System

Abstract

A system is provided.SOLUTION: A system comprising: means for receiving a past material of a company and storing the past material in a database; means for acquiring a material from the database, applying a machine learning algorithm using the material, and learning a company-specific taste; means for automatically generating a material reflecting the company-specific taste using the learned machine learning model based on a summary of material generation received from a user; and means for outputting the generated material to the user.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] In the document creation process for companies, creating documents that reflect a company's unique style and style requires a great deal of effort and time. Existing document generation services are based on generic templates and are unable to reflect the characteristics of individual companies. As a result, when creating company-specific documents, companies have to manually refer to past documents, making it difficult to create documents efficiently while maintaining consistency. [Means for solving the problem]

[0005] The present invention solves the above-mentioned problems by providing a system including: means for receiving a company's past documents and storing them in a database; means for acquiring documents from the database and using the documents to apply a machine learning algorithm to learn the company's unique taste; means for automatically generating documents that reflect the company's unique taste using the trained machine learning model based on a document generation outline received from a user; and means for outputting the generated documents to the user. This enables efficient generation of documents that consistently maintain the company's unique style.

[0006] "Corporate taste" refers to consistent design, formatting, terminology, style, and other characteristics in the materials produced by a company.

[0007] A "database" is a collection of information that systematically manages a company's past documents and allows for efficient search and retrieval.

[0008] A "machine learning algorithm" is a computational method that learns patterns and characteristics from past data and applies them to generating new data.

[0009] "Natural language processing technology" refers to technology that enables computers to understand, generate, and translate human language.

[0010] The "document creation summary" is information that briefly explains the content and purpose of the document that the user wants to create.

[0011] "Text extraction and cleaning" refers to the process of extracting necessary text information from documents and deleting and correcting unnecessary data.

[0012] "Means for automatically generating materials" refers to a method that utilizes tastes learned through machine learning to automatically create new materials without human intervention.

[0013] "Means for outputting to the user" refers to a method for making the generated materials available for display, transmission, or downloading to the user. [Brief explanation of the drawings]

[0014] [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

[0015] 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.

[0016] First, the terms used in the following description will be explained.

[0017] 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).

[0018] 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.

[0019] 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.

[0020] 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.

[0021] 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."

[0022] [First embodiment]

[0023] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0024] 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.

[0025] 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).

[0026] 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.

[0027] 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.

[0028] 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.

[0029] 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.

[0030] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0031] 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.

[0032] 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.

[0033] 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.

[0034] 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."

[0035] The present invention relates to a system for automatically generating materials that reflect a company's unique taste. This system comprises the following means.

[0036] First, it is equipped with a means for receiving a company's past documents and storing them in a database. When a user uploads a company's past documents using their own device, the device sends the documents to the server. Once sent, the server receives the documents and stores them in the database. This allows a company's past documents to be managed systematically and made available for later processes.

[0037] Next, the system is equipped with a means for retrieving documents stored in the database and applying a machine learning algorithm based on the documents to learn the company's unique taste. The server retrieves the documents from the database and performs preprocessing. Preprocessing includes extracting text information from the documents and deleting and correcting unnecessary data. After preprocessing is complete, the documents are analyzed using a machine learning algorithm. This allows the system to learn parameters that characterize the company's unique taste, such as format, design, and terminology selection.

[0038] Furthermore, the system is equipped with a means for automatically generating materials that reflect the company's unique taste using a trained machine learning model based on the outline of the materials received from the user. The user inputs an outline of the materials they want to generate into the terminal and sends that information to the server. The server generates materials using a trained AI model based on the received outline. These materials reflect the company's unique taste, ensuring consistency.

[0039] Finally, the system includes a means for outputting the generated materials to the user. The server transmits the generated materials to the user's terminal. The terminal displays the received materials and provides a link for the user to download them.

[0040] As a concrete example, consider the case where a company is preparing presentation materials for the launch of a new product. A user uploads previously created materials to the system, which then learns from them. After that, when the user enters the description "New product launch presentation," the server generates presentation materials that reflect the company's unique style based on that information and provides them to the user. In this way, new materials can be created efficiently while maintaining the company's consistent brand image.

[0041] The processing flow will be explained below.

[0042] Step 1:

[0043] The user selects the company's past documents on the terminal and clicks the upload button.

[0044] Step 2:

[0045] The terminal sends the selected material to the server as an HTTP request.

[0046] Step 3:

[0047] The server analyzes the received documents and stores them in a database. Metadata such as company ID and user ID for each document are also stored at the same time.

[0048] Step 4:

[0049] The server retrieves all documents related to the target company from the database, preprocesses the documents, extracts text, and cleans them, making the contents of the documents easier to analyze.

[0050] Step 5:

[0051] The server applies machine learning algorithms to the preprocessed materials to learn the company's specific tastes (formatting, terminology, style, etc.), and saves the learned parameters as an AI model.

[0052] Step 6:

[0053] The user inputs an outline of the material they want to generate into the terminal, for example, "New Product Launch Presentation," and clicks the send button.

[0054] Step 7:

[0055] The device sends the summary entered by the user to the server as an HTTP request.

[0056] Step 8:

[0057] Based on the summary received by the server, a pre-trained AI model is used to generate materials that reflect the company's unique style.

[0058] Step 9:

[0059] The server sends the generated document to the user's device in the format specified by the user, such as PDF or PPTX.

[0060] Step 10:

[0061] The device displays the received materials to the user, and also provides a link so the user can download the materials if desired.

[0062] This will create a system that allows companies to create documents efficiently while maintaining a consistent brand image.

[0063] Example 1

[0064] 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."

[0065] When companies create documents, they need a way to efficiently generate them while maintaining a consistent brand image. However, previous methods required a lot of manual work, which took time and effort, making it difficult to create documents efficiently. In addition, utilizing past documents to reflect a company's unique style required advanced technology and specialized knowledge. This has created a need for a new system that can create documents efficiently and consistently.

[0066] 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.

[0067] In this invention, the server includes means for receiving a company's past documents and storing them in a database, means for acquiring documents from the database and using the documents to apply a machine learning algorithm to learn the company's unique taste, means for automatically generating documents that reflect the company's unique taste using the trained machine learning model based on a document generation summary received from a user, means for performing preprocessing to extract text information and deleting and correcting unnecessary data, and means for outputting the generated documents to the user. This enables companies to efficiently automatically generate consistent documents, significantly reducing the effort required while maintaining their brand image.

[0068] "Company archives" are documents and files created or received by a company in the past, including presentation slides, reports, marketing materials, etc.

[0069] A "database" is a system for efficiently storing, managing, and searching information, and is used to systematically store a company's past documents.

[0070] A "machine learning algorithm" is a type of software that finds patterns and features in large amounts of data and makes predictions and classifications based on the data.

[0071] "Corporate style" is a general term for the consistent format, design, and terminology used by a particular company in its materials.

[0072] The "outline of material generation" includes basic information and instructions regarding the purpose and content of the material the user wants to generate.

[0073] A "trained machine learning model" is a machine learning algorithm that has been trained to the point where it is capable of performing a specific task using training data.

[0074] "Preprocessing" refers to a series of operations to convert raw data into an analyzable format, including extracting text information, removing noise data, and standardizing formats.

[0075] "Automatically generated" means that the system creates the materials independently without human intervention.

[0076] "Output" refers to providing the system-generated materials in a format that can be used by the user.

[0077] The present invention relates to a system for automatically generating materials that reflect a company's unique taste. This system comprises the following means.

[0078] First, a user uploads past company documents from their own device to the server, including presentation slides, reports, marketing materials, etc. The device then transmits the uploaded documents to the server.

[0079] The server receives the data sent from the device and stores it in a database. The database uses a data management system such as MySQL or PostgreSQL. When storing data, the server also stores metadata (creation date, creator, etc.) of the data, allowing for systematic management of the data.

[0080] Next, the server retrieves the stored historical documents from the database. If necessary, it filters the documents based on specific criteria (creation date, author, project name, etc.). The server then performs preprocessing on the retrieved documents. This preprocessing involves extracting text information from the documents, removing unnecessary data and noise, and standardizing the format. For example, text can be extracted from PDFs and unnecessary elements such as advertisements and copyright information can be removed.

[0081] The server then applies machine learning algorithms to the preprocessed materials. Specifically, it uses machine learning frameworks such as TensorFlow and PyTorch to learn the company's unique taste. The analysis targets include frequently used fonts, color schemes, and distinctive phrases. This allows the server to learn parameters that characterize the company's unique taste.

[0082] Next, the user inputs a summary of the materials they want to create into their device, for example, by entering specific instructions such as "Please create materials for a presentation to launch a new product." The device then sends the input summary to the server.

[0083] The server generates materials based on the received summary using a trained AI model (e.g., GPT-3, BERT, etc.). The server automatically generates materials that reflect the company's unique style (unique fonts, color schemes, and phrases), and the content reflects the summary entered by the user. For example, in the case of a presentation material for the launch of a new product, the server automatically generates the necessary slides, text content, graphs, etc.

[0084] Finally, the server sends the generated materials to the user's terminal, where they are displayed and a download link is provided for easy access. Users can click the link to download and use the materials.

[0085] As a concrete example, consider the case where a company is preparing presentation materials for the launch of a new product. A user uploads previously created materials to the system, which then learns from them. Then, when the user enters the outline "new product launch presentation" into their terminal, the server generates presentation materials that reflect the company's unique style based on that information and provides them to the user. In this way, new materials can be created efficiently while maintaining the company's consistent brand image.

[0086] An example of a prompt might be, "I would like to create a presentation for the launch of a new product. I would like the design to follow the style of past presentation materials while incorporating the latest trends."

[0087] This system allows companies to efficiently generate materials while maintaining a consistent brand image, resulting in significant savings in time and effort and an improvement in the efficiency of the entire material creation process.

[0088] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0089] Step 1:

[0090] Uploading and saving past documents

[0091] Users upload past company documents from their own devices. These include presentation slides, reports, marketing materials, etc. The documents uploaded by the users are sent from the devices to the server. The server stores the received documents in a database. The input is the documents uploaded by the users, and the output is the documents stored in the database.

[0092] Step 2:

[0093] Material acquisition and preprocessing

[0094] The server retrieves stored past documents from the database. If necessary, it filters the documents using specific criteria (creation date and time, creator, project name, etc.). The server then performs preprocessing on the retrieved documents. Preprocessing includes extracting text information from the documents, removing unnecessary data and noise, and standardizing the format. For example, extracting text from PDFs and removing unnecessary elements such as advertisements and copyright information. The input is the document retrieved from the database, and the output is the document after preprocessing.

[0095] Step 3:

[0096] Learning tastes with machine learning algorithms

[0097] The server applies machine learning algorithms to the preprocessed documents. Specifically, it uses machine learning frameworks such as TensorFlow and PyTorch to learn the company's unique taste. The analysis targets include frequently used fonts, color schemes, and distinctive phrases. The input is the preprocessed documents, and the output is parameters that characterize the company's unique taste.

[0098] Step 4:

[0099] Input and submit outline of document generation

[0100] The user inputs an outline of the material they want to generate into their terminal. For example, they input specific instructions such as "Please create a presentation for the launch of a new product." The terminal then sends the input outline to the server. The input is the outline of the material that the user inputs into the terminal, and the output is the outline sent to the server.

[0101] Step 5:

[0102] Automatic generation of materials that reflect your taste

[0103] The server generates materials based on the received summary using a trained AI model (e.g., GPT-3, BERT, etc.). The server automatically generates materials that reflect the company's unique taste (unique fonts, color schemes, and phrases), and the content of the materials reflects the summary entered by the user. For example, in the case of a presentation material for the launch of a new product, the necessary slides, text content, graphs, etc. are automatically generated. The input is the summary of the material generation sent to the server and the trained taste parameters, and the output is the generated material.

[0104] Step 6:

[0105] Output and download of generated materials

[0106] The server sends the generated material to the user's device. The device displays the received material and provides a download link for easy access by the user. The user can click the link to download and use the material. The input is the material generated by the server, and the output is the download link displayed on the user's device.

[0107] Each of these steps automatically generates materials that reflect the company's unique style, maintaining a consistent brand image and streamlining document creation.

[0108] (Application example 1)

[0109] 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."

[0110] The present invention aims to provide a system that enables sales staff in brick-and-mortar stores to more smoothly communicate with customers and provide information that reflects the company's unique taste in real time. In particular, the system aims to enable sales staff to respond promptly to customers' questions and needs, and provide appropriate information while consistently maintaining the company's brand image.

[0111] 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.

[0112] In this invention, the server includes means for receiving past company information and storing it in a database, means for acquiring information from the database and using the information to apply a machine learning algorithm to learn the company's unique tastes, means for automatically generating information reflecting the company's unique tastes using the trained machine learning model based on an overview of information generation received from a user, means for outputting the generated information to the user, means for extracting customer questions and needs in real time using voice recognition and providing optimal information based on the context, and means for displaying information on a smart device in real time. This enables salespeople to provide appropriate information according to customer needs in real time and interact with customers while maintaining a consistent company brand image.

[0113] "Past information about a company" refers to data such as materials, documents, and presentations created by the company in the past.

[0114] A "database" is a storage system for storing and systematically managing received information.

[0115] A "machine learning algorithm" is a mathematical model that extracts patterns and features from large amounts of data and performs specific tasks.

[0116] "A company's unique taste" is a general term that refers to the brand image that a company maintains, the unique terminology, format, design, etc. that are used.

[0117] "Outline of information generation" refers to an outline of the requirements, purpose, content, etc. of the materials or documents that the user wants to create.

[0118] A "trained machine learning model" is a machine learning algorithm that has been pre-trained using large amounts of data, and is a model that can efficiently perform a specific task.

[0119] "Automatic generation" refers to the process by which a system automatically creates materials or documents without human intervention.

[0120] "Speech recognition" is a technology that converts human speech into digital data and understands its content.

[0121] "Context" refers to the meaning and circumstances behind a customer's questions, needs, and interactions.

[0122] "Smart devices" is a general term for electronic devices that can be worn by users to display and operate information, such as smartphones, smart glasses, and head-mounted displays.

[0123] This invention is a system that provides an "intelligent sales support app" that enables salespeople in brick-and-mortar stores to more smoothly communicate with customers. The system uses smart devices, servers, natural language processing technology, speech recognition technology, and generative AI models.

[0124] First, the server receives the company's past information and stores it in a database. When a user uploads the company's past information using their own device, the device sends the information to the server. Once sent, the server receives the information and stores it in a database. This allows the company's past information to be managed systematically and made available for later processes.

[0125] The server then retrieves the information stored in the database and applies a machine learning algorithm based on that information to learn the company's unique tastes. This process uses natural language processing technology as the machine learning algorithm. The server extracts and cleans text, selecting and formatting the necessary data. Based on the formatted data, the generative AI model learns the company's unique tastes.

[0126] When a user inputs an outline of the document to be generated into the device, the information is sent to the server. Based on the received outline, the server uses a trained generative AI model to automatically generate information that reflects the company's unique taste. This information is displayed on the device in real time. In particular, by utilizing voice recognition technology, questions and needs can be extracted in real time from conversations with customers, and the most appropriate information is displayed based on that context.

[0127] For example, when a salesperson uses smart glasses to interact with a customer, if the customer asks, "What are the features of the new product?", the voice information is sent to the server. The server uses voice recognition technology to convert the customer's question into text, and a generative AI model generates appropriate information. This information is displayed in real time on the smart glasses, allowing the salesperson to provide a prompt explanation to the customer.

[0128] An example prompt is:

[0129] "Create a new product launch presentation based on your past materials: what specific points should you highlight to answer customer questions about the new product's features?"

[0130] This system allows sales staff to provide appropriate information in real time to meet customer needs, enabling them to serve customers while maintaining a consistent company brand image.

[0131] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0132] Step 1:

[0133] The user operates the device to upload the company's past information. The uploaded information is sent from the device to the server. The input is the company's past information, and the output is the information stored on the server. Specifically, the user selects a file using the device interface and clicks the upload button.

[0134] Step 2:

[0135] The server stores the received information in a database. The input is the uploaded information, and the output is the information stored in the database. Specifically, the server analyzes the contents of the received file and writes it to the corresponding database.

[0136] Step 3:

[0137] The server retrieves information from the database and performs text extraction and cleaning. The input is the information stored in the database, and the output is the extracted and cleaned text data. Specifically, the server removes unnecessary parts from the information and extracts the necessary text data.

[0138] Step 4:

[0139] The server applies a machine learning algorithm to the extracted and cleaned text data to learn the company's unique tastes. The input is the extracted and cleaned text data, and the output is the learned taste parameters. Specifically, the server uses natural language processing technology to analyze the text data and extract features.

[0140] Step 5:

[0141] The user inputs a summary of the document generation into the terminal and sends the information to the server. The input is the summary of the document generation, and the output is the summary data sent to the server. Specifically, the user uses the terminal interface to input the summary into the text box and clicks the send button.

[0142] Step 6:

[0143] Based on the received summary data, the server uses a trained generative AI model to automatically generate information that reflects the company's unique taste. The input is the summary data and the trained model, and the output is automatically generated information. Specifically, the server inputs prompts to the generative AI model and obtains the generated information.

[0144] Step 7:

[0145] The server sends the generated information to the user's terminal, which then displays the information. The input is the automatically generated information, and the output is the information displayed on the terminal. Specifically, the server sends the generated information to the terminal via an HTTP request, and the terminal displays the received information on the screen.

[0146] Step 8:

[0147] The device uses voice recognition to extract customer questions and needs in real time and transmit them to the server based on the context. The input is the customer's voice and the output is text-based question data. Specifically, the device's microphone captures the voice and the voice recognition software converts it into text.

[0148] Step 9:

[0149] The server analyzes the text generated by speech recognition and generates the most appropriate information based on the context. The input is the question data in text format, and the output is the most appropriate information. Specifically, the server uses the generative AI model again to generate an appropriate answer to the question.

[0150] Step 10:

[0151] The server displays the generated information on the smart device in real time. The input is the optimal information, and the output is the information displayed on the smart device. Specifically, the server sends the information to a device such as smart glasses, and the device displays the information to the user.

[0152] 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.

