system

Abstract

Provide a system. [Solution] A means for receiving content and target information from a user that the user wants to create a buzz about; A means for retrieving related posts from a database that stores data on past buzz posts; A means for preprocessing the acquired post data and performing text analysis using natural language processing; A means to input the analysis results into a machine learning model to generate optimal post content, A means for presenting the generated post content and recommended keywords and hashtags to the user; 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 today's world, social media is inundated with posts and information, making it extremely difficult to effectively communicate information to specific audiences. It's particularly challenging for marketers and general users to create buzzworthy posts aimed at specific targets. Traditional methods require a great deal of time and effort, as past buzzworthy posts must be individually analyzed to identify effective posting patterns and keywords. Therefore, there is a need for a system that can efficiently and accurately generate optimal posting content and effectively communicate information to specific targets. [Means for solving the problem]

[0005] The present invention provides a system that efficiently and accurately generates optimal post content by having a user input content they want to create buzz and target information. This system includes the following means: a means for accepting the content they want to create buzz and the target information from the user, a means for retrieving related posts from a database storing past buzz post data, a means for preprocessing the retrieved post data and performing text analysis using natural language processing, a means for inputting the analysis results into a machine learning model to generate optimal post content, and a means for presenting the generated post content and recommended keywords and hashtags to the user, thereby enabling the user to effectively communicate information to their target demographic.

[0006] A "user" is an individual or organization that uses the system to input content they want to create buzz.

[0007] "Buzzworthy content" refers to information that you want many people to spread and talk about on social media.

[0008] "Target information" is information about a specific demographic (age, gender, interests, region, etc.) that is likely to be interested in the content you want to create buzz about.

[0009] The "database" is an information storage system that stores past buzz posting data and allows it to be searched and retrieved.

[0010] "Natural language processing" (NLP) is the technology that enables computers to understand, interpret, and generate human language.

[0011] "Preprocessing" refers to the process of removing noise from the acquired post data, deleting stop words, and normalizing the data.

[0012] A "machine learning model" is an algorithm that learns from past data and makes predictions and classifications for new input data.

[0013] "Analysis results" refers to information such as keywords, emotions, and phrases extracted through text analysis using natural language processing.

[0014] "Recommended content" is information including optimal posting patterns, keywords, hashtags, tone, etc. that should be suggested to users based on a machine learning model.

[0015] A "means" refers to a physical or logical device, process, algorithm, or mechanism used to achieve a particular function.

[0016] A "system" is a comprehensive framework that integrates multiple means to efficiently generate and present content that you want to create buzz. [Brief explanation of the drawings]

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

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

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

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

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

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

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

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

[0025] [First embodiment]

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

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

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

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

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

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

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

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

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

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

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

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

[0038] This invention relates to a system for efficiently communicating content that you want to create buzz on social networking sites to specific targets. Below, we will explain the program processing of this system in natural language.

[0039] System configuration

[0040] The system mainly consists of the following means:

[0041] 1. Means of accepting input from users

[0042] 2. How to obtain information from the database

[0043] 3. Natural Language Processing and Text Analysis Methods

[0044] 4. How to apply machine learning models

[0045] 5. How to generate and present recommendations to users

[0046] Program processing

[0047] 1. Means of accepting input from users

[0048] Users access the system's interface using a device (such as a smartphone or PC). They input the content they want to create buzz about and target information (age group, gender, interests, etc.). This information is then sent from the device to the server.

[0049] 2. How to obtain information from the database

[0050] The server receives the information sent by the user and then executes a query to retrieve buzz posts related to the specified target information from a database that stores past buzz post data. Here, the query includes filtering conditions based on the target information.

[0051] 3. Natural Language Processing and Text Analysis Methods

[0052] The server preprocesses the acquired post data, including noise removal, stop word removal, and text normalization. It then uses natural language processing (NLP) techniques to perform morphological analysis of the text and extract important keywords, phrases, and sentiment (positive, negative, neutral).

[0053] 4. How to apply machine learning models

[0054] The server then inputs the data into a machine learning model based on the analysis results. This model learns from past success stories (buzzworthy posts) and predicts the best posting pattern for the target audience. The output generates templates for keywords, hashtags, post tone (emotion), and specific post content to use.

[0055] 5. How to generate and present recommendations to users

[0056] The server then compiles the final recommendations and creates a list of suggested posts for the user, which may include, for example:

[0057] Example keywords

[0058] Hashtag examples

[0059] Post Template

[0060] The device displays the received recommendations to the user, who can then review and edit them and post them directly to the social networking site.

[0061] Specific examples

[0062] For example, if a user wants to create buzz about the opening of a new cafe and specifies the target audience as "women in their 20s who live in urban areas and love cafes," the system will operate as follows:

[0063] 1. Accepting input from the user

[0064] User input: "New cafe opening", "Female in her 20s, urban resident, cafe lover"

[0065] 2. Retrieving information from the database

[0066] The server runs a database query to retrieve past trending posts, such as:

[0067] "Cafe hopping and new discoveries"

[0068] "I went to a new cafe! It's stylish and comfortable - perfect for Instagram."

[0069] "All the cafes in the city are great, but this one was especially great! I want to connect with other cafe lovers."

[0070] 3. Natural Language Processing and Text Analytics

[0071] The server extracts important keywords and phrases from the above submission data:

[0072] Keywords: "cafe," "new," "urban," "Instagrammable"

[0073] Emotion: Positive

[0074] 4. Applying machine learning models

[0075] The server feeds the extracted data into a machine learning model to generate recommendations for:

[0076] Recommended keywords: "new cafe," "Instagrammable," "urban area," "cafe hopping"

[0077] Recommended hashtags: "cafe hopping," "new discoveries," "Instagrammable," "urban areas"

[0078] Post template: "A new cafe has opened! It has a relaxed atmosphere and is sure to be Instagram-worthy! Cafe hopping, urban areas, and connecting with fellow cafe lovers."

[0079] 5. Generating and presenting recommendations to the user

[0080] The device displays the recommended content to the user, who then checks and edits it and posts it to the social networking site.

[0081] The above is a specific embodiment of the present invention, which makes it possible to efficiently and accurately generate and present buzzworthy posts aimed at a target audience.

[0082] The processing flow will be explained below.

[0083] Step 1:

[0084] Users access the system's interface using a device (smartphone or PC). They input the content they want to create buzz about and target information (age group, gender, interests, etc.). This information is then sent from the device to the server.

[0085] Step 2:

[0086] The server receives the information sent by the user and then executes a query from a database of past buzz posts to retrieve posts related to the target information specified by the user. The query includes filtering conditions based on the target information.

[0087] Step 3:

[0088] The server preprocesses the submitted data retrieved from the database by removing noise, removing stop words, and normalizing the text, making the data easier to analyze.

[0089] Step 4:

[0090] The server then applies natural language processing (NLP) to the preprocessed post data. This step involves morphological analysis to extract important keywords, phrases, and sentiment (positive, negative, neutral).

[0091] Step 5:

[0092] The server then inputs the extracted keywords, phrases, and sentiment data into a machine learning model, which learns from past success stories (buzzworthy posts) and predicts the optimal posting pattern for the target audience.

[0093] Step 6:

[0094] The server creates recommended posts for users based on the predictions generated by the machine learning model, including optimal keywords, hashtags, post tone (emotion), and specific post content templates.

[0095] Step 7:

[0096] The server sends the generated recommendations to the device, which then displays them to the user. The user can then review and edit the recommendations and post them directly to the social networking site.

[0097] Specific examples

[0098] Step 1:

[0099] A user enters the following information into the system: "A new cafe is opening" and "A woman in her 20s who loves cafes and lives in an urban area."

[0100] Step 2:

[0101] A server receives the information submitted by the user and performs a database query to retrieve relevant past buzz posts.

[0102] Step 3:

[0103] The server removes noise from the acquired posting data, removes stop words, and normalizes the text.

[0104] Step 4:

[0105] The server applies natural language processing to the preprocessed data to extract important keywords, phrases, and sentiment.

[0106] Step 5:

[0107] The server inputs the extracted data into a machine learning model to predict the optimal posting patterns for the target audience.

[0108] Step 6:

[0109] The server generates recommended posts based on predictions from the machine learning model.

[0110] Step 7:

[0111] The server sends the generated recommendations to the device, which displays them to the user, who then checks and edits them and posts them to the SNS.

[0112] Through these processing steps, the system can effectively generate and present buzz posts to the user's target audience.

[0113] Example 1

[0114] 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."

[0115] Currently, many users are using social networking services (SNS) to spread information, but it is not easy to efficiently create buzz for a specific target audience. Existing methods make it difficult to predict what content and format will create buzz, and require a great deal of time and effort. Furthermore, selecting appropriate keywords and hashtags is difficult, so generating optimal post content requires specialized knowledge. Therefore, there is a need for a system that can efficiently and accurately generate and present buzzworthy posts for specific targets.

[0116] 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.

[0117] In this invention, the server includes means for receiving content and target information from a user that the user wants to create a buzz, means for retrieving related posts from a database storing past buzz post data, means for preprocessing the retrieved post data and performing text analysis using natural language processing, means for inputting the analysis results into a machine learning model to generate optimal post content, means for presenting the generated post content and recommended keywords and hashtags to the user, and means for the user to edit and confirm the generated post content and then post it to an SNS. This allows users to effectively create buzz posts aimed at a specific target demographic.

[0118] "Means for accepting content to be buzzed and target information from users" refers to an interface or function that allows users to input the content they want to buzz and information about the target person, and send it to the server.

[0119] "Means for retrieving related posts from a database storing data on past buzz posts" refers to a function in which the server queries the database, retrieves data on past buzz posts, and retrieves posts related to the target information specified by the user.

[0120] "Means for preprocessing acquired posted data and performing text analysis using natural language processing" refers to a function that performs preprocessing on acquired posted data, such as noise removal, stop word removal, and text normalization, and then analyzes the text using natural language processing technology.

[0121] "A means of inputting the analysis results into a machine learning model to generate optimal post content" refers to a function that inputs data obtained from text analysis into a machine learning model and uses the model, which has learned from past success stories, to generate post content that is most suitable for the target demographic.

[0122] "Means of presenting generated post content and recommended keywords and hashtags to users" refers to the function of organizing post content, keywords, hashtags, etc. generated by machine learning models and displaying and suggesting them to users.

[0123] "Means for users to post generated posts to SNS after editing and reviewing them" refers to a function that allows users to review and edit the suggested content and post it directly to the SNS platform.

[0124] "Preprocessing of posted data retrieved from the database" refers to the process of removing unnecessary information (noise) from the posted data, deleting commonly used words (stop words) that are irrelevant to a specific analysis, and normalizing the strings to maintain consistency.

[0125] "Natural language processing" is a technology that enables computers to understand, interpret, and generate human language, and in the present invention refers to the technology used for morphological analysis of text data and extraction of important keywords and emotions.

[0126] A "machine learning model" refers to an algorithm that learns patterns from given data and makes predictions and classifications for new data. In the context of this invention, it is used to learn from past buzz posts and generate post content that is optimal for the target demographic.

[0127] "Target information" refers to the attribute information (e.g., age group, gender, interests, etc.) of the intended recipients of the post that the user wants to make go viral, and the content of the post is optimized based on that information.

[0128] This invention relates to a system for efficiently communicating content that users want to create buzz on social media to specific targets. This system accepts input from users, acquires and analyzes relevant data from past buzz posts, and generates optimal post content using a machine learning model, which is then presented to the user. The configuration and operation of this system are described in detail below.

[0129] System configuration

[0130] The system mainly consists of the following means:

[0131] 1. Means of accepting input from users

[0132] 2. How to obtain information from the database

[0133] 3. Natural Language Processing and Text Analysis Methods

[0134] 4. How to apply machine learning models

[0135] 5. How to generate and present recommendations to users

[0136] 6. Means of posting to social media

[0137] Processing Details

[0138] 1. Accepting input from the user

[0139] Users access the system's interface using a smartphone or PC. They input the content they want to create buzz and target information (age group, gender, interests, etc.), and this information is sent from the device to the server. The input is in a specific data format, and the server receives it and begins processing.

[0140] 2. Retrieving information from the database

[0141] After receiving the information sent by the user, the server executes a database query based on that information. The database stores past buzz post data and retrieves relevant post data based on specific filtering criteria. For example, a query is executed to retrieve information about "women in their 20s, urban areas, cafe lovers."

[0142] 3. Natural Language Processing and Text Analytics

[0143] The server preprocesses the acquired post data. This preprocessing includes noise removal (removal of special symbols and unnecessary spaces), stopword removal (frequent phrases that are generally ignored), and text normalization (standardization of characters). Then, using natural language processing technology (e.g., the morphological analysis tool MeCab), it performs morphological analysis of the text data and extracts important keywords, phrases, and sentiment (positive, negative, neutral).

[0144] 4. Applying machine learning models

[0145] Based on the analysis results, the server inputs the data into a machine learning model. The machine learning model learns from past successful posting data and predicts the optimal posting pattern for the target demographic. For example, the model predicts the keywords, hashtags, and tone (emotion) to be used in the post, and generates a specific post content template.

[0146] 5. Generating and presenting recommendations to the user

[0147] Finally, the server presents recommendations to the user, including example keywords, example hashtags, and post templates, which the user can review and edit to select the post they deem most appropriate.

[0148] 6. How to post to social media

[0149] The final post content that users have considered can be posted directly to the SNS platform, and users can easily publish the generated post content to the SNS through their device.

[0150] Specific examples

[0151] For example, if a user wants to create buzz about the opening of a new cafe and specifies the target audience as "women in their 20s who live in urban areas and love cafes," the system will operate as follows:

[0152] 1. Accepting input from the user

[0153] User input: "New cafe opening", "Female in her 20s, urban resident, cafe lover"

[0154] 2. Retrieving information from the database

[0155] The server runs a database query to retrieve past viral posts, such as:

[0156] "Cafe hopping and new discoveries"

[0157] "I went to a new cafe! It's stylish and comfortable - perfect for Instagram."

[0158] "All the cafes in the city are great, but this one was especially great! I want to connect with other cafe lovers."

[0159] 3. Natural Language Processing and Text Analytics

[0160] The server extracts important keywords and phrases from the above submission data:

[0161] Keywords: "cafe," "new," "urban," "Instagrammable"

[0162] Emotion: Positive

[0163] 4. Applying machine learning models

[0164] The server feeds the extracted data into a machine learning model to generate the following recommendations:

[0165] Recommended keywords: "new cafe," "Instagrammable," "urban area," "cafe hopping"

[0166] Recommended hashtags: "cafe hopping," "new discoveries," "Instagrammable," "urban areas"

[0167] Post template: "A new cafe has opened! It has a relaxed atmosphere and is sure to be Instagram-worthy! Cafe hopping, urban areas, and connecting with fellow cafe lovers."

[0168] 5. Generating and presenting recommendations to the user

[0169] The device displays the recommended content to the user, who can then review it, edit it if necessary, and post it to social media.

[0170] Examples of prompt statements

[0171] You might enter the following prompts for the system's generated AI model:

[0172] "I want to create a buzz about the opening of a new cafe among women in their 20s who love cafes and live in urban areas. Based on past data on buzzworthy posts, could you recommend some keywords, hashtags, and a post template?"

[0173] The above is a specific embodiment of the present invention, which makes it possible to efficiently and accurately generate and present buzzworthy posts aimed at a target audience.

[0174] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0175] Step 1: Accepting input from the user

[0176] Users access the system's interface using a smartphone or PC. They input the content they want to create buzz about (e.g., "A new cafe has opened") and target information (e.g., "Women in their 20s who live in urban areas and love cafes"). The device receives this information and sends it to the server. The server temporarily stores the input information before proceeding to the next processing step.

[0177] Step 2: Retrieving information from the database

[0178] The server receives target information sent by the user. Based on this information, the server sends a query to the database to retrieve related posts from past buzz post data. For example, it searches the database for posts related to "women in their 20s" and "cafes" and extracts related post data. Input: Target information. Output: Related past buzz post data.

[0179] Step 3: Natural Language Processing and Text Analytics

[0180] The server preprocesses the acquired post data. Preprocessing includes removing noise (removing special symbols and unnecessary spaces), deleting stop words (frequent words that are generally ignored), and normalizing the text (standardizing characters). Natural language processing techniques (such as morphological analysis tools) are then used to perform morphological analysis of the text data, extract important keywords and phrases, and analyze sentiment (positive, negative, neutral). Input: Past buzz post data. Output: Preprocessed text data and analysis results.

[0181] Step 4: Applying the machine learning model

[0182] The server inputs the preprocessed text data and analysis results into a machine learning model. The machine learning model learns from past posting data of successful cases and predicts the optimal posting pattern for the target demographic. This prediction generates templates for the keywords, hashtags, tone (emotion) of the post, and specific post content to be used. Input: Analysis results. Output: Optimal post content, recommended keywords, recommended hashtags.

