews the summary and reevaluates the contract content if necessary.

[0744] Prompt Sentence Examples

[0745] "Please upload the English contract and summarize the key points. Also, please adjust it to take into account sentiment data."

[0746] "I have uploaded a document in French. Please translate it into English and provide a summary."

[0747] In this way, the system can efficiently analyze contracts and important documents and provide summaries tailored to the user's emotional state, reducing the burden on legal departments and speeding up contract closures.

[0748] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0749] Step 1: Upload your documents

[0750] The user uses a terminal to upload a document to the system. The input is a contract file selected by the user (e.g., a PDF file), and the output is the transfer of the file to the system. The terminal is equipped with a camera and microphone, which also collects emotional data from the user's facial expressions and voice. Specifically, the user clicks the "Upload" button and selects a document file.

[0751] Step 2: Receiving and storing documents

[0752] The server receives the uploaded document and temporarily stores it in storage. The input is the uploaded contract file and emotion data, and the output is the file stored on the server and the collected emotion data meta information. Specifically, the server receives the file and stores it in storage with a file name such as "contract.pdf." At this time, basic information such as the document format and emotion data is also saved.

[0753] Step 3: Preprocessing the Documents

[0754] The server performs preprocessing on the received document. The input is the saved contract file, and the output is text data suitable for analysis. Specifically, it converts the PDF file to text and removes noise such as images and overlapping text. Specifically, it uses OCR technology to extract text from the PDF and runs a script to remove unnecessary noise.

[0755] Step 4: Language Determination

[0756] The server uses natural language processing (NLP) techniques to determine the language of a document. The input is preprocessed text data, and the output is the document's language information (e.g., English, Japanese, French, etc.). Specifically, it analyzes the frequency of words and phrases in the document and runs a model to identify the language based on that.

[0757] Step 5: Parsing the Document

[0758] The server then uses NLP technology to analyze the document based on the language it has identified. The input is linguistic information and preprocessed text data, and the output is key keywords and phrases extracted through the analysis. Specifically, it runs an algorithm to extract keywords such as contract terms, payment terms, and liability clauses.

[0759] Step 6: Generate a summary

[0760] The server generates a document summary based on the analysis results. The input is the extracted keywords and phrases, and the output is the document summary. Specifically, it runs a script that generates a document summary based on the extracted keywords and an existing format.

[0761] Step 7: Emotional Engine Alignment

[0762] The server uses an emotion engine to adjust the summary based on the user's emotion data. The input is the generated summary and emotion data, and the output is a summary adjusted to reflect the emotion. Specifically, if the user's facial expression data indicates "stress," the server runs an algorithm to reconstruct the summary into simpler language.

[0763] Step 8: Multilingual

[0764] If the document language is different from the language specified by the user, the server uses an automatic translation function to translate the summary into the specified language. The input is the adjusted summary, and the output is the translated summary. Specifically, the server calls an API to translate the summary generated in French into English.

[0765] Step 9: Sending the results

[0766] The server sends the generated summary and analysis results to the user's device. The input is the adjusted and translated summary, and the output is the summary sent to the user's device. Specifically, the generated summary and analysis results are sent to the user's device in a data format such as JSON, and the device displays them.

[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] Conventional document analysis systems can analyze documents and generate summaries, but they do not take into account the user's emotional state, resulting in the inability to provide optimal information to users. Furthermore, they are unable to provide appropriate support when users feel stressed or confused, and are particularly inadequate in situations requiring quick and accurate judgments, particularly in the security service field. There is a need for a system that can improve this situation by adjusting document analysis results and summaries based on the user's emotional state.

[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 a user to upload a document to the system; means for the server to receive and store the uploaded document; means for the server to preprocess the document and convert it into a format suitable for analysis; means for the server to determine the language of the document; means for the server to analyze each section of the document using natural language processing technology and extract important keywords and phrases; means for the server to generate a summary of the document based on the analysis results; means for the server to recognize the user's emotional state and adjust the summary based on that state; means for the server to send the summary generated by the server to the user's terminal; and means for the user to check the analysis results and the summary through the terminal. This makes it possible to provide a summary adjusted according to the user's emotional state.

[0771] "User" refers to the person who uploads documents to the analysis system and reviews the analysis results and summaries.

[0772] "Document" refers to the documents or information to be analyzed and summarized, and may exist in formats such as PDF or DOCX.

[0773] "System" refers to the entire device and software configuration that performs a series of processes in which a user uploads a document and the server analyzes and summarizes it.

[0774] "Server" refers to a computer and associated software for receiving, storing, and processing documents uploaded by users.

[0775] "Preprocessing" refers to a series of operations to convert a document into a format suitable for analysis, such as converting PDF files to text or removing noise.

[0776] "Natural language processing technology" is a technology in the field of computer science that refers to methods and algorithms that allow machines to understand, analyze, and generate human language.

[0777] "Emotional state" refers to the psychological or emotional state recognized from the user's facial expression or voice, and includes, for example, stress, happiness, surprise, etc.

[0778] A "summary" is a concise summary of the important points from an analyzed document that can be quickly understood by a user.

[0779] "Upload" refers to the operation of a user sending a document from their own terminal to the system.

[0780] "Terminal" refers to a device used by a user, such as a smartphone or PC, that interfaces with the system.

[0781] "Adjustment" refers to making changes or modifications to the content of an already generated summary based on the user's emotional state.

[0782] "Automatic translation function" refers to the function by which the server mechanically translates documents and summaries into a language specified by the user.

[0783] "Important sections" are those parts of the document that require special attention, such as specific matters like contract terms, payment terms, and liability clauses.

[0784] This invention is a system for efficiently analyzing documents and generating summaries in the field of security services, and can further adjust the summaries according to the emotional state of the user. The system is implemented through the following steps:

[0785] composition

[0786] 1. Hardware configuration:

[0787] The user's device is a smartphone or a PC, which is equipped with a camera and a microphone.

[0788] The server is equipped with a high-performance CPU and sufficient memory.

[0789] Data storage requires large amounts of storage for analysis.

[0790] 2. Software configuration:

[0791] For natural language processing technology, Google Cloud Natural Language API is used.

[0792] Amazon Rekognition is used for sentiment analysis.

[0793] The automatic translation function uses the Google Translate API.

[0794] The server software used is Apache HTTP Server and Node.js.

[0795] MySQL is used as the database.

[0796] Explanation of program processing

[0797] The server receives documents uploaded by users, analyzes them, and summarizes them. This process involves the following major steps:

[0798] 1. Upload and save documents:

[0799] Users upload documents to the system via their smartphones or PCs. The server receives and stores them, and the documents are used for analysis and summary generation.

[0800] 2. Document preprocessing:

[0801] The server pre-processes the uploaded documents, which includes converting PDF files to text format and removing unwanted noise.

[0802] 3. Document Language Determination:

[0803] The server uses the Google Cloud Natural Language API to determine the language of the document, which tells it what language it is written in.

[0804] 4. Document Analysis:

[0805] The server uses natural language processing techniques to analyze the content of the document and extract important keywords and phrases.

[0806] 5. Generate a summary:

[0807] The server generates a summary of the document based on the extracted keywords and phrases, which is concise and easy for users to understand.

[0808] 6. Emotional engine regulation:

[0809] The server uses Amazon Rekognition to analyze the user's emotional data and adjusts the generated summary according to the user's emotional state—for example, if the user is feeling stressed, the summary will be more concise and clear.

[0810] 7. Submit and review your abstract:

[0811] The server then sends the adjusted summary to the user's device, where the user can check the summary on their smartphone or computer and take any necessary action.

[0812] Specific examples

[0813] For example, if a lawyer uploads a security services contract via smartphone, the system analyzes the document and summarizes key sections. Furthermore, if the lawyer is under stress, the summary can be adjusted to be concise and clear. This process ensures fast and accurate information delivery.

[0814] Prompt Sentence Examples

[0815] Analyze and summarize the contracts uploaded to the system. Adjust the summary and reply based on the user's sentiment.

[0816] In this way, the present invention can efficiently analyze the content of documents and provide summaries according to the emotional state of the user, thereby improving reliability and efficiency in the field of security services.

[0817] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0818] Step 1:

[0819] The user uploads a document to the system. The user uses a smartphone or PC to upload the document to be analyzed to the system. The input is a document file (PDF, DOCX, etc.), and the output is the data uploaded to the server.

[0820] Step 2:

[0821] The server receives and stores the uploaded document. The server receives the document file sent by the user and stores it in a database. The input is the document file sent by the user, and the output is the document data stored in the database.

[0822] Step 3:

[0823] The server preprocesses the document and converts it into a format suitable for analysis. The server converts the document into a text format and removes unnecessary noise (images, duplicated text, etc.). The input is the document data stored in the database, and the output is the preprocessed text data.

[0824] Step 4:

[0825] The server determines the language of the document. The server determines the language of the document using the Google Cloud Natural Language API. The input is preprocessed text data, and the output is the language information of the document.

[0826] Step 5:

[0827] The server uses natural language processing technology to analyze each section of the document and extract important keywords and phrases. The server uses natural language processing technology (Google Cloud Natural Language API) to analyze the document and extract important keywords and phrases. The input is the text data of the document, and the output is the extracted keywords and phrases.

[0828] Step 6:

[0829] The server generates a document summary based on the analysis results. The server generates a document summary based on the extracted keywords and phrases. The input is the extracted keywords and phrases, and the output is the generated summary.

[0830] Step 7:

[0831] The server recognizes the user's emotional state and adjusts the summary based on that state. The server uses Amazon Rekognition to analyze the user's emotional state and adjust the generated summary. The input is the user's emotional data (facial expressions and voice) and the generated summary, and the output is the adjusted summary.

[0832] Step 8:

[0833] The server sends the generated summary to the user's terminal. The server sends the adjusted summary to the user's terminal. The input is the adjusted summary, and the output is the summary sent to the user's terminal.

[0834] Step 9:

[0835] The user checks the analysis results and summary via their device. The user checks the analysis results and summary sent from the server via their smartphone or PC. The input is the summary sent from the server, and the output is the user's understanding and any necessary actions.

[0836] 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.

[0837] 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.

[0838] 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.

[0839] [Third embodiment]

[0840] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0841] 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.

[0842] 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).

[0843] 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.

[0844] 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.

[0845] 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).

[0846] 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. 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.

[0847] 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.

[0848] 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.

[0849] 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.

[0850] 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.

[0851] 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."

[0852] This invention relates to a method for efficiently analyzing contracts and other important documents and generating summaries using a system, in which a user uploads the document to the system using a terminal, a server analyzes the document, generates a summary, and sends it back to the user's terminal.

[0853] Explanation of the specific operation of the program

[0854] 1. Uploading documents

[0855] Users use their devices to upload documents they wish to analyze and summarize to the system, for example by selecting a PDF file of a contract from a PC, smartphone, or other device, and clicking the upload button.

[0856] 2. Document Receipt and Storage

[0857] The server receives the uploaded document and temporarily stores it, at which point it obtains basic information about the document's format (e.g. PDF, DOCX).

[0858] 3. Document Preprocessing

[0859] The server performs pre-processing on the received documents, which includes converting the PDF files to text format and removing noise (e.g., images and redundant text).

[0860] 4. Language Identification

[0861] The server uses natural language processing (NLP) techniques to automatically determine the language of the document, whether it is written in English, Japanese, French, etc.

[0862] 5. Document Analysis

[0863] Based on the language identified, the server uses NLP technology to analyze the document, extracting important keywords and phrases (e.g., contract terms, payment terms, liability clauses, etc.), and analyzes the document structure, dividing it into sections for analysis.

[0864] 6. Generating a Summary

[0865] The server then generates a summary of the document based on the analysis results. The summary is a concise summary of the document's key points, including the most important information extracted. For example, a contract summary might include information such as "Contract period: one year" and "Payment terms: payment by the end of each month."

[0866] 7. Multilingual support

[0867] If the language of the document is different from the language specified by the user, the server will use an automatic translation function to translate the document and summary into the specified language, allowing the user to view the summary in the language they require.

[0868] 8. Submitting the results

[0869] The server sends the generated summary and analysis results to the user's terminal, where the user can check the summary received and perform additional verification or correction as necessary.