[0153] The present invention relates to a system for automatically generating documents that reflect a company's unique taste, as well as a system for recognizing a user's emotions and reflecting the results in the document generation. This system comprises the following means.

[0154] First, it is equipped with a means for receiving past company documents and storing them in a database. When a user uploads past company documents using their own device, the device sends the documents to the server. Once sent, the server receives the documents and stores them in the database. This step is essential for learning the unique tastes of each company.

[0155] Next, the system retrieves the documents stored in the database and applies a machine learning algorithm to learn the company's unique tastes based on those documents. The server retrieves the documents from the database and performs preprocessing. This preprocessing includes extracting text information from the documents and deleting and correcting unnecessary data. After preprocessing is complete, the machine learning algorithm is used to analyze the documents and learn parameters that characterize the company's unique tastes, such as format, design, and terminology selection.

[0156] Furthermore, the system is equipped with a means for automatically generating materials that reflect the company's unique taste, using a trained machine learning model based on the outline of the materials received from the user. The user inputs an outline of the materials they want to generate into the terminal and sends that information to the server. The server generates the materials using a trained AI model based on the received outline. The generated materials reflect the company's unique taste, maintaining consistency.

[0157] In addition, the system is equipped with an emotion engine that recognizes the user's emotions. The emotion engine analyzes the emotions from the user's input. For example, when a user inputs an outline for document generation, the input is analyzed for emotions using natural language processing technology. The results of this emotion analysis can be reflected in the document generation process, allowing the tone and style of the generated document to be adjusted.

[0158] Specifically, when a user inputs an outline of a "new product launch presentation," for example, if the user enters words that contain strong emotions, such as "this new product will take the market by storm," the emotion engine analyzes the words and recognizes them as emotions indicating excitement or anticipation. The server then reflects the emotion recognition results in the creation of the materials, using bolder, more impactful designs and words to generate materials that reflect the user's emotions.

[0159] Finally, a means for outputting the generated materials to the user is provided. The server transmits the generated materials to the user's terminal, and the terminal displays the received materials to the user. The user can download the materials as needed.

[0160] This invention enables companies to efficiently generate materials that respond to users' emotions and intentions while maintaining a consistent brand image, which significantly improves the efficiency of document creation work and allows companies to provide materials that strongly reflect the company's unique character.

[0161] The processing flow will be explained below.

[0162] The present invention is a system that automatically generates materials that reflect a company's unique taste, and also recognizes the user's emotions and reflects the results. This system operates through the following steps.

[0163] Step 1:

[0164] The user selects the company's past documents on the terminal and clicks the upload button.

[0165] Step 2:

[0166] The terminal sends the selected material to the server as an HTTP request.

[0167] Step 3:

[0168] The server analyzes the received materials and stores them in a database, including metadata such as company ID and user ID.

[0169] Step 4:

[0170] The server retrieves all documents related to the company from the database, preprocesses the documents, extracts text, and cleans unnecessary data.

[0171] Step 5:

[0172] The server applies machine learning algorithms to the preprocessed materials to learn the company's specific tastes (formatting, terminology, style, etc.), and saves the learning results as an AI model.

[0173] Step 6:

[0174] The user inputs an outline of the material they want to generate into the terminal, for example, "New Product Launch Presentation," and clicks the send button.

[0175] Step 7:

[0176] The device sends the summary entered by the user to the server as an HTTP request.

[0177] Step 8:

[0178] The server passes the received summary to the emotion engine, which analyzes the emotion from the content entered by the user. For example, natural language processing technology is used to recognize positive emotion from the input "This new product is revolutionary!"

[0179] Step 9:

[0180] The server uses a trained AI model to generate materials based on the analysis results of the emotion engine. The generated materials include elements that reflect the user's emotions in addition to the company's unique style.

[0181] Step 10:

[0182] The server sends the generated document to the user's device in the format specified by the user, such as PDF or PPTX.

[0183] Step 11:

[0184] The device displays the received materials to the user, who is also provided with a link to download the materials if desired.

[0185] As a concrete example, if a company is preparing presentation materials for the launch of a new product, the user first uploads past materials, which the system then learns from. Next, the user enters a summary such as "New product launch presentation." If the user enters words containing strong emotions, such as "taking the market by storm," the emotion engine analyzes this and recognizes it as an emotion indicating excitement or anticipation. The server then reflects this emotion recognition result in the creation of the materials, generating presentation materials using bolder, more impactful designs and words based on the company's unique taste. As a result, materials that respond to the user's emotions are efficiently created while maintaining the company's consistent brand image.

[0186] Example 2

[0187] 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."

[0188] Companies are required to efficiently create materials while maintaining a consistent brand image. However, manually created materials often fail to reflect the company's unique style and require a great deal of time and effort. Furthermore, it is difficult to reflect the user's emotions and intentions in the materials, which can result in ineffective materials. To solve these issues, there is a need for a system that automatically generates materials that reflect the company's unique style and take the user's emotions into consideration.

[0189] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0190] In this invention, the server includes means for receiving past company documents and storing them in a database, means for acquiring documents from the database and applying a machine learning algorithm to learn the company's unique taste, means for automatically generating documents that reflect the company's unique taste using a trained machine learning model based on a document generation summary received from a user, an emotion engine that recognizes the user's emotions and reflects them in the document generation process, and means for outputting the generated documents to the user. This makes it possible to efficiently generate optimal documents that match the user's emotions and intentions while maintaining the company's unique taste.

[0191] "Corporate historical documents" refer to information assets such as documents, presentations, reports, and data sets that a company has created or held in the past.

[0192] A "database" is a collection of structured data for systematically organizing and managing information, and is a system that allows for various searches and operations to be performed efficiently.

[0193] A "machine learning algorithm" is a set of computational techniques that allow a computer to find patterns and rules in data and make predictions or classifications.

[0194] "Corporate taste" refers to the unique design, format, terminology, style, etc. that a company consistently uses, and is an element that forms the company's brand image.

[0195] A "trained machine learning model" is the output of a machine learning algorithm that has been trained using historical data and is ready to automate a specific task.

[0196] An "emotion engine" is a system that analyzes emotions from user input and text data and adjusts the tone and style of content based on the results.

[0197] "User emotion" refers to the psychological state, such as joy, expectation, excitement, sadness, etc., that the user shows when inputting the outline of the material generation.

[0198] "Natural language processing technology" is a technology that enables computers to understand, interpret, and generate human language, and is used for text analysis and sentiment analysis.

[0199] The "document generation overview" is a specific explanation and requirements such as the purpose, target, and necessary elements of the document the user wants to generate.

[0200] A "document generation process" refers to a series of steps or operations that automatically create documents based on input data.

[0201] This invention relates to a system that automatically generates documents that reflect a company's unique taste, as well as a system that recognizes the user's emotions and reflects the results in the document generation. This system is realized by the cooperation of the server, terminals, and users.

[0202] First, it is equipped with a means to receive a company's past documents and store them in a database. When a user uploads a company's past documents using their own device, the device sends the documents to a server. Once sent, the server receives the documents and stores them in a database such as MySQL. This step allows data to be collected to learn the unique tastes of each company.

[0203] The system then retrieves documents stored in the database and applies machine learning algorithms to learn the company's unique tastes. The server retrieves the documents from the database and performs preprocessing. This preprocessing includes extracting text information using OCR technology and deleting unnecessary data. After preprocessing is complete, the documents are analyzed using machine learning libraries such as Scikit-learn and TensorFlow to learn parameters that characterize the company's unique tastes, such as formatting, design, and terminology selection.

[0204] Furthermore, it is equipped with a means to automatically generate materials that reflect the company's unique taste using a trained machine learning model based on the outline of the materials received from the user. The user inputs the outline of the materials they want to generate into the terminal and sends that information to the server. The server generates the materials based on the received outline using a trained AI model such as BERT or GPT-3. The generated materials reflect the company's unique taste and maintain consistency. Below are some example prompts:

[0205] Example prompt sentence:

[0206] "Please create a presentation for the launch of a new product. This product is expected to take the market by storm. The presentation should be visually impactful and the design should be modern and dynamic."

[0207] In addition, the system is equipped with an emotion engine that recognizes user emotions. The emotion engine is responsible for analyzing emotions from user input. For example, when a user enters an outline for document generation, the input is analyzed for emotion using natural language processing technology (e.g., AWS Comprehend or Google Cloud Natural Language). The results of this emotion analysis can be reflected in the document generation process to adjust the tone and style of the generated document. For example, if a user enters words that contain strong emotions, such as "This new product will take the market by storm," the emotion engine will analyze it and recognize it as an emotion indicating excitement or anticipation. The server then reflects the emotion recognition results in the document generation, using bolder, more impactful designs and words.

[0208] Finally, a means for outputting the generated materials to the user is provided. The server transmits the generated materials to the user's terminal, and the terminal displays the received materials to the user. The user can download the materials as needed.

[0209] This invention enables companies to efficiently generate materials that respond to users' emotions and intentions while maintaining a consistent brand image, which significantly improves the efficiency of document creation work and allows companies to provide materials that strongly reflect the company's unique character.

[0210] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0211] Step 1:

[0212] Input: The user uploads the company's past documents.

[0213] How it works: A user uses their device to upload past company documents (e.g., PDF or Word files) through the interface. The device retrieves these documents and selects the documents to be uploaded through the file selection function.

[0214] Output: The terminal sends the selected material to the server as an HTTP POST request.

[0215] Step 2:

[0216] Input: The data file sent from the terminal.

[0217] Operation: The server receives the file and saves it in a database such as MySQL. The received file is stored in the specified location, and metadata (e.g., upload date and time, user ID) is also recorded.

[0218] Output: Material files and metadata stored in a database.

[0219] Step 3:

[0220] Input: Materials stored in the database.

[0221] How it works: The server periodically or on demand retrieves historical data from the database and performs pre-processing, which involves extracting text from images using OCR technology and removing or correcting unnecessary data (e.g., white space, advertisements).

[0222] Output: The preprocessed material with extracted text information.

[0223] Step 4:

[0224] Input: Preprocessed material.

[0225] How it works: The server uses machine learning algorithms (such as Scikit-learn or TensorFlow) to analyze the pre-processed materials, using them as a training dataset to learn parameters that characterize the company's specific formatting, design, and terminology choices.

[0226] Output: A machine learning model that reflects your company's unique taste.

[0227] Step 5:

[0228] Input: A summary of the generated material entered by the user into the terminal.

[0229] Operation: The user inputs a detailed outline of the material creation into the terminal and sends the information to the server.

[0230] Output: Summary information of the document generation is sent to the server.

[0231] Step 6:

[0232] Input: Summary information for material generation.

[0233] How it works: The server uses trained machine learning models (such as BERT or GPT-3) to automatically create documents based on the summary information. The server then checks whether the generated documents reflect the company's unique taste and makes any necessary adjustments.

[0234] Output: The generated material.

[0235] Step 7:

[0236] Input: A summary of what the user has entered and what materials are generated.

[0237] How it works: The server uses an emotion engine to parse the emotion from the user's input. It uses natural language processing technology (such as AWS Comprehend or Google Cloud Natural Language) to identify the emotion in the input and adjust the tone and style of the material.

[0238] Output: Sentiment analysis results and generated materials adjusted to reflect them.

[0239] Step 8:

[0240] Input: Final adjusted generated material.

[0241] How it works: The server sends the final generated material to the user's device. The material is provided via email or a dedicated download link.

[0242] Output: The generated material sent to the user's device.

[0243] Step 9:

[0244] Input: Generated material sent from the server.

[0245] Operation: The device displays the data received from the server, and the user can review it. If necessary, the data can be downloaded and saved.

[0246] Output: The final generated material displayed on the user's terminal.

[0247] (Application example 2)

[0248] 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."

[0249] When companies create materials such as advertisements, it is extremely difficult to tailor the materials to reflect user emotions and intentions while maintaining a consistent brand image. Furthermore, creating advertisements manually takes time and effort, making efficient operation difficult. Furthermore, there is no method for quickly generating advertisements that reflect user emotions. This leaves companies unable to create effective advertisements that match the emotions of target users, limiting the effectiveness of their advertisements.

[0250] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0251] In this invention, the server includes: means for receiving a company's past materials and storing them in a database; means for acquiring materials from the database and using the materials to apply a machine learning algorithm to learn the company's unique tastes; means for automatically generating materials that reflect the company's unique tastes and the user's emotions using the trained machine learning model based on a material generation summary received from the user; means for analyzing the user's emotions, including an emotion analysis engine for analyzing the user's emotions, and means for outputting the generated materials to the user. This enables a company to quickly and efficiently generate effective advertising materials that reflect the user's emotions while maintaining a consistent brand image.

[0252] "Corporate historical documents" refers to data such as documents, reports, presentation materials, advertising materials, and other documents of any format that a company has created, published, or stored in the past.

[0253] A "database" is a system that organizes and stores collected data, enabling efficient searches and analysis.

[0254] A "machine learning algorithm" is a technology that automatically learns patterns and rules from data and makes predictions and analyses based on that learned knowledge.

[0255] "A company's unique taste" refers to the brand image, style, and design concept that a company aims for, and is a general term for elements that have characteristics unique to that company.

[0256] "User emotion" refers to the emotional state expressed by the user during the process of creating materials, and includes various emotions such as joy, excitement, anticipation, and anxiety.

[0257] An "emotion analysis engine" is a system that analyzes the emotions from user input and identifies specific emotional states.

[0258] The "document generation overview" is a description showing basic information, purpose, theme, and specifications of the document that the user wants to create.

[0259] "Natural language processing technology" is a technology that allows computers to understand and generate human language, and involves analyzing and generating text.

[0260] A "machine learning model" is an AI model that is trained based on data to understand certain patterns and trends and generate responses.

[0261] "Means for outputting to the user" refers to a method or system for providing the generated materials to the user, and includes functions such as display, download, and sharing.

[0262] The present invention is a system that automatically generates advertising materials by analyzing user emotions and reflecting the unique tastes of a company. Specific embodiments for carrying out the present invention will be described below.

[0263] System configuration

[0264] The system mainly consists of a server, the user's smartphone, a database, a machine learning algorithm, a natural language processing engine, and a sentiment analysis engine.

[0265] Hardware

[0266] Server: Receives, stores, analyzes data, generates advertising materials, and outputs them.

[0267] Smartphone: The user inputs the outline of the material generation and receives the generated advertising material.

[0268] software

[0269] Machine learning libraries: TensorFlow and PyTorch are used to learn company-specific tastes.

[0270] Natural language processing engine: spaCy, which uses BERT to analyze user-input text.

[0271] Database: MySQL and MongoDB are used to store corporate historical data.

[0272] Sentiment analysis engine: Analyzes user emotions in real time.

[0273] Process Overview

[0274] 1. Data Collection:

[0275] The smartphone uploads the company's past data (such as advertising data) to the server, which receives the data and stores it in a database.

[0276] 2. Training the machine learning model:

[0277] We retrieve past advertising data stored in a database and preprocess the text information. After text extraction, cleaning, and removing or correcting unnecessary data, we use TensorFlow and PyTorch to train a machine learning model that learns the unique tastes of each company.

[0278] 3. Emotion analysis:

[0279] When a user inputs a summary of the ad they want to generate, the input is analyzed in real time using a natural language processing engine (spaCy, BERT), and the emotional state is identified using a sentiment analysis engine.

[0280] 4. Automatic generation of advertising materials:

[0281] Based on the outline of the materials generated by the user and the results of sentiment analysis, advertising materials are automatically generated using a trained machine learning model. The generated advertisements reflect the company's unique taste and the user's emotions.

[0282] 5. Output to the user:

[0283] The generated advertising materials are sent from the server to the user's smartphone, where the user can preview, edit, and download the generated advertisements.

[0284] Specific examples

[0285] For example, if a user inputs the summary "I want to create a campaign ad for a new eco-product. I want to convey my passion for protecting the earth," the emotion "passion for protecting the earth" will be recognized as a positive and strong emotion from this summary. Below is an example of a prompt to input to the generative AI model:

[0286] Prompt: "We want to create an ad campaign for a new eco-friendly product. We want to convey our passion for protecting the planet. The ad should reflect our company's unique eco-friendly image and have a passionate, positive tone."

[0287] Based on this prompt, the generative AI model automatically creates an ad with a passionate and positive tone that reflects the company's unique taste.

[0288] conclusion

[0289] This system enables companies to efficiently create advertisements that match the emotions of target users while maintaining a consistent brand image. A major feature of this invention is that it combines highly accurate emotion analysis technology with machine learning models to generate effective advertisements that match user needs.

[0290] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0291] Step 1:

[0292] Receive past company documents and store them in a database.

[0293] Input: Upload past company documents (advertising data) from your smartphone.

[0294] Operation: The device sends the data to the server.

[0295] Data processing: The server organizes the received data and stores it in a database.

[0296] Output: Past company information stored in the database.

[0297] Step 2:

[0298] It retrieves data from a database and uses it to apply machine learning algorithms to learn the unique tastes of each company.

[0299] Input: Past advertising materials stored in the database.

[0300] How it works: The server retrieves material from the database and pre-processes it (text extraction, cleaning, removing and modifying unnecessary data).

[0301] Data calculation: Using preprocessed data, machine learning models such as TensorFlow and PyTorch are used to learn the company's specific tastes.

[0302] Output: A trained machine learning model.

[0303] Step 3:

[0304] The user inputs an outline of the document to be generated, and the emotion analysis engine analyzes the user's emotions.

[0305] Input: An overview of the generated materials entered by the user on their smartphone.

[0306] How it works: A natural language processing engine (spaCy, BERT) analyzes user-entered text. A sentiment analysis engine identifies emotional states.

[0307] Data calculation: Identify the user's emotional state based on the emotion analysis results.

[0308] Output: Emotion analysis results.

[0309] Step 4:

[0310] Advertising materials are automatically generated based on the outline of material generation received from the user and the results of sentiment analysis.

[0311] Input: Summary of material generation and sentiment analysis results.

[0312] How it works: The server uses trained machine learning models to automatically generate advertising materials based on your input.

[0313] Data calculation: Generate advertising materials that reflect the company's unique taste and user emotions.

[0314] Output: Generated advertising materials.

[0315] Step 5:

[0316] The generated advertising material is output to the user.

[0317] Input: Generated advertising materials.