[0183] Step 5: Generate recommendations and present them to the user

[0184] The server organizes the final recommendations and creates a list of post content to suggest to the user. The list includes example keywords, example hashtags, and post templates. The device displays the received recommendations to the user. The user can review and edit these recommendations and select the most suitable post content. Input: Best post content, recommended keywords, recommended hashtags. Output: List of posts presented to the user.

[0185] Step 6: How to post to social media

[0186] The user reviews the suggested content and edits it as necessary. Through the device interface, the user can post the final post directly to social media. This allows for efficient spreading of content that is targeted to a specific demographic. Input: Final post content after user editing. Output: Post to social media.

[0187] (Application example 1)

[0188] 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."

[0189] In order to efficiently create buzz on social media for specific content, it is necessary to generate posts that are optimal for the target audience. However, conventional systems that utilize artificial intelligence and machine learning require the manual collection and analysis of data and the generation of post content, which is time-consuming and labor-intensive. In addition, there is uncertainty as to whether the generated post content will actually appeal to the target, so a reliable method is needed to increase the probability of buzz.

[0190] 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.

[0191] In this invention, the server includes means for receiving content and target information from a user that the user wants to create a buzz, means for retrieving related posts from a database storing past buzz post data, means for preprocessing the retrieved post data and performing text analysis using natural language processing, means for inputting the analysis results into a machine learning model to generate optimal post content, means for presenting the generated post content and recommended keywords and hashtags to the user, and means for posting the generated optimal post content to an SNS via an application on a smart device. This allows users to avoid complex manual data collection and analysis work, quickly and easily generate post content that is most effective for their target, and significantly increases the chances of creating a buzz on an SNS.

[0192] "User" refers to any individual or entity that uses the System.

[0193] "Buzzworthy content" refers to information or messages posted on social media with the aim of attracting a lot of attention.

[0194] "Target information" is data that indicates the attributes and interests of the target audience for a particular post, including, for example, age group, gender, place of residence, and interests.

[0195] "Database" refers to a collection of information that stores collected data on past buzz posts and makes it searchable and retrievalable.

[0196] "Preprocessing" refers to the process of processing the data by removing noise, deleting stop words, normalizing text, etc. from the acquired submission data.

[0197] "Natural language processing" refers to the technology that enables computers to understand and analyze human language.

[0198] "Text analysis" refers to the process of using natural language processing techniques to interpret the meaning and structure of a document and extract important information.

[0199] A "machine learning model" refers to a statistical method for learning patterns and rules from past data and making predictions and classifications for future data.

[0200] "Keywords" are particularly important words or phrases in your post that will grab your target's attention.

[0201] A "hashtag" is a type of tagging notation used in social media posts, and refers to a keyword that makes it easier to group posts related to a particular topic or theme.

[0202] "Smart device" refers to a portable electronic device that can connect to a network, such as a smartphone or tablet.

[0203] "Application" refers to a software program that runs on a smart device and provides specific functionality.

[0204] "SNS" is an abbreviation for social networking service, which refers to a platform where users can share information and interact with each other online.

[0205] This invention relates to a system for efficiently communicating content that you want to create buzz on SNS to designated targets, and includes the following means and processes.

[0206] Program processing explanation

[0207] The server receives the content and target information from the user and performs the following processing based on this: First, it retrieves data related to the specified target information from a database of past buzz posts, and then performs preprocessing to remove noise, stop words, and normalize the data.

[0208] The server then performs text analysis using natural language processing (NLP) techniques, including morphological analysis, keyword extraction, and sentiment analysis of the acquired data, using software such as NLTK (Natural Language Toolkit) and Scikit-learn.

[0209] The server then inputs the analyzed data into a machine learning model, which uses classification techniques such as logistic regression to predict the best post content, keywords, and hashtags for the target audience based on past success stories.

[0210] Finally, the generated optimal post content, recommended keywords, and hashtags are presented to the user, who posts it to a social networking site via an application on their smart device.

[0211] Specific examples

[0212] For example, if a user wants to create buzz about the opening of a new cafe and specifies "women in their 20s who love cafes and live in urban areas" as their target, the system will operate as follows.

[0213] User Input

[0214] The user wants to create buzz: "New cafe opening"

[0215] Target information: "Women in their 20s who love cafes and live in urban areas"

[0216] Server Processing

[0217] Get past trending posts related to your target information from the database. Example:

[0218] "Cafe hopping and new discoveries"

[0219] "I went to a new cafe! It's stylish and comfortable - perfect for Instagram."

[0220] "All the cafes in the city are great, but this one was especially great! I want to connect with other cafe lovers."

[0221] Keyword extraction from analysis results:

[0222] Keywords: "cafe," "new," "urban," "Instagrammable"

[0223] Emotion: Positive

[0224] Generate suggested posts

[0225] Recommended keywords: "new cafe," "Instagrammable," "urban area," "cafe hopping"

[0226] Recommended hashtags: "cafe hopping," "new discoveries," "Instagrammable," "urban areas"

[0227] Post template: "A new cafe has opened! It has a relaxed atmosphere and is sure to be Instagram-worthy! Cafe hopping, urban areas, and connecting with fellow cafe lovers."

[0228] Prompt Sentence Examples

[0229] The following sentences can be used as prompts for users to input to a generative AI model:

[0230] "Generate the best social media post content to promote the opening of a new cafe. The target audience is women in their 20s who love cafes and live in urban areas. Past related post data is shown below. Please include keywords and hashtags that should be used in the post."

[0231] This allows users to quickly generate posts that are suitable for their target audience and are likely to create buzz, and then easily post them on social media.

[0232] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0233] Step 1:

[0234] Users enter the content they want to create buzz on social media and target information.

[0235] Input: What you want to create a buzz about (e.g., the opening of a new cafe), target information (e.g., women in their 20s who love cafes and live in urban areas)

[0236] What happens: A user provides input via an application on their smart device, which is sent to the server.

[0237] Step 2:

[0238] The server retrieves post data related to the target information from a database of past buzz posts.

[0239] Input: Target information sent by the user

[0240] Data processing: Perform database queries based on target information and filter relevant post data

[0241] Output: Data of past buzz posts (e.g., "Cafe hopping, new discoveries")

[0242] What happens: The server uses SQL to run a search query against the database to retrieve the relevant data.

[0243] Step 3:

[0244] Preprocess the submitted data received by the server.

[0245] Input: Previous buzz post data obtained

[0246] Data processing: noise removal, stopword removal, text normalization

[0247] Output: Clean post data after preprocessing

[0248] Specific operation: The server uses a natural language processing library such as NLTK to preprocess the text data.

[0249] Step 4:

[0250] The server performs text analysis using natural language processing technology.

[0251] Input: Clean post data after preprocessing

[0252] Data processing: morphological analysis of text, keyword extraction, sentiment analysis

[0253] Output: Key keywords, phrases, and emotional state (positive, negative, neutral)

[0254] Specific operation: The server uses Scikit-learn or similar tools to apply models for morphological analysis and keyword extraction.

[0255] Step 5:

[0256] The server inputs the analysis results into a machine learning model to generate optimal post content.

[0257] Input: Keywords, phrases, and emotional states obtained through natural language processing

[0258] Data Computing: Prediction and Content Generation with Machine Learning Models

[0259] Output: Best post content, recommended keywords, recommended hashtags, post templates

[0260] What it does: The server uses machine learning models such as logistic regression to generate targeted posts.

[0261] Step 6:

[0262] The server presents the generated post content and recommended keywords and hashtags to the user.

[0263] Input: Best post content, suggested keywords, suggested hashtags, post template

[0264] Output: What the user sees on their smart device

[0265] Specific behavior: The server sends the generated content to the user interface so that the user can view it on their smart device.

[0266] Step 7:

[0267] The optimal post content that the user has checked and edited is posted to SNS.

[0268] Input: Post content provided by the server, edits made by the user

[0269] Output: Posts published on social media

[0270] Specific behavior: The user publishes a post using the interface that allows them to post directly to the social networking site from the application.

[0271] 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.

[0272] This invention relates to a system that enables efficient communication of content that you want to create buzz on social media to specific targets. This system achieves even more effective communication by combining it with an emotion engine that recognizes and analyzes user emotions. Below, we will explain the program processing of this system in natural language.

[0273] System configuration

[0274] The system mainly consists of the following means:

[0275] 1. Means of accepting input from users

[0276] 2. How to obtain information from the database

[0277] 3. Natural Language Processing and Text Analysis Methods

[0278] 4. How to apply machine learning models

[0279] 5. User Emotion Analysis Method Using Emotion Engine

[0280] 6. How to generate and present recommendations to users

[0281] Program processing

[0282] 1. Means of accepting input from users

[0283] Users access the system's interface using a device (such as a smartphone or PC). They input the content they want to create buzz about and target information (age group, gender, interests, etc.). This information is then sent from the device to the server.

[0284] 2. How to obtain information from the database

[0285] The server receives the information sent by the user and then executes a query from a database of past buzz posts to retrieve posts related to the target information specified by the user. The query includes filtering conditions based on the target information.

[0286] 3. Natural Language Processing and Text Analysis Methods

[0287] The server preprocesses the submitted data by removing noise, removing stop words, and normalizing the text, making the data easier to analyze.

[0288] 4. How to apply machine learning models

[0289] The server then applies natural language processing (NLP) to the preprocessed post data. This step involves morphological analysis to extract important keywords, phrases, and sentiment (positive, negative, neutral).

[0290] 5. User Emotion Analysis Method Using Emotion Engine

[0291] The server uses an emotion engine to analyze the user's emotions based on the user's facial expressions, tone of voice, or text input. Based on the analysis results, the user's emotions are classified as positive, negative, or neutral.

[0292] 6. How to generate and present recommendations to users

[0293] The server adjusts the predictions generated by the machine learning model based on the user's emotions recognized by the emotion engine to create optimal post content, including keywords, hashtags, post tone (emotion), and specific post content templates to use.

[0294] For example, if the user's emotions are positive, the system generates posts with a bright and cheerful tone. If the emotions are negative, the system emphasizes an encouraging and comforting tone.

[0295] Specific examples

[0296] For example, if a user wants to create buzz about the opening of a new cafe and specifies the target audience as "women in their 20s who live in urban areas and love cafes," the system will operate as follows:

[0297] 1. Accepting input from the user

[0298] User input: "New cafe opening", "Female in her 20s, urban resident, cafe lover"

[0299] 2. Retrieving information from the database

[0300] A server receives the information submitted by the user and performs a database query to retrieve relevant past buzz posts.

[0301] 3. Natural Language Processing and Text Analytics

[0302] The server removes noise from the acquired posting data, removes stop words, and normalizes the text.

[0303] 4. Applying machine learning models

[0304] The server applies natural language processing to the preprocessed data to extract important keywords, phrases, and sentiment.

[0305] 5. User Emotion Analysis Using an Emotion Engine

[0306] The server analyzes facial expression data, voice data, or input text sent from the user's device and recognizes that the user's emotions are positive.

[0307] 6. Generating and Presenting Recommendations to the User

[0308] The server generates posts that are appropriate for positive emotions based on the analysis results of the emotion engine:

[0309] Recommended keywords: "new cafe," "Instagrammable," "urban area," "cafe hopping"

[0310] Recommended hashtags: "cafe hopping," "new discoveries," "Instagrammable," "urban areas"

[0311] Post template: "A new cafe has opened! It has a relaxed atmosphere and is sure to be Instagram-worthy! Cafe hopping, urban areas, and connecting with fellow cafe lovers."

[0312] Through the above processing steps, the system can effectively generate and present buzz posts to the user's target demographic, and provide optimal content based on the user's emotions.

[0313] The processing flow will be explained below.

[0314] Step 1:

[0315] Users access the system's interface using a device (smartphone or PC). They input the content they want to create buzz about (e.g., "A new cafe has opened") and target information (e.g., "Women in their 20s who live in urban areas and love cafes"). This information is then sent from the device to the server.

[0316] Step 2:

[0317] The server receives the input information sent by the user. It then executes a query to retrieve posts related to the specified target information from a database of past buzz posts. For example, it filters past posts related to "women in their 20s" and "cafes" based on the user's target information.

[0318] Step 3:

[0319] The server preprocesses the submitted data retrieved from the database by removing noise, removing stop words, and normalizing the text, making the data easier to analyze.

[0320] Step 4:

[0321] The server then applies natural language processing (NLP) to the preprocessed post data. This step involves morphological analysis to extract important keywords (e.g., "new cafe"), phrases, and sentiment (positive, negative, neutral).

[0322] Step 5:

[0323] The server inputs data into a machine learning model based on keywords and sentiment extracted through natural language processing. This model learns from past success stories (buzzworthy posts) and predicts the best posting pattern for the target audience. For example, it generates recommended hashtags such as "Instagrammable" and "cafe hopping."

[0324] Step 6:

[0325] The server analyzes the user's emotions using an emotion engine. This analysis is based on the user's facial expression recognition, voice analysis, or text input. For example, if the user has a positive expression while typing "new cafe," this emotion is recognized as positive.

[0326] Step 7:

[0327] The server adjusts the predictions generated by the machine learning model based on the user's emotions recognized by the emotion engine. For example, if the user's emotions are positive, it generates posts with a bright tone. On the other hand, if the emotions are negative, it generates posts that contain encouragement and comfort.

[0328] Step 8:

[0329] The server then compiles the final recommendations and creates a list of suggested posts for the user, including the best keywords, hashtags, and post templates, for example:

[0330] Recommended keywords: "new cafe," "Instagrammable," "urban area," "cafe hopping"

[0331] Recommended hashtags: "cafe hopping," "new discoveries," "Instagrammable," "urban areas"

[0332] Post template: "A new cafe has opened! It has a relaxed atmosphere and is sure to be Instagram-worthy! Cafe hopping, urban areas, and connecting with fellow cafe lovers."

[0333] Step 9:

[0334] The server sends the generated recommendations to the device. The device displays the received recommendations to the user. The user can then review and edit the recommendations and post them directly to the SNS.

[0335] Through the above processing steps, the system can effectively generate buzz posts for the user's target demographic and provide optimal content according to the user's emotions.

[0336] Example 2

[0337] 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."

[0338] With conventional SNS posting systems, it was difficult for users to effectively communicate information to their target audience. Also, because only uniform post content could be generated without considering the user's emotions, the effectiveness of the posts was limited. As a result, it was difficult for the posts to go viral, and it was difficult for the posts to reach the intended audience.

[0339] 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.

[0340] In this invention, the server includes means for receiving content to be made buzzworthy and target information from a user, means for retrieving related posts from a database storing past buzz posting data, means for preprocessing the retrieved posting data and performing text analysis using natural language processing, means for recognizing and analyzing user emotions using a sentiment analysis engine, means for inputting the analysis results into a machine learning algorithm to generate optimal posting content, and means for presenting the generated posting content and recommended keywords and hashtags to the user. This makes it possible to generate posting content that is effective for the target demographic and in line with their emotions.

[0341] "Means for receiving content to be buzzed and target information from a user" refers to an interface or process for receiving content to be buzzed and target information input by a user.

[0342] "Means for retrieving related posts from a database storing past buzz posting data" refers to a function for searching and retrieving information related to the user's requirements from stored past buzz posting data.

[0343] "Means for preprocessing acquired posted data and performing text analysis using natural language processing" refers to the process of removing noise, deleting stop words, and normalizing acquired posted data, and then analyzing the content using natural language processing.

[0344] "Means for recognizing and analyzing user emotions using an emotion analysis engine" refers to the function of analyzing emotions from data such as text, voice, and facial expressions entered by the user and classifying them as positive, negative, neutral, etc.

[0345] "Method of inputting analysis results into a machine learning algorithm to generate optimal post content" refers to the process of generating appropriate post content through a machine learning model based on the results of natural language processing and sentiment analysis.

[0346] "Means for presenting generated post content and recommended keywords and hashtags to users" refers to functions and interfaces for providing users with post content generated by the system and effective keyword and hashtag suggestions.

[0347] A specific embodiment for carrying out the present invention will be described below. The system of the present invention realizes effective information transmission by combining users, terminals, and servers.

[0348] System configuration and hardware / software used

[0349] 1. Means of accepting input from users

[0350] Users access the system interface via a web browser on their device (e.g., smartphone, PC). They then access the system and enter the content they want to create buzz about and target information (age group, gender, interests, etc.). This information is sent to the server as an HTTP request.

[0351] 2. How to obtain information from the database

[0352] The server receives the information sent by the user, then executes an SQL query against a database of past buzz posts (e.g., MySQL (registered trademark) or PostgreSQL) to search for and retrieve posts related to the target information specified by the user.

[0353] 3. Natural Language Processing and Text Analysis Methods

[0354] The server uses a Python NLP library (e.g., NLTK, spaCy) to preprocess the submitted data. Specifically, it performs the following steps:

[0355] Noise removal: Remove unnecessary special characters and HTML tags.

[0356] Stopword removal: Removal of common words that have no meaning (e.g. "wa", "o", "ni").

[0357] Text normalization: Convert all text to lower case for consistency.

