system

Abstract

Provide a system. [Solution] a means for collecting conversation data; A means of analyzing collected conversation data to identify frequently used words and phrases, and A means for generating character string data based on the identified frequently occurring words and phrases; means for generating an illustration corresponding to character string data; A means for combining character string data and illustrations to create an emoji set for a communication app; A system including:

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 modern communication, there is no way to automatically generate emojis and stickers that meet individual needs based on frequently used words and phrases. This forces users to rely on ready-made emojis and stickers, limiting the potential for making their communications more personalized and engaging. Furthermore, manually creating customized stickers is labor-intensive and time-consuming, making it difficult for average users to achieve, depriving them of the opportunity to enjoy personalized communication. There is a need to address these issues. [Means for solving the problem]

[0005] The present invention provides a system that collects conversation data, analyzes the collected conversation data, and identifies frequently occurring words and phrases. It also includes a means for generating character string data based on the identified frequently occurring words and phrases, and automatically generating illustrations corresponding to the character string data. This illustration generation uses a machine learning model, and can create emoji sets for communication apps by combining the character string data and illustrations. This allows users to easily obtain original emoji sets based on their own conversation content, without relying on existing stamps.

[0006] "Conversation data" refers to information about text messages and voice messages sent by users using communication apps.

[0007] "Analysis" refers to the processing steps used to find specific patterns or characteristics in collected data.

[0008] "Frequent words" and "phrases" refer to words or sequences of words that appear particularly frequently in the conversation data being analyzed.

[0009] "String data" refers to text information corresponding to frequently occurring words and phrases.

[0010] "Illustration generation" refers to the process of creating graphical elements based on string data.

[0011] A "machine learning model" refers to an algorithm or system that generates or predicts new data based on past data.

[0012] An "emoji set for communication apps" refers to a collection of emojis and stamps that are compiled in a format that can be used in communication applications.

[0013] "Natural language processing" refers to the technology of using computers to understand, interpret, and generate human language.

[0014] "Tokenization" refers to the process of breaking down text data into components (words and phrases).

[0015] "Stop words" refer to frequently occurring but meaningless words that are ignored during analysis (e.g., kedo, however, etc.). [Brief explanation of the drawings]

[0016] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13]FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0017] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

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

[0019] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0020] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0021] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0022] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0023] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0024] [First embodiment]

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

[0026] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0027] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0028] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0029] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0030] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0031] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

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

[0033] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0034] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0035] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0036] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0037] The invention described in this specification relates to a system that collects and analyzes conversation data from users in a communication app to identify frequently occurring words and phrases, and automatically generates an emoji set based on the collected data. This system includes a user terminal, a server, and a communication means for transmitting and receiving data between them.

[0038] Data collection

[0039] The device collects the user's conversation data, including text and voice messages sent and received within the communication app. The device periodically retrieves this conversation data and stores it locally.

[0040] Examples:

[0041] When a user chats with a friend, they exchange messages such as "Good work," "Thank you," and "What are you having for lunch today?" This conversation data is stored on the device.

[0042] Data transmission

[0043] The device sends the saved conversation data to the server, which then converts the conversation data into a database format and uploads it to the server via a network connection.

[0044] Examples:

[0045] The device compiles conversation logs from over a month, converts them into a database format, and sends them to the server.

[0046] Data analysis

[0047] The server analyzes the received conversation data to identify frequently occurring words and phrases, and then uses natural language processing technology to analyze the text, tokenize it, remove stop words, and extract frequently occurring words.

[0048] Examples:

[0049] The server analyzes the conversation data and extracts frequently used phrases such as "good work," "thank you," and "lunch."

[0050] Generating string data and illustrations

[0051] The server generates character strings based on the frequently occurring words and phrases identified, and automatically generates illustrations corresponding to the character strings. The illustration generation utilizes machine learning models.

[0052] Examples:

[0053] The server generates the string data "Thank you for your hard work" and uses a machine learning model to generate a matching illustration of a cat character.

[0054] Creating an emoji set

[0055] The server combines the generated character string data with illustrations to create an emoji set for the communication app, which is then sent to the user's device.

[0056] Examples:

[0057] The server creates a stamp set that combines character strings such as "Good work," "Thank you," and "Lunch" with corresponding illustrations, and provides it to the user.

[0058] Use of stamps

[0059] Users can download the completed emoji set and use it in their communication apps, allowing them to use the new stickers to make their communications more personal and engaging.

[0060] Examples:

[0061] Users can choose a "Thank you for your hard work" stamp from the stamp set they downloaded and send it to a friend to express their gratitude.

[0062] The system allows users to easily create their own original emoji sets based on frequently used phrases and words, making everyday communication more personal and fun.

[0063] The processing flow will be explained below.

[0064] Step 1:

[0065] The user enables the conversation log collection function in the communication app, which starts the collection of conversation data.

[0066] Step 2:

[0067] The device periodically retrieves and locally stores the user's conversation data, which includes text messages, voice messages, and other communication information.

[0068] Step 3:

[0069] The device periodically sends the saved conversation data to the server, where it is converted into a database format and uploaded via a network connection.

[0070] Step 4:

[0071] The server starts processing to analyze the received conversation data. The server performs text analysis using a natural language processing library.

[0072] Step 5:

[0073] The server tokenizes the conversation data and removes stop words, making it easier to analyze.

[0074] Step 6:

[0075] The server uses the tokenized data to calculate frequency and identify frequently occurring words and phrases.

[0076] Step 7:

[0077] The server generates string data based on frequently occurring words and phrases. For example, frequently occurring phrases such as "Thank you for your hard work" are extracted.

[0078] Step 8:

[0079] The server uses machine learning models to automatically generate illustrations corresponding to frequently occurring words and phrases, and combines the generated illustrations with text data.

[0080] Step 9:

[0081] The server combines multiple illustrations and text data to create an emoji set for the communication app. Each emoji contains text data and a corresponding illustration.

[0082] Step 10:

[0083] The server sends the completed emoji set to the user's device, where the user can download it within the communication app.

[0084] Step 11:

[0085] Users can download and install the emoji set and use it within their communication apps. The new emoji set can be used to make communications more personal and engaging.

[0086] The above are the specific processing steps of the system program. Through this process, users can easily obtain and use original emoji sets based on their own conversation content.

[0087] Example 1

[0088] 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."

[0089] In recent years, the use of communication apps has increased, and users are sending and receiving many messages. However, existing emoji and stamp sets are fixed and do not correspond to the individual conversation content of users. Therefore, there is a need for personalized and more engaging means of communication. In particular, there is a need for a method to easily generate emoji that correspond to frequently used phrases and words.

[0090] 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.

[0091] In this invention, the server includes means for analyzing conversation data to identify frequently occurring words and phrases, means for generating character string data based on the identified frequently occurring words and phrases, and means for using a machine learning model to generate illustrations corresponding to the generated character string data. This allows users to easily create original emoji sets based on the content of their own conversations, making everyday communication more personalized and enjoyable.

[0092] A "terminal" is an electronic device for executing a communication app used by a user, and includes a smartphone, tablet, PC, etc.

[0093] A "server" is a central computer system that receives, processes, and stores data sent from terminals.

[0094] "Conversation data" is a data set that includes text and voice messages sent and received by a user within a communication app.

[0095] A "database format" is a format for storing structured data, and generally includes formats such as JSON and CSV.

[0096] "Natural language processing technology" is a technology that enables computers to understand, interpret, and generate natural language, and includes text tokenization, stop word removal, and frequent word extraction.

[0097] "Tokenization" is the process of breaking text into individual words or phrases.

[0098] "Stop words" are frequently used words that contain little information and are excluded from analysis (for example, "no," "ni," and "o" in Japanese).

[0099] "Frequent words" are words or phrases that are used particularly frequently in conversation data.

[0100] A "machine learning model" is an algorithm that learns patterns and rules based on data and makes predictions or generates data based on new data.

[0101] A "prompt" is an instruction sentence input to a generative AI model, containing instructions for generating illustrations or text based on specific conditions or requests.

[0102] An "emoji set" is a collection of emojis and / or stickers for use by a user within a communication app.

[0103] The present invention relates to a system that collects and analyzes conversation data from users in a communication app to identify frequently occurring words and phrases, and automatically generates emoji sets based on the collected data. The system includes a user terminal, a server, and a communication means for transmitting and receiving data between them.

[0104] Data collection

[0105] First, the device collects the user's conversation data. This conversation data includes text messages and voice messages sent and received by the user within the communication app. The device periodically obtains the conversation data using the communication app's API and saves it in local storage. Specifically, the device calls the communication app's API at 2:00 AM every day to obtain the previous day's conversation data. This data includes messages such as "Good work," "Thank you," and "What are you having for lunch today?"

[0106] Data transmission

[0107] Next, the device sends the saved conversation data to the server. The device converts the conversation data into a database format, which is typically JSON or CSV. For example, the device converts the saved conversation data at the end of the month into JSON format, and represents each message as a JSON object. The converted data is then uploaded to the server using the HTTPS protocol.

[0108] Data analysis

[0109] The server receives the data from the device and stores it in a database for analysis. The server uses natural language processing techniques such as Python's NLTK library to tokenize the conversation data, remove stop words, and extract frequently occurring words. For example, it extracts frequently occurring phrases such as "good work," "thank you," and "lunch."

[0110] Generating string data and illustrations

[0111] The server generates character string data based on the frequently occurring words and phrases identified, and automatically generates corresponding illustrations using a machine learning model. At this time, a prompt sentence is input into the generative AI model to generate the illustration. Specifically, the prompt sentence "Please generate an illustration of a cat character saying 'Good job'" is used to generate the corresponding illustration.

[0112] Creating an emoji set

[0113] The server combines the character string data with the generated illustrations to create an emoji set for the communication app. This emoji set is then sent to the user's device. For example, a stamp set is created by combining the character strings "Good work," "Thank you," and "Lunch" with their corresponding illustrations, and sent to the device using the HTTPS protocol.

[0114] Use of stamps

[0115] The user downloads the completed emoji set and uses it in the communication app. The user clicks on the new stamp set to download it, which is then saved in the specified folder in the communication app. For example, a user can select the "Thank you for your hard work" stamp on the chat screen and send it to a friend to express their gratitude.

[0116] In this way, the system allows users to easily create original emoji sets based on frequently used phrases and words, making everyday communication more personal and fun.

[0117] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0118] Program processing steps

[0119] Step 1:

[0120] A user has a conversation within a communication app. This conversation data is collected as it is entered into the device. The device calls the communication app's API and saves the acquired data in local storage. This allows text messages and voice messages to be collected.

[0121] Specific behavior:

[0122] The device calls the communication app's API at 2:00 AM every day, obtains the previous day's conversation data (e.g., "Good work," "Thank you," "What are you having for lunch today?") and stores it in an SQLite database.

[0123] Input: User conversation data (text messages, voice messages)

[0124] Output: Locally stored conversation data

[0125] Step 2:

[0126] The conversation data collected by the device is converted into a database format (JSON). The converted data is then sent from the device to the server. This transmission process uses the HTTPS protocol.

[0127] Specific behavior:

[0128] At the end of the month, the device converts the saved conversation data into JSON format, representing each message as a JSON object, and then uploads it to the server using the HTTPS protocol.

[0129] Input: Locally stored conversation data

[0130] Output: JSON data uploaded to the server

[0131] Step 3:

[0132] The server receives the uploaded JSON data and stores it in a database for analysis. This data is then analyzed using natural language processing techniques to identify frequently occurring words and phrases. Specific analysis steps include tokenization, stop word removal, and frequent word extraction.

[0133] Specific behavior:

[0134] The server uses Python's NLTK library to tokenize the conversation data, remove stop words, count frequently occurring words, and extract frequently occurring phrases such as "good work," "thank you," and "lunch."

[0135] Input: JSON data uploaded to the server

[0136] Output: Identified frequently occurring words and phrases

[0137] Step 4:

[0138] The server generates text data based on the identified frequently occurring words and phrases, which is then used as prompts for the generative AI model to generate illustrations.

[0139] Specific behavior:

[0140] The server generates the phrases "Thank you for your hard work," "Thank you," and "Lunch" as string data and prepares them as a prompt sentence.

[0141] Input: Identified common words and phrases

[0142] Output: String data (prompt statement)

[0143] Step 5:

[0144] The server inputs prompts into the generative AI model, which then automatically generates corresponding illustrations that correspond to the user's conversation.

[0145] Specific behavior:

[0146] The server inputs the prompt "Please generate an illustration of a cat character saying 'Good job'" into the generative AI model and generates the corresponding illustration.

[0147] Input: prompt statement

[0148] Output: Generated illustration

[0149] Step 6:

[0150] The server combines the generated character string data with illustrations to create an emoji set for the communication app, which is then sent to the user's device.

[0151] Specific behavior:

[0152] The server creates a stamp set by combining the strings "Good work," "Thank you," and "Lunch" with corresponding illustrations, and sends it to the terminal using the HTTPS protocol.

[0153] Input: String data and generated illustrations

[0154] Output: The emoji set sent to the user's device

[0155] Step 7:

[0156] Users can download the completed emoji set to their device and use it in communication apps. Specifically, users communicate by selecting and sending new stamps.

[0157] Specific behavior:

[0158] Users can select the "Thank you for your hard work" stamp on the chat screen and send it to their friends to express their gratitude.

[0159] Input: The emoji set sent to the user's device

[0160] Output: Stamps used in communication apps

[0161] (Application example 1)

[0162] 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."

[0163] Conventional advertising methods have difficulty accurately grasping the hobbies and preferences of their target users and personalizing their advertising, which often results in unattractive ads being displayed to users, reducing the effectiveness of advertising and lowering marketing efficiency.

[0164] 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.

[0165] In this invention, the server includes means for collecting conversation data, means for analyzing the collected conversation data to identify frequently occurring words and phrases, means for automatically generating advertising content by combining character string data and illustrations based on the identified frequently occurring words and phrases, and means for displaying the generated advertising content, thereby enabling the generation of personalized advertisements based on user conversation data.

[0166] - "Conversation data" refers to text and voice message data sent and received by a user within a communication app.

[0167] The "analysis means" is a means used to analyze conversation data and identify frequently occurring words and phrases.

[0168] "Frequent words and phrases" are important words and phrases that appear with a certain frequency or more in the collected conversation data.

[0169] "Character string data" is text data generated based on identified frequently occurring words and phrases.

[0170] The "illustration generating means" is a means used to generate an illustration corresponding to character string data.

[0171] "Advertising content" is marketing content generated by combining character string data and illustrations.

[0172] The "generation means" is a means for generating character string data and illustrations based on the identified frequently occurring words and phrases.

[0173] The "display means" is a means for displaying the generated advertising content on a user terminal.

[0174] "Natural language processing technology" is a technology for tokenizing conversational data, removing stop words, and extracting frequently occurring words and phrases.

[0175] A "machine learning model" is an algorithm or model that is trained based on data and used to automatically generate illustrations.

[0176] This invention is a system that automatically generates advertising content by collecting conversation data, analyzing it to identify frequently occurring words and phrases, and generating character string data and illustrations. This system includes a user terminal, a server, and a communication means for transmitting and receiving data between them.

[0177] First, the user's device collects conversation data. This conversation data includes text and voice messages that the user sends daily within communication apps. For example, conversations such as "I want to watch this movie," "I've been wanting to get a new smartphone lately," and "What should we have for lunch today?" are collected. The collected data is temporarily stored on the device, then converted into a database format and sent to the server.

[0178] The server then analyzes the received conversation data using natural language processing (NLP) techniques, tokenizing the conversation data, removing stop words, and extracting frequently occurring words and phrases. For example, the server identifies frequently occurring phrases such as "movie," "smartphone," and "lunch."

[0179] The server then generates character strings based on the frequently occurring words and phrases identified, and generates corresponding illustrations. Machine learning models are used to generate the illustrations, so for example, an illustration of a character holding popcorn is automatically generated for "movie," and an illustration of a character holding a smartphone is automatically generated for "smartphone."

[0180] Next, the server combines the generated text data with illustrations to create advertising content. This advertising content is sent to the user's device, where the user can view it on their smartphone app. By using this new advertising content, users can view personalized ads based on their interests and hobbies. For example, ads such as "Special discount! Great chance to see movies at a great price!" or "New smartphones on sale now!" are displayed.

[0181] The system also automatically generates advertising content by inputting prompts like the following into a generative AI model:

[0182] "You will create a program that extracts frequently used phrases and keywords from user conversation data and generates personalized advertising content based on that data. The collected data will be sent to a server in JSON format, where it will be analyzed. Targeted ads will be generated based on the analysis results, and will ultimately be displayed on the smartphone app."

[0183] As a specific example of use, when a user says something like "I want to watch this movie," the system analyzes it and automatically generates an advertisement that says "Special discount! Movies on sale now!" as completely new content and provides it to the user.

[0184] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0185] Step 1:

[0186] The user's device collects conversation data. In this step, the target is text and voice messages sent and received by the user within the communication app. The input is data from the user's daily chats and messages, and the output is the collected conversation data that is temporarily stored on the device.

[0187] Step 2:

[0188] The conversation data collected by the device is converted into a database format. In this step, the collected text and voice messages are organized and converted into a database format. The input is the collected conversation data, and the output is the converted database format data.

[0189] Step 3:

[0190] The terminal sends the converted database format conversation data to the server. In this step, data is sent from the terminal to the server via the network. The input is the conversation data converted to database format, and the output is the data received on the server side.

[0191] Step 4:

[0192] The server analyzes the received conversation data. In this step, natural language processing techniques are used to tokenize the conversation data, remove stop words, and extract frequently occurring words and phrases. The input is the conversation data, and the output is frequently occurring words and phrases. The specific operation uses the Python sklearn library.

[0193] Step 5:

[0194] The server generates string data based on the identified frequently occurring words and phrases. In this step, text corresponding to the extracted words and phrases is generated. The input is the frequently occurring words and phrases, and the output is the generated string data.

[0195] Step 6:

[0196] The server generates an illustration corresponding to the string data. In this step, the corresponding illustration is automatically generated using a machine learning model. The input is the string data, and the generated illustration is obtained as the output. Specific operations use PyTorch and TENSORFLOW (registered trademark).

[0197] Step 7:

[0198] The server combines the generated text data with the illustration to create advertising content. In this step, marketing content is generated by combining the two. The input is the text data and the corresponding illustration, and the output is the completed advertising content.