[0318] Operation: The server sends the generated materials to the user's smartphone.

[0319] Output: Advertising materials displayed on the user's smartphone, which the user can preview, edit, and download.

[0320] 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.

[0321] 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.

[0322] 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.

[0323] [Second embodiment]

[0324] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0325] 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.

[0326] 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).

[0327] 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.

[0328] 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.

[0329] 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).

[0330] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0331] 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.

[0332] 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.

[0333] 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.

[0334] 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.

[0335] 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."

[0336] The present invention relates to a system for automatically generating materials that reflect a company's unique taste. This system comprises the following means.

[0337] First, it is equipped with a means for receiving a company's past documents and storing them in a database. When a user uploads a company's past documents using their own device, the device sends the documents to the server. Once sent, the server receives the documents and stores them in the database. This allows a company's past documents to be managed systematically and made available for later processes.

[0338] Next, the system is equipped with a means for retrieving documents stored in the database and applying a machine learning algorithm based on the documents to learn the company's unique taste. The server retrieves the documents from the database and performs preprocessing. Preprocessing includes extracting text information from the documents and deleting and correcting unnecessary data. After preprocessing is complete, the documents are analyzed using a machine learning algorithm. This allows the system to learn parameters that characterize the company's unique taste, such as format, design, and terminology selection.

[0339] Furthermore, the system is equipped with a means for automatically generating materials that reflect the company's unique taste using a trained machine learning model based on the outline of the materials received from the user. The user inputs an outline of the materials they want to generate into the terminal and sends that information to the server. The server generates materials using a trained AI model based on the received outline. These materials reflect the company's unique taste, ensuring consistency.

[0340] Finally, the system includes a means for outputting the generated materials to the user. The server transmits the generated materials to the user's terminal. The terminal displays the received materials and provides a link for the user to download them.

[0341] As a concrete example, consider the case where a company is preparing presentation materials for the launch of a new product. A user uploads previously created materials to the system, which then learns from them. After that, when the user enters the description "New product launch presentation," the server generates presentation materials that reflect the company's unique style based on that information and provides them to the user. In this way, new materials can be created efficiently while maintaining the company's consistent brand image.

[0342] The processing flow will be explained below.

[0343] Step 1:

[0344] The user selects the company's past documents on the terminal and clicks the upload button.

[0345] Step 2:

[0346] The terminal sends the selected material to the server as an HTTP request.

[0347] Step 3:

[0348] The server analyzes the received documents and stores them in a database. Metadata such as company ID and user ID for each document are also stored at the same time.

[0349] Step 4:

[0350] The server retrieves all documents related to the target company from the database, preprocesses the documents, extracts text, and cleans them, making the contents of the documents easier to analyze.

[0351] Step 5:

[0352] The server applies machine learning algorithms to the preprocessed materials to learn the company's specific tastes (formatting, terminology, style, etc.), and saves the learned parameters as an AI model.

[0353] Step 6:

[0354] The user inputs an outline of the material they want to generate into the terminal, for example, "New Product Launch Presentation," and clicks the send button.

[0355] Step 7:

[0356] The device sends the summary entered by the user to the server as an HTTP request.

[0357] Step 8:

[0358] Based on the summary received by the server, a pre-trained AI model is used to generate materials that reflect the company's unique style.

[0359] Step 9:

[0360] The server sends the generated document to the user's device in the format specified by the user, such as PDF or PPTX.

[0361] Step 10:

[0362] The device displays the received materials to the user, and also provides a link so the user can download the materials if desired.

[0363] This will create a system that allows companies to create documents efficiently while maintaining a consistent brand image.

[0364] Example 1

[0365] 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."

[0366] When companies create documents, they need a way to efficiently generate them while maintaining a consistent brand image. However, previous methods required a lot of manual work, which took time and effort, making it difficult to create documents efficiently. In addition, utilizing past documents to reflect a company's unique style required advanced technology and specialized knowledge. This has created a need for a new system that can create documents efficiently and consistently.

[0367] 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.

[0368] In this invention, the server includes means for receiving a company's past documents and storing them in a database, means for acquiring documents from the database and using the documents to apply a machine learning algorithm to learn the company's unique taste, means for automatically generating documents that reflect the company's unique taste using the trained machine learning model based on a document generation summary received from a user, means for performing preprocessing to extract text information and deleting and correcting unnecessary data, and means for outputting the generated documents to the user. This enables companies to efficiently automatically generate consistent documents, significantly reducing the effort required while maintaining their brand image.

[0369] "Company archives" are documents and files created or received by a company in the past, including presentation slides, reports, marketing materials, etc.

[0370] A "database" is a system for efficiently storing, managing, and searching information, and is used to systematically store a company's past documents.

[0371] A "machine learning algorithm" is a type of software that finds patterns and features in large amounts of data and makes predictions and classifications based on the data.

[0372] "Corporate style" is a general term for the consistent format, design, and terminology used by a particular company in its materials.

[0373] The "outline of material generation" includes basic information and instructions regarding the purpose and content of the material the user wants to generate.

[0374] A "trained machine learning model" is a machine learning algorithm that has been trained to the point where it is capable of performing a specific task using training data.

[0375] "Preprocessing" refers to a series of operations to convert raw data into an analyzable format, including extracting text information, removing noise data, and standardizing formats.

[0376] "Automatically generated" means that the system creates the materials independently without human intervention.

[0377] "Output" refers to providing the system-generated materials in a format that can be used by the user.

[0378] The present invention relates to a system for automatically generating materials that reflect a company's unique taste. This system comprises the following means.

[0379] First, a user uploads past company documents from their own device to the server, including presentation slides, reports, marketing materials, etc. The device then transmits the uploaded documents to the server.

[0380] The server receives the data sent from the device and stores it in a database. The database uses a data management system such as MySQL or PostgreSQL. When storing data, the server also stores metadata (creation date, creator, etc.) of the data, allowing for systematic management of the data.

[0381] Next, the server retrieves the stored historical documents from the database. If necessary, it filters the documents based on specific criteria (creation date, author, project name, etc.). The server then performs preprocessing on the retrieved documents. This preprocessing involves extracting text information from the documents, removing unnecessary data and noise, and standardizing the format. For example, text can be extracted from PDFs and unnecessary elements such as advertisements and copyright information can be removed.

[0382] The server then applies machine learning algorithms to the preprocessed materials. Specifically, it uses machine learning frameworks such as TensorFlow and PyTorch to learn the company's unique taste. The analysis targets include frequently used fonts, color schemes, and distinctive phrases. This allows the server to learn parameters that characterize the company's unique taste.

[0383] Next, the user inputs a summary of the materials they want to create into their device, for example, by entering specific instructions such as "Please create materials for a presentation to launch a new product." The device then sends the input summary to the server.

[0384] The server generates materials based on the received summary using a trained AI model (e.g., GPT-3, BERT, etc.). The server automatically generates materials that reflect the company's unique style (unique fonts, color schemes, and phrases), and the content reflects the summary entered by the user. For example, in the case of a presentation material for the launch of a new product, the server automatically generates the necessary slides, text content, graphs, etc.

[0385] Finally, the server sends the generated materials to the user's terminal, where they are displayed and a download link is provided for easy access. Users can click the link to download and use the materials.

[0386] As a concrete example, consider the case where a company is preparing presentation materials for the launch of a new product. A user uploads previously created materials to the system, which then learns from them. Then, when the user enters the outline "new product launch presentation" into their terminal, the server generates presentation materials that reflect the company's unique style based on that information and provides them to the user. In this way, new materials can be created efficiently while maintaining the company's consistent brand image.

[0387] An example of a prompt might be, "I would like to create a presentation for the launch of a new product. I would like the design to follow the style of past presentation materials while incorporating the latest trends."

[0388] This system allows companies to efficiently generate materials while maintaining a consistent brand image, resulting in significant savings in time and effort and an improvement in the efficiency of the entire material creation process.

[0389] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0390] Step 1:

[0391] Uploading and saving past documents

[0392] Users upload past company documents from their own devices. These include presentation slides, reports, marketing materials, etc. The documents uploaded by the users are sent from the devices to the server. The server stores the received documents in a database. The input is the documents uploaded by the users, and the output is the documents stored in the database.

[0393] Step 2:

[0394] Material acquisition and preprocessing

[0395] The server retrieves stored past documents from the database. If necessary, it filters the documents using specific criteria (creation date and time, creator, project name, etc.). The server then performs preprocessing on the retrieved documents. Preprocessing includes extracting text information from the documents, removing unnecessary data and noise, and standardizing the format. For example, extracting text from PDFs and removing unnecessary elements such as advertisements and copyright information. The input is the document retrieved from the database, and the output is the document after preprocessing.

[0396] Step 3:

[0397] Learning tastes with machine learning algorithms

[0398] The server applies machine learning algorithms to the preprocessed documents. Specifically, it uses machine learning frameworks such as TensorFlow and PyTorch to learn the company's unique taste. The analysis targets include frequently used fonts, color schemes, and distinctive phrases. The input is the preprocessed documents, and the output is parameters that characterize the company's unique taste.

[0399] Step 4:

[0400] Input and submit outline of document generation

[0401] The user inputs an outline of the material they want to generate into their terminal. For example, they input specific instructions such as "Please create a presentation for the launch of a new product." The terminal then sends the input outline to the server. The input is the outline of the material that the user inputs into the terminal, and the output is the outline sent to the server.

[0402] Step 5:

[0403] Automatic generation of materials that reflect your taste

[0404] The server generates materials based on the received summary using a trained AI model (e.g., GPT-3, BERT, etc.). The server automatically generates materials that reflect the company's unique taste (unique fonts, color schemes, and phrases), and the content of the materials reflects the summary entered by the user. For example, in the case of a presentation material for the launch of a new product, the necessary slides, text content, graphs, etc. are automatically generated. The input is the summary of the material generation sent to the server and the trained taste parameters, and the output is the generated material.

[0405] Step 6:

[0406] Output and download of generated materials

[0407] The server sends the generated material to the user's device. The device displays the received material and provides a download link for easy access by the user. The user can click the link to download and use the material. The input is the material generated by the server, and the output is the download link displayed on the user's device.

[0408] Each of these steps automatically generates materials that reflect the company's unique style, maintaining a consistent brand image and streamlining document creation.

[0409] (Application example 1)

[0410] 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."

[0411] The present invention aims to provide a system that enables sales staff in brick-and-mortar stores to more smoothly communicate with customers and provide information that reflects the company's unique taste in real time. In particular, the system aims to enable sales staff to respond promptly to customers' questions and needs, and provide appropriate information while consistently maintaining the company's brand image.

[0412] 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.

[0413] In this invention, the server includes means for receiving past company information and storing it in a database, means for acquiring information from the database and using the information to apply a machine learning algorithm to learn the company's unique tastes, means for automatically generating information reflecting the company's unique tastes using the trained machine learning model based on an overview of information generation received from a user, means for outputting the generated information to the user, means for extracting customer questions and needs in real time using voice recognition and providing optimal information based on the context, and means for displaying information on a smart device in real time. This enables salespeople to provide appropriate information according to customer needs in real time and interact with customers while maintaining a consistent company brand image.

[0414] "Past information about a company" refers to data such as materials, documents, and presentations created by the company in the past.

[0415] A "database" is a storage system for storing and systematically managing received information.

[0416] A "machine learning algorithm" is a mathematical model that extracts patterns and features from large amounts of data and performs specific tasks.

[0417] "A company's unique taste" is a general term that refers to the brand image that a company maintains, the unique terminology, format, design, etc. that are used.

[0418] "Outline of information generation" refers to an outline of the requirements, purpose, content, etc. of the materials or documents that the user wants to create.

[0419] A "trained machine learning model" is a machine learning algorithm that has been pre-trained using large amounts of data, and is a model that can efficiently perform a specific task.

[0420] "Automatic generation" refers to the process by which a system automatically creates materials or documents without human intervention.

[0421] "Speech recognition" is a technology that converts human speech into digital data and understands its content.

[0422] "Context" refers to the meaning and circumstances behind a customer's questions, needs, and interactions.

[0423] "Smart devices" is a general term for electronic devices that can be worn by users to display and operate information, such as smartphones, smart glasses, and head-mounted displays.

[0424] This invention is a system that provides an "intelligent sales support app" that enables salespeople in brick-and-mortar stores to more smoothly communicate with customers. The system uses smart devices, servers, natural language processing technology, speech recognition technology, and generative AI models.

[0425] First, the server receives the company's past information and stores it in a database. When a user uploads the company's past information using their own device, the device sends the information to the server. Once sent, the server receives the information and stores it in a database. This allows the company's past information to be managed systematically and made available for later processes.

[0426] The server then retrieves the information stored in the database and applies a machine learning algorithm based on that information to learn the company's unique tastes. This process uses natural language processing technology as the machine learning algorithm. The server extracts and cleans text, selecting and formatting the necessary data. Based on the formatted data, the generative AI model learns the company's unique tastes.

[0427] When a user inputs an outline of the document to be generated into the device, the information is sent to the server. Based on the received outline, the server uses a trained generative AI model to automatically generate information that reflects the company's unique taste. This information is displayed on the device in real time. In particular, by utilizing voice recognition technology, questions and needs can be extracted in real time from conversations with customers, and the most appropriate information is displayed based on that context.

[0428] For example, when a salesperson uses smart glasses to interact with a customer, if the customer asks, "What are the features of the new product?", the voice information is sent to the server. The server uses voice recognition technology to convert the customer's question into text, and a generative AI model generates appropriate information. This information is displayed in real time on the smart glasses, allowing the salesperson to provide a prompt explanation to the customer.

[0429] An example prompt is:

[0430] "Create a new product launch presentation based on your past materials: what specific points should you highlight to answer customer questions about the new product's features?"

[0431] This system allows sales staff to provide appropriate information in real time to meet customer needs, enabling them to serve customers while maintaining a consistent company brand image.

[0432] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0433] Step 1:

[0434] The user operates the device to upload the company's past information. The uploaded information is sent from the device to the server. The input is the company's past information, and the output is the information stored on the server. Specifically, the user selects a file using the device interface and clicks the upload button.

[0435] Step 2:

[0436] The server stores the received information in a database. The input is the uploaded information, and the output is the information stored in the database. Specifically, the server analyzes the contents of the received file and writes it to the corresponding database.

[0437] Step 3:

[0438] The server retrieves information from the database and performs text extraction and cleaning. The input is the information stored in the database, and the output is the extracted and cleaned text data. Specifically, the server removes unnecessary parts from the information and extracts the necessary text data.

[0439] Step 4:

[0440] The server applies a machine learning algorithm to the extracted and cleaned text data to learn the company's unique tastes. The input is the extracted and cleaned text data, and the output is the learned taste parameters. Specifically, the server uses natural language processing technology to analyze the text data and extract features.

[0441] Step 5:

[0442] The user inputs a summary of the document generation into the terminal and sends the information to the server. The input is the summary of the document generation, and the output is the summary data sent to the server. Specifically, the user uses the terminal interface to input the summary into the text box and clicks the send button.

[0443] Step 6:

[0444] Based on the received summary data, the server uses a trained generative AI model to automatically generate information that reflects the company's unique taste. The input is the summary data and the trained model, and the output is automatically generated information. Specifically, the server inputs prompts to the generative AI model and obtains the generated information.

[0445] Step 7:

[0446] The server sends the generated information to the user's terminal, which then displays the information. The input is the automatically generated information, and the output is the information displayed on the terminal. Specifically, the server sends the generated information to the terminal via an HTTP request, and the terminal displays the received information on the screen.

[0447] Step 8:

[0448] The device uses voice recognition to extract customer questions and needs in real time and transmit them to the server based on the context. The input is the customer's voice and the output is text-based question data. Specifically, the device's microphone captures the voice and the voice recognition software converts it into text.

[0449] Step 9:

[0450] The server analyzes the text generated by speech recognition and generates the most appropriate information based on the context. The input is the question data in text format, and the output is the most appropriate information. Specifically, the server uses the generative AI model again to generate an appropriate answer to the question.

[0451] Step 10:

[0452] The server displays the generated information on the smart device in real time. The input is the optimal information, and the output is the information displayed on the smart device. Specifically, the server sends the information to a device such as smart glasses, and the device displays the information to the user.

[0453] 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.

[0454] The present invention relates to a system for automatically generating documents that reflect a company's unique taste, as well as a system for recognizing a user's emotions and reflecting the results in the document generation. This system comprises the following means.

[0455] First, it is equipped with a means for receiving past company documents and storing them in a database. When a user uploads past company documents using their own device, the device sends the documents to the server. Once sent, the server receives the documents and stores them in the database. This step is essential for learning the unique tastes of each company.

[0456] Next, the system retrieves the documents stored in the database and applies a machine learning algorithm to learn the company's unique tastes based on those documents. The server retrieves the documents from the database and performs preprocessing. This preprocessing includes extracting text information from the documents and deleting and correcting unnecessary data. After preprocessing is complete, the machine learning algorithm is used to analyze the documents and learn parameters that characterize the company's unique tastes, such as format, design, and terminology selection.

[0457] Furthermore, the system is equipped with a means for automatically generating materials that reflect the company's unique taste, using a trained machine learning model based on the outline of the materials received from the user. The user inputs an outline of the materials they want to generate into the terminal and sends that information to the server. The server generates the materials using a trained AI model based on the received outline. The generated materials reflect the company's unique taste, maintaining consistency.

[0458] In addition, the system is equipped with an emotion engine that recognizes the user's emotions. The emotion engine analyzes the emotions from the user's input. For example, when a user inputs an outline for document generation, the input is analyzed for emotions using natural language processing technology. The results of this emotion analysis can be reflected in the document generation process, allowing the tone and style of the generated document to be adjusted.

[0459] Specifically, when a user inputs an outline of a "new product launch presentation," for example, if the user enters words that contain strong emotions, such as "this new product will take the market by storm," the emotion engine analyzes the words and recognizes them as emotions indicating excitement or anticipation. The server then reflects the emotion recognition results in the creation of the materials, using bolder, more impactful designs and words to generate materials that reflect the user's emotions.

[0460] Finally, a means for outputting the generated materials to the user is provided. The server transmits the generated materials to the user's terminal, and the terminal displays the received materials to the user. The user can download the materials as needed.