[0358] 4. How to apply machine learning models

[0359] The server then performs natural language processing (NLP) on the preprocessed text data, including morphological analysis and sentiment analysis. In this step, it uses Tensorflow® and PyTorch to extract important keywords, phrases, and sentiment (positive, negative, neutral).

[0360] 5. User Emotion Analysis Method Using Emotion Engine

[0361] The server uses an emotion engine (e.g., IBM Watson® Tone Analyzer) to analyze emotions from the user's input data (e.g., text, voice, facial expressions, etc.). Based on the analysis results, the user's emotions are classified as positive, negative, or neutral.

[0362] 6. How to generate and present recommendations to users

[0363] The server combines the results of the machine learning model and the emotion engine to generate optimal posts for the target users. Specifically, it generates the following elements:

[0364] Recommended Keywords

[0365] Recommended hashtags

[0366] The tone (emotion) of the post

[0367] Specific post content template

[0368] The user can review the generated post and make any necessary adjustments.

[0369] Specific examples

[0370] For example, if a user wants to create buzz about the opening of a new cafe and specifies the target audience as "women in their 20s who live in urban areas and love cafes," the system will operate as follows:

[0371] 1. Accepting input from the user

[0372] The user fills in the form with "a new cafe opening" and "a woman in her 20s who loves cafes and lives in an urban area" and submits it.

[0373] 2. Retrieving information from the database

[0374] Based on the input information received by the server, a database query is executed to retrieve related past buzz posts.

[0375] 3. Natural Language Processing and Text Analytics

[0376] The server removes noise from the acquired submission data, removes stop words, and normalizes the text.

[0377] 4. Applying machine learning models

[0378] The server applies NLP to the pre-processed data to extract important keywords, phrases, and sentiment.

[0379] 5. User Emotion Analysis Using an Emotion Engine

[0380] The server analyzes the user's text input and recognizes it as a positive emotion.

[0381] 6. Generating and Presenting Recommendations to the User

[0382] Based on the sentiment and NLP results, the server generates the following post with a positive tone:

[0383] Recommended keywords: "new cafe," "Instagrammable," "urban area," "cafe hopping"

[0384] Recommended hashtags: "cafe hopping," "new discoveries," "Instagrammable," "urban areas"

[0385] Post template: "A new cafe has opened! It has a relaxed atmosphere and is sure to be Instagram-worthy! Cafe hopping, urban areas, and I want to connect with fellow cafe lovers."

[0386] Prompt Sentence Examples

[0387] Example prompts to input to a generative AI model:

[0388] We want to create buzz on social media about the opening of a new cafe. Our target audience is women in their 20s who love cafes and live in urban areas. We will generate effective posts based on data on past buzz posts. Since many users have positive feelings, we recommend creating posts with a bright and fun tone.

[0389] As a result, this system can not only effectively generate and present buzz posts to the user's target demographic, but also provide optimal content that reflects the user's emotions.

[0390] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0391] Divide the processing flow of the system program into processing steps

[0392] Step 1:

[0393] Step 2:

[0394] Step 3:

[0395] Step 4:

[0396] Step 5:

[0397] Step 6:

[0398] Specific explanation of each processing step

[0399] Step 1: Accepting input from the user

[0400] Input: A user accesses the system interface using a terminal and enters the content they want to buzz and target information into a form.

[0401] Processing: The terminal sends the input information to the server in the form of an HTTP request.

[0402] Output: The server temporarily stores the received content and target information in memory.

[0403] What happens: A user opens a browser, fills in a web form with the words "New cafe opening" and "Female, 20s, urban resident, cafe lover," and clicks the submit button.

[0404] Step 2: How to retrieve information from the database

[0405] Input: The server generates a database query based on the buzz content and target information received from the user.

[0406] Processing: The server executes an SQL query against a database of past buzz posts (e.g., MySQL, PostgreSQL) to search and retrieve relevant past posts.

[0407] Output: A dataset containing the retrieved submission data.

[0408] Specific operation: The server queries the database using the target information of "women in their 20s" and "urban areas," and extracts related past buzz posts using an SQL query.

[0409] Step 3: Natural language processing and text analysis tools

[0410] Input: The raw submission data obtained.

[0411] Processing: The server uses Python NLP libraries (e.g., NLTK, spaCy) to preprocess the data, specifically denoising, removing stop words, and normalizing the text.

[0412] Output: Preprocessed text data.

[0413] Specific operations: Noise removal removes HTML tags and special characters, stop word removal removes common words such as "は", "を", and "に", and text normalization standardizes case and removes unnecessary whitespace.

[0414] Step 4: How to apply the machine learning model

[0415] Input: Preprocessed text data.

[0416] Processing: The server performs natural language processing on the preprocessed data, performing morphological and sentiment analysis, specifically extracting key keywords, phrases, and sentiment (positive, negative, neutral) using TensorFlow and PyTorch.

[0417] Output: Extracted keywords, phrases, and sentiment data.

[0418] Specific operation: The server divides the text data into tokens, calculates and classifies the sentiment score of each token, and passes the result on to the next process.

[0419] Step 5: User emotion analysis method using emotion engine

[0420] Input: User text input, voice data, or facial expression data.

[0421] Processing: The server uses an emotion engine (e.g., Tone Analyzer) to analyze the user's emotion and determine whether it is classified as positive, negative, or neutral.

[0422] Output: User sentiment classification result.

[0423] Specific operation: Text and audio files entered by the user are passed to the emotion engine, emotion analysis is performed on each piece of data, and the analysis results are obtained.

[0424] Step 6: How to generate and present recommendations to the user

[0425] Input: Output data from the machine learning model and the sentiment engine.

[0426] Processing: The server integrates this data and generates posts that are optimally tailored to the user's target audience, including recommended keywords, hashtags, post tone, and specific post content templates.

[0427] Output: Generated post content, suggested keywords, suggested hashtags.

[0428] Specific behavior: Keywords to use: "New cafe," "Instagrammable," "Urban area," "Cafe hopping," recommended hashtags: "Cafe hopping," "New discovery," "Instagrammable," "Urban area," post template: "A new cafe has opened! It has a relaxed atmosphere and is sure to be Instagrammable! Cafe hopping, urban area, I want to connect with cafe lovers," and the following will be presented to the user.

[0429] As a result, this system can effectively generate and present buzz posts to the user's target demographic, and can provide optimal content that reflects the user's emotions.

[0430] (Application example 2)

[0431] 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."

[0432] Conventional social media buzz-generating posting systems perform analysis to maximize the impact on the target, but rarely take into account the user's own emotional state. As a result, the content of the post may not match the user's intended emotion or tone, resulting in an ineffective buzz. Another issue is that generating optimal post content to efficiently generate buzz among a specific target takes time and effort.

[0433] 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.

[0434] In this invention, the server includes means for receiving content to be made buzzworthy and target information from a user, means for retrieving related posts from a database storing past buzzworthy post data, means for preprocessing the retrieved post data and performing text analysis using natural language processing, means for inputting the analysis results into a machine learning model to generate optimal post content, means for presenting the generated post content and recommended keywords and hashtags to the user, and means for generating advertising copy based on the user's emotional state using an emotion engine that analyzes the user's emotions. This makes it possible to generate optimal post content according to the user's emotional state and effectively create buzz among the target demographic.

[0435] "Buzzworthy content" refers to content posted on social media that is intended to attract the attention of many users and be spread widely.

[0436] "Target information" refers to information about the target user demographic for buzz posts, such as a specific age group, gender, or interests.

[0437] The "database" is a data storage system that stores past buzz posting data and retrieves related information through queries.

[0438] "Natural language processing" is a technology for analyzing and processing language data, and is used for preprocessing and analyzing text.

[0439] A "machine learning model" is an algorithm that learns patterns from large amounts of data and makes predictions and classifications based on new data.

[0440] An "emotion engine" is a system that analyzes a user's facial expressions and voice to identify the user's current emotional state.

[0441] An "advertising copy" is a post that conveys information about a specific product or service on social media and aims to pique the interest of target users.

[0442] "Noise reduction" is the process of removing irrelevant or false information from the text data being analyzed.

[0443] "Stop word removal" is the process of removing words that are too general to contribute to the analysis (such as "no" and "wa") during text analysis.

[0444] "Normalization" is the process of correcting spelling variations and typos and standardizing data to maintain consistency in text data.

[0445] "Keywords" are important words that represent the main theme or topic of the post content, and increase search and spreading effectiveness on social media.

[0446] A "hashtag" is a linguistic element used on social media to tag posts related to a particular theme or topic, improving the discoverability of the posts.

[0447] The present invention relates to a system that enables efficient communication of content that you want to create buzz on social media to specific targets. By combining this system with an emotion engine that analyzes user emotions, more effective communication of information can be achieved. Specific means for realizing this system are described below.

[0448] This system is implemented primarily using the following hardware and software:

[0449] Hardware

[0450] Smartphone: Used by the user to input information and receive results from the system.

[0451] Server: The central computing resource that stores data, analyzes it, and processes the generated submissions.

[0452] software

[0453] Natural Language Processing (NLP) libraries: Libraries for preprocessing and parsing text (e.g., spaCy, NLTK).

[0454] Machine learning frameworks: Frameworks for learning from large amounts of data and making predictions or classifications (e.g., TensorFlow, PyTorch).

[0455] Emotion analysis engine: Software for analyzing a user's facial expressions and voice to classify their emotional state (e.g., OpenCV, Google® Cloud Speech-to-Text).

[0456] The server is implemented based on a system including the following means:

[0457] 1. Means of accepting input from users

[0458] Users access the system's interface using their smartphones, where they input the content they want to create buzz and target information (age group, gender, interests, etc.). This information is then sent from the smartphone to the server.

[0459] 2. How to obtain information from the database

[0460] The server receives the information sent by the user and then executes a query to retrieve posts related to the user's target information from a database of past buzz posts. The query includes filtering criteria based on the target information.

[0461] 3. Preprocessing and Natural Language Processing Methods

[0462] The server preprocesses the submitted data by removing noise, stop words, and normalizing the text. This step prepares the data for analysis.

[0463] 4. Analysis Methods Using Machine Learning Models

[0464] The server applies natural language processing (NLP) to the preprocessed post data, performing morphological analysis to extract important keywords, phrases, and sentiment (positive, negative, neutral).

[0465] 5. User sentiment analysis using a sentiment analysis engine

[0466] The server uses an emotion analysis engine to analyze the user's emotions. This analysis is performed using the smartphone's camera and microphone to capture the user's facial expressions and voice. Based on the analysis results, the user's emotions are classified as positive, negative, or neutral.

[0467] 6. How to generate and present recommendations to users

[0468] The server adjusts the predictions generated by the machine learning model based on the user's emotions recognized by the emotion analysis engine, and creates optimal post content. Specifically, this includes the keywords, hashtags, tone (emotion) of the post to be used, and specific post content templates. The generated post content is presented to the user on the smartphone screen.

[0469] Specific examples

[0470] For example, if a user wants to "introduce a new product" and specifies the target audience as "20-30 year old women interested in fashion," the system will operate as follows:

[0471] 1. Accepting input from the user

[0472] User entered information: "New product introduction", "Female, 20-30 years old, interested in fashion".

[0473] 2. Retrieving information from the database

[0474] A server receives the information submitted by the user and performs a database query to retrieve relevant past buzz posts.

[0475] 3. Preprocessing and Natural Language Processing Methods

[0476] The server removes noise from the acquired posting data, removes stop words, and normalizes the text.

[0477] 4. Analysis Methods Using Machine Learning Models

[0478] The server applies natural language processing to the preprocessed data to extract important keywords, phrases, and sentiment.

[0479] 5. User Emotion Analysis Method Using an Emotion Engine

[0480] The server analyzes facial expression and voice data sent from the user's device and recognizes that the user's emotions are positive.

[0481] 6. How to generate and present recommendations to users

[0482] The server generates posts that are appropriate for positive emotions based on the analysis results of the emotion engine:

[0483] Recommended keywords: "new products," "trends," "up your style"

[0484] Recommended hashtags: "New products," "Trends," "Fashion," "Style up"

[0485] Post template: "New fashion items have arrived! Don't miss out on these trendy items. Just a glimpse of them will definitely elevate your style! New Products Trends Fashion Style Up"

[0486] Prompt Sentence Examples

[0487] markdown

[0488] New Product Introduction

[0489] Generate positive posts showcasing fashion products targeted at women aged 20-30.

[0490] ---

[0491] Target Information:

[0492] Age range: 20-30 years old

[0493] Gender: Female

[0494] Interests: Fashion

[0495] User's emotional state:

[0496] positive

[0497] Generated content:

[0498] A new fashion item has arrived! This trendy item is a must-see. Just a glimpse of it will definitely enhance your style!

[0499] New products Trend fashion Style up

[0500] In this way, the present invention provides a system that generates optimal posting content based on the user's emotional state and target information, and can efficiently create buzz on SNS.

[0501] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0502] Step 1:

[0503] The content that the user wants to create buzz and target information are accepted.

[0504] Specific behavior:

[0505] Users launch the application on their smartphone and access the interface, where they input the content they want to create buzz about (e.g., "Introducing a new product") and target information (e.g., "Women aged 20-30 who are interested in fashion"). This information is then sent from the smartphone to the server.

[0506] Input: User-entered content and target information they want to create buzz about.

[0507] Output: Information data sent to the server.

[0508] Step 2:

[0509] Retrieve information from a database.

[0510] Specific behavior:

[0511] The server receives the information sent by the user and then executes a database query to retrieve posts related to the specified target information from a database of past buzz posts. The query, including the filtering criteria, is sent to the database.

[0512] Input: User-specified target information.

[0513] Output: Relevant buzz post data retrieved from the database.

[0514] Step 3:

[0515] Preprocess the acquired post data.

[0516] Specific behavior:

[0517] The server preprocesses the submitted data, removing noise, stop words, and normalizing the text to make it easier to analyze. This process uses natural language processing technology.

[0518] Input: Post data retrieved from the database.

[0519] Output: Preprocessed text data.

[0520] Step 4:

[0521] Perform text analysis on the preprocessed data.

[0522] Specific behavior:

[0523] The server applies natural language processing (NLP) to the preprocessed data, performing morphological analysis to extract key keywords, phrases, and sentiment (positive, negative, neutral). NLP libraries (e.g., spaCy, NLTK) are used.

[0524] Input: Preprocessed text data.

[0525] Output: Parsed keywords, phrases and sentiment information.

[0526] Step 5:

[0527] The analysis results are input into a machine learning model to generate optimal post content.

[0528] Specific behavior:

[0529] The server inputs the data analyzed by natural language processing into a machine learning model. The machine learning model predicts the optimal posting pattern for the target based on the results learned from past buzz post data. A machine learning framework (e.g., TensorFlow, PyTorch) is used here.

[0530] Input: Parsed data (keywords, phrases, sentiment information).

[0531] Output: Generated post content, suggested keywords and hashtags.

[0532] Step 6:

[0533] Analyze user sentiment.

[0534] Specific behavior:

[0535] The server analyzes the user's emotions using an emotion engine. It collects and analyzes the user's facial expression and voice data using the smartphone's camera and microphone. The emotion engine classifies the user's emotional state (positive, negative, neutral).

[0536] Input: User's facial expression data, voice data.

[0537] Output: The user's emotional state (positive, negative, neutral).

[0538] Step 7:

[0539] Recommendations are generated and presented to the user.

[0540] Specific behavior:

[0541] The server generates and adjusts optimal post content based on the user's emotional state. It combines the analysis results from the emotion engine with the output of the machine learning model to create templates for keywords, hashtags, post tone (emotion), and specific post content to use. The generated post content is presented to the user on their smartphone screen.

[0542] Input: Sentiment engine analysis results, machine learning model output.

[0543] Output: The post content, suggested keywords, and hashtags presented to the user.

[0544] This allows users to upload content to social media that best suits their emotional state, making it possible to create buzz efficiently and effectively.

[0545] 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.

[0546] 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.

[0547] 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.

[0548] [Second embodiment]

[0549] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0550] 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.

[0551] 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).

[0552] 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.

[0553] 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.

[0554] 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).

[0555] 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.

[0556] 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.

[0557] 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.

[0558] 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.

[0559] 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.

[0560] 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."

[0561] This invention relates to a system for efficiently communicating content that you want to create buzz on social networking sites to specific targets. Below, we will explain the program processing of this system in natural language.

[0562] System configuration

[0563] The system mainly consists of the following means:

[0564] 1. Means of accepting input from users

[0565] 2. How to obtain information from the database

[0566] 3. Natural Language Processing and Text Analysis Methods

[0567] 4. How to apply machine learning models

[0568] 5. How to generate and present recommendations to users

[0569] Program processing

[0570] 1. Means of accepting input from users

[0571] Users access the system's interface using a device (such as a smartphone or PC). They input the content they want to create buzz about and target information (age group, gender, interests, etc.). This information is then sent from the device to the server.

[0572] 2. How to obtain information from the database

[0573] The server receives the information sent by the user and then executes a query to retrieve buzz posts related to the specified target information from a database that stores past buzz post data. Here, the query includes filtering conditions based on the target information.