[0199] Step 8:

[0200] The server transmits the generated advertisement content to the user terminal. In this step, the advertisement content is transmitted to the user terminal via the network. The advertisement content is the input, and the advertisement content received by the user terminal is obtained as the output.

[0201] Step 9:

[0202] The user checks the advertising content on the smartphone app. In this step, the user can view personalized advertising. The input is the received advertising content, and the output is the displayed advertising content. For example, a message such as "Special discount! A great chance to see a movie at a great price!" may be displayed.

[0203] 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.

[0204] The present invention relates to a system that automatically generates emoji sets by collecting and analyzing conversation data from users in a communication app, identifying frequently used words and phrases, and combining this with an emotion engine that recognizes the user's emotions. This system includes a user terminal, a server, and communication means for transmitting and receiving data between them.

[0205] Data collection

[0206] The device collects the user's conversation data, including text and voice messages sent and received within the communication app. The device periodically retrieves this conversation data and stores it locally.

[0207] Examples:

[0208] When a user chats with a friend, they exchange messages such as "Good work," "Thank you," and "What are you having for lunch today?" This conversation data is stored on the device.

[0209] Data transmission

[0210] The device periodically sends the saved conversation data to the server, where it is converted into a database format and uploaded via a network connection.

[0211] Examples:

[0212] The device compiles conversation logs from over a month, converts them into a database format, and sends them to the server.

[0213] Data analysis

[0214] The server analyzes the received conversation data to identify frequently occurring words and phrases, and then uses natural language processing technology to analyze the text, tokenize it, remove stop words, and extract frequently occurring words.

[0215] Examples:

[0216] The server analyzes the conversation data and extracts frequently used phrases such as "good work," "thank you," and "lunch."

[0217] Emotion recognition by emotion engine

[0218] The server further analyzes the conversation data to recognize the user's emotions. The emotion engine uses keywords and contextual analysis to identify the emotions the user is expressing during the conversation.

[0219] Examples:

[0220] The server analyzes phrases like "Otsukaresama" (thank you) and recognizes that it indicates a positive emotion, as well as "arigato" (thank you) as an emotion that conveys gratitude.

[0221] Generating string data and illustrations

[0222] The server generates character strings based on the frequently used words and phrases and the user's emotions, and automatically generates illustrations corresponding to the character strings. The illustration generation utilizes a machine learning model.

[0223] Examples:

[0224] The server associates a positive emotion with the string data "Thank you for your hard work" and generates a matching cat character illustration using a machine learning model.

[0225] Creating an emoji set

[0226] The server combines the generated text data with illustrations to create an emoji set for communication apps, which also includes emotional information.

[0227] Examples:

[0228] The server creates a stamp set that combines character strings such as "good work," "thank you," and "lunch" with corresponding illustrations, providing an original emoticon set that reflects the user's emotions.

[0229] Use of stamps

[0230] Users can download the completed emoji set and use it in their communication apps, allowing them to use the new stickers to communicate more personally and emotionally.

[0231] Examples:

[0232] Users can choose a "Thank you for your hard work" stamp from the stamp set they downloaded and send it to a friend to express their gratitude.

[0233] This system allows users to easily obtain original emoji sets based on their own conversation content and emotions, making everyday communication more personal and emotional.

[0234] The processing flow will be explained below.

[0235] Step 1:

[0236] The user enables the conversation log collection function in the communication app, which starts the collection of conversation data.

[0237] Step 2:

[0238] The device periodically retrieves and locally stores the user's conversation data, which includes text messages, voice messages, and other communication information.

[0239] Step 3:

[0240] The device periodically sends the saved conversation data to the server, where it is converted into a database format and uploaded via a network connection.

[0241] Step 4:

[0242] The server starts processing to analyze the received conversation data. The server performs text analysis using a natural language processing library.

[0243] Step 5:

[0244] The server tokenizes the conversation data and removes stop words, making it easier to analyze.

[0245] Step 6:

[0246] The server uses the tokenized data to calculate frequency and identify frequently occurring words and phrases.

[0247] Step 7:

[0248] The server further analyzes the conversation data and uses an emotion engine to recognize the user's emotions. The emotion engine uses keywords and contextual analysis to identify the user's emotions.

[0249] Step 8:

[0250] The server generates string data based on the identified frequently occurring words and phrases and the recognized emotions.

[0251] Step 9:

[0252] The server uses machine learning models to automatically generate illustrations that correspond to frequently used words and phrases and the user's emotions, and combines the generated illustrations with text data.

[0253] Step 10:

[0254] The server combines multiple illustrations and text data to create an emoji set for the communication app. Each emoji contains text data and a corresponding illustration.

[0255] Step 11:

[0256] The server sends the completed emoji set to the user's device, where the user can download it within the communication app.

[0257] Step 12:

[0258] Users can download and install the emoji set and use it within their communication app. The new emoji set can be used to make communication more personal and emotional.

[0259] Examples:

[0260] When chatting with a friend, a user sends and receives conversations such as "Good work," "Thank you," and "What are we having for lunch today?" This conversation data is saved on the device and sent to a server. The server analyzes the data using natural language processing and an emotion engine, and automatically generates phrases that express positive emotions, such as "Good work," and corresponding illustrations. As a result, the user can download and use stamp sets that include character strings such as "Good work," "Thank you," and "Lunch," along with the corresponding illustrations.

[0261] Example 2

[0262] 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."

[0263] In conventional communication applications, the emoji sets used by users are fixed and do not reflect the user's actual conversation content or emotions. This limits personalized and emotional communication, making it difficult for users to accurately express themselves. Furthermore, the limited selection of existing emojis leads to a standardized user experience and a lack of individuality.

[0264] 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.

[0265] In this invention, the server includes means for collecting conversation data, means for analyzing the conversation data to identify frequently occurring words and phrases, means for analyzing the conversation data to recognize the user's emotions, means for generating character string data based on the identified frequently occurring words and phrases and the recognized emotions, means for generating illustrations corresponding to the character string data, and means for creating an emoji set for a communication application by combining the character string data and illustrations. This enables users to easily obtain an original emoji set based on the content of their own conversations and emotions, making everyday communication more personalized and emotionally rich.

[0266] "Conversation Data" is data including text messages, voice messages, and other communications sent and received by a user within a communication application.

[0267] "Means of collection" refers to the methods and functions by which a user's terminal or system collects conversation data automatically or at the user's direction.

[0268] "Means of analysis" refers to methods and technologies for analyzing conversation data and extracting meaning and information. Specifically, this includes analysis using natural language processing technology.

[0269] "Frequent words and phrases" refer to words and expressions that appear particularly frequently in the collected conversational data.

[0270] "Means for recognizing emotions" refers to methods and technologies for analyzing conversation data and identifying user emotions, including emotion engines and context analysis.

[0271] "String data" refers to text-based information generated from analyzed conversation data.

[0272] "Means for generating illustrations" refers to methods and technologies for automatically creating visual images based on text data and emotional information. Specifically, they use machine learning models.

[0273] "Pictogram set" refers to a collection of graphical representations for communication applications created by combining specified character string data and corresponding illustrations.

[0274] MODE FOR CARRYING OUT THE INVENTION

[0275] The present invention relates to a system that automatically generates emoji sets by collecting and analyzing conversation data from users in a communication app, identifying frequently used words and phrases, and combining this with an emotion engine that recognizes the user's emotions. This system includes a user terminal, a server, and communication means for transmitting and receiving data between them.

[0276] Data collection

[0277] The device collects the user's conversation data, including text and voice messages sent and received within the communication app. The device periodically retrieves this conversation data and stores it locally.

[0278] Examples:

[0279] For example, when a user chats with a friend and exchanges messages such as "Good work," "Thank you," and "What are you having for lunch today?", this conversation data is saved on the device.

[0280] Data transmission

[0281] The device periodically sends the saved conversation data to the server, where it is converted into a database format and uploaded via a network connection.

[0282] Examples:

[0283] The device compiles conversation logs over a month, converts them into a database format, and sends the data to a server using a secure protocol such as SSL / TLS.

[0284] Data analysis

[0285] The server analyzes the received conversation data to identify frequently occurring words and phrases, and uses natural language processing (NLP) technology to tokenize the text, remove stop words, and then extract frequently occurring words.

[0286] Examples:

[0287] The server tokenizes the conversation data using a morphological analysis engine and extracts frequently occurring phrases such as "good work," "thank you," and "lunch."

[0288] Emotion recognition by emotion engine

[0289] The server further analyzes the conversation data to recognize the user's emotions. The emotion engine uses keywords and contextual analysis to identify the emotions the user is showing during the conversation.

[0290] Examples:

[0291] The server analyzes phrases like "Otsukaresama" (thank you) and recognizes that it indicates a positive emotion, as well as "arigato" (thank you) as an emotion that conveys gratitude.

[0292] Generating string data and illustrations

[0293] The server generates text data based on the frequently used words and phrases and the user's emotions, and then automatically generates illustrations corresponding to the text data. The illustration generation utilizes machine learning models (such as GAN and VQ-VAE).

[0294] Examples:

[0295] The server associates a positive emotion with the string data "good work" and generates an illustration of a cat character.

[0296] Creating an emoji set

[0297] The server combines the generated text data with illustrations to create an emoji set for communication apps, which also includes emotional information.

[0298] Examples:

[0299] The server creates a stamp set that combines character strings such as "good work," "thank you," and "lunch" with corresponding illustrations, providing an original emoticon set that reflects the user's emotions.

[0300] Use of stamps

[0301] Users can download the completed emoji set and use it in their communication apps, allowing them to communicate in a personalized and emotional way using the new stickers.

[0302] Examples:

[0303] Users can choose a "Thank you for your hard work" stamp from the stamp set they downloaded and send it to a friend to express their gratitude.

[0304] This allows users to easily obtain original emoji sets based on their own conversation content and emotions, making everyday communication more personal and emotional.

[0305] Examples of prompt statements

[0306] An example of a prompt for a generative AI model is:

[0307] Generate illustrations of stamps that reflect phrases frequently used by users in communication apps, such as "Thank you" and "Thank you," and the emotions associated with them (positive, grateful).

[0308] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0309] Step 1:

[0310] The terminal collects user conversation data.

[0311] Input: Text and voice messages that users send and receive within communication apps.

[0312] How it works: A program runs to periodically save conversation data to the device's local storage, and the user sends and receives messages such as "Good work," "Thank you," and "What are you having for lunch today?" in chats with friends.

[0313] Output: Conversation data stored in the device's local storage.

[0314] Step 2:

[0315] The conversation data stored on the device is periodically sent to the server.

[0316] Input: Conversation data stored in local storage.

[0317] How it works: Data is first converted into a database format and then uploaded to the server over a network connection using SSL / TLS protocol.

[0318] Output: Conversation data uploaded to the server.

[0319] Step 3:

[0320] The server analyzes the conversation data it receives and identifies frequently used words and phrases.

[0321] Input: Conversation data uploaded to the server.

[0322] How it works: It uses natural language processing techniques to tokenize text data and remove stop words, then uses a morphological analysis engine to extract frequently occurring words and phrases.

[0323] Output: A list of frequently occurring words and phrases.

[0324] Step 4:

[0325] The server further analyzes the conversation data and recognizes the user's emotions.

[0326] Input: A list of frequently used words and phrases.

[0327] How it works: Uses a sentiment engine to identify user sentiment based on keywords and contextual analysis, for example recognizing positive sentiment from the phrase "Otsukaresama."

[0328] Output: Emotion recognition result.

[0329] Step 5:

[0330] The server generates character string data based on the identified frequently occurring words and phrases and the recognized emotions.

[0331] Input: list of frequently occurring words and phrases and emotion recognition results.

[0332] How it works: The server combines this information to generate string data. For example, it creates string data associated with the positive emotion "Thank you for your hard work."

[0333] Output: String data.

[0334] Step 6:

[0335] The server generates an illustration corresponding to the character string data.

[0336] Input: String data.

[0337] How it works: Using machine learning models (e.g., GAN or VQ-VAE), the system automatically generates illustrations corresponding to string data. Specifically, it generates an illustration of a cat character that reflects a positive emotion for the string data "Thank you for your hard work."

[0338] Output: Illustration.

[0339] Step 7:

[0340] The server combines the generated string data with illustrations to create an emoji set for communication apps.

[0341] Input: String data and illustrations.

[0342] What it does: It converts text and corresponding illustrations into the correct size and format, and compiles them into an emoji set.

[0343] Output: Emoji set.

[0344] Step 8:

[0345] Users can download the completed emoji set and use it within their communication app.

[0346] Input: A set of emojis provided by the server.

[0347] How it works: A set of emojis is downloaded from a user's device and used within a communication app. For example, a user sends a "Thank you for your hard work" sticker to a friend.

[0348] Output: The emoji set used within the user's communication app.

[0349] (Application example 2)

[0350] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0351] The challenge is to improve customer experience by enriching customer communication in virtual stores and providing reactions that match the customer's emotions. Conventional systems have difficulty in properly reflecting emotions in communication between customer service staff and customers, and lack real-time responses.

[0352] 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.

[0353] In this invention, the server includes means for collecting conversation data, means for analyzing the collected conversation data to identify frequently occurring words and phrases, means for generating character string data based on the identified frequently occurring words and phrases, means for generating illustrations corresponding to the character string data, means for creating an emoji set for a communication application by combining the character string data and illustrations, means for collecting and analyzing customer conversation data in a virtual store to identify customer emotions, and means for generating reaction illustrations based on the identified emotions, thereby enabling communication with customers to be more emotionally rich and in real time.

[0354] "Conversation Data" is a record of text and voice messages sent and received by a user within a communication application.

[0355] "Frequent words and phrases" are words and expressions that appear particularly frequently in conversation data.

[0356] "Character string data" is text data that includes frequently occurring words and phrases extracted from the analyzed conversation data.

[0357] An "illustration" is a graphic representation of an image or pictogram that corresponds to the character string data.

[0358] The "emoticon set for communication applications" is a collection of emoticons that users can use in communication applications, which combines character string data with corresponding illustrations.

[0359] A "virtual store" is an online virtual store space that offers products and services via the Internet.

[0360] "Customer service conversation data" is a record of communication between customer service staff and customers within the virtual store.

[0361] "Emotions" are psychological states and feelings that are analyzed from users' statements and messages.

[0362] "Reaction illustrations" are responsive images or emojis generated based on identified emotions.

[0363] The present invention relates to a system for improving customer service communication in a virtual store. This system is implemented using the following hardware and software.

[0364] Hardware Configuration

[0365] Server: A central computer system that collects and analyzes conversation data, manages the database, and creates the emoji set.

[0366] Device: A device used by a customer or staff member, such as smart glasses or a tablet.

[0367] Software Configuration

[0368] Natural language processing technology: Hugging Face's pipeline is used to perform tokenization, stop word removal, and sentiment analysis.

[0369] Machine learning model: PIL (Python Imaging Library) is used to automatically generate illustrations, which are then given characteristics such as background color based on emotions.

[0370] Database management software: Stores conversation data and manages it so that it can be retrieved when needed.

[0371] Processing flow overview

[0372] 1. Collecting conversation data

[0373] The device collects conversation data in real time within the virtual store. The content of conversations during customer service is automatically recorded and sent to the server at regular intervals.

[0374] 2. Data Analysis

[0375] The server then analyzes the received conversation data using natural language processing techniques, which include tokenization, stop word removal, and extraction of frequently occurring words and phrases.

[0376] 3. Emotion recognition

[0377] The emotion engine recognizes the user's emotions based on frequently occurring words and phrases identified from the conversation data, performs context analysis, and assigns emotion labels.

[0378] 4. Generating reaction illustrations

[0379] Based on the identified emotion, a machine learning model is used to generate reaction illustrations, for example, a light background for positive emotions and a dark background for negative emotions.

[0380] Specific examples

[0381] When a customer says, "I really like this product!" in a virtual store, the conversation data is collected by the device and analyzed by the server. This comment is recognized as a positive emotion, and a corresponding reaction illustration (e.g., a smiling face on a bright background) is generated. This reaction illustration is then displayed to the customer in real time, improving the quality of customer service.

[0382] Prompt Sentence Examples

[0383] The user says "I absolutely love this product!" about a product they're interested in. Analyze the sentiment of this message and generate an illustration for the response based on the positive sentiment.

[0384] This invention is expected to improve customer experience by making customer service in virtual stores more emotional and real-time.

[0385] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0386] Step 1:

[0387] Conversation data collection

[0388] The device collects conversation data in real time within the virtual store, including text messages and voice messages between the user and the staff, and temporarily stores the collected conversation data locally.

[0389] Input: Conversation data (text or voice) in the virtual store

[0390] Output: Locally stored conversation data

[0391] Step 2:

[0392] Sending conversation data

[0393] At specified intervals, the device transmits locally stored conversation data to the server using a secure communication protocol.

[0394] Input: Locally stored conversation data

[0395] Output: Conversation data uploaded to the server

[0396] Step 3:

[0397] Data analysis

[0398] The server analyzes the received conversation data using natural language processing techniques, including tokenization, stop word removal, and extraction of frequently occurring words and phrases. The server then stores the analysis results in a database.

[0399] Input: Conversation data uploaded to the server

[0400] Output: Analysis result data including frequently occurring words and phrases

[0401] Step 4:

[0402] emotion recognition

[0403] The server uses an emotion engine to recognize the user's emotions based on the analyzed frequently used words and phrases. The emotion engine performs contextual analysis and assigns emotion labels. This information is also stored in a database.

[0404] Input: Analysis result data (frequently used words and phrases)

[0405] Output: Data with emotion labels

[0406] Step 5:

[0407] Generate reaction illustrations

[0408] The server uses machine learning models to generate reaction illustrations based on the identified emotion labels. For example, positive emotions are represented by illustrations with bright backgrounds. The generated illustrations are also stored in a database.

[0409] Input: Data with emotion labels

[0410] Output: Reaction illustration

[0411] Step 6:

[0412] Customer feedback

[0413] The device receives the reaction illustrations generated from the server and displays them to the customer in real time, making communication during customer service more emotional and interactive.

[0414] Input: Reaction illustration sent from the server

[0415] Output: Reaction illustrations displayed to the customer

[0416] Step 7:

[0417] Improved data storage and feedback

[0418] The server stores information such as past conversation data, analysis results, emotion labels, and reaction illustrations in a database and periodically retrains the machine learning model to improve the quality of feedback.