[0461] This invention enables companies to efficiently generate materials that respond to users' emotions and intentions while maintaining a consistent brand image, which significantly improves the efficiency of document creation work and allows companies to provide materials that strongly reflect the company's unique character.

[0462] The processing flow will be explained below.

[0463] The present invention is a system that automatically generates materials that reflect a company's unique taste, and also recognizes the user's emotions and reflects the results. This system operates through the following steps.

[0464] Step 1:

[0465] The user selects the company's past documents on the terminal and clicks the upload button.

[0466] Step 2:

[0467] The terminal sends the selected material to the server as an HTTP request.

[0468] Step 3:

[0469] The server analyzes the received materials and stores them in a database, including metadata such as company ID and user ID.

[0470] Step 4:

[0471] The server retrieves all documents related to the company from the database, preprocesses the documents, extracts text, and cleans unnecessary data.

[0472] Step 5:

[0473] The server applies machine learning algorithms to the preprocessed materials to learn the company's specific tastes (formatting, terminology, style, etc.), and saves the learning results as an AI model.

[0474] Step 6:

[0475] The user inputs an outline of the material they want to generate into the terminal, for example, "New Product Launch Presentation," and clicks the send button.

[0476] Step 7:

[0477] The device sends the summary entered by the user to the server as an HTTP request.

[0478] Step 8:

[0479] The server passes the received summary to the emotion engine, which analyzes the emotion from the content entered by the user. For example, natural language processing technology is used to recognize positive emotion from the input "This new product is revolutionary!"

[0480] Step 9:

[0481] The server uses a trained AI model to generate materials based on the analysis results of the emotion engine. The generated materials include elements that reflect the user's emotions in addition to the company's unique style.

[0482] Step 10:

[0483] The server sends the generated document to the user's device in the format specified by the user, such as PDF or PPTX.

[0484] Step 11:

[0485] The device displays the received materials to the user, who is also provided with a link to download the materials if desired.

[0486] As a concrete example, if a company is preparing presentation materials for the launch of a new product, the user first uploads past materials, which the system then learns from. Next, the user enters a summary such as "New product launch presentation." If the user enters words containing strong emotions, such as "taking the market by storm," the emotion engine analyzes this and recognizes it as an emotion indicating excitement or anticipation. The server then reflects this emotion recognition result in the creation of the materials, generating presentation materials using bolder, more impactful designs and words based on the company's unique taste. As a result, materials that respond to the user's emotions are efficiently created while maintaining the company's consistent brand image.

[0487] Example 2

[0488] 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."

[0489] Companies are required to efficiently create materials while maintaining a consistent brand image. However, manually created materials often fail to reflect the company's unique style and require a great deal of time and effort. Furthermore, it is difficult to reflect the user's emotions and intentions in the materials, which can result in ineffective materials. To solve these issues, there is a need for a system that automatically generates materials that reflect the company's unique style and take the user's emotions into consideration.

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

[0491] In this invention, the server includes means for receiving past company documents and storing them in a database, means for acquiring documents from the database and applying a machine learning algorithm to learn the company's unique taste, means for automatically generating documents that reflect the company's unique taste using a trained machine learning model based on a document generation summary received from a user, an emotion engine that recognizes the user's emotions and reflects them in the document generation process, and means for outputting the generated documents to the user. This makes it possible to efficiently generate optimal documents that match the user's emotions and intentions while maintaining the company's unique taste.

[0492] "Corporate historical documents" refer to information assets such as documents, presentations, reports, and data sets that a company has created or held in the past.

[0493] A "database" is a collection of structured data for systematically organizing and managing information, and is a system that allows for various searches and operations to be performed efficiently.

[0494] A "machine learning algorithm" is a set of computational techniques that allow a computer to find patterns and rules in data and make predictions or classifications.

[0495] "Corporate taste" refers to the unique design, format, terminology, style, etc. that a company consistently uses, and is an element that forms the company's brand image.

[0496] A "trained machine learning model" is the output of a machine learning algorithm that has been trained using historical data and is ready to automate a specific task.

[0497] An "emotion engine" is a system that analyzes emotions from user input and text data and adjusts the tone and style of content based on the results.

[0498] "User emotion" refers to the psychological state, such as joy, expectation, excitement, sadness, etc., that the user shows when inputting the outline of the material generation.

[0499] "Natural language processing technology" is a technology that enables computers to understand, interpret, and generate human language, and is used for text analysis and sentiment analysis.

[0500] The "document generation overview" is a specific explanation and requirements such as the purpose, target, and necessary elements of the document the user wants to generate.

[0501] A "document generation process" refers to a series of steps or operations that automatically create documents based on input data.

[0502] This invention relates to a system that automatically generates documents that reflect a company's unique taste, as well as a system that recognizes the user's emotions and reflects the results in the document generation. This system is realized by the cooperation of the server, terminals, and users.

[0503] First, it is equipped with a means to receive a company's past documents and store them in a database. When a user uploads a company's past documents using their own device, the device sends the documents to a server. Once sent, the server receives the documents and stores them in a database such as MySQL. This step allows data to be collected to learn the unique tastes of each company.

[0504] The system then retrieves documents stored in the database and applies machine learning algorithms to learn the company's unique tastes. The server retrieves the documents from the database and performs preprocessing. This preprocessing includes extracting text information using OCR technology and deleting unnecessary data. After preprocessing is complete, the documents are analyzed using machine learning libraries such as Scikit-learn and TensorFlow to learn parameters that characterize the company's unique tastes, such as formatting, design, and terminology selection.

[0505] Furthermore, it is equipped with a means to automatically generate materials that reflect the company's unique taste using a trained machine learning model based on the outline of the materials received from the user. The user inputs the outline of the materials they want to generate into the terminal and sends that information to the server. The server generates the materials based on the received outline using a trained AI model such as BERT or GPT-3. The generated materials reflect the company's unique taste and maintain consistency. Below are some example prompts:

[0506] Example prompt sentence:

[0507] "Please create a presentation for the launch of a new product. This product is expected to take the market by storm. The presentation should be visually impactful and the design should be modern and dynamic."

[0508] In addition, the system is equipped with an emotion engine that recognizes user emotions. The emotion engine is responsible for analyzing emotions from user input. For example, when a user enters an outline for document generation, the input is analyzed for emotion using natural language processing technology (e.g., AWS Comprehend or Google Cloud Natural Language). The results of this emotion analysis can be reflected in the document generation process to adjust the tone and style of the generated document. For example, if a user enters words that contain strong emotions, such as "This new product will take the market by storm," the emotion engine will analyze it and recognize it as an emotion indicating excitement or anticipation. The server then reflects the emotion recognition results in the document generation, using bolder, more impactful designs and words.

[0509] Finally, a means for outputting the generated materials to the user is provided. The server transmits the generated materials to the user's terminal, and the terminal displays the received materials to the user. The user can download the materials as needed.

[0510] This invention enables companies to efficiently generate materials that respond to users' emotions and intentions while maintaining a consistent brand image, which significantly improves the efficiency of document creation work and allows companies to provide materials that strongly reflect the company's unique character.

[0511] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0512] Step 1:

[0513] Input: The user uploads the company's past documents.

[0514] How it works: A user uses their device to upload past company documents (e.g., PDF or Word files) through the interface. The device retrieves these documents and selects the documents to be uploaded through the file selection function.

[0515] Output: The terminal sends the selected material to the server as an HTTP POST request.

[0516] Step 2:

[0517] Input: The data file sent from the terminal.

[0518] Operation: The server receives the file and saves it in a database such as MySQL. The received file is stored in the specified location, and metadata (e.g., upload date and time, user ID) is also recorded.

[0519] Output: Material files and metadata stored in a database.

[0520] Step 3:

[0521] Input: Materials stored in the database.

[0522] How it works: The server periodically or on demand retrieves historical data from the database and performs pre-processing, which involves extracting text from images using OCR technology and removing or correcting unnecessary data (e.g., white space, advertisements).

[0523] Output: The preprocessed material with extracted text information.

[0524] Step 4:

[0525] Input: Preprocessed material.

[0526] How it works: The server uses machine learning algorithms (such as Scikit-learn or TensorFlow) to analyze the pre-processed materials, using them as a training dataset to learn parameters that characterize the company's specific formatting, design, and terminology choices.

[0527] Output: A machine learning model that reflects your company's unique taste.

[0528] Step 5:

[0529] Input: A summary of the generated material entered by the user into the terminal.

[0530] Operation: The user inputs a detailed outline of the material creation into the terminal and sends the information to the server.

[0531] Output: Summary information of the document generation is sent to the server.

[0532] Step 6:

[0533] Input: Summary information for material generation.

[0534] How it works: The server uses trained machine learning models (such as BERT or GPT-3) to automatically create documents based on the summary information. The server then checks whether the generated documents reflect the company's unique taste and makes any necessary adjustments.

[0535] Output: The generated material.

[0536] Step 7:

[0537] Input: A summary of what the user has entered and what materials are generated.

[0538] How it works: The server uses an emotion engine to parse the emotion from the user's input. It uses natural language processing technology (such as AWS Comprehend or Google Cloud Natural Language) to identify the emotion in the input and adjust the tone and style of the material.

[0539] Output: Sentiment analysis results and generated materials adjusted to reflect them.

[0540] Step 8:

[0541] Input: Final adjusted generated material.

[0542] How it works: The server sends the final generated material to the user's device. The material is provided via email or a dedicated download link.

[0543] Output: The generated material sent to the user's device.

[0544] Step 9:

[0545] Input: Generated material sent from the server.

[0546] Operation: The device displays the data received from the server, and the user can review it. If necessary, the data can be downloaded and saved.

[0547] Output: The final generated material displayed on the user's terminal.

[0548] (Application example 2)

[0549] 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."

[0550] When companies create materials such as advertisements, it is extremely difficult to tailor the materials to reflect user emotions and intentions while maintaining a consistent brand image. Furthermore, creating advertisements manually takes time and effort, making efficient operation difficult. Furthermore, there is no method for quickly generating advertisements that reflect user emotions. This leaves companies unable to create effective advertisements that match the emotions of target users, limiting the effectiveness of their advertisements.

[0551] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0552] In this invention, the server includes: means for receiving a company's past materials and storing them in a database; means for acquiring materials from the database and using the materials to apply a machine learning algorithm to learn the company's unique tastes; means for automatically generating materials that reflect the company's unique tastes and the user's emotions using the trained machine learning model based on a material generation summary received from the user; means for analyzing the user's emotions, including an emotion analysis engine for analyzing the user's emotions, and means for outputting the generated materials to the user. This enables a company to quickly and efficiently generate effective advertising materials that reflect the user's emotions while maintaining a consistent brand image.

[0553] "Corporate historical documents" refers to data such as documents, reports, presentation materials, advertising materials, and other documents of any format that a company has created, published, or stored in the past.

[0554] A "database" is a system that organizes and stores collected data, enabling efficient searches and analysis.

[0555] A "machine learning algorithm" is a technology that automatically learns patterns and rules from data and makes predictions and analyses based on that learned knowledge.

[0556] "A company's unique taste" refers to the brand image, style, and design concept that a company aims for, and is a general term for elements that have characteristics unique to that company.

[0557] "User emotion" refers to the emotional state expressed by the user during the process of creating materials, and includes various emotions such as joy, excitement, anticipation, and anxiety.

[0558] An "emotion analysis engine" is a system that analyzes the emotions from user input and identifies specific emotional states.

[0559] The "document generation overview" is a description showing basic information, purpose, theme, and specifications of the document that the user wants to create.

[0560] "Natural language processing technology" is a technology that allows computers to understand and generate human language, and involves analyzing and generating text.

[0561] A "machine learning model" is an AI model that is trained based on data to understand certain patterns and trends and generate responses.

[0562] "Means for outputting to the user" refers to a method or system for providing the generated materials to the user, and includes functions such as display, download, and sharing.

[0563] The present invention is a system that automatically generates advertising materials by analyzing user emotions and reflecting the unique tastes of a company. Specific embodiments for carrying out the present invention will be described below.

[0564] System configuration

[0565] The system mainly consists of a server, the user's smartphone, a database, a machine learning algorithm, a natural language processing engine, and a sentiment analysis engine.

[0566] Hardware

[0567] Server: Receives, stores, analyzes data, generates advertising materials, and outputs them.

[0568] Smartphone: The user inputs the outline of the material generation and receives the generated advertising material.

[0569] software

[0570] Machine learning libraries: TensorFlow and PyTorch are used to learn company-specific tastes.

[0571] Natural language processing engine: spaCy, which uses BERT to analyze user-input text.

[0572] Database: MySQL and MongoDB are used to store corporate historical data.

[0573] Sentiment analysis engine: Analyzes user emotions in real time.

[0574] Process Overview

[0575] 1. Data Collection:

[0576] The smartphone uploads the company's past data (such as advertising data) to the server, which receives the data and stores it in a database.

[0577] 2. Training the machine learning model:

[0578] We retrieve past advertising data stored in a database and preprocess the text information. After text extraction, cleaning, and removing or correcting unnecessary data, we use TensorFlow and PyTorch to train a machine learning model that learns the unique tastes of each company.

[0579] 3. Emotion analysis:

[0580] When a user inputs a summary of the ad they want to generate, the input is analyzed in real time using a natural language processing engine (spaCy, BERT), and the emotional state is identified using a sentiment analysis engine.

[0581] 4. Automatic generation of advertising materials:

[0582] Based on the outline of the materials generated by the user and the results of sentiment analysis, advertising materials are automatically generated using a trained machine learning model. The generated advertisements reflect the company's unique taste and the user's emotions.

[0583] 5. Output to the user:

[0584] The generated advertising materials are sent from the server to the user's smartphone, where the user can preview, edit, and download the generated advertisements.

[0585] Specific examples

[0586] For example, if a user inputs the summary "I want to create a campaign ad for a new eco-product. I want to convey my passion for protecting the earth," the emotion "passion for protecting the earth" will be recognized as a positive and strong emotion from this summary. Below is an example of a prompt to input to the generative AI model:

[0587] Prompt: "We want to create an ad campaign for a new eco-friendly product. We want to convey our passion for protecting the planet. The ad should reflect our company's unique eco-friendly image and have a passionate, positive tone."

[0588] Based on this prompt, the generative AI model automatically creates an ad with a passionate and positive tone that reflects the company's unique taste.

[0589] conclusion

[0590] This system enables companies to efficiently create advertisements that match the emotions of target users while maintaining a consistent brand image. A major feature of this invention is that it combines highly accurate emotion analysis technology with machine learning models to generate effective advertisements that match user needs.

[0591] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0592] Step 1:

[0593] Receive past company documents and store them in a database.

[0594] Input: Upload past company documents (advertising data) from your smartphone.

[0595] Operation: The device sends the data to the server.

[0596] Data processing: The server organizes the received data and stores it in a database.

[0597] Output: Past company information stored in the database.

[0598] Step 2:

[0599] It retrieves data from a database and uses it to apply machine learning algorithms to learn the unique tastes of each company.

[0600] Input: Past advertising materials stored in the database.

[0601] How it works: The server retrieves material from the database and pre-processes it (text extraction, cleaning, removing and modifying unnecessary data).

[0602] Data calculation: Using preprocessed data, machine learning models such as TensorFlow and PyTorch are used to learn the company's specific tastes.

[0603] Output: A trained machine learning model.

[0604] Step 3:

[0605] The user inputs an outline of the document to be generated, and the emotion analysis engine analyzes the user's emotions.

[0606] Input: An overview of the generated materials entered by the user on their smartphone.

[0607] How it works: A natural language processing engine (spaCy, BERT) analyzes user-entered text. A sentiment analysis engine identifies emotional states.

[0608] Data calculation: Identify the user's emotional state based on the emotion analysis results.

[0609] Output: Emotion analysis results.

[0610] Step 4:

[0611] Advertising materials are automatically generated based on the outline of material generation received from the user and the results of sentiment analysis.

[0612] Input: Summary of material generation and sentiment analysis results.

[0613] How it works: The server uses trained machine learning models to automatically generate advertising materials based on your input.

[0614] Data calculation: Generate advertising materials that reflect the company's unique taste and user emotions.

[0615] Output: Generated advertising materials.

[0616] Step 5:

[0617] The generated advertising material is output to the user.

[0618] Input: Generated advertising materials.

[0619] Operation: The server sends the generated materials to the user's smartphone.

[0620] Output: Advertising materials displayed on the user's smartphone, which the user can preview, edit, and download.

[0621] 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.

[0622] 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.

[0623] 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.

[0624] [Third embodiment]

[0625] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0626] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0627] 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).

[0628] 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.

[0629] 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.

[0630] 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).

[0631] 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.

[0632] 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.

[0633] 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.

[0634] 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.

[0635] 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.

[0636] 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."

[0637] The present invention relates to a system for automatically generating materials that reflect a company's unique taste. This system comprises the following means.

[0638] First, it is equipped with a means for receiving a company's past documents and storing them in a database. When a user uploads a company's past documents using their own device, the device sends the documents to the server. Once sent, the server receives the documents and stores them in the database. This allows a company's past documents to be managed systematically and made available for later processes.

[0639] Next, the system is equipped with a means for retrieving documents stored in the database and applying a machine learning algorithm based on the documents to learn the company's unique taste. The server retrieves the documents from the database and performs preprocessing. Preprocessing includes extracting text information from the documents and deleting and correcting unnecessary data. After preprocessing is complete, the documents are analyzed using a machine learning algorithm. This allows the system to learn parameters that characterize the company's unique taste, such as format, design, and terminology selection.

[0640] Furthermore, the system is equipped with a means for automatically generating materials that reflect the company's unique taste using a trained machine learning model based on the outline of the materials received from the user. The user inputs an outline of the materials they want to generate into the terminal and sends that information to the server. The server generates materials using a trained AI model based on the received outline. These materials reflect the company's unique taste, ensuring consistency.

[0641] Finally, the system includes a means for outputting the generated materials to the user. The server transmits the generated materials to the user's terminal. The terminal displays the received materials and provides a link for the user to download them.

[0642] As a concrete example, consider the case where a company is preparing presentation materials for the launch of a new product. A user uploads previously created materials to the system, which then learns from them. After that, when the user enters the description "New product launch presentation," the server generates presentation materials that reflect the company's unique style based on that information and provides them to the user. In this way, new materials can be created efficiently while maintaining the company's consistent brand image.