[0574] 3. Natural Language Processing and Text Analysis Methods

[0575] The server preprocesses the acquired post data, including noise removal, stop word removal, and text normalization. It then uses natural language processing (NLP) techniques to perform morphological analysis of the text and extract important keywords, phrases, and sentiment (positive, negative, neutral).

[0576] 4. How to apply machine learning models

[0577] The server then inputs the data into a machine learning model based on the analysis results. This model learns from past success stories (buzzworthy posts) and predicts the best posting pattern for the target audience. The output generates templates for keywords, hashtags, post tone (emotion), and specific post content to use.

[0578] 5. How to generate and present recommendations to users

[0579] The server then compiles the final recommendations and creates a list of suggested posts for the user, which may include, for example:

[0580] Example keywords

[0581] Hashtag examples

[0582] Post Template

[0583] The device displays the received recommendations to the user, who can then review and edit them and post them directly to the social networking site.

[0584] Specific examples

[0585] For example, if a user wants to create buzz about the opening of a new cafe and specifies the target audience as "women in their 20s who live in urban areas and love cafes," the system will operate as follows:

[0586] 1. Accepting input from the user

[0587] User input: "New cafe opening", "Female in her 20s, urban resident, cafe lover"

[0588] 2. Retrieving information from the database

[0589] The server runs a database query to retrieve past trending posts, such as:

[0590] "Cafe hopping and new discoveries"

[0591] "I went to a new cafe! It's stylish and comfortable - perfect for Instagram."

[0592] "All the cafes in the city are great, but this one was especially great! I want to connect with other cafe lovers."

[0593] 3. Natural Language Processing and Text Analytics

[0594] The server extracts important keywords and phrases from the above submission data:

[0595] Keywords: "cafe," "new," "urban," "Instagrammable"

[0596] Emotion: Positive

[0597] 4. Applying machine learning models

[0598] The server feeds the extracted data into a machine learning model to generate recommendations for:

[0599] Recommended keywords: "new cafe," "Instagrammable," "urban area," "cafe hopping"

[0600] Recommended hashtags: "cafe hopping," "new discoveries," "Instagrammable," "urban areas"

[0601] Post template: "A new cafe has opened! It has a relaxed atmosphere and is sure to be Instagram-worthy! Cafe hopping, urban areas, and connecting with fellow cafe lovers."

[0602] 5. Generating and presenting recommendations to the user

[0603] The device displays the recommended content to the user, who then checks and edits it and posts it to the social networking site.

[0604] The above is a specific embodiment of the present invention, which makes it possible to efficiently and accurately generate and present buzzworthy posts aimed at a target audience.

[0605] The processing flow will be explained below.

[0606] Step 1:

[0607] Users access the system's interface using a device (smartphone or PC). They input the content they want to create buzz about and target information (age group, gender, interests, etc.). This information is then sent from the device to the server.

[0608] Step 2:

[0609] The server receives the information sent by the user and then executes a query from a database of past buzz posts to retrieve posts related to the target information specified by the user. The query includes filtering conditions based on the target information.

[0610] Step 3:

[0611] The server preprocesses the submitted data retrieved from the database by removing noise, removing stop words, and normalizing the text, making the data easier to analyze.

[0612] Step 4:

[0613] The server then applies natural language processing (NLP) to the preprocessed post data. This step involves morphological analysis to extract important keywords, phrases, and sentiment (positive, negative, neutral).

[0614] Step 5:

[0615] The server then inputs the extracted keywords, phrases, and sentiment data into a machine learning model, which learns from past success stories (buzzworthy posts) and predicts the optimal posting pattern for the target audience.

[0616] Step 6:

[0617] The server creates recommended posts for users based on the predictions generated by the machine learning model, including optimal keywords, hashtags, post tone (emotion), and specific post content templates.

[0618] Step 7:

[0619] The server sends the generated recommendations to the device, which then displays them to the user. The user can then review and edit the recommendations and post them directly to the social networking site.

[0620] Specific examples

[0621] Step 1:

[0622] A user enters the following information into the system: "A new cafe is opening" and "A woman in her 20s who loves cafes and lives in an urban area."

[0623] Step 2:

[0624] A server receives the information submitted by the user and performs a database query to retrieve relevant past buzz posts.

[0625] Step 3:

[0626] The server removes noise from the acquired posting data, removes stop words, and normalizes the text.

[0627] Step 4:

[0628] The server applies natural language processing to the preprocessed data to extract important keywords, phrases, and sentiment.

[0629] Step 5:

[0630] The server inputs the extracted data into a machine learning model to predict the optimal posting patterns for the target audience.

[0631] Step 6:

[0632] The server generates recommended posts based on predictions from the machine learning model.

[0633] Step 7:

[0634] The server sends the generated recommendations to the device, which displays them to the user, who then checks and edits them and posts them to the SNS.

[0635] Through these processing steps, the system can effectively generate and present buzz posts to the user's target audience.

[0636] Example 1

[0637] 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."

[0638] Currently, many users are using social networking services (SNS) to spread information, but it is not easy to efficiently create buzz for a specific target audience. Existing methods make it difficult to predict what content and format will create buzz, and require a great deal of time and effort. Furthermore, selecting appropriate keywords and hashtags is difficult, so generating optimal post content requires specialized knowledge. Therefore, there is a need for a system that can efficiently and accurately generate and present buzzworthy posts for specific targets.

[0639] 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.

[0640] In this invention, the server includes means for receiving content and target information from a user that the user wants to create a buzz, means for retrieving related posts from a database storing past buzz post data, means for preprocessing the retrieved post data and performing text analysis using natural language processing, means for inputting the analysis results into a machine learning model to generate optimal post content, means for presenting the generated post content and recommended keywords and hashtags to the user, and means for the user to edit and confirm the generated post content and then post it to an SNS. This allows users to effectively create buzz posts aimed at a specific target demographic.

[0641] "Means for accepting content to be buzzed and target information from users" refers to an interface or function that allows users to input the content they want to buzz and information about the target person, and send it to the server.

[0642] "Means for retrieving related posts from a database storing data on past buzz posts" refers to a function in which the server queries the database, retrieves data on past buzz posts, and retrieves posts related to the target information specified by the user.

[0643] "Means for preprocessing acquired posted data and performing text analysis using natural language processing" refers to a function that performs preprocessing on acquired posted data, such as noise removal, stop word removal, and text normalization, and then analyzes the text using natural language processing technology.

[0644] "A means of inputting the analysis results into a machine learning model to generate optimal post content" refers to a function that inputs data obtained from text analysis into a machine learning model and uses the model, which has learned from past success stories, to generate post content that is most suitable for the target demographic.

[0645] "Means of presenting generated post content and recommended keywords and hashtags to users" refers to the function of organizing post content, keywords, hashtags, etc. generated by machine learning models and displaying and suggesting them to users.

[0646] "Means for users to post generated posts to SNS after editing and reviewing them" refers to a function that allows users to review and edit the suggested content and post it directly to the SNS platform.

[0647] "Preprocessing of posted data retrieved from the database" refers to the process of removing unnecessary information (noise) from the posted data, deleting commonly used words (stop words) that are irrelevant to a specific analysis, and normalizing the strings to maintain consistency.

[0648] "Natural language processing" is a technology that enables computers to understand, interpret, and generate human language, and in the present invention refers to the technology used for morphological analysis of text data and extraction of important keywords and emotions.

[0649] A "machine learning model" refers to an algorithm that learns patterns from given data and makes predictions and classifications for new data. In the context of this invention, it is used to learn from past buzz posts and generate post content that is optimal for the target demographic.

[0650] "Target information" refers to the attribute information (e.g., age group, gender, interests, etc.) of the intended recipients of the post that the user wants to make go viral, and the content of the post is optimized based on that information.

[0651] This invention relates to a system for efficiently communicating content that users want to create buzz on social media to specific targets. This system accepts input from users, acquires and analyzes relevant data from past buzz posts, and generates optimal post content using a machine learning model, which is then presented to the user. The configuration and operation of this system are described in detail below.

[0652] System configuration

[0653] The system mainly consists of the following means:

[0654] 1. Means of accepting input from users

[0655] 2. How to obtain information from the database

[0656] 3. Natural Language Processing and Text Analysis Methods

[0657] 4. How to apply machine learning models

[0658] 5. How to generate and present recommendations to users

[0659] 6. Means of posting to social media

[0660] Processing Details

[0661] 1. Accepting input from the user

[0662] Users access the system's interface using a smartphone or PC. They input the content they want to create buzz and target information (age group, gender, interests, etc.), and this information is sent from the device to the server. The input is in a specific data format, and the server receives it and begins processing.

[0663] 2. Retrieving information from the database

[0664] After receiving the information sent by the user, the server executes a database query based on that information. The database stores past buzz post data and retrieves relevant post data based on specific filtering criteria. For example, a query is executed to retrieve information about "women in their 20s, urban areas, cafe lovers."

[0665] 3. Natural Language Processing and Text Analytics

[0666] The server preprocesses the acquired post data. This preprocessing includes noise removal (removal of special symbols and unnecessary spaces), stopword removal (frequent phrases that are generally ignored), and text normalization (standardization of characters). Then, using natural language processing technology (e.g., the morphological analysis tool MeCab), it performs morphological analysis of the text data and extracts important keywords, phrases, and sentiment (positive, negative, neutral).

[0667] 4. Applying machine learning models

[0668] Based on the analysis results, the server inputs the data into a machine learning model. The machine learning model learns from past successful posting data and predicts the optimal posting pattern for the target demographic. For example, the model predicts the keywords, hashtags, and tone (emotion) to be used in the post, and generates a specific post content template.

[0669] 5. Generating and presenting recommendations to the user

[0670] Finally, the server presents recommendations to the user, including example keywords, example hashtags, and post templates, which the user can review and edit to select the post they deem most appropriate.

[0671] 6. How to post to social media

[0672] The final post content that users have considered can be posted directly to the SNS platform, and users can easily publish the generated post content to the SNS through their device.

[0673] Specific examples

[0674] For example, if a user wants to create buzz about the opening of a new cafe and specifies the target audience as "women in their 20s who live in urban areas and love cafes," the system will operate as follows:

[0675] 1. Accepting input from the user

[0676] User input: "New cafe opening", "Female in her 20s, urban resident, cafe lover"

[0677] 2. Retrieving information from the database

[0678] The server runs a database query to retrieve past viral posts, such as:

[0679] "Cafe hopping and new discoveries"

[0680] "I went to a new cafe! It's stylish and comfortable - perfect for Instagram."

[0681] "All the cafes in the city are great, but this one was especially great! I want to connect with other cafe lovers."

[0682] 3. Natural Language Processing and Text Analytics

[0683] The server extracts important keywords and phrases from the above submission data:

[0684] Keywords: "cafe," "new," "urban," "Instagrammable"

[0685] Emotion: Positive

[0686] 4. Applying machine learning models

[0687] The server feeds the extracted data into a machine learning model to generate the following recommendations:

[0688] Recommended keywords: "new cafe," "Instagrammable," "urban area," "cafe hopping"

[0689] Recommended hashtags: "cafe hopping," "new discoveries," "Instagrammable," "urban areas"

[0690] Post template: "A new cafe has opened! It has a relaxed atmosphere and is sure to be Instagram-worthy! Cafe hopping, urban areas, and connecting with fellow cafe lovers."

[0691] 5. Generating and presenting recommendations to the user

[0692] The device displays the recommended content to the user, who can then review it, edit it if necessary, and post it to social media.

[0693] Examples of prompt statements

[0694] You might enter the following prompts for the system's generated AI model:

[0695] "I want to create a buzz about the opening of a new cafe among women in their 20s who love cafes and live in urban areas. Based on past data on buzzworthy posts, could you recommend some keywords, hashtags, and a post template?"

[0696] The above is a specific embodiment of the present invention, which makes it possible to efficiently and accurately generate and present buzzworthy posts aimed at a target audience.

[0697] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0698] Step 1: Accepting input from the user

[0699] Users access the system's interface using a smartphone or PC. They input the content they want to create buzz about (e.g., "A new cafe has opened") and target information (e.g., "Women in their 20s who live in urban areas and love cafes"). The device receives this information and sends it to the server. The server temporarily stores the input information before proceeding to the next processing step.

[0700] Step 2: Retrieving information from the database

[0701] The server receives target information sent by the user. Based on this information, the server sends a query to the database to retrieve related posts from past buzz post data. For example, it searches the database for posts related to "women in their 20s" and "cafes" and extracts related post data. Input: Target information. Output: Related past buzz post data.

[0702] Step 3: Natural Language Processing and Text Analytics

[0703] The server preprocesses the acquired post data. Preprocessing includes removing noise (removing special symbols and unnecessary spaces), deleting stop words (frequent words that are generally ignored), and normalizing the text (standardizing characters). Natural language processing techniques (such as morphological analysis tools) are then used to perform morphological analysis of the text data, extract important keywords and phrases, and analyze sentiment (positive, negative, neutral). Input: Past buzz post data. Output: Preprocessed text data and analysis results.

[0704] Step 4: Applying the machine learning model

[0705] The server inputs the preprocessed text data and analysis results into a machine learning model. The machine learning model learns from past posting data of successful cases and predicts the optimal posting pattern for the target demographic. This prediction generates templates for the keywords, hashtags, tone (emotion) of the post, and specific post content to be used. Input: Analysis results. Output: Optimal post content, recommended keywords, recommended hashtags.

[0706] Step 5: Generate recommendations and present them to the user

[0707] The server organizes the final recommendations and creates a list of post content to suggest to the user. The list includes example keywords, example hashtags, and post templates. The device displays the received recommendations to the user. The user can review and edit these recommendations and select the most suitable post content. Input: Best post content, recommended keywords, recommended hashtags. Output: List of posts presented to the user.

[0708] Step 6: How to post to social media

[0709] The user reviews the suggested content and edits it as necessary. Through the device interface, the user can post the final post directly to social media. This allows for efficient spreading of content that is targeted to a specific demographic. Input: Final post content after user editing. Output: Post to social media.

[0710] (Application example 1)

[0711] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0712] In order to efficiently create buzz on social media for specific content, it is necessary to generate posts that are optimal for the target audience. However, conventional systems that utilize artificial intelligence and machine learning require the manual collection and analysis of data and the generation of post content, which is time-consuming and labor-intensive. In addition, there is uncertainty as to whether the generated post content will actually appeal to the target, so a reliable method is needed to increase the probability of buzz.

[0713] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0714] In this invention, the server includes means for receiving content and target information from a user that the user wants to create a buzz, means for retrieving related posts from a database storing past buzz post data, means for preprocessing the retrieved post data and performing text analysis using natural language processing, means for inputting the analysis results into a machine learning model to generate optimal post content, means for presenting the generated post content and recommended keywords and hashtags to the user, and means for posting the generated optimal post content to an SNS via an application on a smart device. This allows users to avoid complex manual data collection and analysis work, quickly and easily generate post content that is most effective for their target, and significantly increases the chances of creating a buzz on an SNS.

[0715] "User" refers to any individual or entity that uses the System.

[0716] "Buzzworthy content" refers to information or messages posted on social media with the aim of attracting a lot of attention.

[0717] "Target information" is data that indicates the attributes and interests of the target audience for a particular post, including, for example, age group, gender, place of residence, and interests.

[0718] "Database" refers to a collection of information that stores collected data on past buzz posts and makes it searchable and retrievalable.

[0719] "Preprocessing" refers to the process of processing the data by removing noise, deleting stop words, normalizing text, etc. from the acquired submission data.

[0720] "Natural language processing" refers to the technology that enables computers to understand and analyze human language.

[0721] "Text analysis" refers to the process of using natural language processing techniques to interpret the meaning and structure of a document and extract important information.

[0722] A "machine learning model" refers to a statistical method for learning patterns and rules from past data and making predictions and classifications for future data.

[0723] "Keywords" are particularly important words or phrases in your post that will grab your target's attention.

[0724] A "hashtag" is a type of tagging notation used in social media posts, and refers to a keyword that makes it easier to group posts related to a particular topic or theme.

[0725] "Smart device" refers to a portable electronic device that can connect to a network, such as a smartphone or tablet.

[0726] "Application" refers to a software program that runs on a smart device and provides specific functionality.

[0727] "SNS" is an abbreviation for social networking service, which refers to a platform where users can share information and interact with each other online.

[0728] This invention relates to a system for efficiently communicating content that you want to create buzz on SNS to designated targets, and includes the following means and processes.

[0729] Program processing explanation

[0730] The server receives the content and target information from the user and performs the following processing based on this: First, it retrieves data related to the specified target information from a database of past buzz posts, and then performs preprocessing to remove noise, stop words, and normalize the data.

[0731] The server then performs text analysis using natural language processing (NLP) techniques, including morphological analysis, keyword extraction, and sentiment analysis of the acquired data, using software such as NLTK (Natural Language Toolkit) and Scikit-learn.

[0732] The server then inputs the analyzed data into a machine learning model, which uses classification techniques such as logistic regression to predict the best post content, keywords, and hashtags for the target audience based on past success stories.