[0870] Specific examples

[0871] Example 1: Processing an English contract

[0872] A user uploads an English contract titled "Service Agreement" in PDF format to the system using their PC. The server receives the document, converts it to text format, and removes noise. After determining that the document is in English, it uses NLP technology to extract important information such as the contract period, payment terms, and liability clauses. A summary is generated based on this information and sent to the user's PC. The user then checks the summary on their PC and determines whether the contract contents are reasonable.

[0873] Example 2: Processing multilingual contracts

[0874] A user uploads a contract written in French to the system using their smartphone. The server receives the French document and performs the necessary preprocessing. It determines that the document is in French and uses NLP technology to extract important information. If the user specifies English, the server uses an automatic translation function to translate the French summary into English. The translated summary is sent to the user's smartphone, where the user can review it.

[0875] This allows the system to efficiently analyze the contents of contracts and important documents, generate summaries, and provide them to users, thereby reducing the burden on legal departments and speeding up contract signing processes.

[0876] The processing flow will be explained below.

[0877] Step 1:

[0878] The user uses the terminal to upload the document to be analyzed and summarized to the system. For example, the user selects a PDF file of a contract from the terminal interface and clicks the "Upload" button.

[0879] Step 2:

[0880] The server receives the uploaded document. The received document is temporarily stored on the server and its format information (PDF, DOCX, etc.) is acquired.

[0881] Step 3:

[0882] The server pre-processes the document. If the document is in PDF format, it converts it to text format. It also removes unwanted images, redundant text, and other noise from the document.

[0883] Step 4:

[0884] The server determines the language of the document, automatically detecting the language using natural language processing (NLP) techniques and preparing the next analysis step based on the results.

[0885] Step 5:

[0886] The server analyzes the document. Based on the language identified, it uses NLP techniques to analyze each section of the document and extract important keywords and phrases. For example, in the case of a contract, it identifies important information such as "contract duration," "payment terms," ​​and "liability clauses."

[0887] Step 6:

[0888] The server generates a summary based on the analysis results, organizes the extracted important information, and generates a summary of the document. The summary is concise and easy for users to understand.

[0889] Step 7:

[0890] The server determines whether multilingual support is necessary, and if the document language is different from the language specified by the user, it uses an automatic translation function to translate the document or abstract into the specified language.

[0891] Step 8:

[0892] The server sends the generated summary and analysis results to the user's terminal, where the user receives the results and checks them on the screen.

[0893] Step 9:

[0894] The user checks the analysis results and summary on the terminal, and determines whether the contract contents are valid based on the provided summary, and performs additional verification or corrections as necessary.

[0895] This series of steps results in a system that efficiently analyzes the contents of documents and quickly provides users with summaries, reducing the burden on legal departments and streamlining contract execution.

[0896] Example 1

[0897] 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."

[0898] To efficiently analyze documents and generate summaries, it is necessary to process documents in different formats in a unified manner, accurately extract information using natural language processing technology, and generate summaries that include particularly important sections. Furthermore, when documents are written in different languages, it is necessary to perform appropriate translation. These challenges have been difficult to completely address with conventional systems.

[0899] 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.

[0900] In this invention, the server includes: a means for a user to upload a document to the system; a means for the server to receive and store the uploaded document; a means for the server to preprocess the document and convert it into a format suitable for analysis; a means for the server to determine the language of the document; a means for the server to analyze each section of the document using natural language processing technology and extract important information; a means for the server to generate a summary of the document based on the analysis results; a means for the server to send the summary generated by the server to the user's terminal; a means for the user to check the analysis results and the summary via the terminal; and a means for the server to select an appropriate preprocessing method depending on the document format. This makes it possible to consistently and accurately extract information and generate summaries including important sections for documents of different formats and languages.

[0901] "User" means a person who uses the system to upload documents and check analysis results.

[0902] A "server" is a device or system that receives uploaded documents, stores them, and performs various processes such as preprocessing, analysis, summary generation, and translation.

[0903] "Upload" refers to the act of sending a document that a user possesses to the system.

[0904] "Preprocessing" refers to the process of converting documents into a format suitable for analysis, and includes text data extraction and noise removal.

[0905] "Language determination" is the process of automatically identifying the language in which an uploaded document is written.

[0906] "Natural language processing technology" refers to technology for analyzing documents and extracting information, including technology for extracting noun phrases and identifying important phrases.

[0907] "Summary generation" is the process of creating a summary that succinctly summarizes the important points of a document based on the analysis results.

[0908] The "automatic translation function" is a function that automatically translates documents and summaries in different languages ​​into a language specified by the user.

[0909] "Certain Important Sections" are those parts of the document that contain particularly important information, such as the terms of the contract, payment terms, and liability clauses.

[0910] "Analysis results" refer to important information and keywords extracted using natural language processing technology.

[0911] "Terminal" refers to a device used by a user to upload documents or check analysis results, and includes PCs, smartphones, etc.

[0912] "Noise removal" is the process of eliminating information unnecessary for analysis (such as images and duplicated text).

[0913] "Text data" refers to character string data extracted from a document.

[0914] "Tokenization" is the process of dividing the text in a document into units (words or phrases) that are easier to analyze.

[0915] "Part-of-speech tagging" is the process of automatically assigning each word its part of speech.

[0916] The above terms are defined so that the technical scope and embodiments of the invention can be clearly understood.

[0917] The present invention relates to a method for efficiently analyzing documents and generating summaries using a system, which is implemented through a series of steps in which a user uploads a document to the system using a terminal, a server analyzes the document, generates a summary, and sends it back to the user's terminal.

[0918] Hardware and software used

[0919] Server: Uses a computer with high-performance computing resources and provides functions for receiving, storing, preprocessing, analyzing, summarizing, translating, and transmitting documents.

[0920] Device: A PC or smartphone used by a user to upload documents and receive and check summaries.

[0921] Software tools:

[0922] PDF text extraction tools (pdfminer, PyMuPDF)

[0923] Natural language processing libraries (spaCy, NLTK)

[0924] Language detection library (langdetect)

[0925] Translation API (Google Translate API, DeepL API)

[0926] Explanation of program processing

[0927] 1. Uploading documents

[0928] The user uses a device to upload the document they want to analyze and summarize to the system. For example, they select a PDF file of a contract from the file selection screen on their PC or smartphone and click the "Upload" button. The device then sends the selected document to the server.

[0929] 2. Document Receipt and Storage

[0930] The server receives the document sent from the device and temporarily stores it, identifying the document format (PDF, DOCX, etc.) and obtaining the associated metadata.

[0931] 3. Document Preprocessing

[0932] The server preprocesses the documents. Specifically, it uses tools (such as pdfminer or PyMuPDF) to convert PDF files into text format, extracts the text data, and then performs a filtering process to remove noise (such as images or overlapping text) from the documents.

[0933] 4. Language Identification

[0934] The server automatically determines the language of the document, specifically using libraries such as langdetect or spaCy to determine whether the document is written in English, Japanese, French, etc.

[0935] 5. Document Analysis

[0936] Based on the language identified, the server analyzes the document using natural language processing techniques, such as tokenization and part-of-speech tagging (POS tagging), to extract noun phrases and key phrases (such as contract terms, payment terms, and liability clauses), using libraries such as spaCy and NLTK.

[0937] 6. Generating a Summary

[0938] The server generates a summary of the document based on the analyzed data. The summary is generated using a text summarization algorithm (TextRank or an extractive summarization model). The generated summary includes extracted important information (e.g., contract period: one year, payment terms: payment by the end of each month).

[0939] 7. Multilingual support

[0940] If the language specified by the user is different from the language of the document, the server translates the summary into the specified language using an automatic translation function, for example, using the Google Translate API or DeepL API to translate the summary from English to Japanese.

[0941] 8. Submitting the results

[0942] The server sends the generated summary and analysis results to the user's device, where the user can check the received summary and perform additional verification or corrections as necessary.

[0943] Specific examples

[0944] Example 1: Processing an English contract

[0945] Using a PC, a user selects an English contract in PDF format titled "Service Agreement" and uploads it to the system. The device sends the selected PDF file to the server. The server receives the document and uses pdfminer to extract text data and remove noise. The server uses the langdetect library to determine that the document is written in English. The server uses spaCy to extract important keywords (such as contract term, payment terms, and liability clauses). The server uses the TextRank algorithm to summarize the document. The summary includes phrases such as "Contract term: 1 year" and "Payment terms: Payment by the end of each month." The server then sends the generated summary to the user's PC. The user then reviews the summary on their PC and determines whether the contract terms are reasonable.

[0946] Example 2: Processing multilingual contracts

[0947] A user uploads a French contract titled "Contrat de Service" to the system using a smartphone. The device sends the selected DOCX file to the server. The server receives the document and uses PyMuPDF to extract text data and remove noise. The server uses spaCy's multilingual model to determine that the document is in French. The server uses a French NLP model to extract important keywords. The server generates a summary based on the extracted data. The server uses the DeepL API to translate the French summary into English. The server sends the translated summary to the user's smartphone. The user reviews the summary and understands the contract.

[0948] Prompt Sentence Examples

[0949] Below is an example of a prompt sentence.

[0950] 1. "Generate a summary of the contract written in English."

[0951] 2. "Analyze the uploaded PDF document and provide key keywords and a brief summary."

[0952] 3. "Translate this French document into English and summarize the key points."

[0953] In this way, the present invention improves the efficiency of document analysis and summary generation, thereby improving the efficiency of legal departments and business operations.

[0954] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0955] Step 1:

[0956] A user uses a terminal to upload a document to the system. Specifically, the user selects the document to be processed (e.g., PDF or DOCX format) from the terminal's file selection screen and clicks the "Upload" button. The input of the terminal is the document file selected by the user, and the output is the transmission of the document file to the server.

[0957] Step 2:

[0958] The server receives documents sent from the device and temporarily stores them. The input is the document file sent from the device, and the output is a file stored in the server's storage area. At this time, the server identifies the document format (PDF, DOCX, etc.) and obtains related metadata (file format, file size, etc.).

[0959] Step 3:

[0960] The server preprocesses the documents. Specifically, it uses tools such as pdfminer and PyMuPDF to convert PDF files into text format and extracts the text data. It also performs a filtering process to remove noise (images, overlapping text, etc.) from the documents. The input is the saved document file, and the output is the text data converted into text format and with noise removed.

[0961] Step 4:

[0962] The server automatically determines the language of a document using natural language processing technology. Specifically, it uses libraries such as langdetect and spaCy to determine the language in which the document is written. The input is preprocessed text data, and the output is the document's language information.

[0963] Step 5:

[0964] Based on the determined language, the server uses natural language processing techniques (such as spaCy or NLTK) to analyze the document. Specifically, it performs tokenization and part-of-speech tagging (POS tagging) to extract noun phrases and important phrases (such as contract terms, payment terms, and liability clauses). The input is the document's linguistic information and preprocessed text data, and the output is the extracted important information and keywords.

[0965] Step 6:

[0966] The server generates a document summary based on the extracted data. It uses a text summarization algorithm such as TextRank or an extractive summarization model to generate the summary. The generated summary includes extracted key information (e.g., contract period: one year, payment terms: payment by the end of each month). The input is the extracted key information and keywords, and the output is the generated summary.

[0967] Step 7:

[0968] If the language specified by the user is different from the language of the document, the server translates the summary into the specified language using an automatic translation function (such as Google Translate API or DeepL API). The input is the generated summary and the language information specified by the user, and the output is the translated summary.

[0969] Step 8:

[0970] The server sends the generated summary and analysis results to the user's device. The user checks the received summary on their own device and performs additional verification or correction as necessary. The input is the generated (translated) summary and analysis results, and the output is transmission to the user's device.

[0971] (Application example 1)

[0972] 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."

[0973] In recent years, logistics centers have become required to process large volumes of documents, and there is a demand for the rapid and accurate understanding of the contents of contracts and other important documents. However, manual document analysis and summary generation is time-consuming and places a burden on human resources. Furthermore, when dealing with documents in different languages, translation costs and time increase, resulting in reduced work efficiency. Even with the introduction of systems using smart glasses, the above issues remain unresolved because existing technologies do not automate document content analysis.

[0974] 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.