[0643] The processing flow will be explained below.

[0644] Step 1:

[0645] The user selects the company's past documents on the terminal and clicks the upload button.

[0646] Step 2:

[0647] The terminal sends the selected material to the server as an HTTP request.

[0648] Step 3:

[0649] The server analyzes the received documents and stores them in a database. Metadata such as company ID and user ID for each document are also stored at the same time.

[0650] Step 4:

[0651] The server retrieves all documents related to the target company from the database, preprocesses the documents, extracts text, and cleans them, making the contents of the documents easier to analyze.

[0652] Step 5:

[0653] The server applies machine learning algorithms to the preprocessed materials to learn the company's specific tastes (formatting, terminology, style, etc.), and saves the learned parameters as an AI model.

[0654] Step 6:

[0655] The user inputs an outline of the material they want to generate into the terminal, for example, "New Product Launch Presentation," and clicks the send button.

[0656] Step 7:

[0657] The device sends the summary entered by the user to the server as an HTTP request.

[0658] Step 8:

[0659] Based on the summary received by the server, a pre-trained AI model is used to generate materials that reflect the company's unique style.

[0660] Step 9:

[0661] The server sends the generated document to the user's device in the format specified by the user, such as PDF or PPTX.

[0662] Step 10:

[0663] The device displays the received materials to the user, and also provides a link so the user can download the materials if desired.

[0664] This will create a system that allows companies to create documents efficiently while maintaining a consistent brand image.

[0665] Example 1

[0666] 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."

[0667] When companies create documents, they need a way to efficiently generate them while maintaining a consistent brand image. However, previous methods required a lot of manual work, which took time and effort, making it difficult to create documents efficiently. In addition, utilizing past documents to reflect a company's unique style required advanced technology and specialized knowledge. This has created a need for a new system that can create documents efficiently and consistently.

[0668] 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.

[0669] In this invention, the server includes means for receiving a company's past documents and storing them in a database, means for acquiring documents from the database and using the documents to apply a machine learning algorithm to learn the company's unique taste, means for automatically generating documents that reflect the company's unique taste using the trained machine learning model based on a document generation summary received from a user, means for performing preprocessing to extract text information and deleting and correcting unnecessary data, and means for outputting the generated documents to the user. This enables companies to efficiently automatically generate consistent documents, significantly reducing the effort required while maintaining their brand image.

[0670] "Company archives" are documents and files created or received by a company in the past, including presentation slides, reports, marketing materials, etc.

[0671] A "database" is a system for efficiently storing, managing, and searching information, and is used to systematically store a company's past documents.

[0672] A "machine learning algorithm" is a type of software that finds patterns and features in large amounts of data and makes predictions and classifications based on the data.

[0673] "Corporate style" is a general term for the consistent format, design, and terminology used by a particular company in its materials.

[0674] The "outline of material generation" includes basic information and instructions regarding the purpose and content of the material the user wants to generate.

[0675] A "trained machine learning model" is a machine learning algorithm that has been trained to the point where it is capable of performing a specific task using training data.

[0676] "Preprocessing" refers to a series of operations to convert raw data into an analyzable format, including extracting text information, removing noise data, and standardizing formats.

[0677] "Automatically generated" means that the system creates the materials independently without human intervention.

[0678] "Output" refers to providing the system-generated materials in a format that can be used by the user.

[0679] The present invention relates to a system for automatically generating materials that reflect a company's unique taste. This system comprises the following means.

[0680] First, a user uploads past company documents from their own device to the server, including presentation slides, reports, marketing materials, etc. The device then transmits the uploaded documents to the server.

[0681] The server receives the data sent from the device and stores it in a database. The database uses a data management system such as MySQL or PostgreSQL. When storing data, the server also stores metadata (creation date, creator, etc.) of the data, allowing for systematic management of the data.

[0682] Next, the server retrieves the stored historical documents from the database. If necessary, it filters the documents based on specific criteria (creation date, author, project name, etc.). The server then performs preprocessing on the retrieved documents. This preprocessing involves extracting text information from the documents, removing unnecessary data and noise, and standardizing the format. For example, text can be extracted from PDFs and unnecessary elements such as advertisements and copyright information can be removed.

[0683] The server then applies machine learning algorithms to the preprocessed materials. Specifically, it uses machine learning frameworks such as TensorFlow and PyTorch to learn the company's unique taste. The analysis targets include frequently used fonts, color schemes, and distinctive phrases. This allows the server to learn parameters that characterize the company's unique taste.

[0684] Next, the user inputs a summary of the materials they want to create into their device, for example, by entering specific instructions such as "Please create materials for a presentation to launch a new product." The device then sends the input summary to the server.

[0685] The server generates materials based on the received summary using a trained AI model (e.g., GPT-3, BERT, etc.). The server automatically generates materials that reflect the company's unique style (unique fonts, color schemes, and phrases), and the content reflects the summary entered by the user. For example, in the case of a presentation material for the launch of a new product, the server automatically generates the necessary slides, text content, graphs, etc.

[0686] Finally, the server sends the generated materials to the user's terminal, where they are displayed and a download link is provided for easy access. Users can click the link to download and use the materials.

[0687] As a concrete example, consider the case where a company is preparing presentation materials for the launch of a new product. A user uploads previously created materials to the system, which then learns from them. Then, when the user enters the outline "new product launch presentation" into their terminal, the server generates presentation materials that reflect the company's unique style based on that information and provides them to the user. In this way, new materials can be created efficiently while maintaining the company's consistent brand image.

[0688] An example of a prompt might be, "I would like to create a presentation for the launch of a new product. I would like the design to follow the style of past presentation materials while incorporating the latest trends."

[0689] This system allows companies to efficiently generate materials while maintaining a consistent brand image, resulting in significant savings in time and effort and an improvement in the efficiency of the entire material creation process.

[0690] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0691] Step 1:

[0692] Uploading and saving past documents

[0693] Users upload past company documents from their own devices. These include presentation slides, reports, marketing materials, etc. The documents uploaded by the users are sent from the devices to the server. The server stores the received documents in a database. The input is the documents uploaded by the users, and the output is the documents stored in the database.

[0694] Step 2:

[0695] Material acquisition and preprocessing

[0696] The server retrieves stored past documents from the database. If necessary, it filters the documents using specific criteria (creation date and time, creator, project name, etc.). The server then performs preprocessing on the retrieved documents. Preprocessing includes extracting text information from the documents, removing unnecessary data and noise, and standardizing the format. For example, extracting text from PDFs and removing unnecessary elements such as advertisements and copyright information. The input is the document retrieved from the database, and the output is the document after preprocessing.

[0697] Step 3:

[0698] Learning tastes with machine learning algorithms

[0699] The server applies machine learning algorithms to the preprocessed documents. Specifically, it uses machine learning frameworks such as TensorFlow and PyTorch to learn the company's unique taste. The analysis targets include frequently used fonts, color schemes, and distinctive phrases. The input is the preprocessed documents, and the output is parameters that characterize the company's unique taste.

[0700] Step 4:

[0701] Input and submit outline of document generation

[0702] The user inputs an outline of the material they want to generate into their terminal. For example, they input specific instructions such as "Please create a presentation for the launch of a new product." The terminal then sends the input outline to the server. The input is the outline of the material that the user inputs into the terminal, and the output is the outline sent to the server.

[0703] Step 5:

[0704] Automatic generation of materials that reflect your taste

[0705] The server generates materials based on the received summary using a trained AI model (e.g., GPT-3, BERT, etc.). The server automatically generates materials that reflect the company's unique taste (unique fonts, color schemes, and phrases), and the content of the materials reflects the summary entered by the user. For example, in the case of a presentation material for the launch of a new product, the necessary slides, text content, graphs, etc. are automatically generated. The input is the summary of the material generation sent to the server and the trained taste parameters, and the output is the generated material.

[0706] Step 6:

[0707] Output and download of generated materials

[0708] The server sends the generated material to the user's device. The device displays the received material and provides a download link for easy access by the user. The user can click the link to download and use the material. The input is the material generated by the server, and the output is the download link displayed on the user's device.

[0709] Each of these steps automatically generates materials that reflect the company's unique style, maintaining a consistent brand image and streamlining document creation.

[0710] (Application example 1)

[0711] 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."

[0712] The present invention aims to provide a system that enables sales staff in brick-and-mortar stores to more smoothly communicate with customers and provide information that reflects the company's unique taste in real time. In particular, the system aims to enable sales staff to respond promptly to customers' questions and needs, and provide appropriate information while consistently maintaining the company's brand image.

[0713] 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.

[0714] In this invention, the server includes means for receiving past company information and storing it in a database, means for acquiring information from the database and using the information to apply a machine learning algorithm to learn the company's unique tastes, means for automatically generating information reflecting the company's unique tastes using the trained machine learning model based on an overview of information generation received from a user, means for outputting the generated information to the user, means for extracting customer questions and needs in real time using voice recognition and providing optimal information based on the context, and means for displaying information on a smart device in real time. This enables salespeople to provide appropriate information according to customer needs in real time and interact with customers while maintaining a consistent company brand image.

[0715] "Past information about a company" refers to data such as materials, documents, and presentations created by the company in the past.

[0716] A "database" is a storage system for storing and systematically managing received information.

[0717] A "machine learning algorithm" is a mathematical model that extracts patterns and features from large amounts of data and performs specific tasks.

[0718] "A company's unique taste" is a general term that refers to the brand image that a company maintains, the unique terminology, format, design, etc. that are used.

[0719] "Outline of information generation" refers to an outline of the requirements, purpose, content, etc. of the materials or documents that the user wants to create.

[0720] A "trained machine learning model" is a machine learning algorithm that has been pre-trained using large amounts of data, and is a model that can efficiently perform a specific task.

[0721] "Automatic generation" refers to the process by which a system automatically creates materials or documents without human intervention.

[0722] "Speech recognition" is a technology that converts human speech into digital data and understands its content.

[0723] "Context" refers to the meaning and circumstances behind a customer's questions, needs, and interactions.

[0724] "Smart devices" is a general term for electronic devices that can be worn by users to display and operate information, such as smartphones, smart glasses, and head-mounted displays.

[0725] This invention is a system that provides an "intelligent sales support app" that enables salespeople in brick-and-mortar stores to more smoothly communicate with customers. The system uses smart devices, servers, natural language processing technology, speech recognition technology, and generative AI models.

[0726] First, the server receives the company's past information and stores it in a database. When a user uploads the company's past information using their own device, the device sends the information to the server. Once sent, the server receives the information and stores it in a database. This allows the company's past information to be managed systematically and made available for later processes.

[0727] The server then retrieves the information stored in the database and applies a machine learning algorithm based on that information to learn the company's unique tastes. This process uses natural language processing technology as the machine learning algorithm. The server extracts and cleans text, selecting and formatting the necessary data. Based on the formatted data, the generative AI model learns the company's unique tastes.

[0728] When a user inputs an outline of the document to be generated into the device, the information is sent to the server. Based on the received outline, the server uses a trained generative AI model to automatically generate information that reflects the company's unique taste. This information is displayed on the device in real time. In particular, by utilizing voice recognition technology, questions and needs can be extracted in real time from conversations with customers, and the most appropriate information is displayed based on that context.

[0729] For example, when a salesperson uses smart glasses to interact with a customer, if the customer asks, "What are the features of the new product?", the voice information is sent to the server. The server uses voice recognition technology to convert the customer's question into text, and a generative AI model generates appropriate information. This information is displayed in real time on the smart glasses, allowing the salesperson to provide a prompt explanation to the customer.

[0730] An example prompt is:

[0731] "Create a new product launch presentation based on your past materials: what specific points should you highlight to answer customer questions about the new product's features?"

[0732] This system allows sales staff to provide appropriate information in real time to meet customer needs, enabling them to serve customers while maintaining a consistent company brand image.

[0733] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0734] Step 1:

[0735] The user operates the device to upload the company's past information. The uploaded information is sent from the device to the server. The input is the company's past information, and the output is the information stored on the server. Specifically, the user selects a file using the device interface and clicks the upload button.

[0736] Step 2:

[0737] The server stores the received information in a database. The input is the uploaded information, and the output is the information stored in the database. Specifically, the server analyzes the contents of the received file and writes it to the corresponding database.

[0738] Step 3:

[0739] The server retrieves information from the database and performs text extraction and cleaning. The input is the information stored in the database, and the output is the extracted and cleaned text data. Specifically, the server removes unnecessary parts from the information and extracts the necessary text data.

[0740] Step 4:

[0741] The server applies a machine learning algorithm to the extracted and cleaned text data to learn the company's unique tastes. The input is the extracted and cleaned text data, and the output is the learned taste parameters. Specifically, the server uses natural language processing technology to analyze the text data and extract features.

[0742] Step 5:

[0743] The user inputs a summary of the document generation into the terminal and sends the information to the server. The input is the summary of the document generation, and the output is the summary data sent to the server. Specifically, the user uses the terminal interface to input the summary into the text box and clicks the send button.

[0744] Step 6:

[0745] Based on the received summary data, the server uses a trained generative AI model to automatically generate information that reflects the company's unique taste. The input is the summary data and the trained model, and the output is automatically generated information. Specifically, the server inputs prompts to the generative AI model and obtains the generated information.

[0746] Step 7:

[0747] The server sends the generated information to the user's terminal, which then displays the information. The input is the automatically generated information, and the output is the information displayed on the terminal. Specifically, the server sends the generated information to the terminal via an HTTP request, and the terminal displays the received information on the screen.

[0748] Step 8:

[0749] The device uses voice recognition to extract customer questions and needs in real time and transmit them to the server based on the context. The input is the customer's voice and the output is text-based question data. Specifically, the device's microphone captures the voice and the voice recognition software converts it into text.

[0750] Step 9:

[0751] The server analyzes the text generated by speech recognition and generates the most appropriate information based on the context. The input is the question data in text format, and the output is the most appropriate information. Specifically, the server uses the generative AI model again to generate an appropriate answer to the question.

[0752] Step 10:

[0753] The server displays the generated information on the smart device in real time. The input is the optimal information, and the output is the information displayed on the smart device. Specifically, the server sends the information to a device such as smart glasses, and the device displays the information to the user.

[0754] 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.

[0755] The present invention relates to a system for automatically generating documents that reflect a company's unique taste, as well as a system for recognizing a user's emotions and reflecting the results in the document generation. This system comprises the following means.

[0756] First, it is equipped with a means for receiving past company documents and storing them in a database. When a user uploads past company documents using their own device, the device sends the documents to the server. Once sent, the server receives the documents and stores them in the database. This step is essential for learning the unique tastes of each company.

[0757] Next, the system retrieves the documents stored in the database and applies a machine learning algorithm to learn the company's unique tastes based on those documents. The server retrieves the documents from the database and performs preprocessing. This preprocessing includes extracting text information from the documents and deleting and correcting unnecessary data. After preprocessing is complete, the machine learning algorithm is used to analyze the documents and learn parameters that characterize the company's unique tastes, such as format, design, and terminology selection.

[0758] Furthermore, the system is equipped with a means for automatically generating materials that reflect the company's unique taste, using a trained machine learning model based on the outline of the materials received from the user. The user inputs an outline of the materials they want to generate into the terminal and sends that information to the server. The server generates the materials using a trained AI model based on the received outline. The generated materials reflect the company's unique taste, maintaining consistency.

[0759] In addition, the system is equipped with an emotion engine that recognizes the user's emotions. The emotion engine analyzes the emotions from the user's input. For example, when a user inputs an outline for document generation, the input is analyzed for emotions using natural language processing technology. The results of this emotion analysis can be reflected in the document generation process, allowing the tone and style of the generated document to be adjusted.

[0760] Specifically, when a user inputs an outline of a "new product launch presentation," for example, if the user enters words that contain strong emotions, such as "this new product will take the market by storm," the emotion engine analyzes the words and recognizes them as emotions indicating excitement or anticipation. The server then reflects the emotion recognition results in the creation of the materials, using bolder, more impactful designs and words to generate materials that reflect the user's emotions.

[0761] Finally, a means for outputting the generated materials to the user is provided. The server transmits the generated materials to the user's terminal, and the terminal displays the received materials to the user. The user can download the materials as needed.

[0762] This invention enables companies to efficiently generate materials that respond to users' emotions and intentions while maintaining a consistent brand image, which significantly improves the efficiency of document creation work and allows companies to provide materials that strongly reflect the company's unique character.

[0763] The processing flow will be explained below.

[0764] The present invention is a system that automatically generates materials that reflect a company's unique taste, and also recognizes the user's emotions and reflects the results. This system operates through the following steps.

[0765] Step 1:

[0766] The user selects the company's past documents on the terminal and clicks the upload button.

[0767] Step 2:

[0768] The terminal sends the selected material to the server as an HTTP request.

[0769] Step 3:

[0770] The server analyzes the received materials and stores them in a database, including metadata such as company ID and user ID.

[0771] Step 4:

[0772] The server retrieves all documents related to the company from the database, preprocesses the documents, extracts text, and cleans unnecessary data.

[0773] Step 5:

[0774] The server applies machine learning algorithms to the preprocessed materials to learn the company's specific tastes (formatting, terminology, style, etc.), and saves the learning results as an AI model.

[0775] Step 6:

[0776] The user inputs an outline of the material they want to generate into the terminal, for example, "New Product Launch Presentation," and clicks the send button.

[0777] Step 7:

[0778] The device sends the summary entered by the user to the server as an HTTP request.

[0779] Step 8:

[0780] The server passes the received summary to the emotion engine, which analyzes the emotion from the content entered by the user. For example, natural language processing technology is used to recognize positive emotion from the input "This new product is revolutionary!"

[0781] Step 9:

[0782] The server uses a trained AI model to generate materials based on the analysis results of the emotion engine. The generated materials include elements that reflect the user's emotions in addition to the company's unique style.

[0783] Step 10:

[0784] The server sends the generated document to the user's device in the format specified by the user, such as PDF or PPTX.

[0785] Step 11:

[0786] The device displays the received materials to the user, who is also provided with a link to download the materials if desired.