[0733] Finally, the generated optimal post content, recommended keywords, and hashtags are presented to the user, who posts it to a social networking site via an application on their smart device.

[0734] Specific examples

[0735] For example, if a user wants to create buzz about the opening of a new cafe and specifies "women in their 20s who love cafes and live in urban areas" as their target, the system will operate as follows.

[0736] User Input

[0737] The user wants to create buzz: "New cafe opening"

[0738] Target information: "Women in their 20s who love cafes and live in urban areas"

[0739] Server Processing

[0740] Get past trending posts related to your target information from the database. Example:

[0741] "Cafe hopping and new discoveries"

[0742] "I went to a new cafe! It's stylish and comfortable - perfect for Instagram."

[0743] "All the cafes in the city are great, but this one was especially great! I want to connect with other cafe lovers."

[0744] Keyword extraction from analysis results:

[0745] Keywords: "cafe," "new," "urban," "Instagrammable"

[0746] Emotion: Positive

[0747] Generate suggested posts

[0748] Recommended keywords: "new cafe," "Instagrammable," "urban area," "cafe hopping"

[0749] Recommended hashtags: "cafe hopping," "new discoveries," "Instagrammable," "urban areas"

[0750] Post template: "A new cafe has opened! It has a relaxed atmosphere and is sure to be Instagram-worthy! Cafe hopping, urban areas, and connecting with fellow cafe lovers."

[0751] Prompt Sentence Examples

[0752] The following sentences can be used as prompts for users to input to a generative AI model:

[0753] "Generate the best social media post content to promote the opening of a new cafe. The target audience is women in their 20s who love cafes and live in urban areas. Past related post data is shown below. Please include keywords and hashtags that should be used in the post."

[0754] This allows users to quickly generate posts that are suitable for their target audience and are likely to create buzz, and then easily post them on social media.

[0755] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0756] Step 1:

[0757] Users enter the content they want to create buzz on social media and target information.

[0758] Input: What you want to create a buzz about (e.g., the opening of a new cafe), target information (e.g., women in their 20s who love cafes and live in urban areas)

[0759] What happens: A user provides input via an application on their smart device, which is sent to the server.

[0760] Step 2:

[0761] The server retrieves post data related to the target information from a database of past buzz posts.

[0762] Input: Target information sent by the user

[0763] Data processing: Perform database queries based on target information and filter relevant post data

[0764] Output: Data of past buzz posts (e.g., "Cafe hopping, new discoveries")

[0765] What happens: The server uses SQL to run a search query against the database to retrieve the relevant data.

[0766] Step 3:

[0767] Preprocess the submitted data received by the server.

[0768] Input: Previous buzz post data obtained

[0769] Data processing: noise removal, stopword removal, text normalization

[0770] Output: Clean post data after preprocessing

[0771] Specific operation: The server uses a natural language processing library such as NLTK to preprocess the text data.

[0772] Step 4:

[0773] The server performs text analysis using natural language processing technology.

[0774] Input: Clean post data after preprocessing

[0775] Data processing: morphological analysis of text, keyword extraction, sentiment analysis

[0776] Output: Key keywords, phrases, and emotional state (positive, negative, neutral)

[0777] Specific operation: The server uses Scikit-learn or similar tools to apply models for morphological analysis and keyword extraction.

[0778] Step 5:

[0779] The server inputs the analysis results into a machine learning model to generate optimal post content.

[0780] Input: Keywords, phrases, and emotional states obtained through natural language processing

[0781] Data Computing: Prediction and Content Generation with Machine Learning Models

[0782] Output: Best post content, recommended keywords, recommended hashtags, post templates

[0783] What it does: The server uses machine learning models such as logistic regression to generate targeted posts.

[0784] Step 6:

[0785] The server presents the generated post content and recommended keywords and hashtags to the user.

[0786] Input: Best post content, suggested keywords, suggested hashtags, post template

[0787] Output: What the user sees on their smart device

[0788] Specific behavior: The server sends the generated content to the user interface so that the user can view it on their smart device.

[0789] Step 7:

[0790] The optimal post content that the user has checked and edited is posted to SNS.

[0791] Input: Post content provided by the server, edits made by the user

[0792] Output: Posts published on social media

[0793] Specific behavior: The user publishes a post using the interface that allows them to post directly to the social networking site from the application.

[0794] 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.

[0795] This invention relates to a system that enables efficient communication of content that you want to create buzz on social media to specific targets. This system achieves even more effective communication by combining it with an emotion engine that recognizes and analyzes user emotions. Below, we will explain the program processing of this system in natural language.

[0796] System configuration

[0797] The system mainly consists of the following means:

[0798] 1. Means of accepting input from users

[0799] 2. How to obtain information from the database

[0800] 3. Natural Language Processing and Text Analysis Methods

[0801] 4. How to apply machine learning models

[0802] 5. User Emotion Analysis Method Using Emotion Engine

[0803] 6. How to generate and present recommendations to users

[0804] Program processing

[0805] 1. Means of accepting input from users

[0806] Users access the system's interface using a device (such as a smartphone or PC). They input the content they want to create buzz about and target information (age group, gender, interests, etc.). This information is then sent from the device to the server.

[0807] 2. How to obtain information from the database

[0808] The server receives the information sent by the user and then executes a query from a database of past buzz posts to retrieve posts related to the target information specified by the user. The query includes filtering conditions based on the target information.

[0809] 3. Natural Language Processing and Text Analysis Methods

[0810] The server preprocesses the submitted data by removing noise, removing stop words, and normalizing the text, making the data easier to analyze.

[0811] 4. How to apply machine learning models

[0812] The server then applies natural language processing (NLP) to the preprocessed post data. This step involves morphological analysis to extract important keywords, phrases, and sentiment (positive, negative, neutral).

[0813] 5. User Emotion Analysis Method Using Emotion Engine

[0814] The server uses an emotion engine to analyze the user's emotions based on the user's facial expressions, tone of voice, or text input. Based on the analysis results, the user's emotions are classified as positive, negative, or neutral.

[0815] 6. How to generate and present recommendations to users

[0816] The server adjusts the predictions generated by the machine learning model based on the user's emotions recognized by the emotion engine to create optimal post content, including keywords, hashtags, post tone (emotion), and specific post content templates to use.

[0817] For example, if the user's emotions are positive, the system generates posts with a bright and cheerful tone. If the emotions are negative, the system emphasizes an encouraging and comforting tone.

[0818] Specific examples

[0819] For example, if a user wants to create buzz about the opening of a new cafe and specifies the target audience as "women in their 20s who live in urban areas and love cafes," the system will operate as follows:

[0820] 1. Accepting input from the user

[0821] User input: "New cafe opening", "Female in her 20s, urban resident, cafe lover"

[0822] 2. Retrieving information from the database

[0823] A server receives the information submitted by the user and performs a database query to retrieve relevant past buzz posts.

[0824] 3. Natural Language Processing and Text Analytics

[0825] The server removes noise from the acquired posting data, removes stop words, and normalizes the text.

[0826] 4. Applying machine learning models

[0827] The server applies natural language processing to the preprocessed data to extract important keywords, phrases, and sentiment.

[0828] 5. User Emotion Analysis Using an Emotion Engine

[0829] The server analyzes facial expression data, voice data, or input text sent from the user's device and recognizes that the user's emotions are positive.

[0830] 6. Generating and Presenting Recommendations to the User

[0831] The server generates posts that are appropriate for positive emotions based on the analysis results of the emotion engine:

[0832] Recommended keywords: "new cafe," "Instagrammable," "urban area," "cafe hopping"

[0833] Recommended hashtags: "cafe hopping," "new discoveries," "Instagrammable," "urban areas"

[0834] Post template: "A new cafe has opened! It has a relaxed atmosphere and is sure to be Instagram-worthy! Cafe hopping, urban areas, and connecting with fellow cafe lovers."

[0835] Through the above processing steps, the system can effectively generate and present buzz posts to the user's target demographic, and provide optimal content based on the user's emotions.

[0836] The processing flow will be explained below.

[0837] Step 1:

[0838] Users access the system's interface using a device (smartphone or PC). They input the content they want to create buzz about (e.g., "A new cafe has opened") and target information (e.g., "Women in their 20s who live in urban areas and love cafes"). This information is then sent from the device to the server.

[0839] Step 2:

[0840] The server receives the input information sent by the user. It then executes a query to retrieve posts related to the specified target information from a database of past buzz posts. For example, it filters past posts related to "women in their 20s" and "cafes" based on the user's target information.

[0841] Step 3:

[0842] The server preprocesses the submitted data retrieved from the database by removing noise, removing stop words, and normalizing the text, making the data easier to analyze.

[0843] Step 4:

[0844] The server then applies natural language processing (NLP) to the preprocessed post data. This step involves morphological analysis to extract important keywords (e.g., "new cafe"), phrases, and sentiment (positive, negative, neutral).

[0845] Step 5:

[0846] The server inputs data into a machine learning model based on keywords and sentiment extracted through natural language processing. This model learns from past success stories (buzzworthy posts) and predicts the best posting pattern for the target audience. For example, it generates recommended hashtags such as "Instagrammable" and "cafe hopping."

[0847] Step 6:

[0848] The server analyzes the user's emotions using an emotion engine. This analysis is based on the user's facial expression recognition, voice analysis, or text input. For example, if the user has a positive expression while typing "new cafe," this emotion is recognized as positive.

[0849] Step 7:

[0850] The server adjusts the predictions generated by the machine learning model based on the user's emotions recognized by the emotion engine. For example, if the user's emotions are positive, it generates posts with a bright tone. On the other hand, if the emotions are negative, it generates posts that contain encouragement and comfort.

[0851] Step 8:

[0852] The server then compiles the final recommendations and creates a list of suggested posts for the user, including the best keywords, hashtags, and post templates, for example:

[0853] Recommended keywords: "new cafe," "Instagrammable," "urban area," "cafe hopping"

[0854] Recommended hashtags: "cafe hopping," "new discoveries," "Instagrammable," "urban areas"

[0855] Post template: "A new cafe has opened! It has a relaxed atmosphere and is sure to be Instagram-worthy! Cafe hopping, urban areas, and connecting with fellow cafe lovers."

[0856] Step 9:

[0857] The server sends the generated recommendations to the device. The device displays the received recommendations to the user. The user can then review and edit the recommendations and post them directly to the SNS.

[0858] Through the above processing steps, the system can effectively generate buzz posts for the user's target demographic and provide optimal content according to the user's emotions.

[0859] Example 2

[0860] 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."

[0861] With conventional SNS posting systems, it was difficult for users to effectively communicate information to their target audience. Also, because only uniform post content could be generated without considering the user's emotions, the effectiveness of the posts was limited. As a result, it was difficult for the posts to go viral, and it was difficult for the posts to reach the intended audience.

[0862] 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.

[0863] In this invention, the server includes means for receiving content to be made buzzworthy and target information from a user, means for retrieving related posts from a database storing past buzz posting data, means for preprocessing the retrieved posting data and performing text analysis using natural language processing, means for recognizing and analyzing user emotions using a sentiment analysis engine, means for inputting the analysis results into a machine learning algorithm to generate optimal posting content, and means for presenting the generated posting content and recommended keywords and hashtags to the user. This makes it possible to generate posting content that is effective for the target demographic and in line with their emotions.

[0864] "Means for receiving content to be buzzed and target information from a user" refers to an interface or process for receiving content to be buzzed and target information input by a user.

[0865] "Means for retrieving related posts from a database storing past buzz posting data" refers to a function for searching and retrieving information related to the user's requirements from stored past buzz posting data.

[0866] "Means for preprocessing acquired posted data and performing text analysis using natural language processing" refers to the process of removing noise, deleting stop words, and normalizing acquired posted data, and then analyzing the content using natural language processing.

[0867] "Means for recognizing and analyzing user emotions using an emotion analysis engine" refers to the function of analyzing emotions from data such as text, voice, and facial expressions entered by the user and classifying them as positive, negative, neutral, etc.

[0868] "Method of inputting analysis results into a machine learning algorithm to generate optimal post content" refers to the process of generating appropriate post content through a machine learning model based on the results of natural language processing and sentiment analysis.

[0869] "Means for presenting generated post content and recommended keywords and hashtags to users" refers to functions and interfaces for providing users with post content generated by the system and effective keyword and hashtag suggestions.

[0870] A specific embodiment for carrying out the present invention will be described below. The system of the present invention realizes effective information transmission by combining users, terminals, and servers.

[0871] System configuration and hardware / software used

[0872] 1. Means of accepting input from users

[0873] Users access the system interface via a web browser on their device (e.g., smartphone, PC). They then access the system and enter the content they want to create buzz about and target information (age group, gender, interests, etc.). This information is sent to the server as an HTTP request.

[0874] 2. How to obtain information from the database

[0875] The server receives the information sent by the user, then executes an SQL query against a database of past buzz posts (e.g., MySQL or PostgreSQL) to search for and retrieve posts related to the target information specified by the user.

[0876] 3. Natural Language Processing and Text Analysis Methods

[0877] The server uses a Python NLP library (e.g., NLTK, spaCy) to preprocess the submitted data. Specifically, it performs the following steps:

[0878] Noise removal: Remove unnecessary special characters and HTML tags.

[0879] Stopword removal: Removal of common words that have no meaning (e.g. "wa", "o", "ni").

[0880] Text normalization: Convert all text to lower case for consistency.

[0881] 4. How to apply machine learning models

[0882] The server then performs natural language processing (NLP) on the preprocessed text data, including morphological analysis and sentiment analysis, using TensorFlow and PyTorch to extract key keywords, phrases, and sentiment (positive, negative, neutral).

[0883] 5. User Emotion Analysis Method Using Emotion Engine

[0884] The server uses an emotion engine (e.g., IBM Watson's Tone Analyzer) to analyze emotions from the user's input data (e.g., text, voice, facial expressions, etc.). Based on the analysis results, the user's emotions are classified as positive, negative, or neutral.

[0885] 6. How to generate and present recommendations to users

[0886] The server combines the results of the machine learning model and the emotion engine to generate optimal posts for the target users. Specifically, it generates the following elements:

[0887] Recommended Keywords

[0888] Recommended hashtags

[0889] The tone (emotion) of the post

[0890] Specific post content template

[0891] The user can review the generated post and make any necessary adjustments.

[0892] Specific examples

[0893] For example, if a user wants to create buzz about the opening of a new cafe and specifies the target audience as "women in their 20s who live in urban areas and love cafes," the system will operate as follows:

[0894] 1. Accepting input from the user

[0895] The user fills in the form with "a new cafe opening" and "a woman in her 20s who loves cafes and lives in an urban area" and submits it.

[0896] 2. Retrieving information from the database

[0897] Based on the input information received by the server, a database query is executed to retrieve related past buzz posts.

[0898] 3. Natural Language Processing and Text Analytics

[0899] The server removes noise from the acquired submission data, removes stop words, and normalizes the text.

[0900] 4. Applying machine learning models

[0901] The server applies NLP to the pre-processed data to extract important keywords, phrases, and sentiment.

[0902] 5. User Emotion Analysis Using an Emotion Engine

[0903] The server analyzes the user's text input and recognizes it as a positive emotion.

[0904] 6. Generating and Presenting Recommendations to the User

[0905] Based on the sentiment and NLP results, the server generates the following post with a positive tone:

[0906] Recommended keywords: "new cafe," "Instagrammable," "urban area," "cafe hopping"

[0907] Recommended hashtags: "cafe hopping," "new discoveries," "Instagrammable," "urban areas"

[0908] Post template: "A new cafe has opened! It has a relaxed atmosphere and is sure to be Instagram-worthy! Cafe hopping, urban areas, and I want to connect with fellow cafe lovers."

[0909] Prompt Sentence Examples

[0910] Example prompts to input to a generative AI model:

[0911] We want to create buzz on social media about the opening of a new cafe. Our target audience is women in their 20s who love cafes and live in urban areas. We will generate effective posts based on data on past buzz posts. Since many users have positive feelings, we recommend creating posts with a bright and fun tone.

[0912] As a result, this system can not only effectively generate and present buzz posts to the user's target demographic, but also provide optimal content that reflects the user's emotions.

[0913] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0914] Divide the processing flow of the system program into processing steps

[0915] Step 1:

[0916] Step 2:

[0917] Step 3:

[0918] Step 4:

[0919] Step 5:

[0920] Step 6:

[0921] Specific explanation of each processing step

[0922] Step 1: Accepting input from the user

[0923] Input: A user accesses the system interface using a terminal and enters the content they want to buzz and target information into a form.

[0924] Processing: The terminal sends the input information to the server in the form of an HTTP request.

[0925] Output: The server temporarily stores the received content and target information in memory.

[0926] What happens: A user opens a browser, fills in a web form with the words "New cafe opening" and "Female, 20s, urban resident, cafe lover," and clicks the submit button.

[0927] Step 2: How to retrieve information from the database

[0928] Input: The server generates a database query based on the buzz content and target information received from the user.

[0929] Processing: The server executes an SQL query against a database of past buzz posts (e.g., MySQL, PostgreSQL) to search and retrieve relevant past posts.