[0975] In this invention, the server includes: a means for a user to upload a document to the system; a means for the server to receive and store the uploaded document; a means for the server to preprocess the document and convert it into a format suitable for analysis; a means for the server to determine the language of the document; a means for the server to analyze each section of the document using natural language processing technology and extract important keywords and phrases; a means for the server to generate a summary of the document based on the analysis results; a means for the server to send the summary generated by the server to a user's terminal; a means for the user to check the analysis results and the summary through the terminal; and a means for logistics center staff to wear smart glasses and scan contracts or important documents in packages with a camera, causing the system to automatically analyze the documents and display the summaries. This makes it possible to automatically analyze the contents of contracts and important documents and provide summaries quickly and accurately at logistics centers by using smart glasses. The summaries can automatically determine and translate language differences, allowing documents in different languages ​​to be processed efficiently.

[0976] A "user" is an individual or company employee who accesses the system to upload documents and review analysis results.

[0977] "Document" refers to text data containing contracts and other important information, and may be in formats such as PDF or DOCX.

[0978] "System" refers to the set of servers and associated software that receives documents uploaded by users, processes, analyzes, summarizes, and in some cases translates them, and provides the results to users.

[0979] The "server" is the device that forms the core of the system, and is a computer that processes documents received from users, analyzes them, generates summaries, and finally transmits the results to the users.

[0980] "Preprocessing" refers to the initial stage of processing to convert uploaded documents into a state suitable for analysis, and specifically includes text extraction and noise removal.

[0981] "Natural language processing technology" is a technology that enables computers to understand and use human language, specifically, technology that performs text analysis and extracts important keywords.

[0982] A "summary" is a concise text summary of the important information of a document analyzed by the server, presented in a user-friendly format.

[0983] "Translation" refers to the process of converting a document or abstract into a specified language, and is achieved by using an automatic translation function.

[0984] "Smart glasses" are visual devices that have built-in cameras and displays and can present augmented reality (AR) and information when worn by the user.

[0985] A "logistics center" is a facility where goods and packages are collected and prepared for shipment and delivery.

[0986] "Scanning" is the act of taking an image of a document using the camera in the smart glasses and importing it into the system as digital data.

[0987] The present invention is a system in which staff working at a logistics center wear smart glasses and scan contracts and important documents attached to packages with a camera, allowing the system to automatically analyze the documents and display summaries.

[0988] Explanation of the specific operation of the program

[0989] 1. Scanning documents

[0990] A user wears the smart glasses and uses the camera to take an image of a document, allowing work at the logistics center to proceed quickly and efficiently.

[0991] 2. Image to text conversion

[0992] The server receives the document image sent from the smart glasses and uses Tesseract OCR to extract text from the image, which then becomes the base data for analysis.

[0993] 3. Preprocessing and Language Identification

[0994] The server preprocesses the extracted text, removing noise and converting it into a format suitable for analysis, and then automatically determines the language of the document using natural language processing techniques.

[0995] 4. Text Analysis and Summary Generation

[0996] The server then analyzes the text using natural language processing technology such as SpaCy to extract important keywords and phrases based on the language identified, and generates a summary of the document based on the analysis results.

[0997] 5. Translation and result display

[0998] If the document is not in the language specified by the user, the server uses its automatic translation function to translate the summary into the specified language, and the translated summary is sent back to the smart glasses, where the user can view it on the glasses' display.

[0999] Hardware and software used

[1000] Hardware: Smart glasses (camera, display, CPU, memory) and server

[1001] Software: Tesseract OCR, SpaCy, natural language processing technology, translation API

[1002] Specific examples

[1003] Example 1: Processing an English contract

[1004] A staff member at the logistics center wears smart glasses and uses the camera to scan a PDF English contract titled "Service Agreement." The server receives the document, converts it to text format, and removes noise. After determining that the document is written in English, it uses natural language processing technology to extract important information such as the contract period, payment terms, and liability clauses. A summary is generated based on this information and sent to the user's smart glasses. The user then reviews the summary on the smart glasses and determines whether the contract terms are reasonable.

[1005] Example 2: Processing multilingual contracts

[1006] A staff member at a logistics center uses smart glasses to scan a contract written in French with a camera. The server receives the French document and performs the necessary preprocessing. It determines that the document is in French and uses natural language processing technology to extract key information. If the user specifies English, the server uses an automatic translation function to translate the French summary into English. The translated summary is sent to the user's smart glasses, where the user can review it.

[1007] This means that by using smart glasses, it will be possible to automatically analyze the contents of contracts and important documents quickly and accurately on-site at logistics centers and provide summaries.

[1008] Example prompts to input to the generative AI model

[1009] "When a logistics center staff member scans a contract with smart glasses, the text is automatically extracted and displayed."

[1010] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1011] Step 1:

[1012] A user puts on the smart glasses and uses the camera to take an image of a document.

[1013] (Input) An image of a document captured by smart glasses

[1014] (Operation) A user points the camera at a document and presses the shutter button to capture an image of the document.

[1015] (Output) Document image data

[1016] Step 2:

[1017] The terminal transmits the captured image data to the server.

[1018] (Input) Image data of the photographed document

[1019] (Operation) Image data is transferred to the server using a transmission module in the terminal.

[1020] (Output) Image data sent to the server

[1021] Step 3:

[1022] The server converts the received image data into text data using Tesseract OCR.

[1023] (Input) Image data sent to the server

[1024] (Operation) The server calls Tesseract OCR and performs the process of extracting text from the image.

[1025] (Output) Extracted text data

[1026] Step 4:

[1027] The server preprocesses the extracted text data and converts it into a format suitable for analysis.

[1028] (Input) Extracted text data

[1029] (Operation) The server performs preprocessing to remove noise and unnecessary information from the text data.

[1030] (Output) Preprocessed text data

[1031] Step 5:

[1032] The server determines the language of the preprocessed text data using natural language processing techniques.

[1033] (Input) Preprocessed text data

[1034] (Operation) The server uses a natural language processing algorithm to identify the language of the text data.

[1035] (Output) Language information of text data

[1036] Step 6:

[1037] The server analyzes the text data and extracts important keywords and phrases.

[1038] (Input) Text data and its linguistic information

[1039] (Operation) The server uses natural language processing tools such as SpaCy to analyze the text and extract important keywords and phrases.

[1040] (Output) Extracted keywords and phrases

[1041] Step 7:

[1042] The server generates a summary of the document based on important keywords and phrases.

[1043] (Input) Extracted keywords and phrases

[1044] (Operation) The server constructs a summary based on keywords and phrases, concisely summarizing important information.

[1045] (Output) Generated summary

[1046] Step 8:

[1047] If the document is not in the language specified by the user, the server uses an automatic translation function to translate the summary into the specified language.

[1048] (Input) Generated summary and user-specified language information

[1049] (Operation) The server calls the translation API and converts the summary into the specified language.

[1050] (Output) Translated summary

[1051] Step 9:

[1052] The server transmits the generated summary or the translated summary back to the terminal.

[1053] (Input) (Translated) Generated Summary

[1054] (Operation) The server executes a process to transmit the summary data to the terminal.

[1055] (Output) Summary sent to terminal

[1056] Step 10:

[1057] The terminal displays the received summary on the display of the smart glasses.

[1058] (Input) Summary data sent from the server

[1059] (Operation) The terminal formats the summary data for display and displays it on the smart glasses display.

[1060] (Output) Summary displayed on the smart glasses display

[1061] 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.

[1062] This invention relates to a method for efficiently analyzing contracts and other important documents and generating summaries using a system. It also aims to combine an emotion engine that recognizes the user's emotions to analyze and adjust summaries according to the user's emotional state. Specifically, the system involves a series of steps: a user uploads a document to the system using a terminal; a server analyzes the document, generates a summary, and then transmits the summary, adjusted based on the user's emotion, back to the user's terminal.

[1063] Explanation of the specific operation of the program

[1064] 1. Uploading documents

[1065] Users use a device to upload documents to the system for analysis and summarization. The device is equipped with a camera and microphone to collect emotional data through the user's facial expressions and voice.

[1066] 2. Document Receipt and Storage

[1067] The server receives the uploaded document and temporarily stores it. At the same time, the user's emotional data is also recorded. Basic information such as the document format (PDF, DOCX, etc.) and the emotional data is acquired.

[1068] 3. Document Preprocessing

[1069] The server performs pre-processing on the received documents, which includes converting the PDF files to text format and removing noise (e.g., images and redundant text).

[1070] 4. Language Identification

[1071] The server uses natural language processing (NLP) techniques to determine the language of the document, whether it is written in English, Japanese, French, etc.

[1072] 5. Document Analysis

[1073] Based on the language identified, the server uses NLP techniques to analyze the document, specifically extracting important keywords and phrases (e.g., contract terms, payment terms, liability clauses, etc.).

[1074] 6. Generating a Summary

[1075] Based on the analysis results, the server generates a summary of the document, which is a concise summary of the document's key points, including the extracted key points.

[1076] 7. Emotional Engine Adjustment

[1077] The server uses an emotion engine to adapt the analysis results to the emotional data collected from the user, for example, adjusting the results to produce a more concise and clear summary if the user is feeling stressed or confused.

[1078] 8. Multilingual support

[1079] If the language of the document is different from the language specified by the user, the server will use an automatic translation function to translate the document or abstract into the specified language, allowing the user to view the abstract in the language they require.

[1080] 9. Submitting the results

[1081] The server then sends the generated summary and analysis results to the user's device, including summaries adjusted by the emotion engine. The user receives the results and checks them on the screen.

[1082] Specific examples

[1083] Example 1: Processing an English contract

[1084] When a user uploads a PDF English contract titled "Service Agreement" to the system using their PC, a camera and microphone monitor the user's facial expressions and voice to collect emotional data. The server receives the document, converts it to text format, and removes noise. After determining that the document is written in English, it uses NLP technology to extract important information. A summary is generated based on this information and adjusted according to the user's emotional state. The adjusted summary is then sent to the user's PC. The user then reviews the summary on their PC and determines whether the contract terms are appropriate.

[1085] Example 2: Processing multilingual contracts

[1086] When a user uploads a contract written in French to the system using their smartphone, the smartphone's camera and microphone collect the user's emotions. The server receives the French document and performs the necessary preprocessing. It determines the document's language as French and uses NLP technology to extract key information. If the user specifies English, the French summary is translated into English using an automatic translation function. The translated summary is adjusted by the emotion engine and sent to the user's smartphone. The user then reviews the summary and reevaluates the contract content if necessary.

[1087] In this way, a system is realized that efficiently analyzes the content of documents and provides summaries that correspond to the user's emotional state, thereby reducing the burden on legal departments and speeding up contract signing.

[1088] The processing flow will be explained below.

[1089] Step 1:

[1090] The user uploads a document to the system using a device. From the device interface, the user selects the PDF file of the contract they wish to analyze and summarize, and clicks the "Upload" button. At the same time, the device's camera and microphone collect the user's facial expressions and voice data, which are then recorded as emotion data.

[1091] Step 2:

[1092] The server receives the uploaded document and temporarily stores it. The format information of the saved document (e.g., PDF, DOCX) is acquired. The user's emotion data sent from the device is also stored at the same time.

[1093] Step 3:

[1094] The server pre-processes the document: if the document is in PDF format, it performs operations to convert it to text format, and also removes noise from the document, such as unnecessary images and redundant text.

[1095] Step 4:

[1096] The server determines the language of the document. It uses natural language processing (NLP) techniques to automatically detect the language the document is written in. Once the language is determined, it launches an analysis process for that language.

[1097] Step 5:

[1098] The server analyzes the document. Based on the language identified, it uses NLP techniques to analyze each section of the document and extract important keywords and phrases. Specifically, it identifies important information such as contract terms, payment terms, and liability clauses, and extracts the necessary data.

[1099] Step 6:

[1100] The server generates a summary based on the analysis results, organizes the extracted important information, and generates a summary of the document. The summary is concisely compiled in a format that is easy for users to understand.

[1101] Step 7:

[1102] The server uses an emotion engine to analyze the user's emotion data, specifically through facial expression recognition and voice analysis, to determine the user's current emotional state (e.g., stress, excitement, calm).

[1103] Step 8:

[1104] The server then uses the emotion engine to adjust the summary of the document based on the analyzed emotion data. For example, if the user is feeling confused or stressed, the summary will be adjusted to be more concise and clear.