[0787] As a concrete example, if a company is preparing presentation materials for the launch of a new product, the user first uploads past materials, which the system then learns from. Next, the user enters a summary such as "New product launch presentation." If the user enters words containing strong emotions, such as "taking the market by storm," the emotion engine analyzes this and recognizes it as an emotion indicating excitement or anticipation. The server then reflects this emotion recognition result in the creation of the materials, generating presentation materials using bolder, more impactful designs and words based on the company's unique taste. As a result, materials that respond to the user's emotions are efficiently created while maintaining the company's consistent brand image.

[0788] Example 2

[0789] 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."

[0790] Companies are required to efficiently create materials while maintaining a consistent brand image. However, manually created materials often fail to reflect the company's unique style and require a great deal of time and effort. Furthermore, it is difficult to reflect the user's emotions and intentions in the materials, which can result in ineffective materials. To solve these issues, there is a need for a system that automatically generates materials that reflect the company's unique style and take the user's emotions into consideration.

[0791] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0792] In this invention, the server includes means for receiving past company documents and storing them in a database, means for acquiring documents from the database and applying a machine learning algorithm to learn the company's unique taste, means for automatically generating documents that reflect the company's unique taste using a trained machine learning model based on a document generation summary received from a user, an emotion engine that recognizes the user's emotions and reflects them in the document generation process, and means for outputting the generated documents to the user. This makes it possible to efficiently generate optimal documents that match the user's emotions and intentions while maintaining the company's unique taste.

[0793] "Corporate historical documents" refer to information assets such as documents, presentations, reports, and data sets that a company has created or held in the past.

[0794] A "database" is a collection of structured data for systematically organizing and managing information, and is a system that allows for various searches and operations to be performed efficiently.

[0795] A "machine learning algorithm" is a set of computational techniques that allow a computer to find patterns and rules in data and make predictions or classifications.

[0796] "Corporate taste" refers to the unique design, format, terminology, style, etc. that a company consistently uses, and is an element that forms the company's brand image.

[0797] A "trained machine learning model" is the output of a machine learning algorithm that has been trained using historical data and is ready to automate a specific task.

[0798] An "emotion engine" is a system that analyzes emotions from user input and text data and adjusts the tone and style of content based on the results.

[0799] "User emotion" refers to the psychological state, such as joy, expectation, excitement, sadness, etc., that the user shows when inputting the outline of the material generation.

[0800] "Natural language processing technology" is a technology that enables computers to understand, interpret, and generate human language, and is used for text analysis and sentiment analysis.

[0801] The "document generation overview" is a specific explanation and requirements such as the purpose, target, and necessary elements of the document the user wants to generate.

[0802] A "document generation process" refers to a series of steps or operations that automatically create documents based on input data.

[0803] This invention relates to a system that automatically generates documents that reflect a company's unique taste, as well as a system that recognizes the user's emotions and reflects the results in the document generation. This system is realized by the cooperation of the server, terminals, and users.

[0804] First, it is equipped with a means to receive a company's past documents and store them in a database. When a user uploads a company's past documents using their own device, the device sends the documents to a server. Once sent, the server receives the documents and stores them in a database such as MySQL. This step allows data to be collected to learn the unique tastes of each company.

[0805] The system then retrieves documents stored in the database and applies machine learning algorithms to learn the company's unique tastes. The server retrieves the documents from the database and performs preprocessing. This preprocessing includes extracting text information using OCR technology and deleting unnecessary data. After preprocessing is complete, the documents are analyzed using machine learning libraries such as Scikit-learn and TensorFlow to learn parameters that characterize the company's unique tastes, such as formatting, design, and terminology selection.

[0806] Furthermore, it is equipped with a means to automatically generate materials that reflect the company's unique taste using a trained machine learning model based on the outline of the materials received from the user. The user inputs the outline of the materials they want to generate into the terminal and sends that information to the server. The server generates the materials based on the received outline using a trained AI model such as BERT or GPT-3. The generated materials reflect the company's unique taste and maintain consistency. Below are some example prompts:

[0807] Example prompt sentence:

[0808] "Please create a presentation for the launch of a new product. This product is expected to take the market by storm. The presentation should be visually impactful and the design should be modern and dynamic."

[0809] In addition, the system is equipped with an emotion engine that recognizes user emotions. The emotion engine is responsible for analyzing emotions from user input. For example, when a user enters an outline for document generation, the input is analyzed for emotion using natural language processing technology (e.g., AWS Comprehend or Google Cloud Natural Language). The results of this emotion analysis can be reflected in the document generation process to adjust the tone and style of the generated document. For example, if a user enters words that contain strong emotions, such as "This new product will take the market by storm," the emotion engine will analyze it and recognize it as an emotion indicating excitement or anticipation. The server then reflects the emotion recognition results in the document generation, using bolder, more impactful designs and words.

[0810] Finally, a means for outputting the generated materials to the user is provided. The server transmits the generated materials to the user's terminal, and the terminal displays the received materials to the user. The user can download the materials as needed.

[0811] This invention enables companies to efficiently generate materials that respond to users' emotions and intentions while maintaining a consistent brand image, which significantly improves the efficiency of document creation work and allows companies to provide materials that strongly reflect the company's unique character.

[0812] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0813] Step 1:

[0814] Input: The user uploads the company's past documents.

[0815] How it works: A user uses their device to upload past company documents (e.g., PDF or Word files) through the interface. The device retrieves these documents and selects the documents to be uploaded through the file selection function.

[0816] Output: The terminal sends the selected material to the server as an HTTP POST request.

[0817] Step 2:

[0818] Input: The data file sent from the terminal.

[0819] Operation: The server receives the file and saves it in a database such as MySQL. The received file is stored in the specified location, and metadata (e.g., upload date and time, user ID) is also recorded.

[0820] Output: Material files and metadata stored in a database.

[0821] Step 3:

[0822] Input: Materials stored in the database.

[0823] How it works: The server periodically or on demand retrieves historical data from the database and performs pre-processing, which involves extracting text from images using OCR technology and removing or correcting unnecessary data (e.g., white space, advertisements).

[0824] Output: The preprocessed material with extracted text information.

[0825] Step 4:

[0826] Input: Preprocessed material.

[0827] How it works: The server uses machine learning algorithms (such as Scikit-learn or TensorFlow) to analyze the pre-processed materials, using them as a training dataset to learn parameters that characterize the company's specific formatting, design, and terminology choices.

[0828] Output: A machine learning model that reflects your company's unique taste.

[0829] Step 5:

[0830] Input: A summary of the generated material entered by the user into the terminal.

[0831] Operation: The user inputs a detailed outline of the material creation into the terminal and sends the information to the server.

[0832] Output: Summary information of the document generation is sent to the server.

[0833] Step 6:

[0834] Input: Summary information for material generation.

[0835] How it works: The server uses trained machine learning models (such as BERT or GPT-3) to automatically create documents based on the summary information. The server then checks whether the generated documents reflect the company's unique taste and makes any necessary adjustments.

[0836] Output: The generated material.

[0837] Step 7:

[0838] Input: A summary of what the user has entered and what materials are generated.

[0839] How it works: The server uses an emotion engine to parse the emotion from the user's input. It uses natural language processing technology (such as AWS Comprehend or Google Cloud Natural Language) to identify the emotion in the input and adjust the tone and style of the material.

[0840] Output: Sentiment analysis results and generated materials adjusted to reflect them.

[0841] Step 8:

[0842] Input: Final adjusted generated material.

[0843] How it works: The server sends the final generated material to the user's device. The material is provided via email or a dedicated download link.

[0844] Output: The generated material sent to the user's device.

[0845] Step 9:

[0846] Input: Generated material sent from the server.

[0847] Operation: The device displays the data received from the server, and the user can review it. If necessary, the data can be downloaded and saved.

[0848] Output: The final generated material displayed on the user's terminal.

[0849] (Application example 2)

[0850] 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."

[0851] When companies create materials such as advertisements, it is extremely difficult to tailor the materials to reflect user emotions and intentions while maintaining a consistent brand image. Furthermore, creating advertisements manually takes time and effort, making efficient operation difficult. Furthermore, there is no method for quickly generating advertisements that reflect user emotions. This leaves companies unable to create effective advertisements that match the emotions of target users, limiting the effectiveness of their advertisements.

[0852] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0853] In this invention, the server includes: means for receiving a company's past materials and storing them in a database; means for acquiring materials from the database and using the materials to apply a machine learning algorithm to learn the company's unique tastes; means for automatically generating materials that reflect the company's unique tastes and the user's emotions using the trained machine learning model based on a material generation summary received from the user; means for analyzing the user's emotions, including an emotion analysis engine for analyzing the user's emotions, and means for outputting the generated materials to the user. This enables a company to quickly and efficiently generate effective advertising materials that reflect the user's emotions while maintaining a consistent brand image.

[0854] "Corporate historical documents" refers to data such as documents, reports, presentation materials, advertising materials, and other documents of any format that a company has created, published, or stored in the past.

[0855] A "database" is a system that organizes and stores collected data, enabling efficient searches and analysis.

[0856] A "machine learning algorithm" is a technology that automatically learns patterns and rules from data and makes predictions and analyses based on that learned knowledge.

[0857] "A company's unique taste" refers to the brand image, style, and design concept that a company aims for, and is a general term for elements that have characteristics unique to that company.

[0858] "User emotion" refers to the emotional state expressed by the user during the process of creating materials, and includes various emotions such as joy, excitement, anticipation, and anxiety.

[0859] An "emotion analysis engine" is a system that analyzes the emotions from user input and identifies specific emotional states.

[0860] The "document generation overview" is a description showing basic information, purpose, theme, and specifications of the document that the user wants to create.

[0861] "Natural language processing technology" is a technology that allows computers to understand and generate human language, and involves analyzing and generating text.

[0862] A "machine learning model" is an AI model that is trained based on data to understand certain patterns and trends and generate responses.

[0863] "Means for outputting to the user" refers to a method or system for providing the generated materials to the user, and includes functions such as display, download, and sharing.

[0864] The present invention is a system that automatically generates advertising materials by analyzing user emotions and reflecting the unique tastes of a company. Specific embodiments for carrying out the present invention will be described below.

[0865] System configuration

[0866] The system mainly consists of a server, the user's smartphone, a database, a machine learning algorithm, a natural language processing engine, and a sentiment analysis engine.

[0867] Hardware

[0868] Server: Receives, stores, analyzes data, generates advertising materials, and outputs them.

[0869] Smartphone: The user inputs the outline of the material generation and receives the generated advertising material.

[0870] software

[0871] Machine learning libraries: TensorFlow and PyTorch are used to learn company-specific tastes.

[0872] Natural language processing engine: spaCy, which uses BERT to analyze user-input text.

[0873] Database: MySQL and MongoDB are used to store corporate historical data.

[0874] Sentiment analysis engine: Analyzes user emotions in real time.

[0875] Process Overview

[0876] 1. Data Collection:

[0877] The smartphone uploads the company's past data (such as advertising data) to the server, which receives the data and stores it in a database.

[0878] 2. Training the machine learning model:

[0879] We retrieve past advertising data stored in a database and preprocess the text information. After text extraction, cleaning, and removing or correcting unnecessary data, we use TensorFlow and PyTorch to train a machine learning model that learns the unique tastes of each company.

[0880] 3. Emotion analysis:

[0881] When a user inputs a summary of the ad they want to generate, the input is analyzed in real time using a natural language processing engine (spaCy, BERT), and the emotional state is identified using a sentiment analysis engine.

[0882] 4. Automatic generation of advertising materials:

[0883] Based on the outline of the materials generated by the user and the results of sentiment analysis, advertising materials are automatically generated using a trained machine learning model. The generated advertisements reflect the company's unique taste and the user's emotions.

[0884] 5. Output to the user:

[0885] The generated advertising materials are sent from the server to the user's smartphone, where the user can preview, edit, and download the generated advertisements.

[0886] Specific examples

[0887] For example, if a user inputs the summary "I want to create a campaign ad for a new eco-product. I want to convey my passion for protecting the earth," the emotion "passion for protecting the earth" will be recognized as a positive and strong emotion from this summary. Below is an example of a prompt to input to the generative AI model:

[0888] Prompt: "We want to create an ad campaign for a new eco-friendly product. We want to convey our passion for protecting the planet. The ad should reflect our company's unique eco-friendly image and have a passionate, positive tone."

[0889] Based on this prompt, the generative AI model automatically creates an ad with a passionate and positive tone that reflects the company's unique taste.

[0890] conclusion

[0891] This system enables companies to efficiently create advertisements that match the emotions of target users while maintaining a consistent brand image. A major feature of this invention is that it combines highly accurate emotion analysis technology with machine learning models to generate effective advertisements that match user needs.

[0892] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0893] Step 1:

[0894] Receive past company documents and store them in a database.

[0895] Input: Upload past company documents (advertising data) from your smartphone.

[0896] Operation: The device sends the data to the server.

[0897] Data processing: The server organizes the received data and stores it in a database.

[0898] Output: Past company information stored in the database.

[0899] Step 2:

[0900] It retrieves data from a database and uses it to apply machine learning algorithms to learn the unique tastes of each company.

[0901] Input: Past advertising materials stored in the database.

[0902] How it works: The server retrieves material from the database and pre-processes it (text extraction, cleaning, removing and modifying unnecessary data).

[0903] Data calculation: Using preprocessed data, machine learning models such as TensorFlow and PyTorch are used to learn the company's specific tastes.

[0904] Output: A trained machine learning model.

[0905] Step 3:

[0906] The user inputs an outline of the document to be generated, and the emotion analysis engine analyzes the user's emotions.

[0907] Input: An overview of the generated materials entered by the user on their smartphone.

[0908] How it works: A natural language processing engine (spaCy, BERT) analyzes user-entered text. A sentiment analysis engine identifies emotional states.

[0909] Data calculation: Identify the user's emotional state based on the emotion analysis results.

[0910] Output: Emotion analysis results.

[0911] Step 4:

[0912] Advertising materials are automatically generated based on the outline of material generation received from the user and the results of sentiment analysis.

[0913] Input: Summary of material generation and sentiment analysis results.

[0914] How it works: The server uses trained machine learning models to automatically generate advertising materials based on your input.

[0915] Data calculation: Generate advertising materials that reflect the company's unique taste and user emotions.

[0916] Output: Generated advertising materials.

[0917] Step 5:

[0918] The generated advertising material is output to the user.

[0919] Input: Generated advertising materials.

[0920] Operation: The server sends the generated materials to the user's smartphone.

[0921] Output: Advertising materials displayed on the user's smartphone, which the user can preview, edit, and download.

[0922] 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.

[0923] 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.

[0924] 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.

[0925] [Fourth embodiment]

[0926] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0927] 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.

[0928] 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).

[0929] 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.

[0930] 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.

[0931] 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).

[0932] 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.

[0933] 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.

[0934] 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.

[0935] 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.

[0936] 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.

[0937] 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.

[0938] 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."

[0939] The present invention relates to a system for automatically generating materials that reflect a company's unique taste. This system comprises the following means.

[0940] First, it is equipped with a means for receiving a company's past documents and storing them in a database. When a user uploads a company's past documents using their own device, the device sends the documents to the server. Once sent, the server receives the documents and stores them in the database. This allows a company's past documents to be managed systematically and made available for later processes.

[0941] Next, the system is equipped with a means for retrieving documents stored in the database and applying a machine learning algorithm based on the documents to learn the company's unique taste. The server retrieves the documents from the database and performs preprocessing. Preprocessing includes extracting text information from the documents and deleting and correcting unnecessary data. After preprocessing is complete, the documents are analyzed using a machine learning algorithm. This allows the system to learn parameters that characterize the company's unique taste, such as format, design, and terminology selection.

[0942] Furthermore, the system is equipped with a means for automatically generating materials that reflect the company's unique taste using a trained machine learning model based on the outline of the materials received from the user. The user inputs an outline of the materials they want to generate into the terminal and sends that information to the server. The server generates materials using a trained AI model based on the received outline. These materials reflect the company's unique taste, ensuring consistency.

[0943] Finally, the system includes a means for outputting the generated materials to the user. The server transmits the generated materials to the user's terminal. The terminal displays the received materials and provides a link for the user to download them.

[0944] As a concrete example, consider the case where a company is preparing presentation materials for the launch of a new product. A user uploads previously created materials to the system, which then learns from them. After that, when the user enters the description "New product launch presentation," the server generates presentation materials that reflect the company's unique style based on that information and provides them to the user. In this way, new materials can be created efficiently while maintaining the company's consistent brand image.

[0945] The processing flow will be explained below.

[0946] Step 1:

[0947] The user selects the company's past documents on the terminal and clicks the upload button.

[0948] Step 2:

[0949] The terminal sends the selected material to the server as an HTTP request.

[0950] Step 3:

[0951] The server analyzes the received documents and stores them in a database. Metadata such as company ID and user ID for each document are also stored at the same time.

[0952] Step 4:

[0953] The server retrieves all documents related to the target company from the database, preprocesses the documents, extracts text, and cleans them, making the contents of the documents easier to analyze.

[0954] Step 5:

[0955] The server applies machine learning algorithms to the preprocessed materials to learn the company's specific tastes (formatting, terminology, style, etc.), and saves the learned parameters as an AI model.

[0956] Step 6:

[0957] The user inputs an outline of the material they want to generate into the terminal, for example, "New Product Launch Presentation," and clicks the send button.

[0958] Step 7:

[0959] The device sends the summary entered by the user to the server as an HTTP request.

[0960] Step 8:

[0961] Based on the summary received by the server, a pre-trained AI model is used to generate materials that reflect the company's unique style.

[0962] Step 9:

[0963] The server sends the generated document to the user's device in the format specified by the user, such as PDF or PPTX.

[0964] Step 10:

[0965] The device displays the received materials to the user, and also provides a link so the user can download the materials if desired.

[0966] This will create a system that allows companies to create documents efficiently while maintaining a consistent brand image.

[0967] Example 1

[0968] 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."

[0969] When companies create documents, they need a way to efficiently generate them while maintaining a consistent brand image. However, previous methods required a lot of manual work, which took time and effort, making it difficult to create documents efficiently. In addition, utilizing past documents to reflect a company's unique style required advanced technology and specialized knowledge. This has created a need for a new system that can create documents efficiently and consistently.