[0930] Output: A dataset containing the retrieved submission data.

[0931] Specific operation: The server queries the database using the target information of "women in their 20s" and "urban areas," and extracts related past buzz posts using an SQL query.

[0932] Step 3: Natural language processing and text analysis tools

[0933] Input: The raw submission data obtained.

[0934] Processing: The server uses Python NLP libraries (e.g., NLTK, spaCy) to preprocess the data, specifically denoising, removing stop words, and normalizing the text.

[0935] Output: Preprocessed text data.

[0936] Specific operations: Noise removal removes HTML tags and special characters, stop word removal removes common words such as "は", "を", and "に", and text normalization standardizes case and removes unnecessary whitespace.

[0937] Step 4: How to apply the machine learning model

[0938] Input: Preprocessed text data.

[0939] Processing: The server performs natural language processing on the preprocessed data, performing morphological and sentiment analysis, specifically extracting key keywords, phrases, and sentiment (positive, negative, neutral) using TensorFlow and PyTorch.

[0940] Output: Extracted keywords, phrases, and sentiment data.

[0941] Specific operation: The server divides the text data into tokens, calculates and classifies the sentiment score of each token, and passes the result on to the next process.

[0942] Step 5: User emotion analysis method using emotion engine

[0943] Input: User text input, voice data, or facial expression data.

[0944] Processing: The server uses an emotion engine (e.g., Tone Analyzer) to analyze the user's emotion and determine whether it is classified as positive, negative, or neutral.

[0945] Output: User sentiment classification result.

[0946] Specific operation: Text and audio files entered by the user are passed to the emotion engine, emotion analysis is performed on each piece of data, and the analysis results are obtained.

[0947] Step 6: How to generate and present recommendations to the user

[0948] Input: Output data from the machine learning model and the sentiment engine.

[0949] Processing: The server integrates this data and generates posts that are optimally tailored to the user's target audience, including recommended keywords, hashtags, post tone, and specific post content templates.

[0950] Output: Generated post content, suggested keywords, suggested hashtags.

[0951] Specific behavior: Keywords to use: "New cafe," "Instagrammable," "Urban area," "Cafe hopping," recommended hashtags: "Cafe hopping," "New discovery," "Instagrammable," "Urban area," post template: "A new cafe has opened! It has a relaxed atmosphere and is sure to be Instagrammable! Cafe hopping, urban area, I want to connect with cafe lovers," and the following will be presented to the user.

[0952] As a result, this system can effectively generate and present buzz posts to the user's target demographic, and can provide optimal content that reflects the user's emotions.

[0953] (Application example 2)

[0954] 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."

[0955] Conventional social media buzz-generating posting systems perform analysis to maximize the impact on the target, but rarely take into account the user's own emotional state. As a result, the content of the post may not match the user's intended emotion or tone, resulting in an ineffective buzz. Another issue is that generating optimal post content to efficiently generate buzz among a specific target takes time and effort.

[0956] 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.

[0957] In this invention, the server includes means for receiving content to be made buzzworthy and target information from a user, means for retrieving related posts from a database storing past buzzworthy post data, means for preprocessing the retrieved post data and performing text analysis using natural language processing, means for inputting the analysis results into a machine learning model to generate optimal post content, means for presenting the generated post content and recommended keywords and hashtags to the user, and means for generating advertising copy based on the user's emotional state using an emotion engine that analyzes the user's emotions. This makes it possible to generate optimal post content according to the user's emotional state and effectively create buzz among the target demographic.

[0958] "Buzzworthy content" refers to content posted on social media that is intended to attract the attention of many users and be spread widely.

[0959] "Target information" refers to information about the target user demographic for buzz posts, such as a specific age group, gender, or interests.

[0960] The "database" is a data storage system that stores past buzz posting data and retrieves related information through queries.

[0961] "Natural language processing" is a technology for analyzing and processing language data, and is used for preprocessing and analyzing text.

[0962] A "machine learning model" is an algorithm that learns patterns from large amounts of data and makes predictions and classifications based on new data.

[0963] An "emotion engine" is a system that analyzes a user's facial expressions and voice to identify the user's current emotional state.

[0964] An "advertising copy" is a post that conveys information about a specific product or service on social media and aims to pique the interest of target users.

[0965] "Noise reduction" is the process of removing irrelevant or false information from the text data being analyzed.

[0966] "Stop word removal" is the process of removing words that are too general to contribute to the analysis (such as "no" and "wa") during text analysis.

[0967] "Normalization" is the process of correcting spelling variations and typos and standardizing data to maintain consistency in text data.

[0968] "Keywords" are important words that represent the main theme or topic of the post content, and increase search and spreading effectiveness on social media.

[0969] A "hashtag" is a linguistic element used on social media to tag posts related to a particular theme or topic, improving the discoverability of the posts.

[0970] The present invention relates to a system that enables efficient communication of content that you want to create buzz on social media to specific targets. By combining this system with an emotion engine that analyzes user emotions, more effective communication of information can be achieved. Specific means for realizing this system are described below.

[0971] This system is implemented primarily using the following hardware and software:

[0972] Hardware

[0973] Smartphone: Used by the user to input information and receive results from the system.

[0974] Server: The central computing resource that stores data, analyzes it, and processes the generated submissions.

[0975] software

[0976] Natural Language Processing (NLP) libraries: Libraries for preprocessing and parsing text (e.g., spaCy, NLTK).

[0977] Machine learning frameworks: Frameworks for learning from large amounts of data and making predictions or classifications (e.g., TensorFlow, PyTorch).

[0978] Emotion analysis engine: Software to analyze a user's facial expressions and voice to classify their emotional state (e.g., OpenCV, Google Cloud Speech-to-Text).

[0979] The server is implemented based on a system including the following means:

[0980] 1. Means of accepting input from users

[0981] Users access the system's interface using their smartphones, where they input the content they want to create buzz and target information (age group, gender, interests, etc.). This information is then sent from the smartphone to the server.

[0982] 2. How to obtain information from the database

[0983] The server receives the information sent by the user and then executes a query to retrieve posts related to the user's target information from a database of past buzz posts. The query includes filtering criteria based on the target information.

[0984] 3. Preprocessing and Natural Language Processing Methods

[0985] The server preprocesses the submitted data by removing noise, stop words, and normalizing the text. This step prepares the data for analysis.

[0986] 4. Analysis Methods Using Machine Learning Models

[0987] The server applies natural language processing (NLP) to the preprocessed post data, performing morphological analysis to extract important keywords, phrases, and sentiment (positive, negative, neutral).

[0988] 5. User sentiment analysis using a sentiment analysis engine

[0989] The server uses an emotion analysis engine to analyze the user's emotions. This analysis is performed using the smartphone's camera and microphone to capture the user's facial expressions and voice. Based on the analysis results, the user's emotions are classified as positive, negative, or neutral.

[0990] 6. How to generate and present recommendations to users

[0991] The server adjusts the predictions generated by the machine learning model based on the user's emotions recognized by the emotion analysis engine, and creates optimal post content. Specifically, this includes the keywords, hashtags, tone (emotion) of the post to be used, and specific post content templates. The generated post content is presented to the user on the smartphone screen.

[0992] Specific examples

[0993] For example, if a user wants to "introduce a new product" and specifies the target audience as "20-30 year old women interested in fashion," the system will operate as follows:

[0994] 1. Accepting input from the user

[0995] User entered information: "New product introduction", "Female, 20-30 years old, interested in fashion".

[0996] 2. Retrieving information from the database

[0997] A server receives the information submitted by the user and performs a database query to retrieve relevant past buzz posts.

[0998] 3. Preprocessing and Natural Language Processing Methods

[0999] The server removes noise from the acquired posting data, removes stop words, and normalizes the text.

[1000] 4. Analysis Methods Using Machine Learning Models

[1001] The server applies natural language processing to the preprocessed data to extract important keywords, phrases, and sentiment.

[1002] 5. User Emotion Analysis Method Using an Emotion Engine

[1003] The server analyzes facial expression and voice data sent from the user's device and recognizes that the user's emotions are positive.

[1004] 6. How to generate and present recommendations to users

[1005] The server generates posts that are appropriate for positive emotions based on the analysis results of the emotion engine:

[1006] Recommended keywords: "new products," "trends," "up your style"

[1007] Recommended hashtags: "New products," "Trends," "Fashion," "Style up"

[1008] Post template: "New fashion items have arrived! Don't miss out on these trendy items. Just a glimpse of them will definitely elevate your style! New Products Trends Fashion Style Up"

[1009] Prompt Sentence Examples

[1010] markdown

[1011] New Product Introduction

[1012] Generate positive posts showcasing fashion products targeted at women aged 20-30.

[1013] ---

[1014] Target Information:

[1015] Age range: 20-30 years old

[1016] Gender: Female

[1017] Interests: Fashion

[1018] User's emotional state:

[1019] positive

[1020] Generated content:

[1021] A new fashion item has arrived! This trendy item is a must-see. Just a glimpse of it will definitely enhance your style!

[1022] New products Trend fashion Style up

[1023] In this way, the present invention provides a system that generates optimal posting content based on the user's emotional state and target information, and can efficiently create buzz on SNS.

[1024] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1025] Step 1:

[1026] The content that the user wants to create buzz and target information are accepted.

[1027] Specific behavior:

[1028] Users launch the application on their smartphone and access the interface, where they input the content they want to create buzz about (e.g., "Introducing a new product") and target information (e.g., "Women aged 20-30 who are interested in fashion"). This information is then sent from the smartphone to the server.

[1029] Input: User-entered content and target information they want to create buzz about.

[1030] Output: Information data sent to the server.

[1031] Step 2:

[1032] Retrieve information from a database.

[1033] Specific behavior:

[1034] The server receives the information sent by the user and then executes a database query to retrieve posts related to the specified target information from a database of past buzz posts. The query, including the filtering criteria, is sent to the database.

[1035] Input: User-specified target information.

[1036] Output: Relevant buzz post data retrieved from the database.

[1037] Step 3:

[1038] Preprocess the acquired post data.

[1039] Specific behavior:

[1040] The server preprocesses the submitted data, removing noise, stop words, and normalizing the text to make it easier to analyze. This process uses natural language processing technology.

[1041] Input: Post data retrieved from the database.

[1042] Output: Preprocessed text data.

[1043] Step 4:

[1044] Perform text analysis on the preprocessed data.

[1045] Specific behavior:

[1046] The server applies natural language processing (NLP) to the preprocessed data, performing morphological analysis to extract key keywords, phrases, and sentiment (positive, negative, neutral). NLP libraries (e.g., spaCy, NLTK) are used.

[1047] Input: Preprocessed text data.

[1048] Output: Parsed keywords, phrases and sentiment information.

[1049] Step 5:

[1050] The analysis results are input into a machine learning model to generate optimal post content.

[1051] Specific behavior:

[1052] The server inputs the data analyzed by natural language processing into a machine learning model. The machine learning model predicts the optimal posting pattern for the target based on the results learned from past buzz post data. A machine learning framework (e.g., TensorFlow, PyTorch) is used here.

[1053] Input: Parsed data (keywords, phrases, sentiment information).

[1054] Output: Generated post content, suggested keywords and hashtags.

[1055] Step 6:

[1056] Analyze user sentiment.

[1057] Specific behavior:

[1058] The server analyzes the user's emotions using an emotion engine. It collects and analyzes the user's facial expression and voice data using the smartphone's camera and microphone. The emotion engine classifies the user's emotional state (positive, negative, neutral).

[1059] Input: User's facial expression data, voice data.

[1060] Output: The user's emotional state (positive, negative, neutral).

[1061] Step 7:

[1062] Recommendations are generated and presented to the user.

[1063] Specific behavior:

[1064] The server generates and adjusts optimal post content based on the user's emotional state. It combines the analysis results from the emotion engine with the output of the machine learning model to create templates for keywords, hashtags, post tone (emotion), and specific post content to use. The generated post content is presented to the user on their smartphone screen.

[1065] Input: Sentiment engine analysis results, machine learning model output.

[1066] Output: The post content, suggested keywords, and hashtags presented to the user.

[1067] This allows users to upload content to social media that best suits their emotional state, making it possible to create buzz efficiently and effectively.

[1068] 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.

[1069] 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.

[1070] 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.

[1071] [Third embodiment]

[1072] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1073] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[1074] 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).

[1075] 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.

[1076] 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.

[1077] 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).

[1078] 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.

[1079] 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.

[1080] 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.

[1081] 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.

[1082] 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.

[1083] 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."

[1084] This invention relates to a system for efficiently communicating content that you want to create buzz on social networking sites to specific targets. Below, we will explain the program processing of this system in natural language.

[1085] System configuration

[1086] The system mainly consists of the following means:

[1087] 1. Means of accepting input from users

[1088] 2. How to obtain information from the database

[1089] 3. Natural Language Processing and Text Analysis Methods

[1090] 4. How to apply machine learning models

[1091] 5. How to generate and present recommendations to users

[1092] Program processing

[1093] 1. Means of accepting input from users

[1094] Users access the system's interface using a device (such as a smartphone or PC). They input the content they want to create buzz about and target information (age group, gender, interests, etc.). This information is then sent from the device to the server.

[1095] 2. How to obtain information from the database

[1096] The server receives the information sent by the user and then executes a query to retrieve buzz posts related to the specified target information from a database that stores past buzz post data. Here, the query includes filtering conditions based on the target information.

[1097] 3. Natural Language Processing and Text Analysis Methods

[1098] The server preprocesses the acquired post data, including noise removal, stop word removal, and text normalization. It then uses natural language processing (NLP) techniques to perform morphological analysis of the text and extract important keywords, phrases, and sentiment (positive, negative, neutral).

[1099] 4. How to apply machine learning models

[1100] The server then inputs the data into a machine learning model based on the analysis results. This model learns from past success stories (buzzworthy posts) and predicts the best posting pattern for the target audience. The output generates templates for keywords, hashtags, post tone (emotion), and specific post content to use.

[1101] 5. How to generate and present recommendations to users

[1102] The server then compiles the final recommendations and creates a list of suggested posts for the user, which may include, for example:

[1103] Example keywords

[1104] Hashtag examples

[1105] Post Template

[1106] The device displays the received recommendations to the user, who can then review and edit them and post them directly to the social networking site.

[1107] Specific examples

[1108] For example, if a user wants to create buzz about the opening of a new cafe and specifies the target audience as "women in their 20s who live in urban areas and love cafes," the system will operate as follows:

[1109] 1. Accepting input from the user

[1110] User input: "New cafe opening", "Female in her 20s, urban resident, cafe lover"

[1111] 2. Retrieving information from the database

[1112] The server runs a database query to retrieve past trending posts, such as:

[1113] "Cafe hopping and new discoveries"

[1114] "I went to a new cafe! It's stylish and comfortable - perfect for Instagram."

[1115] "All the cafes in the city are great, but this one was especially great! I want to connect with other cafe lovers."

[1116] 3. Natural Language Processing and Text Analytics

[1117] The server extracts important keywords and phrases from the above submission data:

[1118] Keywords: "cafe," "new," "urban," "Instagrammable"

[1119] Emotion: Positive

[1120] 4. Applying machine learning models

[1121] The server feeds the extracted data into a machine learning model to generate recommendations for:

[1122] Recommended keywords: "new cafe," "Instagrammable," "urban area," "cafe hopping"

[1123] Recommended hashtags: "cafe hopping," "new discoveries," "Instagrammable," "urban areas"

[1124] Post template: "A new cafe has opened! It has a relaxed atmosphere and is sure to be Instagram-worthy! Cafe hopping, urban areas, and connecting with fellow cafe lovers."

[1125] 5. Generating and presenting recommendations to the user

[1126] The device displays the recommended content to the user, who then checks and edits it and posts it to the social networking site.

[1127] The above is a specific embodiment of the present invention, which makes it possible to efficiently and accurately generate and present buzzworthy posts aimed at a target audience.

[1128] The processing flow will be explained below.

[1129] Step 1:

[1130] Users access the system's interface using a device (smartphone or PC). They input the content they want to create buzz about and target information (age group, gender, interests, etc.). This information is then sent from the device to the server.

[1131] Step 2:

[1132] The server receives the information sent by the user and then executes a query from a database of past buzz posts to retrieve posts related to the target information specified by the user. The query includes filtering conditions based on the target information.

[1133] Step 3:

[1134] The server preprocesses the submitted data retrieved from the database by removing noise, removing stop words, and normalizing the text, making the data easier to analyze.

[1135] Step 4:

[1136] The server then applies natural language processing (NLP) to the preprocessed post data. This step involves morphological analysis to extract important keywords, phrases, and sentiment (positive, negative, neutral).

[1137] Step 5:

[1138] The server then inputs the extracted keywords, phrases, and sentiment data into a machine learning model, which learns from past success stories (buzzworthy posts) and predicts the optimal posting pattern for the target audience.

[1139] Step 6:

[1140] The server creates recommended posts for users based on the predictions generated by the machine learning model, including optimal keywords, hashtags, post tone (emotion), and specific post content templates.