[1105] Step 9:

[1106] If the language of the document is different from the language specified by the user, the server uses an automatic translation function to translate the document or abstract into the specified language.

[1107] Step 10:

[1108] The server sends the generated summary and analysis results to the user's device, including summaries adjusted by the emotion engine. The user receives the results and checks them on the screen.

[1109] Step 11:

[1110] The user checks the analysis results and summary on the terminal, and determines whether the contract contents are valid based on the provided summary, and performs additional verification or corrections as necessary.

[1111] This series of steps results in a system that efficiently analyzes the content of documents and provides summaries tailored to the user's emotional state, reducing the burden on legal departments and speeding up contract execution.

[1112] Example 2

[1113] 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."

[1114] Many systems exist for quickly and accurately analyzing contracts and other important documents and generating summaries. However, these systems lack the ability to adjust their analysis results to take into account the user's emotional state. As a result, if a user is stressed or confused, the summary may be difficult to understand, making it difficult for them to make effective decisions. Another issue is that multilingual support is often limited, meaning that the summary cannot be viewed in the user's preferred language.

[1115] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: a means for a user to upload data to the system; a means for the server to receive and store the uploaded data; a means for the server to preprocess the data and convert it into a format suitable for analysis; a means for the server to determine the language of the data; a means for the server to analyze each portion of the data using natural language processing technology and extract important information; a means for the server to generate a summary of the data based on the analysis results; a means for the server to send the summary generated by the server to the user's terminal; a means for the user to check the analysis results and the summary through the terminal; an emotion data collection means including a sensor attached to the terminal for collecting user emotion data; and a means for the server to analyze the user's emotion data and adjust the generated summary. This allows the server to provide a summary adjusted according to the user's emotional state, support multiple languages, and more quickly and accurately understand the content of a document.

[1116] "User" means any person or entity that uses the System to upload documents or receive summaries.

[1117] "Data" refers to information in any form, including contracts and other important documents.

[1118] "System" refers to the collection of devices and software that provides a set of processes for uploading data, analyzing, generating summaries, translating, and adjusting emotion data.

[1119] "Server" refers to a computing device within the system that receives, stores, pre-processes, analyzes, language-determines, summarizes, adjusts, translates, and transmits results.

[1120] "Terminal" refers to a computer device (PC, smartphone, tablet, etc.) used by a user to upload data and check analysis results.

[1121] "Preprocessing" refers to a series of operations that convert data into a format suitable for analysis, including conversion to text format and noise removal.

[1122] "Language determination" refers to the process of identifying the language in which data is written.

[1123] "Natural language processing technology" refers to computer technology used to analyze data and extract important information.

[1124] "Key information" refers to specific keywords or phrases extracted from each piece of data.

[1125] A "summary" is a document that briefly summarizes the key points of the data.

[1126] "Emotion data" refers to data that indicates the emotional state of a user, collected from facial expressions and voice.

[1127] "Emotion data collection means" refers to a device that collects emotional data using sensors, cameras, and microphones attached to a terminal.

[1128] "Tuning" refers to the process of optimizing summaries and analysis results based on the user's emotional state.

[1129] "Automatic translation" refers to technology for translating data or summaries into another language.

[1130] This invention relates to a method for efficiently analyzing contracts and other important documents and generating summaries using a system. Furthermore, it aims to combine this system with an emotion engine that recognizes the user's emotions to adjust the analysis and summaries according to the user's emotional state. Specific embodiments for implementing this invention are described below.

[1131] System configuration

[1132] Document upload

[1133] Users upload documents such as contracts to the system using devices (PCs, smartphones, tablets, etc.). The devices are equipped with cameras and microphones, allowing for the collection of emotional data through the user's facial expressions and voice.

[1134] Document Receipt and Storage

[1135] The server receives the uploaded document and temporarily stores it in storage. At this time, basic information such as the document format (PDF, DOCX, etc.) and emotion data is acquired and recorded as metadata.

[1136] Document Preprocessing

[1137] The server performs pre-processing on the received documents, which includes converting the PDF files to text format and removing noise (e.g., images and overlapping text) using specific document analysis software (e.g., OCR technology).

[1138] Language Identification

[1139] The server uses natural language processing (NLP) techniques to determine the language of the document, whether it is written in English, Japanese, French, etc. The techniques used include, for example, language identification algorithms.

[1140] Document analysis and summary generation

[1141] The server analyzes the document using NLP technology based on the determined language. It extracts important keywords and phrases (e.g., contract terms, payment terms, liability clauses, etc.) and generates a document summary based on the analysis results. The summary is a concise summary of the document's key points. Specific technologies used include keyword extraction algorithms and summary generation models.

[1142] Emotional Engine Adjustment

[1143] The server uses an emotion engine to analyze the emotional data collected from the user and incorporate it into the analysis results. For example, if the user is feeling stressed or confused, the server will adjust the summary to generate a more concise and clear summary. Facial recognition software and voice analysis technology are used to analyze the emotional data.

[1144] Multilingual support and results transmission

[1145] If the document language is different from the language specified by the user, the server uses an automatic translation function to translate the document or summary into the specified language. This function allows the user to view the summary in the language they require. This is achieved using a translation API. The generated summary and analysis results are then sent to the user's device, where the user can view the results.

[1146] Specific examples

[1147] Example 1: Processing an English contract

[1148] When a user uploads a PDF English contract titled "Service Agreement" to the system using their PC, a camera and microphone monitor the user's facial expressions and voice to collect emotional data. The server receives the document, converts it to text format, and removes noise. After determining that the document is written in English, it uses NLP technology to extract key information. A summary is generated based on this information and adjusted according to the user's emotional state. The server then sends the adjusted summary to the user's PC. The user then reviews the summary on their PC and determines whether the contract terms are appropriate.

[1149] Example 2: Processing multilingual contracts

[1150] When a user uploads a contract written in French to the system using their smartphone, the smartphone's camera and microphone collect the user's emotions. The server receives the French document and performs the necessary preprocessing. It determines the document's language as French and uses NLP technology to extract key information. If the user specifies English, the French summary is translated into English using an automatic translation function. The translated summary is then adjusted by the emotion engine and sent to the user's smartphone. The user then reviews the summary and reevaluates the contract content if necessary.

[1151] Prompt Sentence Examples

[1152] "Please upload the English contract and summarize the key points. Also, please adjust it to take into account sentiment data."

[1153] "I have uploaded a document in French. Please translate it into English and provide a summary."

[1154] In this way, the system can efficiently analyze contracts and important documents and provide summaries tailored to the user's emotional state, reducing the burden on legal departments and speeding up contract closures.

[1155] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1156] Step 1: Upload your documents

[1157] The user uses a terminal to upload a document to the system. The input is a contract file selected by the user (e.g., a PDF file), and the output is the transfer of the file to the system. The terminal is equipped with a camera and microphone, which also collects emotional data from the user's facial expressions and voice. Specifically, the user clicks the "Upload" button and selects a document file.

[1158] Step 2: Receiving and storing documents

[1159] The server receives the uploaded document and temporarily stores it in storage. The input is the uploaded contract file and emotion data, and the output is the file stored on the server and the collected emotion data meta information. Specifically, the server receives the file and stores it in storage with a file name such as "contract.pdf." At this time, basic information such as the document format and emotion data is also saved.

[1160] Step 3: Preprocessing the Documents

[1161] The server performs preprocessing on the received document. The input is the saved contract file, and the output is text data suitable for analysis. Specifically, it converts the PDF file to text and removes noise such as images and overlapping text. Specifically, it uses OCR technology to extract text from the PDF and runs a script to remove unnecessary noise.

[1162] Step 4: Language Determination

[1163] The server uses natural language processing (NLP) techniques to determine the language of a document. The input is preprocessed text data, and the output is the document's language information (e.g., English, Japanese, French, etc.). Specifically, it analyzes the frequency of words and phrases in the document and runs a model to identify the language based on that.

[1164] Step 5: Parsing the Document

[1165] The server then uses NLP technology to analyze the document based on the language it has identified. The input is linguistic information and preprocessed text data, and the output is key keywords and phrases extracted through the analysis. Specifically, it runs an algorithm to extract keywords such as contract terms, payment terms, and liability clauses.

[1166] Step 6: Generate a summary

[1167] The server generates a document summary based on the analysis results. The input is the extracted keywords and phrases, and the output is the document summary. Specifically, it runs a script that generates a document summary based on the extracted keywords and an existing format.

[1168] Step 7: Emotional Engine Alignment

[1169] The server uses an emotion engine to adjust the summary based on the user's emotion data. The input is the generated summary and emotion data, and the output is a summary adjusted to reflect the emotion. Specifically, if the user's facial expression data indicates "stress," the server runs an algorithm to reconstruct the summary into simpler language.

[1170] Step 8: Multilingual

[1171] If the document language is different from the language specified by the user, the server uses an automatic translation function to translate the summary into the specified language. The input is the adjusted summary, and the output is the translated summary. Specifically, the server calls an API to translate the summary generated in French into English.

[1172] Step 9: Sending the results

[1173] The server sends the generated summary and analysis results to the user's device. The input is the adjusted and translated summary, and the output is the summary sent to the user's device. Specifically, the generated summary and analysis results are sent to the user's device in a data format such as JSON, and the device displays them.

[1174] (Application example 2)

[1175] 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."

[1176] Conventional document analysis systems can analyze documents and generate summaries, but they do not take into account the user's emotional state, resulting in the inability to provide optimal information to users. Furthermore, they are unable to provide appropriate support when users feel stressed or confused, and are particularly inadequate in situations requiring quick and accurate judgments, particularly in the security service field. There is a need for a system that can improve this situation by adjusting document analysis results and summaries based on the user's emotional state.

[1177] 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 a user to upload a document to the system; means for the server to receive and store the uploaded document; means for the server to preprocess the document and convert it into a format suitable for analysis; means for the server to determine the language of the document; means for the server to analyze each section of the document using natural language processing technology and extract important keywords and phrases; means for the server to generate a summary of the document based on the analysis results; means for the server to recognize the user's emotional state and adjust the summary based on that state; means for the server to send the summary generated by the server to the user's terminal; and means for the user to check the analysis results and the summary through the terminal. This makes it possible to provide a summary adjusted according to the user's emotional state.

[1178] "User" refers to the person who uploads documents to the analysis system and reviews the analysis results and summaries.

[1179] "Document" refers to the documents or information to be analyzed and summarized, and may exist in formats such as PDF or DOCX.

[1180] "System" refers to the entire device and software configuration that performs a series of processes in which a user uploads a document and the server analyzes and summarizes it.

[1181] "Server" refers to a computer and associated software for receiving, storing, and processing documents uploaded by users.

[1182] "Preprocessing" refers to a series of operations to convert a document into a format suitable for analysis, such as converting PDF files to text or removing noise.

[1183] "Natural language processing technology" is a technology in the field of computer science that refers to methods and algorithms that allow machines to understand, analyze, and generate human language.

[1184] "Emotional state" refers to the psychological or emotional state recognized from the user's facial expression or voice, and includes, for example, stress, happiness, surprise, etc.

[1185] A "summary" is a concise summary of the important points from an analyzed document that can be quickly understood by a user.

[1186] "Upload" refers to the operation of a user sending a document from their own terminal to the system.

[1187] "Terminal" refers to a device used by a user, such as a smartphone or PC, that interfaces with the system.

[1188] "Adjustment" refers to making changes or modifications to the content of an already generated summary based on the user's emotional state.

[1189] "Automatic translation function" refers to the function by which the server mechanically translates documents and summaries into a language specified by the user.

[1190] "Important sections" are those parts of the document that require special attention, such as specific matters like contract terms, payment terms, and liability clauses.

[1191] This invention is a system for efficiently analyzing documents and generating summaries in the field of security services, and can further adjust the summaries according to the emotional state of the user. The system is implemented through the following steps:

[1192] composition

[1193] 1. Hardware configuration:

[1194] The user's device is a smartphone or a PC, which is equipped with a camera and a microphone.

[1195] The server is equipped with a high-performance CPU and sufficient memory.

[1196] Data storage requires large amounts of storage for analysis.

[1197] 2. Software configuration:

[1198] For natural language processing technology, Google Cloud Natural Language API is used.

[1199] Amazon Rekognition is used for sentiment analysis.