[0970] 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.

[0971] In this invention, the server includes means for receiving a company's past documents and storing them in a database, means for acquiring documents from the database and using the documents to apply a machine learning algorithm to learn the company's unique taste, means for automatically generating documents that reflect the company's unique taste using the trained machine learning model based on a document generation summary received from a user, means for performing preprocessing to extract text information and deleting and correcting unnecessary data, and means for outputting the generated documents to the user. This enables companies to efficiently automatically generate consistent documents, significantly reducing the effort required while maintaining their brand image.

[0972] "Company archives" are documents and files created or received by a company in the past, including presentation slides, reports, marketing materials, etc.

[0973] A "database" is a system for efficiently storing, managing, and searching information, and is used to systematically store a company's past documents.

[0974] A "machine learning algorithm" is a type of software that finds patterns and features in large amounts of data and makes predictions and classifications based on the data.

[0975] "Corporate style" is a general term for the consistent format, design, and terminology used by a particular company in its materials.

[0976] The "outline of material generation" includes basic information and instructions regarding the purpose and content of the material the user wants to generate.

[0977] A "trained machine learning model" is a machine learning algorithm that has been trained to the point where it is capable of performing a specific task using training data.

[0978] "Preprocessing" refers to a series of operations to convert raw data into an analyzable format, including extracting text information, removing noise data, and standardizing formats.

[0979] "Automatically generated" means that the system creates the materials independently without human intervention.

[0980] "Output" refers to providing the system-generated materials in a format that can be used by the user.

[0981] The present invention relates to a system for automatically generating materials that reflect a company's unique taste. This system comprises the following means.

[0982] First, a user uploads past company documents from their own device to the server, including presentation slides, reports, marketing materials, etc. The device then transmits the uploaded documents to the server.

[0983] The server receives the data sent from the device and stores it in a database. The database uses a data management system such as MySQL or PostgreSQL. When storing data, the server also stores metadata (creation date, creator, etc.) of the data, allowing for systematic management of the data.

[0984] Next, the server retrieves the stored historical documents from the database. If necessary, it filters the documents based on specific criteria (creation date, author, project name, etc.). The server then performs preprocessing on the retrieved documents. This preprocessing involves extracting text information from the documents, removing unnecessary data and noise, and standardizing the format. For example, text can be extracted from PDFs and unnecessary elements such as advertisements and copyright information can be removed.

[0985] The server then applies machine learning algorithms to the preprocessed materials. Specifically, it uses machine learning frameworks such as TensorFlow and PyTorch to learn the company's unique taste. The analysis targets include frequently used fonts, color schemes, and distinctive phrases. This allows the server to learn parameters that characterize the company's unique taste.

[0986] Next, the user inputs a summary of the materials they want to create into their device, for example, by entering specific instructions such as "Please create materials for a presentation to launch a new product." The device then sends the input summary to the server.

[0987] The server generates materials based on the received summary using a trained AI model (e.g., GPT-3, BERT, etc.). The server automatically generates materials that reflect the company's unique style (unique fonts, color schemes, and phrases), and the content reflects the summary entered by the user. For example, in the case of a presentation material for the launch of a new product, the server automatically generates the necessary slides, text content, graphs, etc.

[0988] Finally, the server sends the generated materials to the user's terminal, where they are displayed and a download link is provided for easy access. Users can click the link to download and use the materials.

[0989] As a concrete example, consider the case where a company is preparing presentation materials for the launch of a new product. A user uploads previously created materials to the system, which then learns from them. Then, when the user enters the outline "new product launch presentation" into their terminal, the server generates presentation materials that reflect the company's unique style based on that information and provides them to the user. In this way, new materials can be created efficiently while maintaining the company's consistent brand image.

[0990] An example of a prompt might be, "I would like to create a presentation for the launch of a new product. I would like the design to follow the style of past presentation materials while incorporating the latest trends."

[0991] This system allows companies to efficiently generate materials while maintaining a consistent brand image, resulting in significant savings in time and effort and an improvement in the efficiency of the entire material creation process.

[0992] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0993] Step 1:

[0994] Uploading and saving past documents

[0995] Users upload past company documents from their own devices. These include presentation slides, reports, marketing materials, etc. The documents uploaded by the users are sent from the devices to the server. The server stores the received documents in a database. The input is the documents uploaded by the users, and the output is the documents stored in the database.

[0996] Step 2:

[0997] Material acquisition and preprocessing

[0998] The server retrieves stored past documents from the database. If necessary, it filters the documents using specific criteria (creation date and time, creator, project name, etc.). The server then performs preprocessing on the retrieved documents. Preprocessing includes extracting text information from the documents, removing unnecessary data and noise, and standardizing the format. For example, extracting text from PDFs and removing unnecessary elements such as advertisements and copyright information. The input is the document retrieved from the database, and the output is the document after preprocessing.

[0999] Step 3:

[1000] Learning tastes with machine learning algorithms

[1001] The server applies machine learning algorithms to the preprocessed documents. Specifically, it uses machine learning frameworks such as TensorFlow and PyTorch to learn the company's unique taste. The analysis targets include frequently used fonts, color schemes, and distinctive phrases. The input is the preprocessed documents, and the output is parameters that characterize the company's unique taste.

[1002] Step 4:

[1003] Input and submit outline of document generation

[1004] The user inputs an outline of the material they want to generate into their terminal. For example, they input specific instructions such as "Please create a presentation for the launch of a new product." The terminal then sends the input outline to the server. The input is the outline of the material that the user inputs into the terminal, and the output is the outline sent to the server.

[1005] Step 5:

[1006] Automatic generation of materials that reflect your taste

[1007] The server generates materials based on the received summary using a trained AI model (e.g., GPT-3, BERT, etc.). The server automatically generates materials that reflect the company's unique taste (unique fonts, color schemes, and phrases), and the content of the materials reflects the summary entered by the user. For example, in the case of a presentation material for the launch of a new product, the necessary slides, text content, graphs, etc. are automatically generated. The input is the summary of the material generation sent to the server and the trained taste parameters, and the output is the generated material.

[1008] Step 6:

[1009] Output and download of generated materials

[1010] The server sends the generated material to the user's device. The device displays the received material and provides a download link for easy access by the user. The user can click the link to download and use the material. The input is the material generated by the server, and the output is the download link displayed on the user's device.

[1011] Each of these steps automatically generates materials that reflect the company's unique style, maintaining a consistent brand image and streamlining document creation.

[1012] (Application example 1)

[1013] 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."

[1014] The present invention aims to provide a system that enables sales staff in brick-and-mortar stores to more smoothly communicate with customers and provide information that reflects the company's unique taste in real time. In particular, the system aims to enable sales staff to respond promptly to customers' questions and needs, and provide appropriate information while consistently maintaining the company's brand image.

[1015] 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.

[1016] In this invention, the server includes means for receiving past company information and storing it in a database, means for acquiring information from the database and using the information to apply a machine learning algorithm to learn the company's unique tastes, means for automatically generating information reflecting the company's unique tastes using the trained machine learning model based on an overview of information generation received from a user, means for outputting the generated information to the user, means for extracting customer questions and needs in real time using voice recognition and providing optimal information based on the context, and means for displaying information on a smart device in real time. This enables salespeople to provide appropriate information according to customer needs in real time and interact with customers while maintaining a consistent company brand image.

[1017] "Past information about a company" refers to data such as materials, documents, and presentations created by the company in the past.

[1018] A "database" is a storage system for storing and systematically managing received information.

[1019] A "machine learning algorithm" is a mathematical model that extracts patterns and features from large amounts of data and performs specific tasks.

[1020] "A company's unique taste" is a general term that refers to the brand image that a company maintains, the unique terminology, format, design, etc. that are used.

[1021] "Outline of information generation" refers to an outline of the requirements, purpose, content, etc. of the materials or documents that the user wants to create.

[1022] A "trained machine learning model" is a machine learning algorithm that has been pre-trained using large amounts of data, and is a model that can efficiently perform a specific task.

[1023] "Automatic generation" refers to the process by which a system automatically creates materials or documents without human intervention.

[1024] "Speech recognition" is a technology that converts human speech into digital data and understands its content.

[1025] "Context" refers to the meaning and circumstances behind a customer's questions, needs, and interactions.

[1026] "Smart devices" is a general term for electronic devices that can be worn by users to display and operate information, such as smartphones, smart glasses, and head-mounted displays.

[1027] This invention is a system that provides an "intelligent sales support app" that enables salespeople in brick-and-mortar stores to more smoothly communicate with customers. The system uses smart devices, servers, natural language processing technology, speech recognition technology, and generative AI models.

[1028] First, the server receives the company's past information and stores it in a database. When a user uploads the company's past information using their own device, the device sends the information to the server. Once sent, the server receives the information and stores it in a database. This allows the company's past information to be managed systematically and made available for later processes.

[1029] The server then retrieves the information stored in the database and applies a machine learning algorithm based on that information to learn the company's unique tastes. This process uses natural language processing technology as the machine learning algorithm. The server extracts and cleans text, selecting and formatting the necessary data. Based on the formatted data, the generative AI model learns the company's unique tastes.

[1030] When a user inputs an outline of the document to be generated into the device, the information is sent to the server. Based on the received outline, the server uses a trained generative AI model to automatically generate information that reflects the company's unique taste. This information is displayed on the device in real time. In particular, by utilizing voice recognition technology, questions and needs can be extracted in real time from conversations with customers, and the most appropriate information is displayed based on that context.

[1031] For example, when a salesperson uses smart glasses to interact with a customer, if the customer asks, "What are the features of the new product?", the voice information is sent to the server. The server uses voice recognition technology to convert the customer's question into text, and a generative AI model generates appropriate information. This information is displayed in real time on the smart glasses, allowing the salesperson to provide a prompt explanation to the customer.

[1032] An example prompt is:

[1033] "Create a new product launch presentation based on your past materials: what specific points should you highlight to answer customer questions about the new product's features?"

[1034] This system allows sales staff to provide appropriate information in real time to meet customer needs, enabling them to serve customers while maintaining a consistent company brand image.

[1035] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1036] Step 1:

[1037] The user operates the device to upload the company's past information. The uploaded information is sent from the device to the server. The input is the company's past information, and the output is the information stored on the server. Specifically, the user selects a file using the device interface and clicks the upload button.

[1038] Step 2:

[1039] The server stores the received information in a database. The input is the uploaded information, and the output is the information stored in the database. Specifically, the server analyzes the contents of the received file and writes it to the corresponding database.

[1040] Step 3:

[1041] The server retrieves information from the database and performs text extraction and cleaning. The input is the information stored in the database, and the output is the extracted and cleaned text data. Specifically, the server removes unnecessary parts from the information and extracts the necessary text data.

[1042] Step 4:

[1043] The server applies a machine learning algorithm to the extracted and cleaned text data to learn the company's unique tastes. The input is the extracted and cleaned text data, and the output is the learned taste parameters. Specifically, the server uses natural language processing technology to analyze the text data and extract features.

[1044] Step 5:

[1045] The user inputs a summary of the document generation into the terminal and sends the information to the server. The input is the summary of the document generation, and the output is the summary data sent to the server. Specifically, the user uses the terminal interface to input the summary into the text box and clicks the send button.

[1046] Step 6:

[1047] Based on the received summary data, the server uses a trained generative AI model to automatically generate information that reflects the company's unique taste. The input is the summary data and the trained model, and the output is automatically generated information. Specifically, the server inputs prompts to the generative AI model and obtains the generated information.

[1048] Step 7:

[1049] The server sends the generated information to the user's terminal, which then displays the information. The input is the automatically generated information, and the output is the information displayed on the terminal. Specifically, the server sends the generated information to the terminal via an HTTP request, and the terminal displays the received information on the screen.

[1050] Step 8:

[1051] The device uses voice recognition to extract customer questions and needs in real time and transmit them to the server based on the context. The input is the customer's voice and the output is text-based question data. Specifically, the device's microphone captures the voice and the voice recognition software converts it into text.

[1052] Step 9:

[1053] The server analyzes the text generated by speech recognition and generates the most appropriate information based on the context. The input is the question data in text format, and the output is the most appropriate information. Specifically, the server uses the generative AI model again to generate an appropriate answer to the question.

[1054] Step 10:

[1055] The server displays the generated information on the smart device in real time. The input is the optimal information, and the output is the information displayed on the smart device. Specifically, the server sends the information to a device such as smart glasses, and the device displays the information to the user.

[1056] 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.

[1057] The present invention relates to a system for automatically generating documents that reflect a company's unique taste, as well as a system for recognizing a user's emotions and reflecting the results in the document generation. This system comprises the following means.

[1058] First, it is equipped with a means for receiving past company documents and storing them in a database. When a user uploads past company documents using their own device, the device sends the documents to the server. Once sent, the server receives the documents and stores them in the database. This step is essential for learning the unique tastes of each company.

[1059] Next, the system retrieves the documents stored in the database and applies a machine learning algorithm to learn the company's unique tastes based on those documents. The server retrieves the documents from the database and performs preprocessing. This preprocessing includes extracting text information from the documents and deleting and correcting unnecessary data. After preprocessing is complete, the machine learning algorithm is used to analyze the documents and learn parameters that characterize the company's unique tastes, such as format, design, and terminology selection.

[1060] Furthermore, the system is equipped with a means for automatically generating materials that reflect the company's unique taste, using a trained machine learning model based on the outline of the materials received from the user. The user inputs an outline of the materials they want to generate into the terminal and sends that information to the server. The server generates the materials using a trained AI model based on the received outline. The generated materials reflect the company's unique taste, maintaining consistency.

[1061] In addition, the system is equipped with an emotion engine that recognizes the user's emotions. The emotion engine analyzes the emotions from the user's input. For example, when a user inputs an outline for document generation, the input is analyzed for emotions using natural language processing technology. The results of this emotion analysis can be reflected in the document generation process, allowing the tone and style of the generated document to be adjusted.

[1062] Specifically, when a user inputs an outline of a "new product launch presentation," for example, if the user enters words that contain strong emotions, such as "this new product will take the market by storm," the emotion engine analyzes the words and recognizes them as emotions indicating excitement or anticipation. The server then reflects the emotion recognition results in the creation of the materials, using bolder, more impactful designs and words to generate materials that reflect the user's emotions.

[1063] Finally, a means for outputting the generated materials to the user is provided. The server transmits the generated materials to the user's terminal, and the terminal displays the received materials to the user. The user can download the materials as needed.

[1064] This invention enables companies to efficiently generate materials that respond to users' emotions and intentions while maintaining a consistent brand image, which significantly improves the efficiency of document creation work and allows companies to provide materials that strongly reflect the company's unique character.

[1065] The processing flow will be explained below.

[1066] The present invention is a system that automatically generates materials that reflect a company's unique taste, and also recognizes the user's emotions and reflects the results. This system operates through the following steps.

[1067] Step 1:

[1068] The user selects the company's past documents on the terminal and clicks the upload button.

[1069] Step 2:

[1070] The terminal sends the selected material to the server as an HTTP request.

[1071] Step 3:

[1072] The server analyzes the received materials and stores them in a database, including metadata such as company ID and user ID.

[1073] Step 4:

[1074] The server retrieves all documents related to the company from the database, preprocesses the documents, extracts text, and cleans unnecessary data.

[1075] Step 5:

[1076] The server applies machine learning algorithms to the preprocessed materials to learn the company's specific tastes (formatting, terminology, style, etc.), and saves the learning results as an AI model.

[1077] Step 6:

[1078] The user inputs an outline of the material they want to generate into the terminal, for example, "New Product Launch Presentation," and clicks the send button.

[1079] Step 7:

[1080] The device sends the summary entered by the user to the server as an HTTP request.

[1081] Step 8:

[1082] The server passes the received summary to the emotion engine, which analyzes the emotion from the content entered by the user. For example, natural language processing technology is used to recognize positive emotion from the input "This new product is revolutionary!"

[1083] Step 9:

[1084] The server uses a trained AI model to generate materials based on the analysis results of the emotion engine. The generated materials include elements that reflect the user's emotions in addition to the company's unique style.

[1085] Step 10:

[1086] The server sends the generated document to the user's device in the format specified by the user, such as PDF or PPTX.

[1087] Step 11:

[1088] The device displays the received materials to the user, who is also provided with a link to download the materials if desired.

[1089] As a concrete example, if a company is preparing presentation materials for the launch of a new product, the user first uploads past materials, which the system then learns from. Next, the user enters a summary such as "New product launch presentation." If the user enters words containing strong emotions, such as "taking the market by storm," the emotion engine analyzes this and recognizes it as an emotion indicating excitement or anticipation. The server then reflects this emotion recognition result in the creation of the materials, generating presentation materials using bolder, more impactful designs and words based on the company's unique taste. As a result, materials that respond to the user's emotions are efficiently created while maintaining the company's consistent brand image.

[1090] Example 2

[1091] 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."

[1092] Companies are required to efficiently create materials while maintaining a consistent brand image. However, manually created materials often fail to reflect the company's unique style and require a great deal of time and effort. Furthermore, it is difficult to reflect the user's emotions and intentions in the materials, which can result in ineffective materials. To solve these issues, there is a need for a system that automatically generates materials that reflect the company's unique style and take the user's emotions into consideration.

[1093] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1094] In this invention, the server includes means for receiving past company documents and storing them in a database, means for acquiring documents from the database and applying a machine learning algorithm to learn the company's unique taste, means for automatically generating documents that reflect the company's unique taste using a trained machine learning model based on a document generation summary received from a user, an emotion engine that recognizes the user's emotions and reflects them in the document generation process, and means for outputting the generated documents to the user. This makes it possible to efficiently generate optimal documents that match the user's emotions and intentions while maintaining the company's unique taste.

[1095] "Corporate historical documents" refer to information assets such as documents, presentations, reports, and data sets that a company has created or held in the past.

[1096] A "database" is a collection of structured data for systematically organizing and managing information, and is a system that allows for various searches and operations to be performed efficiently.

[1097] A "machine learning algorithm" is a set of computational techniques that allow a computer to find patterns and rules in data and make predictions or classifications.

[1098] "Corporate taste" refers to the unique design, format, terminology, style, etc. that a company consistently uses, and is an element that forms the company's brand image.