[1141] Step 7:

[1142] The server sends the generated recommendations to the device, which then displays them to the user. The user can then review and edit the recommendations and post them directly to the social networking site.

[1143] Specific examples

[1144] Step 1:

[1145] A user enters the following information into the system: "A new cafe is opening" and "A woman in her 20s who loves cafes and lives in an urban area."

[1146] Step 2:

[1147] A server receives the information submitted by the user and performs a database query to retrieve relevant past buzz posts.

[1148] Step 3:

[1149] The server removes noise from the acquired posting data, removes stop words, and normalizes the text.

[1150] Step 4:

[1151] The server applies natural language processing to the preprocessed data to extract important keywords, phrases, and sentiment.

[1152] Step 5:

[1153] The server inputs the extracted data into a machine learning model to predict the optimal posting patterns for the target audience.

[1154] Step 6:

[1155] The server generates recommended posts based on predictions from the machine learning model.

[1156] Step 7:

[1157] The server sends the generated recommendations to the device, which displays them to the user, who then checks and edits them and posts them to the SNS.

[1158] Through these processing steps, the system can effectively generate and present buzz posts to the user's target audience.

[1159] Example 1

[1160] 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."

[1161] Currently, many users are using social networking services (SNS) to spread information, but it is not easy to efficiently create buzz for a specific target audience. Existing methods make it difficult to predict what content and format will create buzz, and require a great deal of time and effort. Furthermore, selecting appropriate keywords and hashtags is difficult, so generating optimal post content requires specialized knowledge. Therefore, there is a need for a system that can efficiently and accurately generate and present buzzworthy posts for specific targets.

[1162] 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.

[1163] In this invention, the server includes means for receiving content and target information from a user that the user wants to create a buzz, means for retrieving related posts from a database storing past buzz post data, means for preprocessing the retrieved post data and performing text analysis using natural language processing, means for inputting the analysis results into a machine learning model to generate optimal post content, means for presenting the generated post content and recommended keywords and hashtags to the user, and means for the user to edit and confirm the generated post content and then post it to an SNS. This allows users to effectively create buzz posts aimed at a specific target demographic.

[1164] "Means for accepting content to be buzzed and target information from users" refers to an interface or function that allows users to input the content they want to buzz and information about the target person, and send it to the server.

[1165] "Means for retrieving related posts from a database storing data on past buzz posts" refers to a function in which the server queries the database, retrieves data on past buzz posts, and retrieves posts related to the target information specified by the user.

[1166] "Means for preprocessing acquired posted data and performing text analysis using natural language processing" refers to a function that performs preprocessing on acquired posted data, such as noise removal, stop word removal, and text normalization, and then analyzes the text using natural language processing technology.

[1167] "A means of inputting the analysis results into a machine learning model to generate optimal post content" refers to a function that inputs data obtained from text analysis into a machine learning model and uses the model, which has learned from past success stories, to generate post content that is most suitable for the target demographic.

[1168] "Means of presenting generated post content and recommended keywords and hashtags to users" refers to the function of organizing post content, keywords, hashtags, etc. generated by machine learning models and displaying and suggesting them to users.

[1169] "Means for users to post generated posts to SNS after editing and reviewing them" refers to a function that allows users to review and edit the suggested content and post it directly to the SNS platform.

[1170] "Preprocessing of posted data retrieved from the database" refers to the process of removing unnecessary information (noise) from the posted data, deleting commonly used words (stop words) that are irrelevant to a specific analysis, and normalizing the strings to maintain consistency.

[1171] "Natural language processing" is a technology that enables computers to understand, interpret, and generate human language, and in the present invention refers to the technology used for morphological analysis of text data and extraction of important keywords and emotions.

[1172] A "machine learning model" refers to an algorithm that learns patterns from given data and makes predictions and classifications for new data. In the context of this invention, it is used to learn from past buzz posts and generate post content that is optimal for the target demographic.

[1173] "Target information" refers to the attribute information (e.g., age group, gender, interests, etc.) of the intended recipients of the post that the user wants to make go viral, and the content of the post is optimized based on that information.

[1174] This invention relates to a system for efficiently communicating content that users want to create buzz on social media to specific targets. This system accepts input from users, acquires and analyzes relevant data from past buzz posts, and generates optimal post content using a machine learning model, which is then presented to the user. The configuration and operation of this system are described in detail below.

[1175] System configuration

[1176] The system mainly consists of the following means:

[1177] 1. Means of accepting input from users

[1178] 2. How to obtain information from the database

[1179] 3. Natural Language Processing and Text Analysis Methods

[1180] 4. How to apply machine learning models

[1181] 5. How to generate and present recommendations to users

[1182] 6. Means of posting to social media

[1183] Processing Details

[1184] 1. Accepting input from the user

[1185] Users access the system's interface using a smartphone or PC. They input the content they want to create buzz and target information (age group, gender, interests, etc.), and this information is sent from the device to the server. The input is in a specific data format, and the server receives it and begins processing.

[1186] 2. Retrieving information from the database

[1187] After receiving the information sent by the user, the server executes a database query based on that information. The database stores past buzz post data and retrieves relevant post data based on specific filtering criteria. For example, a query is executed to retrieve information about "women in their 20s, urban areas, cafe lovers."

[1188] 3. Natural Language Processing and Text Analytics

[1189] The server preprocesses the acquired post data. This preprocessing includes noise removal (removal of special symbols and unnecessary spaces), stopword removal (frequent phrases that are generally ignored), and text normalization (standardization of characters). Then, using natural language processing technology (e.g., the morphological analysis tool MeCab), it performs morphological analysis of the text data and extracts important keywords, phrases, and sentiment (positive, negative, neutral).

[1190] 4. Applying machine learning models

[1191] Based on the analysis results, the server inputs the data into a machine learning model. The machine learning model learns from past successful posting data and predicts the optimal posting pattern for the target demographic. For example, the model predicts the keywords, hashtags, and tone (emotion) to be used in the post, and generates a specific post content template.

[1192] 5. Generating and presenting recommendations to the user

[1193] Finally, the server presents recommendations to the user, including example keywords, example hashtags, and post templates, which the user can review and edit to select the post they deem most appropriate.

[1194] 6. How to post to social media

[1195] The final post content that users have considered can be posted directly to the SNS platform, and users can easily publish the generated post content to the SNS through their device.

[1196] Specific examples

[1197] For example, if a user wants to create buzz about the opening of a new cafe and specifies the target audience as "women in their 20s who live in urban areas and love cafes," the system will operate as follows:

[1198] 1. Accepting input from the user

[1199] User input: "New cafe opening", "Female in her 20s, urban resident, cafe lover"

[1200] 2. Retrieving information from the database

[1201] The server runs a database query to retrieve past viral posts, such as:

[1202] "Cafe hopping and new discoveries"

[1203] "I went to a new cafe! It's stylish and comfortable - perfect for Instagram."

[1204] "All the cafes in the city are great, but this one was especially great! I want to connect with other cafe lovers."

[1205] 3. Natural Language Processing and Text Analytics

[1206] The server extracts important keywords and phrases from the above submission data:

[1207] Keywords: "cafe," "new," "urban," "Instagrammable"

[1208] Emotion: Positive

[1209] 4. Applying machine learning models

[1210] The server feeds the extracted data into a machine learning model to generate the following recommendations:

[1211] Recommended keywords: "new cafe," "Instagrammable," "urban area," "cafe hopping"

[1212] Recommended hashtags: "cafe hopping," "new discoveries," "Instagrammable," "urban areas"

[1213] Post template: "A new cafe has opened! It has a relaxed atmosphere and is sure to be Instagram-worthy! Cafe hopping, urban areas, and connecting with fellow cafe lovers."

[1214] 5. Generating and presenting recommendations to the user

[1215] The device displays the recommended content to the user, who can then review it, edit it if necessary, and post it to social media.

[1216] Examples of prompt statements

[1217] You might enter the following prompts for the system's generated AI model:

[1218] "I want to create a buzz about the opening of a new cafe among women in their 20s who love cafes and live in urban areas. Based on past data on buzzworthy posts, could you recommend some keywords, hashtags, and a post template?"

[1219] The above is a specific embodiment of the present invention, which makes it possible to efficiently and accurately generate and present buzzworthy posts aimed at a target audience.

[1220] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1221] Step 1: Accepting input from the user

[1222] Users access the system's interface using a smartphone or PC. They input the content they want to create buzz about (e.g., "A new cafe has opened") and target information (e.g., "Women in their 20s who live in urban areas and love cafes"). The device receives this information and sends it to the server. The server temporarily stores the input information before proceeding to the next processing step.

[1223] Step 2: Retrieving information from the database

[1224] The server receives target information sent by the user. Based on this information, the server sends a query to the database to retrieve related posts from past buzz post data. For example, it searches the database for posts related to "women in their 20s" and "cafes" and extracts related post data. Input: Target information. Output: Related past buzz post data.

[1225] Step 3: Natural Language Processing and Text Analytics

[1226] The server preprocesses the acquired post data. Preprocessing includes removing noise (removing special symbols and unnecessary spaces), deleting stop words (frequent words that are generally ignored), and normalizing the text (standardizing characters). Natural language processing techniques (such as morphological analysis tools) are then used to perform morphological analysis of the text data, extract important keywords and phrases, and analyze sentiment (positive, negative, neutral). Input: Past buzz post data. Output: Preprocessed text data and analysis results.

[1227] Step 4: Applying the machine learning model

[1228] The server inputs the preprocessed text data and analysis results into a machine learning model. The machine learning model learns from past posting data of successful cases and predicts the optimal posting pattern for the target demographic. This prediction generates templates for the keywords, hashtags, tone (emotion) of the post, and specific post content to be used. Input: Analysis results. Output: Optimal post content, recommended keywords, recommended hashtags.

[1229] Step 5: Generate recommendations and present them to the user

[1230] The server organizes the final recommendations and creates a list of post content to suggest to the user. The list includes example keywords, example hashtags, and post templates. The device displays the received recommendations to the user. The user can review and edit these recommendations and select the most suitable post content. Input: Best post content, recommended keywords, recommended hashtags. Output: List of posts presented to the user.

[1231] Step 6: How to post to social media

[1232] The user reviews the suggested content and edits it as necessary. Through the device interface, the user can post the final post directly to social media. This allows for efficient spreading of content that is targeted to a specific demographic. Input: Final post content after user editing. Output: Post to social media.

[1233] (Application example 1)

[1234] 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."

[1235] In order to efficiently create buzz on social media for specific content, it is necessary to generate posts that are optimal for the target audience. However, conventional systems that utilize artificial intelligence and machine learning require the manual collection and analysis of data and the generation of post content, which is time-consuming and labor-intensive. In addition, there is uncertainty as to whether the generated post content will actually appeal to the target, so a reliable method is needed to increase the probability of buzz.

[1236] 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.

[1237] In this invention, the server includes means for receiving content and target information from a user that the user wants to create a buzz, means for retrieving related posts from a database storing past buzz post data, means for preprocessing the retrieved post data and performing text analysis using natural language processing, means for inputting the analysis results into a machine learning model to generate optimal post content, means for presenting the generated post content and recommended keywords and hashtags to the user, and means for posting the generated optimal post content to an SNS via an application on a smart device. This allows users to avoid complex manual data collection and analysis work, quickly and easily generate post content that is most effective for their target, and significantly increases the chances of creating a buzz on an SNS.

[1238] "User" refers to any individual or entity that uses the System.

[1239] "Buzzworthy content" refers to information or messages posted on social media with the aim of attracting a lot of attention.

[1240] "Target information" is data that indicates the attributes and interests of the target audience for a particular post, including, for example, age group, gender, place of residence, and interests.

[1241] "Database" refers to a collection of information that stores collected data on past buzz posts and makes it searchable and retrievalable.

[1242] "Preprocessing" refers to the process of processing the data by removing noise, deleting stop words, normalizing text, etc. from the acquired submission data.

[1243] "Natural language processing" refers to the technology that enables computers to understand and analyze human language.

[1244] "Text analysis" refers to the process of using natural language processing techniques to interpret the meaning and structure of a document and extract important information.

[1245] A "machine learning model" refers to a statistical method for learning patterns and rules from past data and making predictions and classifications for future data.

[1246] "Keywords" are particularly important words or phrases in your post that will grab your target's attention.

[1247] A "hashtag" is a type of tagging notation used in social media posts, and refers to a keyword that makes it easier to group posts related to a particular topic or theme.

[1248] "Smart device" refers to a portable electronic device that can connect to a network, such as a smartphone or tablet.

[1249] "Application" refers to a software program that runs on a smart device and provides specific functionality.

[1250] "SNS" is an abbreviation for social networking service, which refers to a platform where users can share information and interact with each other online.

[1251] This invention relates to a system for efficiently communicating content that you want to create buzz on SNS to designated targets, and includes the following means and processes.

[1252] Program processing explanation

[1253] The server receives the content and target information from the user and performs the following processing based on this: First, it retrieves data related to the specified target information from a database of past buzz posts, and then performs preprocessing to remove noise, stop words, and normalize the data.

[1254] The server then performs text analysis using natural language processing (NLP) techniques, including morphological analysis, keyword extraction, and sentiment analysis of the acquired data, using software such as NLTK (Natural Language Toolkit) and Scikit-learn.

[1255] The server then inputs the analyzed data into a machine learning model, which uses classification techniques such as logistic regression to predict the best post content, keywords, and hashtags for the target audience based on past success stories.

[1256] Finally, the generated optimal post content, recommended keywords, and hashtags are presented to the user, who posts it to a social networking site via an application on their smart device.

[1257] Specific examples

[1258] For example, if a user wants to create buzz about the opening of a new cafe and specifies "women in their 20s who love cafes and live in urban areas" as their target, the system will operate as follows.

[1259] User Input

[1260] The user wants to create buzz: "New cafe opening"

[1261] Target information: "Women in their 20s who love cafes and live in urban areas"

[1262] Server Processing

[1263] Get past trending posts related to your target information from the database. Example:

[1264] "Cafe hopping and new discoveries"

[1265] "I went to a new cafe! It's stylish and comfortable - perfect for Instagram."

[1266] "All the cafes in the city are great, but this one was especially great! I want to connect with other cafe lovers."

[1267] Keyword extraction from analysis results:

[1268] Keywords: "cafe," "new," "urban," "Instagrammable"

[1269] Emotion: Positive

[1270] Generate suggested posts

[1271] Recommended keywords: "new cafe," "Instagrammable," "urban area," "cafe hopping"

[1272] Recommended hashtags: "cafe hopping," "new discoveries," "Instagrammable," "urban areas"

[1273] Post template: "A new cafe has opened! It has a relaxed atmosphere and is sure to be Instagram-worthy! Cafe hopping, urban areas, and connecting with fellow cafe lovers."

[1274] Prompt Sentence Examples

[1275] The following sentences can be used as prompts for users to input to a generative AI model:

[1276] "Generate the best social media post content to promote the opening of a new cafe. The target audience is women in their 20s who love cafes and live in urban areas. Past related post data is shown below. Please include keywords and hashtags that should be used in the post."

[1277] This allows users to quickly generate posts that are suitable for their target audience and are likely to create buzz, and then easily post them on social media.

[1278] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1279] Step 1:

[1280] Users enter the content they want to create buzz on social media and target information.

[1281] Input: What you want to create a buzz about (e.g., the opening of a new cafe), target information (e.g., women in their 20s who love cafes and live in urban areas)

[1282] What happens: A user provides input via an application on their smart device, which is sent to the server.

[1283] Step 2:

[1284] The server retrieves post data related to the target information from a database of past buzz posts.

[1285] Input: Target information sent by the user

[1286] Data processing: Perform database queries based on target information and filter relevant post data

[1287] Output: Data of past buzz posts (e.g., "Cafe hopping, new discoveries")

[1288] What happens: The server uses SQL to run a search query against the database to retrieve the relevant data.

[1289] Step 3:

[1290] Preprocess the submitted data received by the server.

[1291] Input: Previous buzz post data obtained

[1292] Data processing: noise removal, stopword removal, text normalization

[1293] Output: Clean post data after preprocessing

[1294] Specific operation: The server uses a natural language processing library such as NLTK to preprocess the text data.

[1295] Step 4:

[1296] The server performs text analysis using natural language processing technology.

[1297] Input: Clean post data after preprocessing

[1298] Data processing: morphological analysis of text, keyword extraction, sentiment analysis

[1299] Output: Key keywords, phrases, and emotional state (positive, negative, neutral)

[1300] Specific operation: The server uses Scikit-learn or similar tools to apply models for morphological analysis and keyword extraction.

[1301] Step 5:

[1302] The server inputs the analysis results into a machine learning model to generate optimal post content.

[1303] Input: Keywords, phrases, and emotional states obtained through natural language processing

[1304] Data Computing: Prediction and Content Generation with Machine Learning Models

[1305] Output: Best post content, recommended keywords, recommended hashtags, post templates

[1306] What it does: The server uses machine learning models such as logistic regression to generate targeted posts.

[1307] Step 6:

[1308] The server presents the generated post content and recommended keywords and hashtags to the user.