[1200] The automatic translation function uses the Google Translate API.

[1201] The server software used is Apache HTTP Server and Node.js.

[1202] MySQL is used as the database.

[1203] Explanation of program processing

[1204] The server receives documents uploaded by users, analyzes them, and summarizes them. This process involves the following major steps:

[1205] 1. Upload and save documents:

[1206] Users upload documents to the system via their smartphones or PCs. The server receives and stores them, and the documents are used for analysis and summary generation.

[1207] 2. Document preprocessing:

[1208] The server pre-processes the uploaded documents, which includes converting PDF files to text format and removing unwanted noise.

[1209] 3. Document Language Determination:

[1210] The server uses the Google Cloud Natural Language API to determine the language of the document, which tells it what language it is written in.

[1211] 4. Document Analysis:

[1212] The server uses natural language processing techniques to analyze the content of the document and extract important keywords and phrases.

[1213] 5. Generate a summary:

[1214] The server generates a summary of the document based on the extracted keywords and phrases, which is concise and easy for users to understand.

[1215] 6. Emotional engine regulation:

[1216] The server uses Amazon Rekognition to analyze the user's emotional data and adjusts the generated summary according to the user's emotional state—for example, if the user is feeling stressed, the summary will be more concise and clear.

[1217] 7. Submit and review your abstract:

[1218] The server then sends the adjusted summary to the user's device, where the user can check the summary on their smartphone or computer and take any necessary action.

[1219] Specific examples

[1220] For example, if a lawyer uploads a security services contract via smartphone, the system analyzes the document and summarizes key sections. Furthermore, if the lawyer is under stress, the summary can be adjusted to be concise and clear. This process ensures fast and accurate information delivery.

[1221] Prompt Sentence Examples

[1222] Analyze and summarize the contracts uploaded to the system. Adjust the summary and reply based on the user's sentiment.

[1223] In this way, the present invention can efficiently analyze the content of documents and provide summaries according to the emotional state of the user, thereby improving reliability and efficiency in the field of security services.

[1224] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1225] Step 1:

[1226] The user uploads a document to the system. The user uses a smartphone or PC to upload the document to be analyzed to the system. The input is a document file (PDF, DOCX, etc.), and the output is the data uploaded to the server.

[1227] Step 2:

[1228] The server receives and stores the uploaded document. The server receives the document file sent by the user and stores it in a database. The input is the document file sent by the user, and the output is the document data stored in the database.

[1229] Step 3:

[1230] The server preprocesses the document and converts it into a format suitable for analysis. The server converts the document into a text format and removes unnecessary noise (images, duplicated text, etc.). The input is the document data stored in the database, and the output is the preprocessed text data.

[1231] Step 4:

[1232] The server determines the language of the document. The server determines the language of the document using the Google Cloud Natural Language API. The input is preprocessed text data, and the output is the language information of the document.

[1233] Step 5:

[1234] The server uses natural language processing technology to analyze each section of the document and extract important keywords and phrases. The server uses natural language processing technology (Google Cloud Natural Language API) to analyze the document and extract important keywords and phrases. The input is the text data of the document, and the output is the extracted keywords and phrases.

[1235] Step 6:

[1236] The server generates a document summary based on the analysis results. The server generates a document summary based on the extracted keywords and phrases. The input is the extracted keywords and phrases, and the output is the generated summary.

[1237] Step 7:

[1238] The server recognizes the user's emotional state and adjusts the summary based on that state. The server uses Amazon Rekognition to analyze the user's emotional state and adjust the generated summary. The input is the user's emotional data (facial expressions and voice) and the generated summary, and the output is the adjusted summary.

[1239] Step 8:

[1240] The server sends the generated summary to the user's terminal. The server sends the adjusted summary to the user's terminal. The input is the adjusted summary, and the output is the summary sent to the user's terminal.

[1241] Step 9:

[1242] The user checks the analysis results and summary via their device. The user checks the analysis results and summary sent from the server via their smartphone or PC. The input is the summary sent from the server, and the output is the user's understanding and any necessary actions.

[1243] 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.

[1244] 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.

[1245] 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.

[1246] [Fourth embodiment]

[1247] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1248] 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.

[1249] 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).

[1250] 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.

[1251] 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.

[1252] 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).

[1253] 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. 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.

[1254] 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.

[1255] 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.

[1256] 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.

[1257] 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.

[1258] 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.

[1259] 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."

[1260] This invention relates to a method for efficiently analyzing contracts and other important documents and generating summaries using a system, in which a user uploads the document to the system using a terminal, a server analyzes the document, generates a summary, and sends it back to the user's terminal.

[1261] Explanation of the specific operation of the program

[1262] 1. Uploading documents

[1263] Users use their devices to upload documents they wish to analyze and summarize to the system, for example by selecting a PDF file of a contract from a PC, smartphone, or other device, and clicking the upload button.

[1264] 2. Document Receipt and Storage

[1265] The server receives the uploaded document and temporarily stores it, at which point it obtains basic information about the document's format (e.g. PDF, DOCX).

[1266] 3. Document Preprocessing

[1267] The server performs pre-processing on the received documents, which includes converting the PDF files to text format and removing noise (e.g., images and redundant text).

[1268] 4. Language Identification

[1269] The server uses natural language processing (NLP) techniques to automatically determine the language of the document, whether it is written in English, Japanese, French, etc.

[1270] 5. Document Analysis

[1271] Based on the language identified, the server uses NLP technology to analyze the document, extracting important keywords and phrases (e.g., contract terms, payment terms, liability clauses, etc.), and analyzes the document structure, dividing it into sections for analysis.

[1272] 6. Generating a Summary

[1273] The server then generates a summary of the document based on the analysis results. The summary is a concise summary of the document's key points, including the most important information extracted. For example, a contract summary might include information such as "Contract period: one year" and "Payment terms: payment by the end of each month."

[1274] 7. Multilingual support

[1275] If the language of the document is different from the language specified by the user, the server will use an automatic translation function to translate the document and summary into the specified language, allowing the user to view the summary in the language they require.

[1276] 8. Submitting the results

[1277] The server sends the generated summary and analysis results to the user's terminal, where the user can check the summary received and perform additional verification or correction as necessary.

[1278] Specific examples

[1279] Example 1: Processing an English contract

[1280] A user uploads an English contract titled "Service Agreement" in PDF format to the system using their PC. The server receives the document, converts it to text format, and removes noise. After determining that the document is in English, it uses NLP technology to extract important information such as the contract period, payment terms, and liability clauses. A summary is generated based on this information and sent to the user's PC. The user then checks the summary on their PC and determines whether the contract contents are reasonable.

[1281] Example 2: Processing multilingual contracts

[1282] A user uploads a contract written in French to the system using their smartphone. The server receives the French document and performs the necessary preprocessing. It determines that the document is in French and uses NLP technology to extract important information. If the user specifies English, the server uses an automatic translation function to translate the French summary into English. The translated summary is sent to the user's smartphone, where the user can review it.

[1283] This allows the system to efficiently analyze the contents of contracts and important documents, generate summaries, and provide them to users, thereby reducing the burden on legal departments and speeding up contract signing processes.

[1284] The processing flow will be explained below.

[1285] Step 1:

[1286] The user uses the terminal to upload the document to be analyzed and summarized to the system. For example, the user selects a PDF file of a contract from the terminal interface and clicks the "Upload" button.

[1287] Step 2:

[1288] The server receives the uploaded document. The received document is temporarily stored on the server and its format information (PDF, DOCX, etc.) is acquired.

[1289] Step 3:

[1290] The server pre-processes the document. If the document is in PDF format, it converts it to text format. It also removes unwanted images, redundant text, and other noise from the document.

[1291] Step 4:

[1292] The server determines the language of the document, automatically detecting the language using natural language processing (NLP) techniques and preparing the next analysis step based on the results.

[1293] Step 5:

[1294] The server analyzes the document. Based on the language identified, it uses NLP techniques to analyze each section of the document and extract important keywords and phrases. For example, in the case of a contract, it identifies important information such as "contract duration," "payment terms," ​​and "liability clauses."

[1295] Step 6:

[1296] The server generates a summary based on the analysis results, organizes the extracted important information, and generates a summary of the document. The summary is concise and easy for users to understand.

[1297] Step 7:

[1298] The server determines whether multilingual support is necessary, and if the document language is different from the language specified by the user, it uses an automatic translation function to translate the document or abstract into the specified language.

[1299] Step 8:

[1300] The server sends the generated summary and analysis results to the user's terminal, where the user receives the results and checks them on the screen.

[1301] Step 9:

[1302] The user checks the analysis results and summary on the terminal, and determines whether the contract contents are valid based on the provided summary, and performs additional verification or corrections as necessary.

[1303] This series of steps results in a system that efficiently analyzes the contents of documents and quickly provides users with summaries, reducing the burden on legal departments and streamlining contract execution.

[1304] Example 1

[1305] 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."

[1306] To efficiently analyze documents and generate summaries, it is necessary to process documents in different formats in a unified manner, accurately extract information using natural language processing technology, and generate summaries that include particularly important sections. Furthermore, when documents are written in different languages, it is necessary to perform appropriate translation. These challenges have been difficult to completely address with conventional systems.

[1307] 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.

[1308] In this invention, the server includes: a means for a user to upload a document to the system; a means for the server to receive and store the uploaded document; a means for the server to preprocess the document and convert it into a format suitable for analysis; a means for the server to determine the language of the document; a means for the server to analyze each section of the document using natural language processing technology and extract important information; a means for the server to generate a summary of the document based on the analysis results; a means for the server to send the summary generated by the server to the user's terminal; a means for the user to check the analysis results and the summary via the terminal; and a means for the server to select an appropriate preprocessing method depending on the document format. This makes it possible to consistently and accurately extract information and generate summaries including important sections for documents of different formats and languages.

[1309] "User" means a person who uses the system to upload documents and check analysis results.

[1310] A "server" is a device or system that receives uploaded documents, stores them, and performs various processes such as preprocessing, analysis, summary generation, and translation.

[1311] "Upload" refers to the act of sending a document that a user possesses to the system.

[1312] "Preprocessing" refers to the process of converting documents into a format suitable for analysis, and includes text data extraction and noise removal.

[1313] "Language determination" is the process of automatically identifying the language in which an uploaded document is written.

[1314] "Natural language processing technology" refers to technology for analyzing documents and extracting information, including technology for extracting noun phrases and identifying important phrases.

[1315] "Summary generation" is the process of creating a summary that succinctly summarizes the important points of a document based on the analysis results.

[1316] The "automatic translation function" is a function that automatically translates documents and summaries in different languages ​​into a language specified by the user.

[1317] "Certain Important Sections" are those parts of the document that contain particularly important information, such as the terms of the contract, payment terms, and liability clauses.

[1318] "Analysis results" refer to important information and keywords extracted using natural language processing technology.

[1319] "Terminal" refers to a device used by a user to upload documents or check analysis results, and includes PCs, smartphones, etc.

[1320] "Noise removal" is the process of eliminating information unnecessary for analysis (such as images and duplicated text).

[1321] "Text data" refers to character string data extracted from a document.

[1322] "Tokenization" is the process of dividing the text in a document into units (words or phrases) that are easier to analyze.

[1323] "Part-of-speech tagging" is the process of automatically assigning each word its part of speech.

[1324] The above terms are defined so that the technical scope and embodiments of the invention can be clearly understood.

[1325] The present invention relates to a method for efficiently analyzing documents and generating summaries using a system, which is implemented through a series of steps in which a user uploads a document to the system using a terminal, a server analyzes the document, generates a summary, and sends it back to the user's terminal.

[1326] Hardware and software used

[1327] Server: Uses a computer with high-performance computing resources and provides functions for receiving, storing, preprocessing, analyzing, summarizing, translating, and transmitting documents.

[1328] Device: A PC or smartphone used by a user to upload documents and receive and check summaries.

[1329] Software tools:

[1330] PDF text extraction tools (pdfminer, PyMuPDF)

[1331] Natural language processing libraries (spaCy, NLTK)

[1332] Language detection library (langdetect)

[1333] Translation API (Google Translate API, DeepL API)

[1334] Explanation of program processing

[1335] 1. Uploading documents

[1336] The user uses a device to upload the document they want to analyze and summarize to the system. For example, they select a PDF file of a contract from the file selection screen on their PC or smartphone and click the "Upload" button. The device then sends the selected document to the server.