[0419] Input: All collected and analyzed data (conversation data, emotion labels, reaction illustrations)

[0420] Output: Improved feedback system

[0421] Through these steps, communication with customers in the virtual store becomes more emotional and real-time, improving the customer experience.

[0422] 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.

[0423] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0424] 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.

[0425] [Second embodiment]

[0426] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0427] 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.

[0428] 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).

[0429] 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.

[0430] 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.

[0431] 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).

[0432] 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.

[0433] 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.

[0434] 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.

[0435] 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.

[0436] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0437] 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."

[0438] The invention described in this specification relates to a system that collects and analyzes conversation data from users in a communication app to identify frequently occurring words and phrases, and automatically generates an emoji set based on the collected data. This system includes a user terminal, a server, and a communication means for transmitting and receiving data between them.

[0439] Data collection

[0440] The device collects the user's conversation data, including text and voice messages sent and received within the communication app. The device periodically retrieves this conversation data and stores it locally.

[0441] Examples:

[0442] When a user chats with a friend, they exchange messages such as "Good work," "Thank you," and "What are you having for lunch today?" This conversation data is stored on the device.

[0443] Data transmission

[0444] The device sends the saved conversation data to the server, which then converts the conversation data into a database format and uploads it to the server via a network connection.

[0445] Examples:

[0446] The device compiles conversation logs from over a month, converts them into a database format, and sends them to the server.

[0447] Data analysis

[0448] The server analyzes the received conversation data to identify frequently occurring words and phrases, and then uses natural language processing technology to analyze the text, tokenize it, remove stop words, and extract frequently occurring words.

[0449] Examples:

[0450] The server analyzes the conversation data and extracts frequently used phrases such as "good work," "thank you," and "lunch."

[0451] Generating string data and illustrations

[0452] The server generates character strings based on the frequently occurring words and phrases identified, and automatically generates illustrations corresponding to the character strings. The illustration generation utilizes machine learning models.

[0453] Examples:

[0454] The server generates the string data "Thank you for your hard work" and uses a machine learning model to generate a matching illustration of a cat character.

[0455] Creating an emoji set

[0456] The server combines the generated character string data with illustrations to create an emoji set for the communication app, which is then sent to the user's device.

[0457] Examples:

[0458] The server creates a stamp set that combines character strings such as "Good work," "Thank you," and "Lunch" with corresponding illustrations, and provides it to the user.

[0459] Use of stamps

[0460] Users can download the completed emoji set and use it in their communication apps, allowing them to use the new stickers to make their communications more personal and engaging.

[0461] Examples:

[0462] Users can choose a "Thank you for your hard work" stamp from the stamp set they downloaded and send it to a friend to express their gratitude.

[0463] The system allows users to easily create their own original emoji sets based on frequently used phrases and words, making everyday communication more personal and fun.

[0464] The processing flow will be explained below.

[0465] Step 1:

[0466] The user enables the conversation log collection function in the communication app, which starts the collection of conversation data.

[0467] Step 2:

[0468] The device periodically retrieves and locally stores the user's conversation data, which includes text messages, voice messages, and other communication information.

[0469] Step 3:

[0470] The device periodically sends the saved conversation data to the server, where it is converted into a database format and uploaded via a network connection.

[0471] Step 4:

[0472] The server starts processing to analyze the received conversation data. The server performs text analysis using a natural language processing library.

[0473] Step 5:

[0474] The server tokenizes the conversation data and removes stop words, making it easier to analyze.

[0475] Step 6:

[0476] The server uses the tokenized data to calculate frequency and identify frequently occurring words and phrases.

[0477] Step 7:

[0478] The server generates string data based on frequently occurring words and phrases. For example, frequently occurring phrases such as "Thank you for your hard work" are extracted.

[0479] Step 8:

[0480] The server uses machine learning models to automatically generate illustrations corresponding to frequently occurring words and phrases, and combines the generated illustrations with text data.

[0481] Step 9:

[0482] The server combines multiple illustrations and text data to create an emoji set for the communication app. Each emoji contains text data and a corresponding illustration.

[0483] Step 10:

[0484] The server sends the completed emoji set to the user's device, where the user can download it within the communication app.

[0485] Step 11:

[0486] Users can download and install the emoji set and use it within their communication apps. The new emoji set can be used to make communications more personal and engaging.

[0487] The above are the specific processing steps of the system program. Through this process, users can easily obtain and use original emoji sets based on their own conversation content.

[0488] Example 1

[0489] 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."

[0490] In recent years, the use of communication apps has increased, and users are sending and receiving many messages. However, existing emoji and stamp sets are fixed and do not correspond to the individual conversation content of users. Therefore, there is a need for personalized and more engaging means of communication. In particular, there is a need for a method to easily generate emoji that correspond to frequently used phrases and words.

[0491] 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.

[0492] In this invention, the server includes means for analyzing conversation data to identify frequently occurring words and phrases, means for generating character string data based on the identified frequently occurring words and phrases, and means for using a machine learning model to generate illustrations corresponding to the generated character string data. This allows users to easily create original emoji sets based on the content of their own conversations, making everyday communication more personalized and enjoyable.

[0493] A "terminal" is an electronic device for executing a communication app used by a user, and includes a smartphone, tablet, PC, etc.

[0494] A "server" is a central computer system that receives, processes, and stores data sent from terminals.

[0495] "Conversation data" is a data set that includes text and voice messages sent and received by a user within a communication app.

[0496] A "database format" is a format for storing structured data, and generally includes formats such as JSON and CSV.

[0497] "Natural language processing technology" is a technology that enables computers to understand, interpret, and generate natural language, and includes text tokenization, stop word removal, and frequent word extraction.

[0498] "Tokenization" is the process of breaking text into individual words or phrases.

[0499] "Stop words" are frequently used words that contain little information and are excluded from analysis (for example, "no," "ni," and "o" in Japanese).

[0500] "Frequent words" are words or phrases that are used particularly frequently in conversation data.

[0501] A "machine learning model" is an algorithm that learns patterns and rules based on data and makes predictions or generates data based on new data.

[0502] A "prompt" is an instruction sentence input to a generative AI model, containing instructions for generating illustrations or text based on specific conditions or requests.

[0503] An "emoji set" is a collection of emojis and / or stickers for use by a user within a communication app.

[0504] The present invention relates to a system that collects and analyzes conversation data from users in a communication app to identify frequently occurring words and phrases, and automatically generates emoji sets based on the collected data. The system includes a user terminal, a server, and a communication means for transmitting and receiving data between them.

[0505] Data collection

[0506] First, the device collects the user's conversation data. This conversation data includes text messages and voice messages sent and received by the user within the communication app. The device periodically obtains the conversation data using the communication app's API and saves it in local storage. Specifically, the device calls the communication app's API at 2:00 AM every day to obtain the previous day's conversation data. This data includes messages such as "Good work," "Thank you," and "What are you having for lunch today?"

[0507] Data transmission

[0508] Next, the device sends the saved conversation data to the server. The device converts the conversation data into a database format, which is typically JSON or CSV. For example, the device converts the saved conversation data at the end of the month into JSON format, and represents each message as a JSON object. The converted data is then uploaded to the server using the HTTPS protocol.

[0509] Data analysis

[0510] The server receives the data from the device and stores it in a database for analysis. The server uses natural language processing techniques such as Python's NLTK library to tokenize the conversation data, remove stop words, and extract frequently occurring words. For example, it extracts frequently occurring phrases such as "good work," "thank you," and "lunch."

[0511] Generating string data and illustrations

[0512] The server generates character string data based on the frequently occurring words and phrases identified, and automatically generates corresponding illustrations using a machine learning model. At this time, a prompt sentence is input into the generative AI model to generate the illustration. Specifically, the prompt sentence "Please generate an illustration of a cat character saying 'Good job'" is used to generate the corresponding illustration.

[0513] Creating an emoji set

[0514] The server combines the character string data with the generated illustrations to create an emoji set for the communication app. This emoji set is then sent to the user's device. For example, a stamp set is created by combining the character strings "Good work," "Thank you," and "Lunch" with their corresponding illustrations, and sent to the device using the HTTPS protocol.

[0515] Use of stamps

[0516] The user downloads the completed emoji set and uses it in the communication app. The user clicks on the new stamp set to download it, which is then saved in the specified folder in the communication app. For example, a user can select the "Thank you for your hard work" stamp on the chat screen and send it to a friend to express their gratitude.

[0517] In this way, the system allows users to easily create original emoji sets based on frequently used phrases and words, making everyday communication more personal and fun.

[0518] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0519] Program processing steps

[0520] Step 1:

[0521] A user has a conversation within a communication app. This conversation data is collected as it is entered into the device. The device calls the communication app's API and saves the acquired data in local storage. This allows text messages and voice messages to be collected.

[0522] Specific behavior:

[0523] The device calls the communication app's API at 2:00 AM every day, obtains the previous day's conversation data (e.g., "Good work," "Thank you," "What are you having for lunch today?") and stores it in an SQLite database.

[0524] Input: User conversation data (text messages, voice messages)

[0525] Output: Locally stored conversation data

[0526] Step 2:

[0527] The conversation data collected by the device is converted into a database format (JSON). The converted data is then sent from the device to the server. This transmission process uses the HTTPS protocol.

[0528] Specific behavior:

[0529] At the end of the month, the device converts the saved conversation data into JSON format, representing each message as a JSON object, and then uploads it to the server using the HTTPS protocol.

[0530] Input: Locally stored conversation data

[0531] Output: JSON data uploaded to the server

[0532] Step 3:

[0533] The server receives the uploaded JSON data and stores it in a database for analysis. This data is then analyzed using natural language processing techniques to identify frequently occurring words and phrases. Specific analysis steps include tokenization, stop word removal, and frequent word extraction.

[0534] Specific behavior:

[0535] The server uses Python's NLTK library to tokenize the conversation data, remove stop words, count frequently occurring words, and extract frequently occurring phrases such as "good work," "thank you," and "lunch."

[0536] Input: JSON data uploaded to the server

[0537] Output: Identified frequently occurring words and phrases

[0538] Step 4:

[0539] The server generates text data based on the identified frequently occurring words and phrases, which is then used as prompts for the generative AI model to generate illustrations.

[0540] Specific behavior:

[0541] The server generates the phrases "Thank you for your hard work," "Thank you," and "Lunch" as string data and prepares them as a prompt sentence.

[0542] Input: Identified common words and phrases

[0543] Output: String data (prompt statement)

[0544] Step 5:

[0545] The server inputs prompts into the generative AI model, which then automatically generates corresponding illustrations that correspond to the user's conversation.

[0546] Specific behavior:

[0547] The server inputs the prompt "Please generate an illustration of a cat character saying 'Good job'" into the generative AI model and generates the corresponding illustration.

[0548] Input: prompt statement

[0549] Output: Generated illustration

[0550] Step 6:

[0551] The server combines the generated character string data with illustrations to create an emoji set for the communication app, which is then sent to the user's device.

[0552] Specific behavior:

[0553] The server creates a stamp set by combining the strings "Good work," "Thank you," and "Lunch" with corresponding illustrations, and sends it to the terminal using the HTTPS protocol.

[0554] Input: String data and generated illustrations

[0555] Output: The emoji set sent to the user's device

[0556] Step 7:

[0557] Users can download the completed emoji set to their device and use it in communication apps. Specifically, users communicate by selecting and sending new stamps.

[0558] Specific behavior:

[0559] Users can select the "Thank you for your hard work" stamp on the chat screen and send it to their friends to express their gratitude.

[0560] Input: The emoji set sent to the user's device

[0561] Output: Stamps used in communication apps

[0562] (Application example 1)

[0563] 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."

[0564] Conventional advertising methods have difficulty accurately grasping the hobbies and preferences of their target users and personalizing their advertising, which often results in unattractive ads being displayed to users, reducing the effectiveness of advertising and lowering marketing efficiency.

[0565] 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.

[0566] In this invention, the server includes means for collecting conversation data, means for analyzing the collected conversation data to identify frequently occurring words and phrases, means for automatically generating advertising content by combining character string data and illustrations based on the identified frequently occurring words and phrases, and means for displaying the generated advertising content, thereby enabling the generation of personalized advertisements based on user conversation data.

[0567] - "Conversation data" refers to text and voice message data sent and received by a user within a communication app.

[0568] The "analysis means" is a means used to analyze conversation data and identify frequently occurring words and phrases.

[0569] "Frequent words and phrases" are important words and phrases that appear with a certain frequency or more in the collected conversation data.

[0570] "Character string data" is text data generated based on identified frequently occurring words and phrases.

[0571] The "illustration generating means" is a means used to generate an illustration corresponding to character string data.

[0572] "Advertising content" is marketing content generated by combining character string data and illustrations.

[0573] The "generation means" is a means for generating character string data and illustrations based on the identified frequently occurring words and phrases.

[0574] The "display means" is a means for displaying the generated advertising content on a user terminal.

[0575] "Natural language processing technology" is a technology for tokenizing conversational data, removing stop words, and extracting frequently occurring words and phrases.

[0576] A "machine learning model" is an algorithm or model that is trained based on data and used to automatically generate illustrations.

[0577] This invention is a system that automatically generates advertising content by collecting conversation data, analyzing it to identify frequently occurring words and phrases, and generating character string data and illustrations. This system includes a user terminal, a server, and a communication means for transmitting and receiving data between them.

[0578] First, the user's device collects conversation data. This conversation data includes text and voice messages that the user sends daily within communication apps. For example, conversations such as "I want to watch this movie," "I've been wanting to get a new smartphone lately," and "What should we have for lunch today?" are collected. The collected data is temporarily stored on the device, then converted into a database format and sent to the server.

[0579] The server then analyzes the received conversation data using natural language processing (NLP) techniques, tokenizing the conversation data, removing stop words, and extracting frequently occurring words and phrases. For example, the server identifies frequently occurring phrases such as "movie," "smartphone," and "lunch."

[0580] The server then generates character strings based on the frequently occurring words and phrases identified, and generates corresponding illustrations. Machine learning models are used to generate the illustrations, so for example, an illustration of a character holding popcorn is automatically generated for "movie," and an illustration of a character holding a smartphone is automatically generated for "smartphone."

[0581] Next, the server combines the generated text data with illustrations to create advertising content. This advertising content is sent to the user's device, where the user can view it on their smartphone app. By using this new advertising content, users can view personalized ads based on their interests and hobbies. For example, ads such as "Special discount! Great chance to see movies at a great price!" or "New smartphones on sale now!" are displayed.

[0582] The system also automatically generates advertising content by inputting prompts like the following into a generative AI model:

[0583] "You will create a program that extracts frequently used phrases and keywords from user conversation data and generates personalized advertising content based on that data. The collected data will be sent to a server in JSON format, where it will be analyzed. Targeted ads will be generated based on the analysis results, and will ultimately be displayed on the smartphone app."

[0584] As a specific example of use, when a user says something like "I want to watch this movie," the system analyzes it and automatically generates an advertisement that says "Special discount! Movies on sale now!" as completely new content and provides it to the user.

[0585] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0586] Step 1:

[0587] The user's device collects conversation data. In this step, the target is text and voice messages sent and received by the user within the communication app. The input is data from the user's daily chats and messages, and the output is the collected conversation data that is temporarily stored on the device.

[0588] Step 2:

[0589] The conversation data collected by the device is converted into a database format. In this step, the collected text and voice messages are organized and converted into a database format. The input is the collected conversation data, and the output is the converted database format data.

[0590] Step 3:

[0591] The terminal sends the converted database format conversation data to the server. In this step, data is sent from the terminal to the server via the network. The input is the conversation data converted to database format, and the output is the data received on the server side.

[0592] Step 4:

[0593] The server analyzes the received conversation data. In this step, natural language processing techniques are used to tokenize the conversation data, remove stop words, and extract frequently occurring words and phrases. The input is the conversation data, and the output is frequently occurring words and phrases. The specific operation uses the Python sklearn library.

[0594] Step 5:

[0595] The server generates string data based on the identified frequently occurring words and phrases. In this step, text corresponding to the extracted words and phrases is generated. The input is the frequently occurring words and phrases, and the output is the generated string data.

[0596] Step 6:

[0597] The server generates an illustration corresponding to the string data. In this step, the corresponding illustration is automatically generated using a machine learning model. The input is the string data, and the generated illustration is obtained as the output. Specific operations use PyTorch and TensorFlow.

[0598] Step 7:

[0599] The server combines the generated text data with the illustration to create advertising content. In this step, marketing content is generated by combining the two. The input is the text data and the corresponding illustration, and the output is the completed advertising content.

[0600] Step 8:

[0601] The server transmits the generated advertisement content to the user terminal. In this step, the advertisement content is transmitted to the user terminal via the network. The advertisement content is the input, and the advertisement content received by the user terminal is obtained as the output.

[0602] Step 9:

[0603] The user checks the advertising content on the smartphone app. In this step, the user can view personalized advertising. The input is the received advertising content, and the output is the displayed advertising content. For example, a message such as "Special discount! A great chance to see a movie at a great price!" may be displayed.

[0604] 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.

[0605] The present invention relates to a system that automatically generates emoji sets by collecting and analyzing conversation data from users in a communication app, identifying frequently used words and phrases, and combining this with an emotion engine that recognizes the user's emotions. This system includes a user terminal, a server, and communication means for transmitting and receiving data between them.

[0606] Data collection

[0607] The device collects the user's conversation data, including text and voice messages sent and received within the communication app. The device periodically retrieves this conversation data and stores it locally.

[0608] Examples:

[0609] When a user chats with a friend, they exchange messages such as "Good work," "Thank you," and "What are you having for lunch today?" This conversation data is stored on the device.

[0610] Data transmission

[0611] The device periodically sends the saved conversation data to the server, where it is converted into a database format and uploaded via a network connection.

[0612] Examples:

[0613] The device compiles conversation logs from over a month, converts them into a database format, and sends them to the server.

[0614] Data analysis

[0615] The server analyzes the received conversation data to identify frequently occurring words and phrases, and then uses natural language processing technology to analyze the text, tokenize it, remove stop words, and extract frequently occurring words.

[0616] Examples:

[0617] The server analyzes the conversation data and extracts frequently used phrases such as "good work," "thank you," and "lunch."

[0618] Emotion recognition by emotion engine

[0619] The server further analyzes the conversation data to recognize the user's emotions. The emotion engine uses keywords and contextual analysis to identify the emotions the user is expressing during the conversation.