[1099] A "trained machine learning model" is the output of a machine learning algorithm that has been trained using historical data and is ready to automate a specific task.

[1100] An "emotion engine" is a system that analyzes emotions from user input and text data and adjusts the tone and style of content based on the results.

[1101] "User emotion" refers to the psychological state, such as joy, expectation, excitement, sadness, etc., that the user shows when inputting the outline of the material generation.

[1102] "Natural language processing technology" is a technology that enables computers to understand, interpret, and generate human language, and is used for text analysis and sentiment analysis.

[1103] The "document generation overview" is a specific explanation and requirements such as the purpose, target, and necessary elements of the document the user wants to generate.

[1104] A "document generation process" refers to a series of steps or operations that automatically create documents based on input data.

[1105] This invention relates to a system that automatically generates documents that reflect a company's unique taste, as well as a system that recognizes the user's emotions and reflects the results in the document generation. This system is realized by the cooperation of the server, terminals, and users.

[1106] First, it is equipped with a means to receive a company's past documents and store them in a database. When a user uploads a company's past documents using their own device, the device sends the documents to a server. Once sent, the server receives the documents and stores them in a database such as MySQL. This step allows data to be collected to learn the unique tastes of each company.

[1107] The system then retrieves documents stored in the database and applies machine learning algorithms to learn the company's unique tastes. The server retrieves the documents from the database and performs preprocessing. This preprocessing includes extracting text information using OCR technology and deleting unnecessary data. After preprocessing is complete, the documents are analyzed using machine learning libraries such as Scikit-learn and TensorFlow to learn parameters that characterize the company's unique tastes, such as formatting, design, and terminology selection.

[1108] Furthermore, it is equipped with a means to automatically generate materials that reflect the company's unique taste using a trained machine learning model based on the outline of the materials received from the user. The user inputs the outline of the materials they want to generate into the terminal and sends that information to the server. The server generates the materials based on the received outline using a trained AI model such as BERT or GPT-3. The generated materials reflect the company's unique taste and maintain consistency. Below are some example prompts:

[1109] Example prompt sentence:

[1110] "Please create a presentation for the launch of a new product. This product is expected to take the market by storm. The presentation should be visually impactful and the design should be modern and dynamic."

[1111] In addition, the system is equipped with an emotion engine that recognizes user emotions. The emotion engine is responsible for analyzing emotions from user input. For example, when a user enters an outline for document generation, the input is analyzed for emotion using natural language processing technology (e.g., AWS Comprehend or Google Cloud Natural Language). The results of this emotion analysis can be reflected in the document generation process to adjust the tone and style of the generated document. For example, if a user enters words that contain strong emotions, such as "This new product will take the market by storm," the emotion engine will analyze it and recognize it as an emotion indicating excitement or anticipation. The server then reflects the emotion recognition results in the document generation, using bolder, more impactful designs and words.

[1112] Finally, a means for outputting the generated materials to the user is provided. The server transmits the generated materials to the user's terminal, and the terminal displays the received materials to the user. The user can download the materials as needed.

[1113] This invention enables companies to efficiently generate materials that respond to users' emotions and intentions while maintaining a consistent brand image, which significantly improves the efficiency of document creation work and allows companies to provide materials that strongly reflect the company's unique character.

[1114] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1115] Step 1:

[1116] Input: The user uploads the company's past documents.

[1117] How it works: A user uses their device to upload past company documents (e.g., PDF or Word files) through the interface. The device retrieves these documents and selects the documents to be uploaded through the file selection function.

[1118] Output: The terminal sends the selected material to the server as an HTTP POST request.

[1119] Step 2:

[1120] Input: The data file sent from the terminal.

[1121] Operation: The server receives the file and saves it in a database such as MySQL. The received file is stored in the specified location, and metadata (e.g., upload date and time, user ID) is also recorded.

[1122] Output: Material files and metadata stored in a database.

[1123] Step 3:

[1124] Input: Materials stored in the database.

[1125] How it works: The server periodically or on demand retrieves historical data from the database and performs pre-processing, which involves extracting text from images using OCR technology and removing or correcting unnecessary data (e.g., white space, advertisements).

[1126] Output: The preprocessed material with extracted text information.

[1127] Step 4:

[1128] Input: Preprocessed material.

[1129] How it works: The server uses machine learning algorithms (such as Scikit-learn or TensorFlow) to analyze the pre-processed materials, using them as a training dataset to learn parameters that characterize the company's specific formatting, design, and terminology choices.

[1130] Output: A machine learning model that reflects your company's unique taste.

[1131] Step 5:

[1132] Input: A summary of the generated material entered by the user into the terminal.

[1133] Operation: The user inputs a detailed outline of the material creation into the terminal and sends the information to the server.

[1134] Output: Summary information of the document generation is sent to the server.

[1135] Step 6:

[1136] Input: Summary information for material generation.

[1137] How it works: The server uses trained machine learning models (such as BERT or GPT-3) to automatically create documents based on the summary information. The server then checks whether the generated documents reflect the company's unique taste and makes any necessary adjustments.

[1138] Output: The generated material.

[1139] Step 7:

[1140] Input: A summary of what the user has entered and what materials are generated.

[1141] How it works: The server uses an emotion engine to parse the emotion from the user's input. It uses natural language processing technology (such as AWS Comprehend or Google Cloud Natural Language) to identify the emotion in the input and adjust the tone and style of the material.

[1142] Output: Sentiment analysis results and generated materials adjusted to reflect them.

[1143] Step 8:

[1144] Input: Final adjusted generated material.

[1145] How it works: The server sends the final generated material to the user's device. The material is provided via email or a dedicated download link.

[1146] Output: The generated material sent to the user's device.

[1147] Step 9:

[1148] Input: Generated material sent from the server.

[1149] Operation: The device displays the data received from the server, and the user can review it. If necessary, the data can be downloaded and saved.

[1150] Output: The final generated material displayed on the user's terminal.

[1151] (Application example 2)

[1152] 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."

[1153] When companies create materials such as advertisements, it is extremely difficult to tailor the materials to reflect user emotions and intentions while maintaining a consistent brand image. Furthermore, creating advertisements manually takes time and effort, making efficient operation difficult. Furthermore, there is no method for quickly generating advertisements that reflect user emotions. This leaves companies unable to create effective advertisements that match the emotions of target users, limiting the effectiveness of their advertisements.

[1154] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1155] In this invention, the server includes: means for receiving a company's past materials and storing them in a database; means for acquiring materials from the database and using the materials to apply a machine learning algorithm to learn the company's unique tastes; means for automatically generating materials that reflect the company's unique tastes and the user's emotions using the trained machine learning model based on a material generation summary received from the user; means for analyzing the user's emotions, including an emotion analysis engine for analyzing the user's emotions, and means for outputting the generated materials to the user. This enables a company to quickly and efficiently generate effective advertising materials that reflect the user's emotions while maintaining a consistent brand image.

[1156] "Corporate historical documents" refers to data such as documents, reports, presentation materials, advertising materials, and other documents of any format that a company has created, published, or stored in the past.

[1157] A "database" is a system that organizes and stores collected data, enabling efficient searches and analysis.

[1158] A "machine learning algorithm" is a technology that automatically learns patterns and rules from data and makes predictions and analyses based on that learned knowledge.

[1159] "A company's unique taste" refers to the brand image, style, and design concept that a company aims for, and is a general term for elements that have characteristics unique to that company.

[1160] "User emotion" refers to the emotional state expressed by the user during the process of creating materials, and includes various emotions such as joy, excitement, anticipation, and anxiety.

[1161] An "emotion analysis engine" is a system that analyzes the emotions from user input and identifies specific emotional states.

[1162] The "document generation overview" is a description showing basic information, purpose, theme, and specifications of the document that the user wants to create.

[1163] "Natural language processing technology" is a technology that allows computers to understand and generate human language, and involves analyzing and generating text.

[1164] A "machine learning model" is an AI model that is trained based on data to understand certain patterns and trends and generate responses.

[1165] "Means for outputting to the user" refers to a method or system for providing the generated materials to the user, and includes functions such as display, download, and sharing.

[1166] The present invention is a system that automatically generates advertising materials by analyzing user emotions and reflecting the unique tastes of a company. Specific embodiments for carrying out the present invention will be described below.

[1167] System configuration

[1168] The system mainly consists of a server, the user's smartphone, a database, a machine learning algorithm, a natural language processing engine, and a sentiment analysis engine.

[1169] Hardware

[1170] Server: Receives, stores, analyzes data, generates advertising materials, and outputs them.

[1171] Smartphone: The user inputs the outline of the material generation and receives the generated advertising material.

[1172] software

[1173] Machine learning libraries: TensorFlow and PyTorch are used to learn company-specific tastes.

[1174] Natural language processing engine: spaCy, which uses BERT to analyze user-input text.

[1175] Database: MySQL and MongoDB are used to store corporate historical data.

[1176] Sentiment analysis engine: Analyzes user emotions in real time.

[1177] Process Overview

[1178] 1. Data Collection:

[1179] The smartphone uploads the company's past data (such as advertising data) to the server, which receives the data and stores it in a database.

[1180] 2. Training the machine learning model:

[1181] We retrieve past advertising data stored in a database and preprocess the text information. After text extraction, cleaning, and removing or correcting unnecessary data, we use TensorFlow and PyTorch to train a machine learning model that learns the unique tastes of each company.

[1182] 3. Emotion analysis:

[1183] When a user inputs a summary of the ad they want to generate, the input is analyzed in real time using a natural language processing engine (spaCy, BERT), and the emotional state is identified using a sentiment analysis engine.

[1184] 4. Automatic generation of advertising materials:

[1185] Based on the outline of the materials generated by the user and the results of sentiment analysis, advertising materials are automatically generated using a trained machine learning model. The generated advertisements reflect the company's unique taste and the user's emotions.

[1186] 5. Output to the user:

[1187] The generated advertising materials are sent from the server to the user's smartphone, where the user can preview, edit, and download the generated advertisements.

[1188] Specific examples

[1189] For example, if a user inputs the summary "I want to create a campaign ad for a new eco-product. I want to convey my passion for protecting the earth," the emotion "passion for protecting the earth" will be recognized as a positive and strong emotion from this summary. Below is an example of a prompt to input to the generative AI model:

[1190] Prompt: "We want to create an ad campaign for a new eco-friendly product. We want to convey our passion for protecting the planet. The ad should reflect our company's unique eco-friendly image and have a passionate, positive tone."

[1191] Based on this prompt, the generative AI model automatically creates an ad with a passionate and positive tone that reflects the company's unique taste.

[1192] conclusion

[1193] This system enables companies to efficiently create advertisements that match the emotions of target users while maintaining a consistent brand image. A major feature of this invention is that it combines highly accurate emotion analysis technology with machine learning models to generate effective advertisements that match user needs.

[1194] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1195] Step 1:

[1196] Receive past company documents and store them in a database.

[1197] Input: Upload past company documents (advertising data) from your smartphone.

[1198] Operation: The device sends the data to the server.

[1199] Data processing: The server organizes the received data and stores it in a database.

[1200] Output: Past company information stored in the database.

[1201] Step 2:

[1202] It retrieves data from a database and uses it to apply machine learning algorithms to learn the unique tastes of each company.

[1203] Input: Past advertising materials stored in the database.

[1204] How it works: The server retrieves material from the database and pre-processes it (text extraction, cleaning, removing and modifying unnecessary data).

[1205] Data calculation: Using preprocessed data, machine learning models such as TensorFlow and PyTorch are used to learn the company's specific tastes.

[1206] Output: A trained machine learning model.

[1207] Step 3:

[1208] The user inputs an outline of the document to be generated, and the emotion analysis engine analyzes the user's emotions.

[1209] Input: An overview of the generated materials entered by the user on their smartphone.

[1210] How it works: A natural language processing engine (spaCy, BERT) analyzes user-entered text. A sentiment analysis engine identifies emotional states.

[1211] Data calculation: Identify the user's emotional state based on the emotion analysis results.

[1212] Output: Emotion analysis results.

[1213] Step 4:

[1214] Advertising materials are automatically generated based on the outline of material generation received from the user and the results of sentiment analysis.

[1215] Input: Summary of material generation and sentiment analysis results.

[1216] How it works: The server uses trained machine learning models to automatically generate advertising materials based on your input.

[1217] Data calculation: Generate advertising materials that reflect the company's unique taste and user emotions.

[1218] Output: Generated advertising materials.

[1219] Step 5:

[1220] The generated advertising material is output to the user.

[1221] Input: Generated advertising materials.

[1222] Operation: The server sends the generated materials to the user's smartphone.

[1223] Output: Advertising materials displayed on the user's smartphone, which the user can preview, edit, and download.

[1224] 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.

[1225] 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.

[1226] 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.

[1227] 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.

[1228] 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.

[1229] 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.

[1230] 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).

[1231] 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.

[1232] 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."

[1233] 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.

[1234] 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).

[1235] 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.

[1236] 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.

[1237] 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.

[1238] 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.

[1239] 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.

[1240] 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.

[1241] 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.

[1242] 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.

[1243] 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.

[1244] 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.

[1245] The following is further disclosed regarding the above embodiment.

[1246] (Claim 1)

[1247] A means of receiving past company documents and storing them in a database;

[1248] A means for acquiring data from the database, applying a machine learning algorithm using the data, and learning the tastes specific to the company;

[1249] A means for automatically generating materials that reflect the tastes specific to the company using the trained machine learning model based on an outline of the materials generation received from the user;

[1250] means for outputting the generated materials to a user;

[1251] A system including:

[1252] (Claim 2)

[1253] 10. The system of claim 1, wherein the machine learning algorithm uses natural language processing techniques to learn company-specific terminology and formats.

[1254] (Claim 3)

[1255] 10. The system of claim 1, further comprising means for text extraction and cleaning of the material to analyze corporate taste.

[1256]

[1257] "Example 1"

[1258] (Claim 1)

[1259] A means of receiving past company documents and storing them in a database;

[1260] A means for acquiring data from the database, applying a machine learning algorithm using the data, and learning the tastes specific to the company;

[1261] A means for automatically generating materials that reflect the tastes specific to the company using the trained machine learning model based on an outline of the materials generation received from the user;

[1262] A means of preprocessing to extract text information, and deleting and correcting unnecessary data;

[1263] means for outputting the generated materials to a user;

[1264] A system including:

[1265] (Claim 2)

[1266] 10. The system of claim 1, wherein the machine learning algorithm uses natural language processing techniques to learn company-specific terminology and formats.

[1267] (Claim 3)

[1268] 10. The system of claim 1, further comprising means for text extraction and cleaning of the material to analyze corporate taste.

[1269] "Application Example 1"

[1270] (Claim 1)

[1271] a means for receiving and storing historical information about the company in a database;

[1272] a means for acquiring information from the database and using the information to apply a machine learning algorithm to learn the tastes specific to the company;

[1273] A means for automatically generating information that reflects the tastes specific to a company using the trained machine learning model based on an outline of information generation received from a user;

[1274] means for outputting the generated information to a user;

[1275] A means of extracting customer questions and needs in real time using voice recognition and providing the most appropriate information based on that context;

[1276] A means for displaying information in real time to a smart device;

[1277] A system including:

[1278] (Claim 2)

[1279] 10. The system of claim 1, wherein the machine learning algorithm uses natural language processing techniques to learn company-specific terminology and formats.

[1280] (Claim 3)

[1281] 10. The system of claim 1, further comprising means for text extraction and cleaning of information for analyzing company specific tastes.

[1282] "Example 2: Combining Emotion Engines"

[1283] (Claim 1)

[1284] A means of receiving past company documents and storing them in a database;

[1285] A means for acquiring data from the database, applying a machine learning algorithm using the data, and learning the tastes specific to the company;

[1286] A means for automatically generating materials that reflect the tastes specific to the company using the trained machine learning model based on an outline of the materials generation received from the user;

[1287] An emotion engine that recognizes the user's emotions and reflects them in the document generation process;

[1288] means for outputting the generated materials to a user;

[1289] A system including:

[1290] (Claim 2)

[1291] 10. The system of claim 1, wherein the machine learning algorithm uses natural language processing techniques to learn company-specific terminology and formats.

[1292] (Claim 3)

[1293] 10. The system of claim 1, further comprising means for text extraction and cleaning of the material to analyze corporate taste.

[1294] (Claim 4)

[1295] 10. The system of claim 1, wherein the system analyzes emotions from user input using natural language processing technology and adjusts the tone and style of the document based on the emotion analysis results.

[1296] (Claim 5)

[1297] The system of claim 1, wherein the emotion engine reflects the results of emotion analysis in document generation, and uses bolder and more impactful designs and words when emotions such as excitement or anticipation are expressed.

[1298] "Application example 2 when combining emotion engines"

[1299] (Claim 1)

[1300] A means of receiving past company documents and storing them in a database;

[1301] A means for acquiring data from the database, applying a machine learning algorithm using the data, and learning the tastes specific to the company;

[1302] A means for automatically generating materials that reflect the company's unique taste and the user's emotions using the trained machine learning model based on the outline of the materials generated received from the user;

[1303] a means for analyzing emotions from user input, the means comprising an emotion analysis engine for analyzing user emotions;

[1304] means for outputting the generated materials to a user;

[1305] A system including:

[1306] (Claim 2)

[1307] 10. The system of claim 1, wherein the machine learning algorithm uses natural language processing techniques to learn company-specific terminology and formats.

[1308] (Claim 3)

[1309] 10. The system of claim 1, further comprising means for text extraction and cleaning of the material to analyze corporate taste.

[1310] (Claim 4)

[1311] 10. The system of claim 1, further comprising means for adjusting the style and tone of the material based on the results of the user's sentiment analysis. [Explanation of symbols]

[1312] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of receiving past company documents and storing them in a database; A means for acquiring data from the database, applying a machine learning algorithm using the data, and learning the tastes specific to the company; A means for automatically generating materials that reflect the tastes specific to the company using the trained machine learning model based on an outline of the materials generation received from the user; means for outputting the generated materials to a user; A system including:

2. The system of claim 1 , wherein the machine learning algorithm uses natural language processing techniques to learn company-specific terminology and formats.

3. 10. The system of claim 1, further comprising means for text extraction and cleaning of the material to analyze corporate taste.

Images