[1337] 2. Document Receipt and Storage

[1338] The server receives the document sent from the device and temporarily stores it, identifying the document format (PDF, DOCX, etc.) and obtaining the associated metadata.

[1339] 3. Document Preprocessing

[1340] The server preprocesses the documents. Specifically, it uses tools (such as pdfminer or PyMuPDF) to convert PDF files into text format, extracts the text data, and then performs a filtering process to remove noise (such as images or overlapping text) from the documents.

[1341] 4. Language Identification

[1342] The server automatically determines the language of the document, specifically using libraries such as langdetect or spaCy to determine whether the document is written in English, Japanese, French, etc.

[1343] 5. Document Analysis

[1344] Based on the language identified, the server analyzes the document using natural language processing techniques, such as tokenization and part-of-speech tagging (POS tagging), to extract noun phrases and key phrases (such as contract terms, payment terms, and liability clauses), using libraries such as spaCy and NLTK.

[1345] 6. Generating a Summary

[1346] The server generates a summary of the document based on the analyzed data. The summary is generated using a text summarization algorithm (TextRank or an extractive summarization model). The generated summary includes extracted important information (e.g., contract period: one year, payment terms: payment by the end of each month).

[1347] 7. Multilingual support

[1348] If the language specified by the user is different from the language of the document, the server translates the summary into the specified language using an automatic translation function, for example, using the Google Translate API or DeepL API to translate the summary from English to Japanese.

[1349] 8. Submitting the results

[1350] The server sends the generated summary and analysis results to the user's device, where the user can check the received summary and perform additional verification or corrections as necessary.

[1351] Specific examples

[1352] Example 1: Processing an English contract

[1353] Using a PC, a user selects an English contract in PDF format titled "Service Agreement" and uploads it to the system. The device sends the selected PDF file to the server. The server receives the document and uses pdfminer to extract text data and remove noise. The server uses the langdetect library to determine that the document is written in English. The server uses spaCy to extract important keywords (such as contract term, payment terms, and liability clauses). The server uses the TextRank algorithm to summarize the document. The summary includes phrases such as "Contract term: 1 year" and "Payment terms: Payment by the end of each month." The server then sends the generated summary to the user's PC. The user then reviews the summary on their PC and determines whether the contract terms are reasonable.

[1354] Example 2: Processing multilingual contracts

[1355] A user uploads a French contract titled "Contrat de Service" to the system using a smartphone. The device sends the selected DOCX file to the server. The server receives the document and uses PyMuPDF to extract text data and remove noise. The server uses spaCy's multilingual model to determine that the document is in French. The server uses a French NLP model to extract important keywords. The server generates a summary based on the extracted data. The server uses the DeepL API to translate the French summary into English. The server sends the translated summary to the user's smartphone. The user reviews the summary and understands the contract.

[1356] Prompt Sentence Examples

[1357] Below is an example of a prompt sentence.

[1358] 1. "Generate a summary of the contract written in English."

[1359] 2. "Analyze the uploaded PDF document and provide key keywords and a brief summary."

[1360] 3. "Translate this French document into English and summarize the key points."

[1361] In this way, the present invention improves the efficiency of document analysis and summary generation, thereby improving the efficiency of legal departments and business operations.

[1362] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1363] Step 1:

[1364] A user uses a terminal to upload a document to the system. Specifically, the user selects the document to be processed (e.g., PDF or DOCX format) from the terminal's file selection screen and clicks the "Upload" button. The input of the terminal is the document file selected by the user, and the output is the transmission of the document file to the server.

[1365] Step 2:

[1366] The server receives documents sent from the device and temporarily stores them. The input is the document file sent from the device, and the output is a file stored in the server's storage area. At this time, the server identifies the document format (PDF, DOCX, etc.) and obtains related metadata (file format, file size, etc.).

[1367] Step 3:

[1368] The server preprocesses the documents. Specifically, it uses tools such as pdfminer and PyMuPDF to convert PDF files into text format and extracts the text data. It also performs a filtering process to remove noise (images, overlapping text, etc.) from the documents. The input is the saved document file, and the output is the text data converted into text format and with noise removed.

[1369] Step 4:

[1370] The server automatically determines the language of a document using natural language processing technology. Specifically, it uses libraries such as langdetect and spaCy to determine the language in which the document is written. The input is preprocessed text data, and the output is the document's language information.

[1371] Step 5:

[1372] Based on the determined language, the server uses natural language processing techniques (such as spaCy or NLTK) to analyze the document. Specifically, it performs tokenization and part-of-speech tagging (POS tagging) to extract noun phrases and important phrases (such as contract terms, payment terms, and liability clauses). The input is the document's linguistic information and preprocessed text data, and the output is the extracted important information and keywords.

[1373] Step 6:

[1374] The server generates a document summary based on the extracted data. It uses a text summarization algorithm such as TextRank or an extractive summarization model to generate the summary. The generated summary includes extracted key information (e.g., contract period: one year, payment terms: payment by the end of each month). The input is the extracted key information and keywords, and the output is the generated summary.

[1375] Step 7:

[1376] If the language specified by the user is different from the language of the document, the server translates the summary into the specified language using an automatic translation function (such as Google Translate API or DeepL API). The input is the generated summary and the language information specified by the user, and the output is the translated summary.

[1377] Step 8:

[1378] The server sends the generated summary and analysis results to the user's device. The user checks the received summary on their own device and performs additional verification or correction as necessary. The input is the generated (translated) summary and analysis results, and the output is transmission to the user's device.

[1379] (Application example 1)

[1380] 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."

[1381] In recent years, logistics centers have become required to process large volumes of documents, and there is a demand for the rapid and accurate understanding of the contents of contracts and other important documents. However, manual document analysis and summary generation is time-consuming and places a burden on human resources. Furthermore, when dealing with documents in different languages, translation costs and time increase, resulting in reduced work efficiency. Even with the introduction of systems using smart glasses, the above issues remain unresolved because existing technologies do not automate document content analysis.

[1382] 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.

[1383] In this invention, the server includes: a means for a user to upload a document to the system; a means for the server to receive and store the uploaded document; a means for the server to preprocess the document and convert it into a format suitable for analysis; a means for the server to determine the language of the document; a means for the server to analyze each section of the document using natural language processing technology and extract important keywords and phrases; a means for the server to generate a summary of the document based on the analysis results; a means for the server to send the summary generated by the server to a user's terminal; a means for the user to check the analysis results and the summary through the terminal; and a means for logistics center staff to wear smart glasses and scan contracts or important documents in packages with a camera, causing the system to automatically analyze the documents and display the summaries. This makes it possible to automatically analyze the contents of contracts and important documents and provide summaries quickly and accurately at logistics centers by using smart glasses. The summaries can automatically determine and translate language differences, allowing documents in different languages ​​to be processed efficiently.

[1384] A "user" is an individual or company employee who accesses the system to upload documents and review analysis results.

[1385] "Document" refers to text data containing contracts and other important information, and may be in formats such as PDF or DOCX.

[1386] "System" refers to the set of servers and associated software that receives documents uploaded by users, processes, analyzes, summarizes, and in some cases translates them, and provides the results to users.

[1387] The "server" is the device that forms the core of the system, and is a computer that processes documents received from users, analyzes them, generates summaries, and finally transmits the results to the users.

[1388] "Preprocessing" refers to the initial stage of processing to convert uploaded documents into a state suitable for analysis, and specifically includes text extraction and noise removal.

[1389] "Natural language processing technology" is a technology that enables computers to understand and use human language, specifically, technology that performs text analysis and extracts important keywords.

[1390] A "summary" is a concise text summary of the important information of a document analyzed by the server, presented in a user-friendly format.

[1391] "Translation" refers to the process of converting a document or abstract into a specified language, and is achieved by using an automatic translation function.

[1392] "Smart glasses" are visual devices that have built-in cameras and displays and can present augmented reality (AR) and information when worn by the user.

[1393] A "logistics center" is a facility where goods and packages are collected and prepared for shipment and delivery.

[1394] "Scanning" is the act of taking an image of a document using the camera in the smart glasses and importing it into the system as digital data.

[1395] The present invention is a system in which staff working at a logistics center wear smart glasses and scan contracts and important documents attached to packages with a camera, allowing the system to automatically analyze the documents and display summaries.

[1396] Explanation of the specific operation of the program

[1397] 1. Scanning documents

[1398] A user wears the smart glasses and uses the camera to take an image of a document, allowing work at the logistics center to proceed quickly and efficiently.

[1399] 2. Image to text conversion

[1400] The server receives the document image sent from the smart glasses and uses Tesseract OCR to extract text from the image, which then becomes the base data for analysis.

[1401] 3. Preprocessing and Language Identification

[1402] The server preprocesses the extracted text, removing noise and converting it into a format suitable for analysis, and then automatically determines the language of the document using natural language processing techniques.

[1403] 4. Text Analysis and Summary Generation

[1404] The server then analyzes the text using natural language processing technology such as SpaCy to extract important keywords and phrases based on the language identified, and generates a summary of the document based on the analysis results.

[1405] 5. Translation and result display

[1406] If the document is not in the language specified by the user, the server uses its automatic translation function to translate the summary into the specified language, and the translated summary is sent back to the smart glasses, where the user can view it on the glasses' display.

[1407] Hardware and software used

[1408] Hardware: Smart glasses (camera, display, CPU, memory) and server

[1409] Software: Tesseract OCR, SpaCy, natural language processing technology, translation API

[1410] Specific examples

[1411] Example 1: Processing an English contract

[1412] A staff member at the logistics center wears smart glasses and uses the camera to scan a PDF English contract titled "Service Agreement." The server receives the document, converts it to text format, and removes noise. After determining that the document is written in English, it uses natural language processing technology to extract important information such as the contract period, payment terms, and liability clauses. A summary is generated based on this information and sent to the user's smart glasses. The user then reviews the summary on the smart glasses and determines whether the contract terms are reasonable.

[1413] Example 2: Processing multilingual contracts

[1414] A staff member at a logistics center uses smart glasses to scan a contract written in French with a camera. The server receives the French document and performs the necessary preprocessing. It determines that the document is in French and uses natural language processing technology to extract key information. If the user specifies English, the server uses an automatic translation function to translate the French summary into English. The translated summary is sent to the user's smart glasses, where the user can review it.

[1415] This means that by using smart glasses, it will be possible to automatically analyze the contents of contracts and important documents quickly and accurately on-site at logistics centers and provide summaries.

[1416] Example prompts to input to the generative AI model

[1417] "When a logistics center staff member scans a contract with smart glasses, the text is automatically extracted and displayed."

[1418] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1419] Step 1:

[1420] A user puts on the smart glasses and uses the camera to take an image of a document.

[1421] (Input) An image of a document captured by smart glasses

[1422] (Operation) A user points the camera at a document and presses the shutter button to capture an image of the document.

[1423] (Output) Document image data

[1424] Step 2:

[1425] The terminal transmits the captured image data to the server.

[1426] (Input) Image data of the photographed document

[1427] (Operation) Image data is transferred to the server using a transmission module in the terminal.

[1428] (Output) Image data sent to the server

[1429] Step 3:

[1430] The server converts the received image data into text data using Tesseract OCR.

[1431] (Input) Image data sent to the server

[1432] (Operation) The server calls Tesseract OCR and performs the process of extracting text from the image.

[1433] (Output) Extracted text data

[1434] Step 4:

[1435] The server preprocesses the extracted text data and converts it into a format suitable for analysis.

[1436] (Input) Extracted text data

[1437] (Operation) The server performs preprocessing to remove noise and unnecessary information from the text data.

[1438] (Output) Preprocessed text data

[1439] Step 5:

[1440] The server determines the language of the preprocessed text data using natural language processing techniques.

[1441] (Input) Preprocessed text data

[1442] (Operation) The server uses a natural language processing algorithm to identify the language of the text data.

[1443] (Output) Language information of text data

[1444] Step 6:

[1445] The server analyzes the text data and extracts important keywords and phrases.

[1446] (Input) Text data and its linguistic information

[1447] (Operation) The server uses natural language processing tools such as SpaCy to analyze the text and extract important keywords and phrases.