[0620] Examples:

[0621] The server analyzes phrases like "Otsukaresama" (thank you) and recognizes that it indicates a positive emotion, as well as "arigato" (thank you) as an emotion that conveys gratitude.

[0622] Generating string data and illustrations

[0623] The server generates character strings based on the frequently used words and phrases and the user's emotions, and automatically generates illustrations corresponding to the character strings. The illustration generation utilizes a machine learning model.

[0624] Examples:

[0625] The server associates a positive emotion with the string data "Thank you for your hard work" and generates a matching cat character illustration using a machine learning model.

[0626] Creating an emoji set

[0627] The server combines the generated text data with illustrations to create an emoji set for communication apps, which also includes emotional information.

[0628] Examples:

[0629] The server creates a stamp set that combines character strings such as "good work," "thank you," and "lunch" with corresponding illustrations, providing an original emoticon set that reflects the user's emotions.

[0630] Use of stamps

[0631] Users can download the completed emoji set and use it in their communication apps, allowing them to use the new stickers to communicate more personally and emotionally.

[0632] Examples:

[0633] Users can choose a "Thank you for your hard work" stamp from the stamp set they downloaded and send it to a friend to express their gratitude.

[0634] This system allows users to easily obtain original emoji sets based on their own conversation content and emotions, making everyday communication more personal and emotional.

[0635] The processing flow will be explained below.

[0636] Step 1:

[0637] The user enables the conversation log collection function in the communication app, which starts the collection of conversation data.

[0638] Step 2:

[0639] The device periodically retrieves and locally stores the user's conversation data, which includes text messages, voice messages, and other communication information.

[0640] Step 3:

[0641] The device periodically sends the saved conversation data to the server, where it is converted into a database format and uploaded via a network connection.

[0642] Step 4:

[0643] The server starts processing to analyze the received conversation data. The server performs text analysis using a natural language processing library.

[0644] Step 5:

[0645] The server tokenizes the conversation data and removes stop words, making it easier to analyze.

[0646] Step 6:

[0647] The server uses the tokenized data to calculate frequency and identify frequently occurring words and phrases.

[0648] Step 7:

[0649] The server further analyzes the conversation data and uses an emotion engine to recognize the user's emotions. The emotion engine uses keywords and contextual analysis to identify the user's emotions.

[0650] Step 8:

[0651] The server generates string data based on the identified frequently occurring words and phrases and the recognized emotions.

[0652] Step 9:

[0653] The server uses machine learning models to automatically generate illustrations that correspond to frequently used words and phrases and the user's emotions, and combines the generated illustrations with text data.

[0654] Step 10:

[0655] The server combines multiple illustrations and text data to create an emoji set for the communication app. Each emoji contains text data and a corresponding illustration.

[0656] Step 11:

[0657] The server sends the completed emoji set to the user's device, where the user can download it within the communication app.

[0658] Step 12:

[0659] Users can download and install the emoji set and use it within their communication app. The new emoji set can be used to make communication more personal and emotional.

[0660] Examples:

[0661] When chatting with a friend, a user sends and receives conversations such as "Good work," "Thank you," and "What are we having for lunch today?" This conversation data is saved on the device and sent to a server. The server analyzes the data using natural language processing and an emotion engine, and automatically generates phrases that express positive emotions, such as "Good work," and corresponding illustrations. As a result, the user can download and use stamp sets that include character strings such as "Good work," "Thank you," and "Lunch," along with the corresponding illustrations.

[0662] Example 2

[0663] 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."

[0664] In conventional communication applications, the emoji sets used by users are fixed and do not reflect the user's actual conversation content or emotions. This limits personalized and emotional communication, making it difficult for users to accurately express themselves. Furthermore, the limited selection of existing emojis leads to a standardized user experience and a lack of individuality.

[0665] 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.

[0666] In this invention, the server includes means for collecting conversation data, means for analyzing the conversation data to identify frequently occurring words and phrases, means for analyzing the conversation data to recognize the user's emotions, means for generating character string data based on the identified frequently occurring words and phrases and the recognized emotions, means for generating illustrations corresponding to the character string data, and means for creating an emoji set for a communication application by combining the character string data and illustrations. This enables users to easily obtain an original emoji set based on the content of their own conversations and emotions, making everyday communication more personalized and emotionally rich.

[0667] "Conversation Data" is data including text messages, voice messages, and other communications sent and received by a user within a communication application.

[0668] "Means of collection" refers to the methods and functions by which a user's terminal or system collects conversation data automatically or at the user's direction.

[0669] "Means of analysis" refers to methods and technologies for analyzing conversation data and extracting meaning and information. Specifically, this includes analysis using natural language processing technology.

[0670] "Frequent words and phrases" refer to words and expressions that appear particularly frequently in the collected conversational data.

[0671] "Means for recognizing emotions" refers to methods and technologies for analyzing conversation data and identifying user emotions, including emotion engines and context analysis.

[0672] "String data" refers to text-based information generated from analyzed conversation data.

[0673] "Means for generating illustrations" refers to methods and technologies for automatically creating visual images based on text data and emotional information. Specifically, they use machine learning models.

[0674] "Pictogram set" refers to a collection of graphical representations for communication applications created by combining specified character string data and corresponding illustrations.

[0675] MODE FOR CARRYING OUT THE INVENTION

[0676] The present invention relates to a system that automatically generates emoji sets by collecting and analyzing conversation data from users in a communication app, identifying frequently used words and phrases, and combining this with an emotion engine that recognizes the user's emotions. This system includes a user terminal, a server, and communication means for transmitting and receiving data between them.

[0677] Data collection

[0678] The device collects the user's conversation data, including text and voice messages sent and received within the communication app. The device periodically retrieves this conversation data and stores it locally.

[0679] Examples:

[0680] For example, when a user chats with a friend and exchanges messages such as "Good work," "Thank you," and "What are you having for lunch today?", this conversation data is saved on the device.

[0681] Data transmission

[0682] The device periodically sends the saved conversation data to the server, where it is converted into a database format and uploaded via a network connection.

[0683] Examples:

[0684] The device compiles conversation logs over a month, converts them into a database format, and sends the data to a server using a secure protocol such as SSL / TLS.

[0685] Data analysis

[0686] The server analyzes the received conversation data to identify frequently occurring words and phrases, and uses natural language processing (NLP) technology to tokenize the text, remove stop words, and then extract frequently occurring words.

[0687] Examples:

[0688] The server tokenizes the conversation data using a morphological analysis engine and extracts frequently occurring phrases such as "good work," "thank you," and "lunch."

[0689] Emotion recognition by emotion engine

[0690] The server further analyzes the conversation data to recognize the user's emotions. The emotion engine uses keywords and contextual analysis to identify the emotions the user is showing during the conversation.

[0691] Examples:

[0692] The server analyzes phrases like "Otsukaresama" (thank you) and recognizes that it indicates a positive emotion, as well as "arigato" (thank you) as an emotion that conveys gratitude.

[0693] Generating string data and illustrations

[0694] The server generates text data based on the frequently used words and phrases and the user's emotions, and then automatically generates illustrations corresponding to the text data. The illustration generation utilizes machine learning models (such as GAN and VQ-VAE).

[0695] Examples:

[0696] The server associates a positive emotion with the string data "good work" and generates an illustration of a cat character.

[0697] Creating an emoji set

[0698] The server combines the generated text data with illustrations to create an emoji set for communication apps, which also includes emotional information.

[0699] Examples:

[0700] The server creates a stamp set that combines character strings such as "good work," "thank you," and "lunch" with corresponding illustrations, providing an original emoticon set that reflects the user's emotions.

[0701] Use of stamps

[0702] Users can download the completed emoji set and use it in their communication apps, allowing them to communicate in a personalized and emotional way using the new stickers.

[0703] Examples:

[0704] Users can choose a "Thank you for your hard work" stamp from the stamp set they downloaded and send it to a friend to express their gratitude.

[0705] This allows users to easily obtain original emoji sets based on their own conversation content and emotions, making everyday communication more personal and emotional.

[0706] Examples of prompt statements

[0707] An example of a prompt for a generative AI model is:

[0708] Generate illustrations of stamps that reflect phrases frequently used by users in communication apps, such as "Thank you" and "Thank you," and the emotions associated with them (positive, grateful).

[0709] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0710] Step 1:

[0711] The terminal collects user conversation data.

[0712] Input: Text and voice messages that users send and receive within communication apps.

[0713] How it works: A program runs to periodically save conversation data to the device's local storage, and the user sends and receives messages such as "Good work," "Thank you," and "What are you having for lunch today?" in chats with friends.

[0714] Output: Conversation data stored in the device's local storage.

[0715] Step 2:

[0716] The conversation data stored on the device is periodically sent to the server.

[0717] Input: Conversation data stored in local storage.

[0718] How it works: Data is first converted into a database format and then uploaded to the server over a network connection using SSL / TLS protocol.

[0719] Output: Conversation data uploaded to the server.

[0720] Step 3:

[0721] The server analyzes the conversation data it receives and identifies frequently used words and phrases.

[0722] Input: Conversation data uploaded to the server.

[0723] How it works: It uses natural language processing techniques to tokenize text data and remove stop words, then uses a morphological analysis engine to extract frequently occurring words and phrases.

[0724] Output: A list of frequently occurring words and phrases.

[0725] Step 4:

[0726] The server further analyzes the conversation data and recognizes the user's emotions.

[0727] Input: A list of frequently used words and phrases.

[0728] How it works: Uses a sentiment engine to identify user sentiment based on keywords and contextual analysis, for example recognizing positive sentiment from the phrase "Otsukaresama."

[0729] Output: Emotion recognition result.

[0730] Step 5:

[0731] The server generates character string data based on the identified frequently occurring words and phrases and the recognized emotions.

[0732] Input: list of frequently occurring words and phrases and emotion recognition results.

[0733] How it works: The server combines this information to generate string data. For example, it creates string data associated with the positive emotion "Thank you for your hard work."

[0734] Output: String data.

[0735] Step 6:

[0736] The server generates an illustration corresponding to the character string data.

[0737] Input: String data.

[0738] How it works: Using machine learning models (e.g., GAN or VQ-VAE), the system automatically generates illustrations corresponding to string data. Specifically, it generates an illustration of a cat character that reflects a positive emotion for the string data "Thank you for your hard work."

[0739] Output: Illustration.

[0740] Step 7:

[0741] The server combines the generated string data with illustrations to create an emoji set for communication apps.

[0742] Input: String data and illustrations.

[0743] What it does: It converts text and corresponding illustrations into the correct size and format, and compiles them into an emoji set.

[0744] Output: Emoji set.

[0745] Step 8:

[0746] Users can download the completed emoji set and use it within their communication app.

[0747] Input: A set of emojis provided by the server.

[0748] How it works: A set of emojis is downloaded from a user's device and used within a communication app. For example, a user sends a "Thank you for your hard work" sticker to a friend.

[0749] Output: The emoji set used within the user's communication app.

[0750] (Application example 2)

[0751] 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."

[0752] The challenge is to improve customer experience by enriching customer communication in virtual stores and providing reactions that match the customer's emotions. Conventional systems have difficulty in properly reflecting emotions in communication between customer service staff and customers, and lack real-time responses.

[0753] 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.

[0754] In this invention, the server includes means for collecting conversation data, means for analyzing the collected conversation data to identify frequently occurring words and phrases, means for generating character string data based on the identified frequently occurring words and phrases, means for generating illustrations corresponding to the character string data, means for creating an emoji set for a communication application by combining the character string data and illustrations, means for collecting and analyzing customer conversation data in a virtual store to identify customer emotions, and means for generating reaction illustrations based on the identified emotions, thereby enabling communication with customers to be more emotionally rich and in real time.

[0755] "Conversation Data" is a record of text and voice messages sent and received by a user within a communication application.

[0756] "Frequent words and phrases" are words and expressions that appear particularly frequently in conversation data.

[0757] "Character string data" is text data that includes frequently occurring words and phrases extracted from the analyzed conversation data.

[0758] An "illustration" is a graphic representation of an image or pictogram that corresponds to the character string data.

[0759] The "emoticon set for communication applications" is a collection of emoticons that users can use in communication applications, which combines character string data with corresponding illustrations.

[0760] A "virtual store" is an online virtual store space that offers products and services via the Internet.

[0761] "Customer service conversation data" is a record of communication between customer service staff and customers within the virtual store.

[0762] "Emotions" are psychological states and feelings that are analyzed from users' statements and messages.

[0763] "Reaction illustrations" are responsive images or emojis generated based on identified emotions.

[0764] The present invention relates to a system for improving customer service communication in a virtual store. This system is implemented using the following hardware and software.

[0765] Hardware Configuration

[0766] Server: A central computer system that collects and analyzes conversation data, manages the database, and creates the emoji set.

[0767] Device: A device used by a customer or staff member, such as smart glasses or a tablet.

[0768] Software Configuration

[0769] Natural language processing technology: Hugging Face's pipeline is used to perform tokenization, stop word removal, and sentiment analysis.

[0770] Machine learning model: PIL (Python Imaging Library) is used to automatically generate illustrations, which are then given characteristics such as background color based on emotions.

[0771] Database management software: Stores conversation data and manages it so that it can be retrieved when needed.

[0772] Processing flow overview

[0773] 1. Collecting conversation data

[0774] The device collects conversation data in real time within the virtual store. The content of conversations during customer service is automatically recorded and sent to the server at regular intervals.

[0775] 2. Data Analysis

[0776] The server then analyzes the received conversation data using natural language processing techniques, which include tokenization, stop word removal, and extraction of frequently occurring words and phrases.

[0777] 3. Emotion recognition

[0778] The emotion engine recognizes the user's emotions based on frequently occurring words and phrases identified from the conversation data, performs context analysis, and assigns emotion labels.

[0779] 4. Generating reaction illustrations

[0780] Based on the identified emotion, a machine learning model is used to generate reaction illustrations, for example, a light background for positive emotions and a dark background for negative emotions.

[0781] Specific examples

[0782] When a customer says, "I really like this product!" in a virtual store, the conversation data is collected by the device and analyzed by the server. This comment is recognized as a positive emotion, and a corresponding reaction illustration (e.g., a smiling face on a bright background) is generated. This reaction illustration is then displayed to the customer in real time, improving the quality of customer service.

[0783] Prompt Sentence Examples

[0784] The user says "I absolutely love this product!" about a product they're interested in. Analyze the sentiment of this message and generate an illustration for the response based on the positive sentiment.

[0785] This invention is expected to improve customer experience by making customer service in virtual stores more emotional and real-time.

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

[0787] Step 1:

[0788] Conversation data collection

[0789] The device collects conversation data in real time within the virtual store, including text messages and voice messages between the user and the staff, and temporarily stores the collected conversation data locally.

[0790] Input: Conversation data (text or voice) in the virtual store

[0791] Output: Locally stored conversation data

[0792] Step 2:

[0793] Sending conversation data

[0794] At specified intervals, the device transmits locally stored conversation data to the server using a secure communication protocol.

[0795] Input: Locally stored conversation data

[0796] Output: Conversation data uploaded to the server

[0797] Step 3:

[0798] Data analysis

[0799] The server analyzes the received conversation data using natural language processing techniques, including tokenization, stop word removal, and extraction of frequently occurring words and phrases. The server then stores the analysis results in a database.

[0800] Input: Conversation data uploaded to the server

[0801] Output: Analysis result data including frequently occurring words and phrases

[0802] Step 4:

[0803] emotion recognition

[0804] The server uses an emotion engine to recognize the user's emotions based on the analyzed frequently used words and phrases. The emotion engine performs contextual analysis and assigns emotion labels. This information is also stored in a database.

[0805] Input: Analysis result data (frequently used words and phrases)

[0806] Output: Data with emotion labels

[0807] Step 5:

[0808] Generate reaction illustrations

[0809] The server uses machine learning models to generate reaction illustrations based on the identified emotion labels. For example, positive emotions are represented by illustrations with bright backgrounds. The generated illustrations are also stored in a database.

[0810] Input: Data with emotion labels

[0811] Output: Reaction illustration

[0812] Step 6:

[0813] Customer feedback

[0814] The device receives the reaction illustrations generated from the server and displays them to the customer in real time, making communication during customer service more emotional and interactive.

[0815] Input: Reaction illustration sent from the server

[0816] Output: Reaction illustrations displayed to the customer

[0817] Step 7:

[0818] Improved data storage and feedback

[0819] The server stores information such as past conversation data, analysis results, emotion labels, and reaction illustrations in a database and periodically retrains the machine learning model to improve the quality of feedback.

[0820] Input: All collected and analyzed data (conversation data, emotion labels, reaction illustrations)

[0821] Output: Improved feedback system

[0822] Through these steps, communication with customers in the virtual store becomes more emotional and real-time, improving the customer experience.

[0823] 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.

[0824] 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.

[0825] 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.

[0826] [Third embodiment]

[0827] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0828] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0829] 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).

[0830] 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.

[0831] 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.

[0832] 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).

[0833] 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.

[0834] 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.

[0835] 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.

[0836] 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.

[0837] 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.

[0838] 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."

[0839] The invention described in this specification relates to a system that collects and analyzes conversation data from users in a communication app to identify frequently occurring words and phrases, and automatically generates an emoji set based on the collected data. This system includes a user terminal, a server, and a communication means for transmitting and receiving data between them.

[0840] Data collection

[0841] The device collects the user's conversation data, including text and voice messages sent and received within the communication app. The device periodically retrieves this conversation data and stores it locally.

[0842] Examples:

[0843] When a user chats with a friend, they exchange messages such as "Good work," "Thank you," and "What are you having for lunch today?" This conversation data is stored on the device.

[0844] Data transmission

[0845] The device sends the saved conversation data to the server, which then converts the conversation data into a database format and uploads it to the server via a network connection.

[0846] Examples:

[0847] The device compiles conversation logs from over a month, converts them into a database format, and sends them to the server.

[0848] Data analysis

[0849] The server analyzes the received conversation data to identify frequently occurring words and phrases, and then uses natural language processing technology to analyze the text, tokenize it, remove stop words, and extract frequently occurring words.

[0850] Examples:

[0851] The server analyzes the conversation data and extracts frequently used phrases such as "good work," "thank you," and "lunch."

[0852] Generating string data and illustrations

[0853] The server generates character strings based on the frequently occurring words and phrases identified, and automatically generates illustrations corresponding to the character strings. The illustration generation utilizes machine learning models.

[0854] Examples:

[0855] The server generates the string data "Thank you for your hard work" and uses a machine learning model to generate a matching illustration of a cat character.

[0856] Creating an emoji set

[0857] The server combines the generated character string data with illustrations to create an emoji set for the communication app, which is then sent to the user's device.

[0858] Examples:

[0859] The server creates a stamp set that combines character strings such as "Good work," "Thank you," and "Lunch" with corresponding illustrations, and provides it to the user.

[0860] Use of stamps

[0861] Users can download the completed emoji set and use it in their communication apps, allowing them to use the new stickers to make their communications more personal and engaging.

[0862] Examples:

[0863] Users can choose a "Thank you for your hard work" stamp from the stamp set they downloaded and send it to a friend to express their gratitude.

[0864] The system allows users to easily create their own original emoji sets based on frequently used phrases and words, making everyday communication more personal and fun.

[0865] The processing flow will be explained below.

[0866] Step 1:

[0867] The user enables the conversation log collection function in the communication app, which starts the collection of conversation data.

[0868] Step 2:

[0869] The device periodically retrieves and locally stores the user's conversation data, which includes text messages, voice messages, and other communication information.

[0870] Step 3:

[0871] The device periodically sends the saved conversation data to the server, where it is converted into a database format and uploaded via a network connection.

[0872] Step 4:

[0873] The server starts processing to analyze the received conversation data. The server performs text analysis using a natural language processing library.

[0874] Step 5:

[0875] The server tokenizes the conversation data and removes stop words, making it easier to analyze.

[0876] Step 6:

[0877] The server uses the tokenized data to calculate frequency and identify frequently occurring words and phrases.

[0878] Step 7:

[0879] The server generates string data based on frequently occurring words and phrases. For example, frequently occurring phrases such as "Thank you for your hard work" are extracted.

[0880] Step 8:

[0881] The server uses machine learning models to automatically generate illustrations corresponding to frequently occurring words and phrases, and combines the generated illustrations with text data.

[0882] Step 9:

[0883] The server combines multiple illustrations and text data to create an emoji set for the communication app. Each emoji contains text data and a corresponding illustration.

[0884] Step 10:

[0885] The server sends the completed emoji set to the user's device, where the user can download it within the communication app.

[0886] Step 11:

[0887] Users can download and install the emoji set and use it within their communication apps. The new emoji set can be used to make communications more personal and engaging.

[0888] The above are the specific processing steps of the system program. Through this process, users can easily obtain and use original emoji sets based on their own conversation content.

[0889] Example 1

[0890] 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."

[0891] In recent years, the use of communication apps has increased, and users are sending and receiving many messages. However, existing emoji and stamp sets are fixed and do not correspond to the individual conversation content of users. Therefore, there is a need for personalized and more engaging means of communication. In particular, there is a need for a method to easily generate emoji that correspond to frequently used phrases and words.

[0892] 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.

[0893] In this invention, the server includes means for analyzing conversation data to identify frequently occurring words and phrases, means for generating character string data based on the identified frequently occurring words and phrases, and means for using a machine learning model to generate illustrations corresponding to the generated character string data. This allows users to easily create original emoji sets based on the content of their own conversations, making everyday communication more personalized and enjoyable.

[0894] A "terminal" is an electronic device for executing a communication app used by a user, and includes a smartphone, tablet, PC, etc.

[0895] A "server" is a central computer system that receives, processes, and stores data sent from terminals.

[0896] "Conversation data" is a data set that includes text and voice messages sent and received by a user within a communication app.

[0897] A "database format" is a format for storing structured data, and generally includes formats such as JSON and CSV.

[0898] "Natural language processing technology" is a technology that enables computers to understand, interpret, and generate natural language, and includes text tokenization, stop word removal, and frequent word extraction.

[0899] "Tokenization" is the process of breaking text into individual words or phrases.

[0900] "Stop words" are frequently used words that contain little information and are excluded from analysis (for example, "no," "ni," and "o" in Japanese).

[0901] "Frequent words" are words or phrases that are used particularly frequently in conversation data.

[0902] A "machine learning model" is an algorithm that learns patterns and rules based on data and makes predictions or generates data based on new data.

[0903] A "prompt" is an instruction sentence input to a generative AI model, containing instructions for generating illustrations or text based on specific conditions or requests.

[0904] An "emoji set" is a collection of emojis and / or stickers for use by a user within a communication app.

[0905] The present invention relates to a system that collects and analyzes conversation data from users in a communication app to identify frequently occurring words and phrases, and automatically generates emoji sets based on the collected data. The system includes a user terminal, a server, and a communication means for transmitting and receiving data between them.

[0906] Data collection

[0907] First, the device collects the user's conversation data. This conversation data includes text messages and voice messages sent and received by the user within the communication app. The device periodically obtains the conversation data using the communication app's API and saves it in local storage. Specifically, the device calls the communication app's API at 2:00 AM every day to obtain the previous day's conversation data. This data includes messages such as "Good work," "Thank you," and "What are you having for lunch today?"

[0908] Data transmission

[0909] Next, the device sends the saved conversation data to the server. The device converts the conversation data into a database format, which is typically JSON or CSV. For example, the device converts the saved conversation data at the end of the month into JSON format, and represents each message as a JSON object. The converted data is then uploaded to the server using the HTTPS protocol.

[0910] Data analysis

[0911] The server receives the data from the device and stores it in a database for analysis. The server uses natural language processing techniques such as Python's NLTK library to tokenize the conversation data, remove stop words, and extract frequently occurring words. For example, it extracts frequently occurring phrases such as "good work," "thank you," and "lunch."

[0912] Generating string data and illustrations

[0913] The server generates character string data based on the frequently occurring words and phrases identified, and automatically generates corresponding illustrations using a machine learning model. At this time, a prompt sentence is input into the generative AI model to generate the illustration. Specifically, the prompt sentence "Please generate an illustration of a cat character saying 'Good job'" is used to generate the corresponding illustration.

[0914] Creating an emoji set

[0915] The server combines the character string data with the generated illustrations to create an emoji set for the communication app. This emoji set is then sent to the user's device. For example, a stamp set is created by combining the character strings "Good work," "Thank you," and "Lunch" with their corresponding illustrations, and sent to the device using the HTTPS protocol.

[0916] Use of stamps

[0917] The user downloads the completed emoji set and uses it in the communication app. The user clicks on the new stamp set to download it, which is then saved in the specified folder in the communication app. For example, a user can select the "Thank you for your hard work" stamp on the chat screen and send it to a friend to express their gratitude.

[0918] In this way, the system allows users to easily create original emoji sets based on frequently used phrases and words, making everyday communication more personal and fun.

[0919] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0920] Program processing steps

[0921] Step 1:

[0922] A user has a conversation within a communication app. This conversation data is collected as it is entered into the device. The device calls the communication app's API and saves the acquired data in local storage. This allows text messages and voice messages to be collected.

[0923] Specific behavior:

[0924] The device calls the communication app's API at 2:00 AM every day, obtains the previous day's conversation data (e.g., "Good work," "Thank you," "What are you having for lunch today?") and stores it in an SQLite database.

[0925] Input: User conversation data (text messages, voice messages)

[0926] Output: Locally stored conversation data

[0927] Step 2:

[0928] The conversation data collected by the device is converted into a database format (JSON). The converted data is then sent from the device to the server. This transmission process uses the HTTPS protocol.

[0929] Specific behavior:

[0930] At the end of the month, the device converts the saved conversation data into JSON format, representing each message as a JSON object, and then uploads it to the server using the HTTPS protocol.

[0931] Input: Locally stored conversation data

[0932] Output: JSON data uploaded to the server

[0933] Step 3:

[0934] The server receives the uploaded JSON data and stores it in a database for analysis. This data is then analyzed using natural language processing techniques to identify frequently occurring words and phrases. Specific analysis steps include tokenization, stop word removal, and frequent word extraction.

[0935] Specific behavior:

[0936] The server uses Python's NLTK library to tokenize the conversation data, remove stop words, count frequently occurring words, and extract frequently occurring phrases such as "good work," "thank you," and "lunch."

[0937] Input: JSON data uploaded to the server

[0938] Output: Identified frequently occurring words and phrases

[0939] Step 4:

[0940] The server generates text data based on the identified frequently occurring words and phrases, which is then used as prompts for the generative AI model to generate illustrations.

[0941] Specific behavior:

[0942] The server generates the phrases "Thank you for your hard work," "Thank you," and "Lunch" as string data and prepares them as a prompt sentence.

[0943] Input: Identified common words and phrases

[0944] Output: String data (prompt statement)

[0945] Step 5:

[0946] The server inputs prompts into the generative AI model, which then automatically generates corresponding illustrations that correspond to the user's conversation.

[0947] Specific behavior:

[0948] The server inputs the prompt "Please generate an illustration of a cat character saying 'Good job'" into the generative AI model and generates the corresponding illustration.

[0949] Input: prompt statement

[0950] Output: Generated illustration

[0951] Step 6:

[0952] The server combines the generated character string data with illustrations to create an emoji set for the communication app, which is then sent to the user's device.

[0953] Specific behavior:

[0954] The server creates a stamp set by combining the strings "Good work," "Thank you," and "Lunch" with corresponding illustrations, and sends it to the terminal using the HTTPS protocol.

[0955] Input: String data and generated illustrations

[0956] Output: The emoji set sent to the user's device

[0957] Step 7:

[0958] Users can download the completed emoji set to their device and use it in communication apps. Specifically, users communicate by selecting and sending new stamps.

[0959] Specific behavior:

[0960] Users can select the "Thank you for your hard work" stamp on the chat screen and send it to their friends to express their gratitude.

[0961] Input: The emoji set sent to the user's device

[0962] Output: Stamps used in communication apps

[0963] (Application example 1)

[0964] 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."

[0965] Conventional advertising methods have difficulty accurately grasping the hobbies and preferences of their target users and personalizing their advertising, which often results in unattractive ads being displayed to users, reducing the effectiveness of advertising and lowering marketing efficiency.

[0966] 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.

[0967] In this invention, the server includes means for collecting conversation data, means for analyzing the collected conversation data to identify frequently occurring words and phrases, means for automatically generating advertising content by combining character string data and illustrations based on the identified frequently occurring words and phrases, and means for displaying the generated advertising content, thereby enabling the generation of personalized advertisements based on user conversation data.

[0968] - "Conversation data" refers to text and voice message data sent and received by a user within a communication app.

[0969] The "analysis means" is a means used to analyze conversation data and identify frequently occurring words and phrases.

[0970] "Frequent words and phrases" are important words and phrases that appear with a certain frequency or more in the collected conversation data.

[0971] "Character string data" is text data generated based on identified frequently occurring words and phrases.

[0972] The "illustration generating means" is a means used to generate an illustration corresponding to character string data.

[0973] "Advertising content" is marketing content generated by combining character string data and illustrations.

[0974] The "generation means" is a means for generating character string data and illustrations based on the identified frequently occurring words and phrases.

[0975] The "display means" is a means for displaying the generated advertising content on a user terminal.

[0976] "Natural language processing technology" is a technology for tokenizing conversational data, removing stop words, and extracting frequently occurring words and phrases.

[0977] A "machine learning model" is an algorithm or model that is trained based on data and used to automatically generate illustrations.

[0978] This invention is a system that automatically generates advertising content by collecting conversation data, analyzing it to identify frequently occurring words and phrases, and generating character string data and illustrations. This system includes a user terminal, a server, and a communication means for transmitting and receiving data between them.

[0979] First, the user's device collects conversation data. This conversation data includes text and voice messages that the user sends daily within communication apps. For example, conversations such as "I want to watch this movie," "I've been wanting to get a new smartphone lately," and "What should we have for lunch today?" are collected. The collected data is temporarily stored on the device, then converted into a database format and sent to the server.

[0980] The server then analyzes the received conversation data using natural language processing (NLP) techniques, tokenizing the conversation data, removing stop words, and extracting frequently occurring words and phrases. For example, the server identifies frequently occurring phrases such as "movie," "smartphone," and "lunch."

[0981] The server then generates character strings based on the frequently occurring words and phrases identified, and generates corresponding illustrations. Machine learning models are used to generate the illustrations, so for example, an illustration of a character holding popcorn is automatically generated for "movie," and an illustration of a character holding a smartphone is automatically generated for "smartphone."

[0982] Next, the server combines the generated text data with illustrations to create advertising content. This advertising content is sent to the user's device, where the user can view it on their smartphone app. By using this new advertising content, users can view personalized ads based on their interests and hobbies. For example, ads such as "Special discount! Great chance to see movies at a great price!" or "New smartphones on sale now!" are displayed.

[0983] The system also automatically generates advertising content by inputting prompts like the following into a generative AI model:

[0984] "You will create a program that extracts frequently used phrases and keywords from user conversation data and generates personalized advertising content based on that data. The collected data will be sent to a server in JSON format, where it will be analyzed. Targeted ads will be generated based on the analysis results, and will ultimately be displayed on the smartphone app."

[0985] As a specific example of use, when a user says something like "I want to watch this movie," the system analyzes it and automatically generates an advertisement that says "Special discount! Movies on sale now!" as completely new content and provides it to the user.

[0986] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0987] Step 1:

[0988] The user's device collects conversation data. In this step, the target is text and voice messages sent and received by the user within the communication app. The input is data from the user's daily chats and messages, and the output is the collected conversation data that is temporarily stored on the device.

[0989] Step 2:

[0990] The conversation data collected by the device is converted into a database format. In this step, the collected text and voice messages are organized and converted into a database format. The input is the collected conversation data, and the output is the converted database format data.

[0991] Step 3:

[0992] The terminal sends the converted database format conversation data to the server. In this step, data is sent from the terminal to the server via the network. The input is the conversation data converted to database format, and the output is the data received on the server side.

[0993] Step 4:

[0994] The server analyzes the received conversation data. In this step, natural language processing techniques are used to tokenize the conversation data, remove stop words, and extract frequently occurring words and phrases. The input is the conversation data, and the output is frequently occurring words and phrases. The specific operation uses the Python sklearn library.

[0995] Step 5:

[0996] The server generates string data based on the identified frequently occurring words and phrases. In this step, text corresponding to the extracted words and phrases is generated. The input is the frequently occurring words and phrases, and the output is the generated string data.

[0997] Step 6:

[0998] The server generates an illustration corresponding to the string data. In this step, the corresponding illustration is automatically generated using a machine learning model. The input is the string data, and the generated illustration is obtained as the output. Specific operations use PyTorch and TensorFlow.

[0999] Step 7:

[1000] The server combines the generated text data with the illustration to create advertising content. In this step, marketing content is generated by combining the two. The input is the text data and the corresponding illustration, and the output is the completed advertising content.

[1001] Step 8:

[1002] The server transmits the generated advertisement content to the user terminal. In this step, the advertisement content is transmitted to the user terminal via the network. The advertisement content is the input, and the advertisement content received by the user terminal is obtained as the output.

[1003] Step 9:

[1004] The user checks the advertising content on the smartphone app. In this step, the user can view personalized advertising. The input is the received advertising content, and the output is the displayed advertising content. For example, a message such as "Special discount! A great chance to see a movie at a great price!" may be displayed.

[1005] 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.

[1006] The present invention relates to a system that automatically generates emoji sets by collecting and analyzing conversation data from users in a communication app, identifying frequently used words and phrases, and combining this with an emotion engine that recognizes the user's emotions. This system includes a user terminal, a server, and communication means for transmitting and receiving data between them.

[1007] Data collection

[1008] The device collects the user's conversation data, including text and voice messages sent and received within the communication app. The device periodically retrieves this conversation data and stores it locally.

[1009] Examples:

[1010] When a user chats with a friend, they exchange messages such as "Good work," "Thank you," and "What are you having for lunch today?" This conversation data is stored on the device.

[1011] Data transmission

[1012] The device periodically sends the saved conversation data to the server, where it is converted into a database format and uploaded via a network connection.

[1013] Examples:

[1014] The device compiles conversation logs from over a month, converts them into a database format, and sends them to the server.

[1015] Data analysis

[1016] The server analyzes the received conversation data to identify frequently occurring words and phrases, and then uses natural language processing technology to analyze the text, tokenize it, remove stop words, and extract frequently occurring words.

[1017] Examples:

[1018] The server analyzes the conversation data and extracts frequently used phrases such as "good work," "thank you," and "lunch."

[1019] Emotion recognition by emotion engine

[1020] The server further analyzes the conversation data to recognize the user's emotions. The emotion engine uses keywords and contextual analysis to identify the emotions the user is expressing during the conversation.

[1021] Examples:

[1022] The server analyzes phrases like "Otsukaresama" (thank you) and recognizes that it indicates a positive emotion, as well as "arigato" (thank you) as an emotion that conveys gratitude.

[1023] Generating string data and illustrations

[1024] The server generates character strings based on the frequently used words and phrases and the user's emotions, and automatically generates illustrations corresponding to the character strings. The illustration generation utilizes a machine learning model.

[1025] Examples:

[1026] The server associates a positive emotion with the string data "Thank you for your hard work" and generates a matching cat character illustration using a machine learning model.

[1027] Creating an emoji set

[1028] The server combines the generated text data with illustrations to create an emoji set for communication apps, which also includes emotional information.

[1029] Examples:

[1030] The server creates a stamp set that combines character strings such as "good work," "thank you," and "lunch" with corresponding illustrations, providing an original emoticon set that reflects the user's emotions.

[1031] Use of stamps

[1032] Users can download the completed emoji set and use it in their communication apps, allowing them to use the new stickers to communicate more personally and emotionally.

[1033] Examples:

[1034] Users can choose a "Thank you for your hard work" stamp from the stamp set they downloaded and send it to a friend to express their gratitude.

[1035] This system allows users to easily obtain original emoji sets based on their own conversation content and emotions, making everyday communication more personal and emotional.

[1036] The processing flow will be explained below.

[1037] Step 1:

[1038] The user enables the conversation log collection function in the communication app, which starts the collection of conversation data.

[1039] Step 2:

[1040] The device periodically retrieves and locally stores the user's conversation data, which includes text messages, voice messages, and other communication information.

[1041] Step 3:

[1042] The device periodically sends the saved conversation data to the server, where it is converted into a database format and uploaded via a network connection.

[1043] Step 4:

[1044] The server starts processing to analyze the received conversation data. The server performs text analysis using a natural language processing library.