[1448] (Output) Extracted keywords and phrases

[1449] Step 7:

[1450] The server generates a summary of the document based on important keywords and phrases.

[1451] (Input) Extracted keywords and phrases

[1452] (Operation) The server constructs a summary based on keywords and phrases, concisely summarizing important information.

[1453] (Output) Generated summary

[1454] Step 8:

[1455] If the document is not in the language specified by the user, the server uses an automatic translation function to translate the summary into the specified language.

[1456] (Input) Generated summary and user-specified language information

[1457] (Operation) The server calls the translation API and converts the summary into the specified language.

[1458] (Output) Translated summary

[1459] Step 9:

[1460] The server transmits the generated summary or the translated summary back to the terminal.

[1461] (Input) (Translated) Generated Summary

[1462] (Operation) The server executes a process to transmit the summary data to the terminal.

[1463] (Output) Summary sent to terminal

[1464] Step 10:

[1465] The terminal displays the received summary on the display of the smart glasses.

[1466] (Input) Summary data sent from the server

[1467] (Operation) The terminal formats the summary data for display and displays it on the smart glasses display.

[1468] (Output) Summary displayed on the smart glasses display

[1469] 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.

[1470] This invention relates to a method for efficiently analyzing contracts and other important documents and generating summaries using a system. It also aims to combine an emotion engine that recognizes the user's emotions to analyze and adjust summaries according to the user's emotional state. Specifically, the system involves a series of steps: a user uploads a document to the system using a terminal; a server analyzes the document, generates a summary, and then transmits the summary, adjusted based on the user's emotion, back to the user's terminal.

[1471] Explanation of the specific operation of the program

[1472] 1. Uploading documents

[1473] Users use a device to upload documents to the system for analysis and summarization. The device is equipped with a camera and microphone to collect emotional data through the user's facial expressions and voice.

[1474] 2. Document Receipt and Storage

[1475] The server receives the uploaded document and temporarily stores it. At the same time, the user's emotional data is also recorded. Basic information such as the document format (PDF, DOCX, etc.) and the emotional data is acquired.

[1476] 3. Document Preprocessing

[1477] The server performs pre-processing on the received documents, which includes converting the PDF files to text format and removing noise (e.g., images and redundant text).

[1478] 4. Language Identification

[1479] The server uses natural language processing (NLP) techniques to determine the language of the document, whether it is written in English, Japanese, French, etc.

[1480] 5. Document Analysis

[1481] Based on the language identified, the server uses NLP techniques to analyze the document, specifically extracting important keywords and phrases (e.g., contract terms, payment terms, liability clauses, etc.).

[1482] 6. Generating a Summary

[1483] Based on the analysis results, the server generates a summary of the document, which is a concise summary of the document's key points, including the extracted key points.

[1484] 7. Emotional Engine Adjustment

[1485] The server uses an emotion engine to adapt the analysis results to the emotional data collected from the user, for example, adjusting the results to produce a more concise and clear summary if the user is feeling stressed or confused.

[1486] 8. Multilingual support

[1487] If the language of the document is different from the language specified by the user, the server will use an automatic translation function to translate the document or abstract into the specified language, allowing the user to view the abstract in the language they require.

[1488] 9. Submitting the results

[1489] The server then sends the generated summary and analysis results to the user's device, including summaries adjusted by the emotion engine. The user receives the results and checks them on the screen.

[1490] Specific examples

[1491] Example 1: Processing an English contract

[1492] When a user uploads a PDF English contract titled "Service Agreement" to the system using their PC, a camera and microphone monitor the user's facial expressions and voice to collect emotional data. The server receives the document, converts it to text format, and removes noise. After determining that the document is written in English, it uses NLP technology to extract important information. A summary is generated based on this information and adjusted according to the user's emotional state. The adjusted summary is then sent to the user's PC. The user then reviews the summary on their PC and determines whether the contract terms are appropriate.

[1493] Example 2: Processing multilingual contracts

[1494] When a user uploads a contract written in French to the system using their smartphone, the smartphone's camera and microphone collect the user's emotions. The server receives the French document and performs the necessary preprocessing. It determines the document's language as French and uses NLP technology to extract key information. If the user specifies English, the French summary is translated into English using an automatic translation function. The translated summary is adjusted by the emotion engine and sent to the user's smartphone. The user then reviews the summary and reevaluates the contract content if necessary.

[1495] In this way, a system is realized that efficiently analyzes the content of documents and provides summaries that correspond to the user's emotional state, thereby reducing the burden on legal departments and speeding up contract signing.

[1496] The processing flow will be explained below.

[1497] Step 1:

[1498] The user uploads a document to the system using a device. From the device interface, the user selects the PDF file of the contract they wish to analyze and summarize, and clicks the "Upload" button. At the same time, the device's camera and microphone collect the user's facial expressions and voice data, which are then recorded as emotion data.

[1499] Step 2:

[1500] The server receives the uploaded document and temporarily stores it. The format information of the saved document (e.g., PDF, DOCX) is acquired. The user's emotion data sent from the device is also stored at the same time.

[1501] Step 3:

[1502] The server pre-processes the document: if the document is in PDF format, it performs operations to convert it to text format, and also removes noise from the document, such as unnecessary images and redundant text.

[1503] Step 4:

[1504] The server determines the language of the document. It uses natural language processing (NLP) techniques to automatically detect the language the document is written in. Once the language is determined, it launches an analysis process for that language.

[1505] Step 5:

[1506] The server analyzes the document. Based on the language identified, it uses NLP techniques to analyze each section of the document and extract important keywords and phrases. Specifically, it identifies important information such as contract terms, payment terms, and liability clauses, and extracts the necessary data.

[1507] Step 6:

[1508] The server generates a summary based on the analysis results, organizes the extracted important information, and generates a summary of the document. The summary is concisely compiled in a format that is easy for users to understand.

[1509] Step 7:

[1510] The server uses an emotion engine to analyze the user's emotion data, specifically through facial expression recognition and voice analysis, to determine the user's current emotional state (e.g., stress, excitement, calm).

[1511] Step 8:

[1512] The server then uses the emotion engine to adjust the summary of the document based on the analyzed emotion data. For example, if the user is feeling confused or stressed, the summary will be adjusted to be more concise and clear.

[1513] Step 9:

[1514] If the language of the document is different from the language specified by the user, the server uses an automatic translation function to translate the document or abstract into the specified language.

[1515] Step 10:

[1516] The server sends the generated summary and analysis results to the user's device, including summaries adjusted by the emotion engine. The user receives the results and checks them on the screen.

[1517] Step 11:

[1518] The user checks the analysis results and summary on the terminal, and determines whether the contract contents are valid based on the provided summary, and performs additional verification or corrections as necessary.

[1519] This series of steps results in a system that efficiently analyzes the content of documents and provides summaries tailored to the user's emotional state, reducing the burden on legal departments and speeding up contract execution.

[1520] Example 2

[1521] 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."

[1522] Many systems exist for quickly and accurately analyzing contracts and other important documents and generating summaries. However, these systems lack the ability to adjust their analysis results to take into account the user's emotional state. As a result, if a user is stressed or confused, the summary may be difficult to understand, making it difficult for them to make effective decisions. Another issue is that multilingual support is often limited, meaning that the summary cannot be viewed in the user's preferred language.

[1523] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: a means for a user to upload data to the system; a means for the server to receive and store the uploaded data; a means for the server to preprocess the data and convert it into a format suitable for analysis; a means for the server to determine the language of the data; a means for the server to analyze each portion of the data using natural language processing technology and extract important information; a means for the server to generate a summary of the data based on the analysis results; a means for the server to send the summary generated by the server to the user's terminal; a means for the user to check the analysis results and the summary through the terminal; an emotion data collection means including a sensor attached to the terminal for collecting user emotion data; and a means for the server to analyze the user's emotion data and adjust the generated summary. This allows the server to provide a summary adjusted according to the user's emotional state, support multiple languages, and more quickly and accurately understand the content of a document.

[1524] "User" means any person or entity that uses the System to upload documents or receive summaries.

[1525] "Data" refers to information in any form, including contracts and other important documents.

[1526] "System" refers to the collection of devices and software that provides a set of processes for uploading data, analyzing, generating summaries, translating, and adjusting emotion data.

[1527] "Server" refers to a computing device within the system that receives, stores, pre-processes, analyzes, language-determines, summarizes, adjusts, translates, and transmits results.

[1528] "Terminal" refers to a computer device (PC, smartphone, tablet, etc.) used by a user to upload data and check analysis results.

[1529] "Preprocessing" refers to a series of operations that convert data into a format suitable for analysis, including conversion to text format and noise removal.

[1530] "Language determination" refers to the process of identifying the language in which data is written.

[1531] "Natural language processing technology" refers to computer technology used to analyze data and extract important information.

[1532] "Key information" refers to specific keywords or phrases extracted from each piece of data.

[1533] A "summary" is a document that briefly summarizes the key points of the data.

[1534] "Emotion data" refers to data that indicates the emotional state of a user, collected from facial expressions and voice.

[1535] "Emotion data collection means" refers to a device that collects emotional data using sensors, cameras, and microphones attached to a terminal.

[1536] "Tuning" refers to the process of optimizing summaries and analysis results based on the user's emotional state.

[1537] "Automatic translation" refers to technology for translating data or summaries into another language.

[1538] This invention relates to a method for efficiently analyzing contracts and other important documents and generating summaries using a system. Furthermore, it aims to combine this system with an emotion engine that recognizes the user's emotions to adjust the analysis and summaries according to the user's emotional state. Specific embodiments for implementing this invention are described below.

[1539] System configuration

[1540] Document upload

[1541] Users upload documents such as contracts to the system using devices (PCs, smartphones, tablets, etc.). The devices are equipped with cameras and microphones, allowing for the collection of emotional data through the user's facial expressions and voice.

[1542] Document Receipt and Storage

[1543] The server receives the uploaded document and temporarily stores it in storage. At this time, basic information such as the document format (PDF, DOCX, etc.) and emotion data is acquired and recorded as metadata.

[1544] Document Preprocessing

[1545] The server performs pre-processing on the received documents, which includes converting the PDF files to text format and removing noise (e.g., images and overlapping text) using specific document analysis software (e.g., OCR technology).

[1546] Language Identification

[1547] The server uses natural language processing (NLP) techniques to determine the language of the document, whether it is written in English, Japanese, French, etc. The techniques used include, for example, language identification algorithms.

[1548] Document analysis and summary generation

[1549] The server analyzes the document using NLP technology based on the determined language. It extracts important keywords and phrases (e.g., contract terms, payment terms, liability clauses, etc.) and generates a document summary based on the analysis results. The summary is a concise summary of the document's key points. Specific technologies used include keyword extraction algorithms and summary generation models.

[1550] Emotional Engine Adjustment

[1551] The server uses an emotion engine to analyze the emotional data collected from the user and incorporate it into the analysis results. For example, if the user is feeling stressed or confused, the server will adjust the summary to generate a more concise and clear summary. Facial recognition software and voice analysis technology are used to analyze the emotional data.

[1552] Multilingual support and results transmission

[1553] If the document language is different from the language specified by the user, the server uses an automatic translation function to translate the document or summary into the specified language. This function allows the user to view the summary in the language they require. This is achieved using a translation API. The generated summary and analysis results are then sent to the user's device, where the user can view the results.

[1554] Specific examples

[1555] Example 1: Processing an English contract

[1556] When a user uploads a PDF English contract titled "Service Agreement" to the system using their PC, a camera and microphone monitor the user's facial expressions and voice to collect emotional data. The server receives the document, converts it to text format, and removes noise. After determining that the document is written in English, it uses NLP technology to extract key information. A summary is generated based on this information and adjusted according to the user's emotional state. The server then sends the adjusted summary to the user's PC. The user then reviews the summary on their PC and determines whether the contract terms are appropriate.

[1557] Example 2: Processing multilingual contracts

[1558] When a user uploads a contract written in French to the system using their smartphone, the smartphone's camera and microphone collect the user's emotions. The server receives the French document and performs the necessary preprocessing. It determines the document's language as French and uses NLP technology to extract key information. If the user specifies English, the French summary is translated into English using an automatic translation function. The translated summary is then adjusted by the emotion engine and sent to the user's smartphone. The user then reviews the summary and reevaluates the contract content if necessary.