[1045] Step 5:

[1046] The server tokenizes the conversation data and removes stop words, making it easier to analyze.

[1047] Step 6:

[1048] The server uses the tokenized data to calculate frequency and identify frequently occurring words and phrases.

[1049] Step 7:

[1050] The server further analyzes the conversation data and uses an emotion engine to recognize the user's emotions. The emotion engine uses keywords and contextual analysis to identify the user's emotions.

[1051] Step 8:

[1052] The server generates string data based on the identified frequently occurring words and phrases and the recognized emotions.

[1053] Step 9:

[1054] The server uses machine learning models to automatically generate illustrations that correspond to frequently used words and phrases and the user's emotions, and combines the generated illustrations with text data.

[1055] Step 10:

[1056] The server combines multiple illustrations and text data to create an emoji set for the communication app. Each emoji contains text data and a corresponding illustration.

[1057] Step 11:

[1058] The server sends the completed emoji set to the user's device, where the user can download it within the communication app.

[1059] Step 12:

[1060] Users can download and install the emoji set and use it within their communication app. The new emoji set can be used to make communication more personal and emotional.

[1061] Examples:

[1062] When chatting with a friend, a user sends and receives conversations such as "Good work," "Thank you," and "What are we having for lunch today?" This conversation data is saved on the device and sent to a server. The server analyzes the data using natural language processing and an emotion engine, and automatically generates phrases that express positive emotions, such as "Good work," and corresponding illustrations. As a result, the user can download and use stamp sets that include character strings such as "Good work," "Thank you," and "Lunch," along with the corresponding illustrations.

[1063] Example 2

[1064] 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."

[1065] In conventional communication applications, the emoji sets used by users are fixed and do not reflect the user's actual conversation content or emotions. This limits personalized and emotional communication, making it difficult for users to accurately express themselves. Furthermore, the limited selection of existing emojis leads to a standardized user experience and a lack of individuality.

[1066] 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.

[1067] In this invention, the server includes means for collecting conversation data, means for analyzing the conversation data to identify frequently occurring words and phrases, means for analyzing the conversation data to recognize the user's emotions, means for generating character string data based on the identified frequently occurring words and phrases and the recognized emotions, means for generating illustrations corresponding to the character string data, and means for creating an emoji set for a communication application by combining the character string data and illustrations. This enables users to easily obtain an original emoji set based on the content of their own conversations and emotions, making everyday communication more personalized and emotionally rich.

[1068] "Conversation Data" is data including text messages, voice messages, and other communications sent and received by a user within a communication application.

[1069] "Means of collection" refers to the methods and functions by which a user's terminal or system collects conversation data automatically or at the user's direction.

[1070] "Means of analysis" refers to methods and technologies for analyzing conversation data and extracting meaning and information. Specifically, this includes analysis using natural language processing technology.

[1071] "Frequent words and phrases" refer to words and expressions that appear particularly frequently in the collected conversational data.

[1072] "Means for recognizing emotions" refers to methods and technologies for analyzing conversation data and identifying user emotions, including emotion engines and context analysis.

[1073] "String data" refers to text-based information generated from analyzed conversation data.

[1074] "Means for generating illustrations" refers to methods and technologies for automatically creating visual images based on text data and emotional information. Specifically, they use machine learning models.

[1075] "Pictogram set" refers to a collection of graphical representations for communication applications created by combining specified character string data and corresponding illustrations.

[1076] MODE FOR CARRYING OUT THE INVENTION

[1077] The present invention relates to a system that automatically generates emoji sets by collecting and analyzing conversation data from users in a communication app, identifying frequently used words and phrases, and combining this with an emotion engine that recognizes the user's emotions. This system includes a user terminal, a server, and communication means for transmitting and receiving data between them.

[1078] Data collection

[1079] The device collects the user's conversation data, including text and voice messages sent and received within the communication app. The device periodically retrieves this conversation data and stores it locally.

[1080] Examples:

[1081] For example, when a user chats with a friend and exchanges messages such as "Good work," "Thank you," and "What are you having for lunch today?", this conversation data is saved on the device.

[1082] Data transmission

[1083] The device periodically sends the saved conversation data to the server, where it is converted into a database format and uploaded via a network connection.

[1084] Examples:

[1085] The device compiles conversation logs over a month, converts them into a database format, and sends the data to a server using a secure protocol such as SSL / TLS.

[1086] Data analysis

[1087] The server analyzes the received conversation data to identify frequently occurring words and phrases, and uses natural language processing (NLP) technology to tokenize the text, remove stop words, and then extract frequently occurring words.

[1088] Examples:

[1089] The server tokenizes the conversation data using a morphological analysis engine and extracts frequently occurring phrases such as "good work," "thank you," and "lunch."

[1090] Emotion recognition by emotion engine

[1091] The server further analyzes the conversation data to recognize the user's emotions. The emotion engine uses keywords and contextual analysis to identify the emotions the user is showing during the conversation.

[1092] Examples:

[1093] The server analyzes phrases like "Otsukaresama" (thank you) and recognizes that it indicates a positive emotion, as well as "arigato" (thank you) as an emotion that conveys gratitude.

[1094] Generating string data and illustrations

[1095] The server generates text data based on the frequently used words and phrases and the user's emotions, and then automatically generates illustrations corresponding to the text data. The illustration generation utilizes machine learning models (such as GAN and VQ-VAE).

[1096] Examples:

[1097] The server associates a positive emotion with the string data "good work" and generates an illustration of a cat character.

[1098] Creating an emoji set

[1099] The server combines the generated text data with illustrations to create an emoji set for communication apps, which also includes emotional information.

[1100] Examples:

[1101] The server creates a stamp set that combines character strings such as "good work," "thank you," and "lunch" with corresponding illustrations, providing an original emoticon set that reflects the user's emotions.

[1102] Use of stamps

[1103] Users can download the completed emoji set and use it in their communication apps, allowing them to communicate in a personalized and emotional way using the new stickers.

[1104] Examples:

[1105] Users can choose a "Thank you for your hard work" stamp from the stamp set they downloaded and send it to a friend to express their gratitude.

[1106] This allows users to easily obtain original emoji sets based on their own conversation content and emotions, making everyday communication more personal and emotional.

[1107] Examples of prompt statements

[1108] An example of a prompt for a generative AI model is:

[1109] Generate illustrations of stamps that reflect phrases frequently used by users in communication apps, such as "Thank you" and "Thank you," and the emotions associated with them (positive, grateful).

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

[1111] Step 1:

[1112] The terminal collects user conversation data.

[1113] Input: Text and voice messages that users send and receive within communication apps.

[1114] How it works: A program runs to periodically save conversation data to the device's local storage, and the user sends and receives messages such as "Good work," "Thank you," and "What are you having for lunch today?" in chats with friends.

[1115] Output: Conversation data stored in the device's local storage.

[1116] Step 2:

[1117] The conversation data stored on the device is periodically sent to the server.

[1118] Input: Conversation data stored in local storage.

[1119] How it works: Data is first converted into a database format and then uploaded to the server over a network connection using SSL / TLS protocol.

[1120] Output: Conversation data uploaded to the server.

[1121] Step 3:

[1122] The server analyzes the conversation data it receives and identifies frequently used words and phrases.

[1123] Input: Conversation data uploaded to the server.

[1124] How it works: It uses natural language processing techniques to tokenize text data and remove stop words, then uses a morphological analysis engine to extract frequently occurring words and phrases.

[1125] Output: A list of frequently occurring words and phrases.

[1126] Step 4:

[1127] The server further analyzes the conversation data and recognizes the user's emotions.

[1128] Input: A list of frequently used words and phrases.

[1129] How it works: Uses a sentiment engine to identify user sentiment based on keywords and contextual analysis, for example recognizing positive sentiment from the phrase "Otsukaresama."

[1130] Output: Emotion recognition result.

[1131] Step 5:

[1132] The server generates character string data based on the identified frequently occurring words and phrases and the recognized emotions.

[1133] Input: list of frequently occurring words and phrases and emotion recognition results.

[1134] How it works: The server combines this information to generate string data. For example, it creates string data associated with the positive emotion "Thank you for your hard work."

[1135] Output: String data.

[1136] Step 6:

[1137] The server generates an illustration corresponding to the character string data.

[1138] Input: String data.

[1139] How it works: Using machine learning models (e.g., GAN or VQ-VAE), the system automatically generates illustrations corresponding to string data. Specifically, it generates an illustration of a cat character that reflects a positive emotion for the string data "Thank you for your hard work."

[1140] Output: Illustration.

[1141] Step 7:

[1142] The server combines the generated string data with illustrations to create an emoji set for communication apps.

[1143] Input: String data and illustrations.

[1144] What it does: It converts text and corresponding illustrations into the correct size and format, and compiles them into an emoji set.

[1145] Output: Emoji set.

[1146] Step 8:

[1147] Users can download the completed emoji set and use it within their communication app.

[1148] Input: A set of emojis provided by the server.

[1149] How it works: A set of emojis is downloaded from a user's device and used within a communication app. For example, a user sends a "Thank you for your hard work" sticker to a friend.

[1150] Output: The emoji set used within the user's communication app.

[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 headset type terminal 314 will be referred to as a "terminal."

[1153] The challenge is to improve customer experience by enriching customer communication in virtual stores and providing reactions that match the customer's emotions. Conventional systems have difficulty in properly reflecting emotions in communication between customer service staff and customers, and lack real-time responses.

[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 collecting conversation data, means for analyzing the collected conversation data to identify frequently occurring words and phrases, means for generating character string data based on the identified frequently occurring words and phrases, means for generating illustrations corresponding to the character string data, means for creating an emoji set for a communication application by combining the character string data and illustrations, means for collecting and analyzing customer conversation data in a virtual store to identify customer emotions, and means for generating reaction illustrations based on the identified emotions, thereby enabling communication with customers to be more emotionally rich and in real time.

[1156] "Conversation Data" is a record of text and voice messages sent and received by a user within a communication application.

[1157] "Frequent words and phrases" are words and expressions that appear particularly frequently in conversation data.

[1158] "Character string data" is text data that includes frequently occurring words and phrases extracted from the analyzed conversation data.

[1159] An "illustration" is a graphic representation of an image or pictogram that corresponds to the character string data.

[1160] The "emoticon set for communication applications" is a collection of emoticons that users can use in communication applications, which combines character string data with corresponding illustrations.

[1161] A "virtual store" is an online virtual store space that offers products and services via the Internet.

[1162] "Customer service conversation data" is a record of communication between customer service staff and customers within the virtual store.

[1163] "Emotions" are psychological states and feelings that are analyzed from users' statements and messages.

[1164] "Reaction illustrations" are responsive images or emojis generated based on identified emotions.

[1165] The present invention relates to a system for improving customer service communication in a virtual store. This system is implemented using the following hardware and software.

[1166] Hardware Configuration

[1167] Server: A central computer system that collects and analyzes conversation data, manages the database, and creates the emoji set.

[1168] Device: A device used by a customer or staff member, such as smart glasses or a tablet.

[1169] Software Configuration

[1170] Natural language processing technology: Hugging Face's pipeline is used to perform tokenization, stop word removal, and sentiment analysis.

[1171] Machine learning model: PIL (Python Imaging Library) is used to automatically generate illustrations, which are then given characteristics such as background color based on emotions.

[1172] Database management software: Stores conversation data and manages it so that it can be retrieved when needed.

[1173] Processing flow overview

[1174] 1. Collecting conversation data

[1175] The device collects conversation data in real time within the virtual store. The content of conversations during customer service is automatically recorded and sent to the server at regular intervals.

[1176] 2. Data Analysis

[1177] The server then analyzes the received conversation data using natural language processing techniques, which include tokenization, stop word removal, and extraction of frequently occurring words and phrases.

[1178] 3. Emotion recognition

[1179] The emotion engine recognizes the user's emotions based on frequently occurring words and phrases identified from the conversation data, performs context analysis, and assigns emotion labels.

[1180] 4. Generating reaction illustrations

[1181] Based on the identified emotion, a machine learning model is used to generate reaction illustrations, for example, a light background for positive emotions and a dark background for negative emotions.

[1182] Specific examples

[1183] When a customer says, "I really like this product!" in a virtual store, the conversation data is collected by the device and analyzed by the server. This comment is recognized as a positive emotion, and a corresponding reaction illustration (e.g., a smiling face on a bright background) is generated. This reaction illustration is then displayed to the customer in real time, improving the quality of customer service.

[1184] Prompt Sentence Examples

[1185] The user says "I absolutely love this product!" about a product they're interested in. Analyze the sentiment of this message and generate an illustration for the response based on the positive sentiment.

[1186] This invention is expected to improve customer experience by making customer service in virtual stores more emotional and real-time.

[1187] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1188] Step 1:

[1189] Conversation data collection

[1190] The device collects conversation data in real time within the virtual store, including text messages and voice messages between the user and the staff, and temporarily stores the collected conversation data locally.

[1191] Input: Conversation data (text or voice) in the virtual store

[1192] Output: Locally stored conversation data

[1193] Step 2:

[1194] Sending conversation data

[1195] At specified intervals, the device transmits locally stored conversation data to the server using a secure communication protocol.

[1196] Input: Locally stored conversation data

[1197] Output: Conversation data uploaded to the server

[1198] Step 3:

[1199] Data analysis

[1200] The server analyzes the received conversation data using natural language processing techniques, including tokenization, stop word removal, and extraction of frequently occurring words and phrases. The server then stores the analysis results in a database.

[1201] Input: Conversation data uploaded to the server

[1202] Output: Analysis result data including frequently occurring words and phrases

[1203] Step 4:

[1204] emotion recognition

[1205] The server uses an emotion engine to recognize the user's emotions based on the analyzed frequently used words and phrases. The emotion engine performs contextual analysis and assigns emotion labels. This information is also stored in a database.

[1206] Input: Analysis result data (frequently used words and phrases)

[1207] Output: Data with emotion labels

[1208] Step 5:

[1209] Generate reaction illustrations

[1210] The server uses machine learning models to generate reaction illustrations based on the identified emotion labels. For example, positive emotions are represented by illustrations with bright backgrounds. The generated illustrations are also stored in a database.

[1211] Input: Data with emotion labels

[1212] Output: Reaction illustration

[1213] Step 6:

[1214] Customer feedback

[1215] The device receives the reaction illustrations generated from the server and displays them to the customer in real time, making communication during customer service more emotional and interactive.

[1216] Input: Reaction illustration sent from the server

[1217] Output: Reaction illustrations displayed to the customer

[1218] Step 7:

[1219] Improved data storage and feedback

[1220] The server stores information such as past conversation data, analysis results, emotion labels, and reaction illustrations in a database and periodically retrains the machine learning model to improve the quality of feedback.

[1221] Input: All collected and analyzed data (conversation data, emotion labels, reaction illustrations)

[1222] Output: Improved feedback system

[1223] Through these steps, communication with customers in the virtual store becomes more emotional and real-time, improving the customer experience.

[1224] 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.

[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 headset type terminal 314.

[1227] [Fourth embodiment]

[1228] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1229] 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.

[1230] 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).

[1231] 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.

[1232] 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.

[1233] 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).

[1234] 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.

[1235] 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.

[1236] 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.

[1237] 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.

[1238] 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.

[1239] 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.

[1240] 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."

[1241] The invention described in this specification relates to a system that collects and analyzes conversation data from users in a communication app to identify frequently occurring words and phrases, and automatically generates an emoji set based on the collected data. This system includes a user terminal, a server, and a communication means for transmitting and receiving data between them.

[1242] Data collection

[1243] The device collects the user's conversation data, including text and voice messages sent and received within the communication app. The device periodically retrieves this conversation data and stores it locally.

[1244] Examples:

[1245] When a user chats with a friend, they exchange messages such as "Good work," "Thank you," and "What are you having for lunch today?" This conversation data is stored on the device.

[1246] Data transmission

[1247] The device sends the saved conversation data to the server, which then converts the conversation data into a database format and uploads it to the server via a network connection.

[1248] Examples:

[1249] The device compiles conversation logs from over a month, converts them into a database format, and sends them to the server.

[1250] Data analysis

[1251] The server analyzes the received conversation data to identify frequently occurring words and phrases, and then uses natural language processing technology to analyze the text, tokenize it, remove stop words, and extract frequently occurring words.

[1252] Examples:

[1253] The server analyzes the conversation data and extracts frequently used phrases such as "good work," "thank you," and "lunch."

[1254] Generating string data and illustrations

[1255] The server generates character strings based on the frequently occurring words and phrases identified, and automatically generates illustrations corresponding to the character strings. The illustration generation utilizes machine learning models.

[1256] Examples:

[1257] The server generates the string data "Thank you for your hard work" and uses a machine learning model to generate a matching illustration of a cat character.

[1258] Creating an emoji set

[1259] The server combines the generated character string data with illustrations to create an emoji set for the communication app, which is then sent to the user's device.

[1260] Examples:

[1261] The server creates a stamp set that combines character strings such as "Good work," "Thank you," and "Lunch" with corresponding illustrations, and provides it to the user.

[1262] Use of stamps

[1263] Users can download the completed emoji set and use it in their communication apps, allowing them to use the new stickers to make their communications more personal and engaging.

[1264] Examples:

[1265] Users can choose a "Thank you for your hard work" stamp from the stamp set they downloaded and send it to a friend to express their gratitude.

[1266] The system allows users to easily create their own original emoji sets based on frequently used phrases and words, making everyday communication more personal and fun.

[1267] The processing flow will be explained below.

[1268] Step 1:

[1269] The user enables the conversation log collection function in the communication app, which starts the collection of conversation data.

[1270] Step 2:

[1271] The device periodically retrieves and locally stores the user's conversation data, which includes text messages, voice messages, and other communication information.

[1272] Step 3:

[1273] The device periodically sends the saved conversation data to the server, where it is converted into a database format and uploaded via a network connection.

[1274] Step 4:

[1275] The server starts processing to analyze the received conversation data. The server performs text analysis using a natural language processing library.

[1276] Step 5:

[1277] The server tokenizes the conversation data and removes stop words, making it easier to analyze.

[1278] Step 6:

[1279] The server uses the tokenized data to calculate frequency and identify frequently occurring words and phrases.

[1280] Step 7:

[1281] The server generates string data based on frequently occurring words and phrases. For example, frequently occurring phrases such as "Thank you for your hard work" are extracted.

[1282] Step 8:

[1283] The server uses machine learning models to automatically generate illustrations corresponding to frequently occurring words and phrases, and combines the generated illustrations with text data.

[1284] Step 9:

[1285] The server combines multiple illustrations and text data to create an emoji set for the communication app. Each emoji contains text data and a corresponding illustration.

[1286] Step 10:

[1287] The server sends the completed emoji set to the user's device, where the user can download it within the communication app.

[1288] Step 11:

[1289] Users can download and install the emoji set and use it within their communication apps. The new emoji set can be used to make communications more personal and engaging.

[1290] The above are the specific processing steps of the system program. Through this process, users can easily obtain and use original emoji sets based on their own conversation content.

[1291] Example 1

[1292] 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."

[1293] In recent years, the use of communication apps has increased, and users are sending and receiving many messages. However, existing emoji and stamp sets are fixed and do not correspond to the individual conversation content of users. Therefore, there is a need for personalized and more engaging means of communication. In particular, there is a need for a method to easily generate emoji that correspond to frequently used phrases and words.

[1294] 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.

[1295] In this invention, the server includes means for analyzing conversation data to identify frequently occurring words and phrases, means for generating character string data based on the identified frequently occurring words and phrases, and means for using a machine learning model to generate illustrations corresponding to the generated character string data. This allows users to easily create original emoji sets based on the content of their own conversations, making everyday communication more personalized and enjoyable.

[1296] A "terminal" is an electronic device for executing a communication app used by a user, and includes a smartphone, tablet, PC, etc.

[1297] A "server" is a central computer system that receives, processes, and stores data sent from terminals.

[1298] "Conversation data" is a data set that includes text and voice messages sent and received by a user within a communication app.

[1299] A "database format" is a format for storing structured data, and generally includes formats such as JSON and CSV.

[1300] "Natural language processing technology" is a technology that enables computers to understand, interpret, and generate natural language, and includes text tokenization, stop word removal, and frequent word extraction.

[1301] "Tokenization" is the process of breaking text into individual words or phrases.

[1302] "Stop words" are frequently used words that contain little information and are excluded from analysis (for example, "no," "ni," and "o" in Japanese).

[1303] "Frequent words" are words or phrases that are used particularly frequently in conversation data.

[1304] A "machine learning model" is an algorithm that learns patterns and rules based on data and makes predictions or generates data based on new data.

[1305] A "prompt" is an instruction sentence input to a generative AI model, containing instructions for generating illustrations or text based on specific conditions or requests.

[1306] An "emoji set" is a collection of emojis and / or stickers for use by a user within a communication app.

[1307] The present invention relates to a system that collects and analyzes conversation data from users in a communication app to identify frequently occurring words and phrases, and automatically generates emoji sets based on the collected data. The system includes a user terminal, a server, and a communication means for transmitting and receiving data between them.

[1308] Data collection

[1309] First, the device collects the user's conversation data. This conversation data includes text messages and voice messages sent and received by the user within the communication app. The device periodically obtains the conversation data using the communication app's API and saves it in local storage. Specifically, the device calls the communication app's API at 2:00 AM every day to obtain the previous day's conversation data. This data includes messages such as "Good work," "Thank you," and "What are you having for lunch today?"

[1310] Data transmission

[1311] Next, the device sends the saved conversation data to the server. The device converts the conversation data into a database format, which is typically JSON or CSV. For example, the device converts the saved conversation data at the end of the month into JSON format, and represents each message as a JSON object. The converted data is then uploaded to the server using the HTTPS protocol.

[1312] Data analysis

[1313] The server receives the data from the device and stores it in a database for analysis. The server uses natural language processing techniques such as Python's NLTK library to tokenize the conversation data, remove stop words, and extract frequently occurring words. For example, it extracts frequently occurring phrases such as "good work," "thank you," and "lunch."

[1314] Generating string data and illustrations

[1315] The server generates character string data based on the frequently occurring words and phrases identified, and automatically generates corresponding illustrations using a machine learning model. At this time, a prompt sentence is input into the generative AI model to generate the illustration. Specifically, the prompt sentence "Please generate an illustration of a cat character saying 'Good job'" is used to generate the corresponding illustration.

[1316] Creating an emoji set

[1317] The server combines the character string data with the generated illustrations to create an emoji set for the communication app. This emoji set is then sent to the user's device. For example, a stamp set is created by combining the character strings "Good work," "Thank you," and "Lunch" with their corresponding illustrations, and sent to the device using the HTTPS protocol.

[1318] Use of stamps

[1319] The user downloads the completed emoji set and uses it in the communication app. The user clicks on the new stamp set to download it, which is then saved in the specified folder in the communication app. For example, a user can select the "Thank you for your hard work" stamp on the chat screen and send it to a friend to express their gratitude.

[1320] In this way, the system allows users to easily create original emoji sets based on frequently used phrases and words, making everyday communication more personal and fun.

[1321] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1322] Program processing steps

[1323] Step 1:

[1324] A user has a conversation within a communication app. This conversation data is collected as it is entered into the device. The device calls the communication app's API and saves the acquired data in local storage. This allows text messages and voice messages to be collected.

[1325] Specific behavior:

[1326] The device calls the communication app's API at 2:00 AM every day, obtains the previous day's conversation data (e.g., "Good work," "Thank you," "What are you having for lunch today?") and stores it in an SQLite database.

[1327] Input: User conversation data (text messages, voice messages)

[1328] Output: Locally stored conversation data

[1329] Step 2:

[1330] The conversation data collected by the device is converted into a database format (JSON). The converted data is then sent from the device to the server. This transmission process uses the HTTPS protocol.

[1331] Specific behavior:

[1332] At the end of the month, the device converts the saved conversation data into JSON format, representing each message as a JSON object, and then uploads it to the server using the HTTPS protocol.

[1333] Input: Locally stored conversation data

[1334] Output: JSON data uploaded to the server

[1335] Step 3:

[1336] The server receives the uploaded JSON data and stores it in a database for analysis. This data is then analyzed using natural language processing techniques to identify frequently occurring words and phrases. Specific analysis steps include tokenization, stop word removal, and frequent word extraction.

[1337] Specific behavior:

[1338] The server uses Python's NLTK library to tokenize the conversation data, remove stop words, count frequently occurring words, and extract frequently occurring phrases such as "good work," "thank you," and "lunch."

[1339] Input: JSON data uploaded to the server

[1340] Output: Identified frequently occurring words and phrases

[1341] Step 4:

[1342] The server generates text data based on the identified frequently occurring words and phrases, which is then used as prompts for the generative AI model to generate illustrations.

[1343] Specific behavior:

[1344] The server generates the phrases "Thank you for your hard work," "Thank you," and "Lunch" as string data and prepares them as a prompt sentence.

[1345] Input: Identified common words and phrases

[1346] Output: String data (prompt statement)

[1347] Step 5:

[1348] The server inputs prompts into the generative AI model, which then automatically generates corresponding illustrations that correspond to the user's conversation.

[1349] Specific behavior:

[1350] The server inputs the prompt "Please generate an illustration of a cat character saying 'Good job'" into the generative AI model and generates the corresponding illustration.

[1351] Input: prompt statement

[1352] Output: Generated illustration

[1353] Step 6:

[1354] The server combines the generated character string data with illustrations to create an emoji set for the communication app, which is then sent to the user's device.

[1355] Specific behavior:

[1356] The server creates a stamp set by combining the strings "Good work," "Thank you," and "Lunch" with corresponding illustrations, and sends it to the terminal using the HTTPS protocol.

[1357] Input: String data and generated illustrations

[1358] Output: The emoji set sent to the user's device

[1359] Step 7:

[1360] Users can download the completed emoji set to their device and use it in communication apps. Specifically, users communicate by selecting and sending new stamps.

[1361] Specific behavior:

[1362] Users can select the "Thank you for your hard work" stamp on the chat screen and send it to their friends to express their gratitude.

[1363] Input: The emoji set sent to the user's device

[1364] Output: Stamps used in communication apps

[1365] (Application example 1)

[1366] 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."

[1367] Conventional advertising methods have difficulty accurately grasping the hobbies and preferences of their target users and personalizing their advertising, which often results in unattractive ads being displayed to users, reducing the effectiveness of advertising and lowering marketing efficiency.

[1368] 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.

[1369] In this invention, the server includes means for collecting conversation data, means for analyzing the collected conversation data to identify frequently occurring words and phrases, means for automatically generating advertising content by combining character string data and illustrations based on the identified frequently occurring words and phrases, and means for displaying the generated advertising content, thereby enabling the generation of personalized advertisements based on user conversation data.

[1370] - "Conversation data" refers to text and voice message data sent and received by a user within a communication app.

[1371] The "analysis means" is a means used to analyze conversation data and identify frequently occurring words and phrases.

[1372] "Frequent words and phrases" are important words and phrases that appear with a certain frequency or more in the collected conversation data.

[1373] "Character string data" is text data generated based on identified frequently occurring words and phrases.

[1374] The "illustration generating means" is a means used to generate an illustration corresponding to character string data.

[1375] "Advertising content" is marketing content generated by combining character string data and illustrations.

[1376] The "generation means" is a means for generating character string data and illustrations based on the identified frequently occurring words and phrases.

[1377] The "display means" is a means for displaying the generated advertising content on a user terminal.

[1378] "Natural language processing technology" is a technology for tokenizing conversational data, removing stop words, and extracting frequently occurring words and phrases.

[1379] A "machine learning model" is an algorithm or model that is trained based on data and used to automatically generate illustrations.

[1380] This invention is a system that automatically generates advertising content by collecting conversation data, analyzing it to identify frequently occurring words and phrases, and generating character string data and illustrations. This system includes a user terminal, a server, and a communication means for transmitting and receiving data between them.

[1381] First, the user's device collects conversation data. This conversation data includes text and voice messages that the user sends daily within communication apps. For example, conversations such as "I want to watch this movie," "I've been wanting to get a new smartphone lately," and "What should we have for lunch today?" are collected. The collected data is temporarily stored on the device, then converted into a database format and sent to the server.

[1382] The server then analyzes the received conversation data using natural language processing (NLP) techniques, tokenizing the conversation data, removing stop words, and extracting frequently occurring words and phrases. For example, the server identifies frequently occurring phrases such as "movie," "smartphone," and "lunch."

[1383] The server then generates character strings based on the frequently occurring words and phrases identified, and generates corresponding illustrations. Machine learning models are used to generate the illustrations, so for example, an illustration of a character holding popcorn is automatically generated for "movie," and an illustration of a character holding a smartphone is automatically generated for "smartphone."

[1384] Next, the server combines the generated text data with illustrations to create advertising content. This advertising content is sent to the user's device, where the user can view it on their smartphone app. By using this new advertising content, users can view personalized ads based on their interests and hobbies. For example, ads such as "Special discount! Great chance to see movies at a great price!" or "New smartphones on sale now!" are displayed.

[1385] The system also automatically generates advertising content by inputting prompts like the following into a generative AI model:

[1386] "You will create a program that extracts frequently used phrases and keywords from user conversation data and generates personalized advertising content based on that data. The collected data will be sent to a server in JSON format, where it will be analyzed. Targeted ads will be generated based on the analysis results, and will ultimately be displayed on the smartphone app."

[1387] As a specific example of use, when a user says something like "I want to watch this movie," the system analyzes it and automatically generates an advertisement that says "Special discount! Movies on sale now!" as completely new content and provides it to the user.

[1388] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1389] Step 1:

[1390] The user's device collects conversation data. In this step, the target is text and voice messages sent and received by the user within the communication app. The input is data from the user's daily chats and messages, and the output is the collected conversation data that is temporarily stored on the device.

[1391] Step 2:

[1392] The conversation data collected by the device is converted into a database format. In this step, the collected text and voice messages are organized and converted into a database format. The input is the collected conversation data, and the output is the converted database format data.

[1393] Step 3:

[1394] The terminal sends the converted database format conversation data to the server. In this step, data is sent from the terminal to the server via the network. The input is the conversation data converted to database format, and the output is the data received on the server side.

[1395] Step 4:

[1396] The server analyzes the received conversation data. In this step, natural language processing techniques are used to tokenize the conversation data, remove stop words, and extract frequently occurring words and phrases. The input is the conversation data, and the output is frequently occurring words and phrases. The specific operation uses the Python sklearn library.

[1397] Step 5:

[1398] The server generates string data based on the identified frequently occurring words and phrases. In this step, text corresponding to the extracted words and phrases is generated. The input is the frequently occurring words and phrases, and the output is the generated string data.

[1399] Step 6:

[1400] The server generates an illustration corresponding to the string data. In this step, the corresponding illustration is automatically generated using a machine learning model. The input is the string data, and the generated illustration is obtained as the output. Specific operations use PyTorch and TensorFlow.

[1401] Step 7:

[1402] The server combines the generated text data with the illustration to create advertising content. In this step, marketing content is generated by combining the two. The input is the text data and the corresponding illustration, and the output is the completed advertising content.

[1403] Step 8:

[1404] The server transmits the generated advertisement content to the user terminal. In this step, the advertisement content is transmitted to the user terminal via the network. The advertisement content is the input, and the advertisement content received by the user terminal is obtained as the output.

[1405] Step 9:

[1406] The user checks the advertising content on the smartphone app. In this step, the user can view personalized advertising. The input is the received advertising content, and the output is the displayed advertising content. For example, a message such as "Special discount! A great chance to see a movie at a great price!" may be displayed.

[1407] 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.

[1408] The present invention relates to a system that automatically generates emoji sets by collecting and analyzing conversation data from users in a communication app, identifying frequently used words and phrases, and combining this with an emotion engine that recognizes the user's emotions. This system includes a user terminal, a server, and communication means for transmitting and receiving data between them.

[1409] Data collection

[1410] The device collects the user's conversation data, including text and voice messages sent and received within the communication app. The device periodically retrieves this conversation data and stores it locally.

[1411] Examples:

[1412] When a user chats with a friend, they exchange messages such as "Good work," "Thank you," and "What are you having for lunch today?" This conversation data is stored on the device.

[1413] Data transmission

[1414] The device periodically sends the saved conversation data to the server, where it is converted into a database format and uploaded via a network connection.

[1415] Examples:

[1416] The device compiles conversation logs from over a month, converts them into a database format, and sends them to the server.

[1417] Data analysis

[1418] The server analyzes the received conversation data to identify frequently occurring words and phrases, and then uses natural language processing technology to analyze the text, tokenize it, remove stop words, and extract frequently occurring words.

[1419] Examples:

[1420] The server analyzes the conversation data and extracts frequently used phrases such as "good work," "thank you," and "lunch."

[1421] Emotion recognition by emotion engine

[1422] The server further analyzes the conversation data to recognize the user's emotions. The emotion engine uses keywords and contextual analysis to identify the emotions the user is expressing during the conversation.

[1423] Examples:

[1424] The server analyzes phrases like "Otsukaresama" (thank you) and recognizes that it indicates a positive emotion, as well as "arigato" (thank you) as an emotion that conveys gratitude.

[1425] Generating string data and illustrations

[1426] The server generates character strings based on the frequently used words and phrases and the user's emotions, and automatically generates illustrations corresponding to the character strings. The illustration generation utilizes a machine learning model.

[1427] Examples:

[1428] The server associates a positive emotion with the string data "Thank you for your hard work" and generates a matching cat character illustration using a machine learning model.

[1429] Creating an emoji set

[1430] The server combines the generated text data with illustrations to create an emoji set for communication apps, which also includes emotional information.

[1431] Examples:

[1432] The server creates a stamp set that combines character strings such as "good work," "thank you," and "lunch" with corresponding illustrations, providing an original emoticon set that reflects the user's emotions.

[1433] Use of stamps

[1434] Users can download the completed emoji set and use it in their communication apps, allowing them to use the new stickers to communicate more personally and emotionally.

[1435] Examples:

[1436] Users can choose a "Thank you for your hard work" stamp from the stamp set they downloaded and send it to a friend to express their gratitude.

[1437] This system allows users to easily obtain original emoji sets based on their own conversation content and emotions, making everyday communication more personal and emotional.

[1438] The processing flow will be explained below.

[1439] Step 1:

[1440] The user enables the conversation log collection function in the communication app, which starts the collection of conversation data.

[1441] Step 2:

[1442] The device periodically retrieves and locally stores the user's conversation data, which includes text messages, voice messages, and other communication information.

[1443] Step 3:

[1444] The device periodically sends the saved conversation data to the server, where it is converted into a database format and uploaded via a network connection.

[1445] Step 4:

[1446] The server starts processing to analyze the received conversation data. The server performs text analysis using a natural language processing library.

[1447] Step 5:

[1448] The server tokenizes the conversation data and removes stop words, making it easier to analyze.

[1449] Step 6:

[1450] The server uses the tokenized data to calculate frequency and identify frequently occurring words and phrases.

[1451] Step 7:

[1452] The server further analyzes the conversation data and uses an emotion engine to recognize the user's emotions. The emotion engine uses keywords and contextual analysis to identify the user's emotions.

[1453] Step 8:

[1454] The server generates string data based on the identified frequently occurring words and phrases and the recognized emotions.

[1455] Step 9:

[1456] The server uses machine learning models to automatically generate illustrations that correspond to frequently used words and phrases and the user's emotions, and combines the generated illustrations with text data.

[1457] Step 10:

[1458] The server combines multiple illustrations and text data to create an emoji set for the communication app. Each emoji contains text data and a corresponding illustration.

[1459] Step 11:

[1460] The server sends the completed emoji set to the user's device, where the user can download it within the communication app.

[1461] Step 12:

[1462] Users can download and install the emoji set and use it within their communication app. The new emoji set can be used to make communication more personal and emotional.

[1463] Examples:

[1464] When chatting with a friend, a user sends and receives conversations such as "Good work," "Thank you," and "What are we having for lunch today?" This conversation data is saved on the device and sent to a server. The server analyzes the data using natural language processing and an emotion engine, and automatically generates phrases that express positive emotions, such as "Good work," and corresponding illustrations. As a result, the user can download and use stamp sets that include character strings such as "Good work," "Thank you," and "Lunch," along with the corresponding illustrations.

[1465] Example 2

[1466] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1467] In conventional communication applications, the emoji sets used by users are fixed and do not reflect the user's actual conversation content or emotions. This limits personalized and emotional communication, making it difficult for users to accurately express themselves. Furthermore, the limited selection of existing emojis leads to a standardized user experience and a lack of individuality.