[1559] Prompt Sentence Examples

[1560] "Please upload the English contract and summarize the key points. Also, please adjust it to take into account sentiment data."

[1561] "I have uploaded a document in French. Please translate it into English and provide a summary."

[1562] In this way, the system can efficiently analyze contracts and important documents and provide summaries tailored to the user's emotional state, reducing the burden on legal departments and speeding up contract closures.

[1563] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1564] Step 1: Upload your documents

[1565] The user uses a terminal to upload a document to the system. The input is a contract file selected by the user (e.g., a PDF file), and the output is the transfer of the file to the system. The terminal is equipped with a camera and microphone, which also collects emotional data from the user's facial expressions and voice. Specifically, the user clicks the "Upload" button and selects a document file.

[1566] Step 2: Receiving and storing documents

[1567] The server receives the uploaded document and temporarily stores it in storage. The input is the uploaded contract file and emotion data, and the output is the file stored on the server and the collected emotion data meta information. Specifically, the server receives the file and stores it in storage with a file name such as "contract.pdf." At this time, basic information such as the document format and emotion data is also saved.

[1568] Step 3: Preprocessing the Documents

[1569] The server performs preprocessing on the received document. The input is the saved contract file, and the output is text data suitable for analysis. Specifically, it converts the PDF file to text and removes noise such as images and overlapping text. Specifically, it uses OCR technology to extract text from the PDF and runs a script to remove unnecessary noise.

[1570] Step 4: Language Determination

[1571] The server uses natural language processing (NLP) techniques to determine the language of a document. The input is preprocessed text data, and the output is the document's language information (e.g., English, Japanese, French, etc.). Specifically, it analyzes the frequency of words and phrases in the document and runs a model to identify the language based on that.

[1572] Step 5: Parsing the Document

[1573] The server then uses NLP technology to analyze the document based on the language it has identified. The input is linguistic information and preprocessed text data, and the output is key keywords and phrases extracted through the analysis. Specifically, it runs an algorithm to extract keywords such as contract terms, payment terms, and liability clauses.

[1574] Step 6: Generate a summary

[1575] The server generates a document summary based on the analysis results. The input is the extracted keywords and phrases, and the output is the document summary. Specifically, it runs a script that generates a document summary based on the extracted keywords and an existing format.

[1576] Step 7: Emotional Engine Alignment

[1577] The server uses an emotion engine to adjust the summary based on the user's emotion data. The input is the generated summary and emotion data, and the output is a summary adjusted to reflect the emotion. Specifically, if the user's facial expression data indicates "stress," the server runs an algorithm to reconstruct the summary into simpler language.

[1578] Step 8: Multilingual

[1579] If the document language is different from the language specified by the user, the server uses an automatic translation function to translate the summary into the specified language. The input is the adjusted summary, and the output is the translated summary. Specifically, the server calls an API to translate the summary generated in French into English.

[1580] Step 9: Sending the results

[1581] The server sends the generated summary and analysis results to the user's device. The input is the adjusted and translated summary, and the output is the summary sent to the user's device. Specifically, the generated summary and analysis results are sent to the user's device in a data format such as JSON, and the device displays them.

[1582] (Application example 2)

[1583] 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."

[1584] Conventional document analysis systems can analyze documents and generate summaries, but they do not take into account the user's emotional state, resulting in the inability to provide optimal information to users. Furthermore, they are unable to provide appropriate support when users feel stressed or confused, and are particularly inadequate in situations requiring quick and accurate judgments, particularly in the security service field. There is a need for a system that can improve this situation by adjusting document analysis results and summaries based on the user's emotional state.

[1585] 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 a user to upload a document to the system; means for the server to receive and store the uploaded document; means for the server to preprocess the document and convert it into a format suitable for analysis; means for the server to determine the language of the document; means for the server to analyze each section of the document using natural language processing technology and extract important keywords and phrases; means for the server to generate a summary of the document based on the analysis results; means for the server to recognize the user's emotional state and adjust the summary based on that state; means for the server to send the summary generated by the server to the user's terminal; and means for the user to check the analysis results and the summary through the terminal. This makes it possible to provide a summary adjusted according to the user's emotional state.

[1586] "User" refers to the person who uploads documents to the analysis system and reviews the analysis results and summaries.

[1587] "Document" refers to the documents or information to be analyzed and summarized, and may exist in formats such as PDF or DOCX.

[1588] "System" refers to the entire device and software configuration that performs a series of processes in which a user uploads a document and the server analyzes and summarizes it.

[1589] "Server" refers to a computer and associated software for receiving, storing, and processing documents uploaded by users.

[1590] "Preprocessing" refers to a series of operations to convert a document into a format suitable for analysis, such as converting PDF files to text or removing noise.

[1591] "Natural language processing technology" is a technology in the field of computer science that refers to methods and algorithms that allow machines to understand, analyze, and generate human language.

[1592] "Emotional state" refers to the psychological or emotional state recognized from the user's facial expression or voice, and includes, for example, stress, happiness, surprise, etc.

[1593] A "summary" is a concise summary of the important points from an analyzed document that can be quickly understood by a user.

[1594] "Upload" refers to the operation of a user sending a document from their own terminal to the system.

[1595] "Terminal" refers to a device used by a user, such as a smartphone or PC, that interfaces with the system.

[1596] "Adjustment" refers to making changes or modifications to the content of an already generated summary based on the user's emotional state.

[1597] "Automatic translation function" refers to the function by which the server mechanically translates documents and summaries into a language specified by the user.

[1598] "Important sections" are those parts of the document that require special attention, such as specific matters like contract terms, payment terms, and liability clauses.

[1599] This invention is a system for efficiently analyzing documents and generating summaries in the field of security services, and can further adjust the summaries according to the emotional state of the user. The system is implemented through the following steps:

[1600] composition

[1601] 1. Hardware configuration:

[1602] The user's device is a smartphone or a PC, which is equipped with a camera and a microphone.

[1603] The server is equipped with a high-performance CPU and sufficient memory.

[1604] Data storage requires large amounts of storage for analysis.

[1605] 2. Software configuration:

[1606] For natural language processing technology, Google Cloud Natural Language API is used.

[1607] Amazon Rekognition is used for sentiment analysis.

[1608] The automatic translation function uses the Google Translate API.

[1609] The server software used is Apache HTTP Server and Node.js.

[1610] MySQL is used as the database.

[1611] Explanation of program processing

[1612] The server receives documents uploaded by users, analyzes them, and summarizes them. This process involves the following major steps:

[1613] 1. Upload and save documents:

[1614] Users upload documents to the system via their smartphones or PCs. The server receives and stores them, and the documents are used for analysis and summary generation.

[1615] 2. Document preprocessing:

[1616] The server pre-processes the uploaded documents, which includes converting PDF files to text format and removing unwanted noise.

[1617] 3. Document Language Determination:

[1618] The server uses the Google Cloud Natural Language API to determine the language of the document, which tells it what language it is written in.

[1619] 4. Document Analysis:

[1620] The server uses natural language processing techniques to analyze the content of the document and extract important keywords and phrases.

[1621] 5. Generate a summary:

[1622] The server generates a summary of the document based on the extracted keywords and phrases, which is concise and easy for users to understand.

[1623] 6. Emotional engine regulation:

[1624] The server uses Amazon Rekognition to analyze the user's emotional data and adjusts the generated summary according to the user's emotional state—for example, if the user is feeling stressed, the summary will be more concise and clear.

[1625] 7. Submit and review your abstract:

[1626] The server then sends the adjusted summary to the user's device, where the user can check the summary on their smartphone or computer and take any necessary action.

[1627] Specific examples

[1628] For example, if a lawyer uploads a security services contract via smartphone, the system analyzes the document and summarizes key sections. Furthermore, if the lawyer is under stress, the summary can be adjusted to be concise and clear. This process ensures fast and accurate information delivery.

[1629] Prompt Sentence Examples

[1630] Analyze and summarize the contracts uploaded to the system. Adjust the summary and reply based on the user's sentiment.

[1631] In this way, the present invention can efficiently analyze the content of documents and provide summaries according to the emotional state of the user, thereby improving reliability and efficiency in the field of security services.

[1632] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1633] Step 1:

[1634] The user uploads a document to the system. The user uses a smartphone or PC to upload the document to be analyzed to the system. The input is a document file (PDF, DOCX, etc.), and the output is the data uploaded to the server.

[1635] Step 2:

[1636] The server receives and stores the uploaded document. The server receives the document file sent by the user and stores it in a database. The input is the document file sent by the user, and the output is the document data stored in the database.

[1637] Step 3:

[1638] The server preprocesses the document and converts it into a format suitable for analysis. The server converts the document into a text format and removes unnecessary noise (images, duplicated text, etc.). The input is the document data stored in the database, and the output is the preprocessed text data.

[1639] Step 4:

[1640] The server determines the language of the document. The server determines the language of the document using the Google Cloud Natural Language API. The input is preprocessed text data, and the output is the language information of the document.

[1641] Step 5:

[1642] The server uses natural language processing technology to analyze each section of the document and extract important keywords and phrases. The server uses natural language processing technology (Google Cloud Natural Language API) to analyze the document and extract important keywords and phrases. The input is the text data of the document, and the output is the extracted keywords and phrases.

[1643] Step 6:

[1644] The server generates a document summary based on the analysis results. The server generates a document summary based on the extracted keywords and phrases. The input is the extracted keywords and phrases, and the output is the generated summary.

[1645] Step 7:

[1646] The server recognizes the user's emotional state and adjusts the summary based on that state. The server uses Amazon Rekognition to analyze the user's emotional state and adjust the generated summary. The input is the user's emotional data (facial expressions and voice) and the generated summary, and the output is the adjusted summary.

[1647] Step 8:

[1648] The server sends the generated summary to the user's terminal. The server sends the adjusted summary to the user's terminal. The input is the adjusted summary, and the output is the summary sent to the user's terminal.

[1649] Step 9:

[1650] The user checks the analysis results and summary via their device. The user checks the analysis results and summary sent from the server via their smartphone or PC. The input is the summary sent from the server, and the output is the user's understanding and any necessary actions.

[1651] 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.

[1652] 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.

[1653] 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.

[1654] 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.

[1655] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1656] 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.

[1657] 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).

[1658] 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.

[1659] 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."

[1660] 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.

[1661] 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).

[1662] 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.

[1663] 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.

[1664] 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.

[1665] 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.

[1666] 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.

[1667] 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.

[1668] 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.

[1669] 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.

[1670] 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.

[1671] 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.

[1672] The following is further disclosed regarding the above embodiment.

[1673] (Claim 1)

[1674] a means for users to upload documents to the system;

[1675] a means for the server to receive and store the uploaded documents;

[1676] A means for the server to pre-process the document and convert it into a format suitable for analysis;

[1677] a means for the server to determine the language of the document;

[1678] The server uses natural language processing techniques to analyze each section of the document and extract key keywords and phrases;

[1679] A means for the server to generate a summary of the document based on the analysis result;

[1680] means for transmitting the generated digest to a user's terminal;

[1681] A means for users to check the analysis results and summaries through their terminals;

[1682] A system including:

[1683] (Claim 2)

[1684] 2. The system of claim 1, further comprising means for the server to translate the document or abstract into a specified language using an automatic translation function if the language of the document is not a language specified by the user.

[1685] (Claim 3)

[1686] 10. The system of claim 1, wherein the server has means for automatically detecting certain important sections (e.g., contract terms, payment terms, liability clauses, etc.) based on the con...

Claims

1. a means for users to upload documents to the system; a means for the server to receive and store the uploaded documents; A means for the server to pre-process the document and convert it into a format suitable for analysis; a means for the server to determine the language of the document; The server uses natural language processing techniques to analyze each section of the document and extract key keywords and phrases; A means for the server to generate a summary of the document based on the analysis result; means for transmitting the generated digest to a user's terminal; A means for users to check the analysis results and summaries through their terminals; A system including:

2. 2. The system of claim 1, further comprising means for the server to translate the document or abstract into the specified language using an automatic translation function if the language of the document is not the language specified by the user.

3. 2. The system of claim 1, wherein the server includes means for automatically detecting certain important sections (e.g., contract terms, payment terms, liability clauses, etc.) based on the content of the document and including these sections in the summary.

Images