[1468] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1469] In this invention, the server includes means for collecting conversation data, means for analyzing the conversation data to identify frequently occurring words and phrases, means for analyzing the conversation data to recognize the user's emotions, means for generating character string data based on the identified frequently occurring words and phrases and the recognized emotions, means for generating illustrations corresponding to the character string data, and means for creating an emoji set for a communication application by combining the character string data and illustrations. This enables users to easily obtain an original emoji set based on the content of their own conversations and emotions, making everyday communication more personalized and emotionally rich.

[1470] "Conversation Data" is data including text messages, voice messages, and other communications sent and received by a user within a communication application.

[1471] "Means of collection" refers to the methods and functions by which a user's terminal or system collects conversation data automatically or at the user's direction.

[1472] "Means of analysis" refers to methods and technologies for analyzing conversation data and extracting meaning and information. Specifically, this includes analysis using natural language processing technology.

[1473] "Frequent words and phrases" refer to words and expressions that appear particularly frequently in the collected conversational data.

[1474] "Means for recognizing emotions" refers to methods and technologies for analyzing conversation data and identifying user emotions, including emotion engines and context analysis.

[1475] "String data" refers to text-based information generated from analyzed conversation data.

[1476] "Means for generating illustrations" refers to methods and technologies for automatically creating visual images based on text data and emotional information. Specifically, they use machine learning models.

[1477] "Pictogram set" refers to a collection of graphical representations for communication applications created by combining specified character string data and corresponding illustrations.

[1478] MODE FOR CARRYING OUT THE INVENTION

[1479] The present invention relates to a system that automatically generates emoji sets by collecting and analyzing conversation data from users in a communication app, identifying frequently used words and phrases, and combining this with an emotion engine that recognizes the user's emotions. This system includes a user terminal, a server, and communication means for transmitting and receiving data between them.

[1480] Data collection

[1481] The device collects the user's conversation data, including text and voice messages sent and received within the communication app. The device periodically retrieves this conversation data and stores it locally.

[1482] Examples:

[1483] For example, when a user chats with a friend and exchanges messages such as "Good work," "Thank you," and "What are you having for lunch today?", this conversation data is saved on the device.

[1484] Data transmission

[1485] The device periodically sends the saved conversation data to the server, where it is converted into a database format and uploaded via a network connection.

[1486] Examples:

[1487] The device compiles conversation logs over a month, converts them into a database format, and sends the data to a server using a secure protocol such as SSL / TLS.

[1488] Data analysis

[1489] The server analyzes the received conversation data to identify frequently occurring words and phrases, and uses natural language processing (NLP) technology to tokenize the text, remove stop words, and then extract frequently occurring words.

[1490] Examples:

[1491] The server tokenizes the conversation data using a morphological analysis engine and extracts frequently occurring phrases such as "good work," "thank you," and "lunch."

[1492] Emotion recognition by emotion engine

[1493] The server further analyzes the conversation data to recognize the user's emotions. The emotion engine uses keywords and contextual analysis to identify the emotions the user is showing during the conversation.

[1494] Examples:

[1495] The server analyzes phrases like "Otsukaresama" (thank you) and recognizes that it indicates a positive emotion, as well as "arigato" (thank you) as an emotion that conveys gratitude.

[1496] Generating string data and illustrations

[1497] The server generates text data based on the frequently used words and phrases and the user's emotions, and then automatically generates illustrations corresponding to the text data. The illustration generation utilizes machine learning models (such as GAN and VQ-VAE).

[1498] Examples:

[1499] The server associates a positive emotion with the string data "good work" and generates an illustration of a cat character.

[1500] Creating an emoji set

[1501] The server combines the generated text data with illustrations to create an emoji set for communication apps, which also includes emotional information.

[1502] Examples:

[1503] The server creates a stamp set that combines character strings such as "good work," "thank you," and "lunch" with corresponding illustrations, providing an original emoticon set that reflects the user's emotions.

[1504] Use of stamps

[1505] Users can download the completed emoji set and use it in their communication apps, allowing them to communicate in a personalized and emotional way using the new stickers.

[1506] Examples:

[1507] Users can choose a "Thank you for your hard work" stamp from the stamp set they downloaded and send it to a friend to express their gratitude.

[1508] This allows users to easily obtain original emoji sets based on their own conversation content and emotions, making everyday communication more personal and emotional.

[1509] Examples of prompt statements

[1510] An example of a prompt for a generative AI model is:

[1511] Generate illustrations of stamps that reflect phrases frequently used by users in communication apps, such as "Thank you" and "Thank you," and the emotions associated with them (positive, grateful).

[1512] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1513] Step 1:

[1514] The terminal collects user conversation data.

[1515] Input: Text and voice messages that users send and receive within communication apps.

[1516] How it works: A program runs to periodically save conversation data to the device's local storage, and the user sends and receives messages such as "Good work," "Thank you," and "What are you having for lunch today?" in chats with friends.

[1517] Output: Conversation data stored in the device's local storage.

[1518] Step 2:

[1519] The conversation data stored on the device is periodically sent to the server.

[1520] Input: Conversation data stored in local storage.

[1521] How it works: Data is first converted into a database format and then uploaded to the server over a network connection using SSL / TLS protocol.

[1522] Output: Conversation data uploaded to the server.

[1523] Step 3:

[1524] The server analyzes the conversation data it receives and identifies frequently used words and phrases.

[1525] Input: Conversation data uploaded to the server.

[1526] How it works: It uses natural language processing techniques to tokenize text data and remove stop words, then uses a morphological analysis engine to extract frequently occurring words and phrases.

[1527] Output: A list of frequently occurring words and phrases.

[1528] Step 4:

[1529] The server further analyzes the conversation data and recognizes the user's emotions.

[1530] Input: A list of frequently used words and phrases.

[1531] How it works: Uses a sentiment engine to identify user sentiment based on keywords and contextual analysis, for example recognizing positive sentiment from the phrase "Otsukaresama."

[1532] Output: Emotion recognition result.

[1533] Step 5:

[1534] The server generates character string data based on the identified frequently occurring words and phrases and the recognized emotions.

[1535] Input: list of frequently occurring words and phrases and emotion recognition results.

[1536] How it works: The server combines this information to generate string data. For example, it creates string data associated with the positive emotion "Thank you for your hard work."

[1537] Output: String data.

[1538] Step 6:

[1539] The server generates an illustration corresponding to the character string data.

[1540] Input: String data.

[1541] How it works: Using machine learning models (e.g., GAN or VQ-VAE), the system automatically generates illustrations corresponding to string data. Specifically, it generates an illustration of a cat character that reflects a positive emotion for the string data "Thank you for your hard work."

[1542] Output: Illustration.

[1543] Step 7:

[1544] The server combines the generated string data with illustrations to create an emoji set for communication apps.

[1545] Input: String data and illustrations.

[1546] What it does: It converts text and corresponding illustrations into the correct size and format, and compiles them into an emoji set.

[1547] Output: Emoji set.

[1548] Step 8:

[1549] Users can download the completed emoji set and use it within their communication app.

[1550] Input: A set of emojis provided by the server.

[1551] How it works: A set of emojis is downloaded from a user's device and used within a communication app. For example, a user sends a "Thank you for your hard work" sticker to a friend.

[1552] Output: The emoji set used within the user's communication app.

[1553] (Application example 2)

[1554] 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."

[1555] The challenge is to improve customer experience by enriching customer communication in virtual stores and providing reactions that match the customer's emotions. Conventional systems have difficulty in properly reflecting emotions in communication between customer service staff and customers, and lack real-time responses.

[1556] 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.

[1557] In this invention, the server includes means for collecting conversation data, means for analyzing the collected conversation data to identify frequently occurring words and phrases, means for generating character string data based on the identified frequently occurring words and phrases, means for generating illustrations corresponding to the character string data, means for creating an emoji set for a communication application by combining the character string data and illustrations, means for collecting and analyzing customer conversation data in a virtual store to identify customer emotions, and means for generating reaction illustrations based on the identified emotions, thereby enabling communication with customers to be more emotionally rich and in real time.

[1558] "Conversation Data" is a record of text and voice messages sent and received by a user within a communication application.

[1559] "Frequent words and phrases" are words and expressions that appear particularly frequently in conversation data.

[1560] "Character string data" is text data that includes frequently occurring words and phrases extracted from the analyzed conversation data.

[1561] An "illustration" is a graphic representation of an image or pictogram that corresponds to the character string data.

[1562] The "emoticon set for communication applications" is a collection of emoticons that users can use in communication applications, which combines character string data with corresponding illustrations.

[1563] A "virtual store" is an online virtual store space that offers products and services via the Internet.

[1564] "Customer service conversation data" is a record of communication between customer service staff and customers within the virtual store.

[1565] "Emotions" are psychological states and feelings that are analyzed from users' statements and messages.

[1566] "Reaction illustrations" are responsive images or emojis generated based on identified emotions.

[1567] The present invention relates to a system for improving customer service communication in a virtual store. This system is implemented using the following hardware and software.

[1568] Hardware Configuration

[1569] Server: A central computer system that collects and analyzes conversation data, manages the database, and creates the emoji set.

[1570] Device: A device used by a customer or staff member, such as smart glasses or a tablet.

[1571] Software Configuration

[1572] Natural language processing technology: Hugging Face's pipeline is used to perform tokenization, stop word removal, and sentiment analysis.

[1573] Machine learning model: PIL (Python Imaging Library) is used to automatically generate illustrations, which are then given characteristics such as background color based on emotions.

[1574] Database management software: Stores conversation data and manages it so that it can be retrieved when needed.

[1575] Processing flow overview

[1576] 1. Collecting conversation data

[1577] The device collects conversation data in real time within the virtual store. The content of conversations during customer service is automatically recorded and sent to the server at regular intervals.

[1578] 2. Data Analysis

[1579] The server then analyzes the received conversation data using natural language processing techniques, which include tokenization, stop word removal, and extraction of frequently occurring words and phrases.

[1580] 3. Emotion recognition

[1581] The emotion engine recognizes the user's emotions based on frequently occurring words and phrases identified from the conversation data, performs context analysis, and assigns emotion labels.

[1582] 4. Generating reaction illustrations

[1583] Based on the identified emotion, a machine learning model is used to generate reaction illustrations, for example, a light background for positive emotions and a dark background for negative emotions.

[1584] Specific examples

[1585] When a customer says, "I really like this product!" in a virtual store, the conversation data is collected by the device and analyzed by the server. This comment is recognized as a positive emotion, and a corresponding reaction illustration (e.g., a smiling face on a bright background) is generated. This reaction illustration is then displayed to the customer in real time, improving the quality of customer service.

[1586] Prompt Sentence Examples

[1587] The user says "I absolutely love this product!" about a product they're interested in. Analyze the sentiment of this message and generate an illustration for the response based on the positive sentiment.

[1588] This invention is expected to improve customer experience by making customer service in virtual stores more emotional and real-time.

[1589] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1590] Step 1:

[1591] Conversation data collection

[1592] The device collects conversation data in real time within the virtual store, including text messages and voice messages between the user and the staff, and temporarily stores the collected conversation data locally.

[1593] Input: Conversation data (text or voice) in the virtual store

[1594] Output: Locally stored conversation data

[1595] Step 2:

[1596] Sending conversation data

[1597] At specified intervals, the device transmits locally stored conversation data to the server using a secure communication protocol.

[1598] Input: Locally stored conversation data

[1599] Output: Conversation data uploaded to the server

[1600] Step 3:

[1601] Data analysis

[1602] The server analyzes the received conversation data using natural language processing techniques, including tokenization, stop word removal, and extraction of frequently occurring words and phrases. The server then stores the analysis results in a database.

[1603] Input: Conversation data uploaded to the server

[1604] Output: Analysis result data including frequently occurring words and phrases

[1605] Step 4:

[1606] emotion recognition

[1607] The server uses an emotion engine to recognize the user's emotions based on the analyzed frequently used words and phrases. The emotion engine performs contextual analysis and assigns emotion labels. This information is also stored in a database.

[1608] Input: Analysis result data (frequently used words and phrases)

[1609] Output: Data with emotion labels

[1610] Step 5:

[1611] Generate reaction illustrations

[1612] The server uses machine learning models to generate reaction illustrations based on the identified emotion labels. For example, positive emotions are represented by illustrations with bright backgrounds. The generated illustrations are also stored in a database.

[1613] Input: Data with emotion labels

[1614] Output: Reaction illustration

[1615] Step 6:

[1616] Customer feedback

[1617] The device receives the reaction illustrations generated from the server and displays them to the customer in real time, making communication during customer service more emotional and interactive.

[1618] Input: Reaction illustration sent from the server

[1619] Output: Reaction illustrations displayed to the customer

[1620] Step 7:

[1621] Improved data storage and feedback

[1622] The server stores information such as past conversation data, analysis results, emotion labels, and reaction illustrations in a database and periodically retrains the machine learning model to improve the quality of feedback.

[1623] Input: All collected and analyzed data (conversation data, emotion labels, reaction illustrations)

[1624] Output: Improved feedback system

[1625] Through these steps, communication with customers in the virtual store becomes more emotional and real-time, improving the customer experience.

[1626] 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.

[1627] 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.

[1628] 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.

[1629] 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.

[1630] FIG. 9 illustrates an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and behaviors arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1631] 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.

[1632] 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).

[1633] 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.

[1634] 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."

[1635] 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.

[1636] 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).

[1637] 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.

[1638] 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.

[1639] 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.

[1640] 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.

[1641] The hardware resource for executing a specific process can be any of the following types of processors: A processor, for example, is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another 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.

[1642] 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.

[1643] 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.

[1644] 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.

[1645] 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.

[1646] 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.

[1647] The following is further disclosed regarding the above embodiment.

[1648] (Claim 1)

[1649] a means for collecting conversation data;

[1650] A means of analyzing collected conversation data to identify frequently used words and phrases, and

[1651] A means for generating character string data based on the identified frequently occurring words and phrases;

[1652] means for generating an illustration corresponding to character string data;

[1653] A means for combining character string data and illustrations to create an emoji set for a communication app;

[1654] A system including:

[1655] (Claim 2)

[1656] 2. The system according to claim 1, wherein the analysis means uses natural language processing techniques to tokenize the conversation data, remove stop words, and extract frequently occurring words and phrases.

[1657] (Claim 3)

[1658] The system of claim 1, wherein the illustration generation means automatically generates illustrations using a machine learning model.

[1659]

[1660] "Example 1"

[1661] (Claim 1)

[1662] A means for a user's terminal to collect conversation data;

[1663] A means for converting the collected conversation data into a database format and transmitting it to a server;

[1664] A means for analyzing the conversation data received by the server to identify frequently used words and phrases;

[1665] A means for generating character string data based on the identified frequently occurring words and phrases;

[1666] a means for using a machine learning model to generate illustrations corresponding to the generated string data;

[1667] A means for combining the generated character string data and illustrations to create an emoji set for a communication application and transmitting the set to a user's terminal;

[1668] A system including:

[1669] (Claim 2)

[1670] 2. The system according to claim 1, wherein the analysis means uses natural language processing techniques to tokenize the conversation data, remove stop words, and extract frequently occurring words and phrases.

[1671] (Claim 3)

[1672] The system according to claim 1, wherein the illustration generation means inputs a prompt sentence into a machine learning model to automatically generate an illustration.

[1673] "Application Example 1"

[1674] (Claim 1)

[1675] a means for collecting conversation data;

[1676] A means of analyzing collected conversation data to identify frequently used words and phrases, and

[1677] A means for generating character string data based on the identified frequently occurring words and phrases;

[1678] means for generating an illustration corresponding to character string data;

[1679] A means for automatically generating advertising content by combining character string data and illustrations;

[1680] means for displaying the generated advertising content;

[1681] A system including:

[1682] (Claim 2)

[1683] 2. The system according to claim 1, wherein the analysis means uses natural language processing techniques to tokenize the conversation data, remove stop words, and extract frequently occurring words and phrases.

[1684] (Claim 3)

[1685] The system of claim 1, wherein the illustration generation means automatically generates illustrations using a machine learning model.

[1686] "Example 2: Combining Emotion Engines"

[1687] (Claim 1)

[1688] a means for collecting conversation data;

[1689] A means of analyzing collected conversation data to identify frequently used words and phrases, and

[1690] A means for analyzing conversation data to recognize user emotions;

[1691] A means for generating character string data based on the identified frequently occurring words and phrases and the recognized emotions;

[1692] means for generating an illustration corresponding to character string data;

[1693] A means for combining character string data and illustrations to create an emoji set for a communication app;

[1694] A system including:

[1695] (Claim 2)

[1696] 2. The system according to claim 1, wherein the analysis means uses natural language processing techniques to tokenize the conversation data, remove stop words, and extract frequently occurring words and phrases.

[1697] (Claim 3)

[1698] The system of claim 1, wherein the illustration generation means automatically generates illustrations using a machine learning model.

[1699] "Application example 2 when combining emotion engines"

[1700] (Claim 1)

[1701] a means for collecting conversation data;

[1702] A means of analyzing collected conversation data to identify frequently used words and phrases, and

[1703] A means for generating character string data based on the identified frequently occurring words and phrases;

[1704] means for generating an illustration corresponding to character string data;

[1705] A means for combining character string data and illustrations to create an emoji set for a communication app;

[1706] A means of collecting and analyzing customer conversation data in a virtual store to identify customer emotions;

[1707] A means for generating a reaction illustration based on the identified emotion;

[1708] A system including:

[1709] (Claim 2)

[1710] 2. The system according to claim 1, wherein the analysis means uses natural language processing techniques to tokenize the conversation data, remove stop words, and extract frequently occurring words and phrases.

[1711] (Claim 3)

[1712] The system of claim 1, wherein the illustration generation means automatically generates illustrations using a machine learning model. [Explanation of symbols]

[1713] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a means for collecting conversation data; A means of analyzing collected conversation data to identify frequently used words and phrases, and A means for generating character string data based on the identified frequently occurring words and phrases; means for generating an illustration corresponding to character string data; A means for combining character string data and illustrations to create an emoji set for a communication app; A system including:

2. 2. The system according to claim 1, wherein the analyzing means uses natural language processing techniques to tokenize the conversation data, remove stop words, and extract frequently occurring words and phrases.

3. The system according to claim 1 , wherein the illustration generation means automatically generates illustrations using a machine learning model.

Images