[1309] Input: Best post content, suggested keywords, suggested hashtags, post template

[1310] Output: What the user sees on their smart device

[1311] Specific behavior: The server sends the generated content to the user interface so that the user can view it on their smart device.

[1312] Step 7:

[1313] The optimal post content that the user has checked and edited is posted to SNS.

[1314] Input: Post content provided by the server, edits made by the user

[1315] Output: Posts published on social media

[1316] Specific behavior: The user publishes a post using the interface that allows them to post directly to the social networking site from the application.

[1317] 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.

[1318] This invention relates to a system that enables efficient communication of content that you want to create buzz on social media to specific targets. This system achieves even more effective communication by combining it with an emotion engine that recognizes and analyzes user emotions. Below, we will explain the program processing of this system in natural language.

[1319] System configuration

[1320] The system mainly consists of the following means:

[1321] 1. Means of accepting input from users

[1322] 2. How to obtain information from the database

[1323] 3. Natural Language Processing and Text Analysis Methods

[1324] 4. How to apply machine learning models

[1325] 5. User Emotion Analysis Method Using Emotion Engine

[1326] 6. How to generate and present recommendations to users

[1327] Program processing

[1328] 1. Means of accepting input from users

[1329] Users access the system's interface using a device (such as a smartphone or PC). They input the content they want to create buzz about and target information (age group, gender, interests, etc.). This information is then sent from the device to the server.

[1330] 2. How to obtain information from the database

[1331] The server receives the information sent by the user and then executes a query from a database of past buzz posts to retrieve posts related to the target information specified by the user. The query includes filtering conditions based on the target information.

[1332] 3. Natural Language Processing and Text Analysis Methods

[1333] The server preprocesses the submitted data by removing noise, removing stop words, and normalizing the text, making the data easier to analyze.

[1334] 4. How to apply machine learning models

[1335] The server then applies natural language processing (NLP) to the preprocessed post data. This step involves morphological analysis to extract important keywords, phrases, and sentiment (positive, negative, neutral).

[1336] 5. User Emotion Analysis Method Using Emotion Engine

[1337] The server uses an emotion engine to analyze the user's emotions based on the user's facial expressions, tone of voice, or text input. Based on the analysis results, the user's emotions are classified as positive, negative, or neutral.

[1338] 6. How to generate and present recommendations to users

[1339] The server adjusts the predictions generated by the machine learning model based on the user's emotions recognized by the emotion engine to create optimal post content, including keywords, hashtags, post tone (emotion), and specific post content templates to use.

[1340] For example, if the user's emotions are positive, the system generates posts with a bright and cheerful tone. If the emotions are negative, the system emphasizes an encouraging and comforting tone.

[1341] Specific examples

[1342] For example, if a user wants to create buzz about the opening of a new cafe and specifies the target audience as "women in their 20s who live in urban areas and love cafes," the system will operate as follows:

[1343] 1. Accepting input from the user

[1344] User input: "New cafe opening", "Female in her 20s, urban resident, cafe lover"

[1345] 2. Retrieving information from the database

[1346] A server receives the information submitted by the user and performs a database query to retrieve relevant past buzz posts.

[1347] 3. Natural Language Processing and Text Analytics

[1348] The server removes noise from the acquired posting data, removes stop words, and normalizes the text.

[1349] 4. Applying machine learning models

[1350] The server applies natural language processing to the preprocessed data to extract important keywords, phrases, and sentiment.

[1351] 5. User Emotion Analysis Using an Emotion Engine

[1352] The server analyzes facial expression data, voice data, or input text sent from the user's device and recognizes that the user's emotions are positive.

[1353] 6. Generating and Presenting Recommendations to the User

[1354] The server generates posts that are appropriate for positive emotions based on the analysis results of the emotion engine:

[1355] Recommended keywords: "new cafe," "Instagrammable," "urban area," "cafe hopping"

[1356] Recommended hashtags: "cafe hopping," "new discoveries," "Instagrammable," "urban areas"

[1357] Post template: "A new cafe has opened! It has a relaxed atmosphere and is sure to be Instagram-worthy! Cafe hopping, urban areas, and connecting with fellow cafe lovers."

[1358] Through the above processing steps, the system can effectively generate and present buzz posts to the user's target demographic, and provide optimal content based on the user's emotions.

[1359] The processing flow will be explained below.

[1360] Step 1:

[1361] Users access the system's interface using a device (smartphone or PC). They input the content they want to create buzz about (e.g., "A new cafe has opened") and target information (e.g., "Women in their 20s who live in urban areas and love cafes"). This information is then sent from the device to the server.

[1362] Step 2:

[1363] The server receives the input information sent by the user. It then executes a query to retrieve posts related to the specified target information from a database of past buzz posts. For example, it filters past posts related to "women in their 20s" and "cafes" based on the user's target information.

[1364] Step 3:

[1365] The server preprocesses the submitted data retrieved from the database by removing noise, removing stop words, and normalizing the text, making the data easier to analyze.

[1366] Step 4:

[1367] The server then applies natural language processing (NLP) to the preprocessed post data. This step involves morphological analysis to extract important keywords (e.g., "new cafe"), phrases, and sentiment (positive, negative, neutral).

[1368] Step 5:

[1369] The server inputs data into a machine learning model based on keywords and sentiment extracted through natural language processing. This model learns from past success stories (buzzworthy posts) and predicts the best posting pattern for the target audience. For example, it generates recommended hashtags such as "Instagrammable" and "cafe hopping."

[1370] Step 6:

[1371] The server analyzes the user's emotions using an emotion engine. This analysis is based on the user's facial expression recognition, voice analysis, or text input. For example, if the user has a positive expression while typing "new cafe," this emotion is recognized as positive.

[1372] Step 7:

[1373] The server adjusts the predictions generated by the machine learning model based on the user's emotions recognized by the emotion engine. For example, if the user's emotions are positive, it generates posts with a bright tone. On the other hand, if the emotions are negative, it generates posts that contain encouragement and comfort.

[1374] Step 8:

[1375] The server then compiles the final recommendations and creates a list of suggested posts for the user, including the best keywords, hashtags, and post templates, for example:

[1376] Recommended keywords: "new cafe," "Instagrammable," "urban area," "cafe hopping"

[1377] Recommended hashtags: "cafe hopping," "new discoveries," "Instagrammable," "urban areas"

[1378] Post template: "A new cafe has opened! It has a relaxed atmosphere and is sure to be Instagram-worthy! Cafe hopping, urban areas, and connecting with fellow cafe lovers."

[1379] Step 9:

[1380] The server sends the generated recommendations to the device. The device displays the received recommendations to the user. The user can then review and edit the recommendations and post them directly to the SNS.

[1381] Through the above processing steps, the system can effectively generate buzz posts for the user's target demographic and provide optimal content according to the user's emotions.

[1382] Example 2

[1383] 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."

[1384] With conventional SNS posting systems, it was difficult for users to effectively communicate information to their target audience. Also, because only uniform post content could be generated without considering the user's emotions, the effectiveness of the posts was limited. As a result, it was difficult for the posts to go viral, and it was difficult for the posts to reach the intended audience.

[1385] 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.

[1386] In this invention, the server includes means for receiving content to be made buzzworthy and target information from a user, means for retrieving related posts from a database storing past buzz posting data, means for preprocessing the retrieved posting data and performing text analysis using natural language processing, means for recognizing and analyzing user emotions using a sentiment analysis engine, means for inputting the analysis results into a machine learning algorithm to generate optimal posting content, and means for presenting the generated posting content and recommended keywords and hashtags to the user. This makes it possible to generate posting content that is effective for the target demographic and in line with their emotions.

[1387] "Means for receiving content to be buzzed and target information from a user" refers to an interface or process for receiving content to be buzzed and target information input by a user.

[1388] "Means for retrieving related posts from a database storing past buzz posting data" refers to a function for searching and retrieving information related to the user's requirements from stored past buzz posting data.

[1389] "Means for preprocessing acquired posted data and performing text analysis using natural language processing" refers to the process of removing noise, deleting stop words, and normalizing acquired posted data, and then analyzing the content using natural language processing.

[1390] "Means for recognizing and analyzing user emotions using an emotion analysis engine" refers to the function of analyzing emotions from data such as text, voice, and facial expressions entered by the user and classifying them as positive, negative, neutral, etc.

[1391] "Method of inputting analysis results into a machine learning algorithm to generate optimal post content" refers to the process of generating appropriate post content through a machine learning model based on the results of natural language processing and sentiment analysis.

[1392] "Means for presenting generated post content and recommended keywords and hashtags to users" refers to functions and interfaces for providing users with post content generated by the system and effective keyword and hashtag suggestions.

[1393] A specific embodiment for carrying out the present invention will be described below. The system of the present invention realizes effective information transmission by combining users, terminals, and servers.

[1394] System configuration and hardware / software used

[1395] 1. Means of accepting input from users

[1396] Users access the system interface via a web browser on their device (e.g., smartphone, PC). They then access the system and enter the content they want to create buzz about and target information (age group, gender, interests, etc.). This information is sent to the server as an HTTP request.

[1397] 2. How to obtain information from the database

[1398] The server receives the information sent by the user, then executes an SQL query against a database of past buzz posts (e.g., MySQL or PostgreSQL) to search for and retrieve posts related to the target information specified by the user.

[1399] 3. Natural Language Processing and Text Analysis Methods

[1400] The server uses a Python NLP library (e.g., NLTK, spaCy) to preprocess the submitted data. Specifically, it performs the following steps:

[1401] Noise removal: Remove unnecessary special characters and HTML tags.

[1402] Stopword removal: Removal of common words that have no meaning (e.g. "wa", "o", "ni").

[1403] Text normalization: Convert all text to lower case for consistency.

[1404] 4. How to apply machine learning models

[1405] The server then performs natural language processing (NLP) on the preprocessed text data, including morphological analysis and sentiment analysis, using TensorFlow and PyTorch to extract key keywords, phrases, and sentiment (positive, negative, neutral).

[1406] 5. User Emotion Analysis Method Using Emotion Engine

[1407] The server uses an emotion engine (e.g., IBM Watson's Tone Analyzer) to analyze emotions from the user's input data (e.g., text, voice, facial expressions, etc.). Based on the analysis results, the user's emotions are classified as positive, negative, or neutral.

[1408] 6. How to generate and present recommendations to users

[1409] The server combines the results of the machine learning model and the emotion engine to generate optimal posts for the target users. Specifically, it generates the following elements:

[1410] Recommended Keywords

[1411] Recommended hashtags

[1412] The tone (emotion) of the post

[1413] Specific post content template

[1414] The user can review the generated post and make any necessary adjustments.

[1415] Specific examples

[1416] For example, if a user wants to create buzz about the opening of a new cafe and specifies the target audience as "women in their 20s who live in urban areas and love cafes," the system will operate as follows:

[1417] 1. Accepting input from the user

[1418] The user fills in the form with "a new cafe opening" and "a woman in her 20s who loves cafes and lives in an urban area" and submits it.

[1419] 2. Retrieving information from the database

[1420] Based on the input information received by the server, a database query is executed to retrieve related past buzz posts.

[1421] 3. Natural Language Processing and Text Analytics

[1422] The server removes noise from the acquired submission data, removes stop words, and normalizes the text.

[1423] 4. Applying machine learning models

[1424] The server applies NLP to the pre-processed data to extract important keywords, phrases, and sentiment.

[1425] 5. User Emotion Analysis Using an Emotion Engine

[1426] The server analyzes the user's text input and recognizes it as a positive emotion.

[1427] 6. Generating and Presenting Recommendations to the User

[1428] Based on the sentiment and NLP results, the server generates the following post with a positive tone:

[1429] Recommended keywords: "new cafe," "Instagrammable," "urban area," "cafe hopping"

[1430] Recommended hashtags: "cafe hopping," "new discoveries," "Instagrammable," "urban areas"

[1431] Post template: "A new cafe has opened! It has a relaxed atmosphere and is sure to be Instagram-worthy! Cafe hopping, urban areas, and I want to connect with fellow cafe lovers."

[1432] Prompt Sentence Examples

[1433] Example prompts to input to a generative AI model:

[1434] We want to create buzz on social media about the opening of a new cafe. Our target audience is women in their 20s who love cafes and live in urban areas. We will generate effective posts based on data on past buzz posts. Since many users have positive feelings, we recommend creating posts with a bright and fun tone.

[1435] As a result, this system can not only effectively generate and present buzz posts to the user's target demographic, but also provide optimal content that reflects the user's emotions.

[1436] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1437] Divide the processing flow of the system program into processing steps

[1438] Step 1:

[1439] Step 2:

[1440] Step 3:

[1441] Step 4:

[1442] Step 5:

[1443] Step 6:

[1444] Specific explanation of each processing step

[1445] Step 1: Accepting input from the user

[1446] Input: A user accesses the system interface using a terminal and enters the content they want to buzz and target information into a form.

[1447] Processing: The terminal sends the input information to the server in the form of an HTTP request.

[1448] Output: The server temporarily stores the received content and target information in memory.

[1449] What happens: A user opens a browser, fills in a web form with the words "New cafe opening" and "Female, 20s, urban resident, cafe lover," and clicks the submit button.

[1450] Step 2: How to retrieve information from the database

[1451] Input: The server generates a database query based on the buzz content and target information received from the user.

[1452] Processing: The server executes an SQL query against a database of past buzz posts (e.g., MySQL, PostgreSQL) to search and retrieve relevant past posts.

[1453] Output: A dataset containing the retrieved submission data.

[1454] Specific operation: The server queries the database using the target information of "women in their 20s" and "urban areas," and extracts related past buzz posts using an SQL query.

[1455] Step 3: Natural language processing and text analysis tools

[1456] Input: The raw submission data obtained.

[1457] Processing: The server uses Python NLP libraries (e.g., NLTK, spaCy) to preprocess the data, specifically denoising, removing stop words, and normalizing the text.

[1458] Output: Preprocessed text data.

[1459] Specific operations: Noise removal removes HTML tags and special characters, stop word removal removes common words such as "は", "を", and "に", and text normalization standardizes case and removes unnecessary whitespace.

[1460] Step 4: How to apply the machine learning model

[1461] Input: Preprocessed text data.

[1462] Processing: The server performs natural language processing on the preprocessed data, performing morphological and sentiment analysis, specifically extracting key keywords, phrases, and sentiment (positive, negative, neutral) using TensorFlow and PyTorch.

[1463] Output: Extracted keywords, phrases, and sentiment data.

[1464] Specific operation: The server divides the text data into tokens, calculates and classifies the sentiment score of each token, and passes the result on to the next process.

[1465] Step 5: User emotion analysis method using emotion engine

[1466] Input: User text input, voice data, or facial expression data.

[1467] Processing: The server uses an emotion engine (e.g., Tone Analyzer) to analyze the user's emotion and determine whether it is classified as positive, negative, or neutral.

[1468] Output: User sentiment classification result.

[1469] Specific operation: Text and audio files entered by the user are passed to the emotion engine, emotion analysis is performed on each piece of data, and the analysis results are obtained.

[1470] Step 6: How to generate and present recommendations to the user

[1471] Input: Output data from the machine learning model and the sentiment engine.

[1472] Processing: The server integrates this data and generates posts that are optimally tailored to the user's target audience, including recommended keywords, hashtags, post tone, and specific post content templates.

[1473] Output: Generated post content, suggested keywords, suggested hashtags.

[1474] Specific behavior: Keywords to use: "New cafe," "Instagrammable," "Urban area," "Cafe hopping," recommended hashtags: "Cafe hopping," "New discovery," "Instagrammable," "Urban area," post template: "A new cafe has opened! It has a relaxed atmosphere and is sure to be Instagrammable! Cafe hopping, urban area, I want to connect with cafe lovers," and the following will be presented to the user.

[1475] As a result, this system can effectively generate and present buzz posts to the user's target demographic, and can provide optimal content that reflects the user's emotions.

[1476] (Application example 2)

[1477] 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."

[1478] Conventional social media buzz-generating posting systems perform analysis to maximize the impact on the target, but rarely take into account the user's own emotional state. As a result, the content of the post may not match the user's intended emotion or tone, resulting in an ineffective buzz. Another issue is that generating optimal post content to efficiently generate buzz among a specific target takes time and effort.

[1479] 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.

[1480] In this invention, the server includes means for receiving content to be made buzzworthy and target information from a user, means for retrieving related posts from a database storing past buzzworthy post data, means for preprocessing the retrieved post data and performing text analysis using natural language processing, means for inputting the analysis results into a machine learning model to generate optimal post content, means for presenting the generated post content and recommended keywords and hashtags to the user, and means for generating advertising copy based on the user's emotional state using an emotion engine that analyzes the user's emotions. This makes it possible to generate optimal post content according to the user's emotional state and effectively create buzz among the target demographic.

[1481] "Buzzworthy content" refers to content posted on social media that is intended to attract the attention of many users and be spread widely.

[1482] "Target information" refers to information about the target user demographic for buzz posts, such as a specific age group, gender, or interests.

[1483] The "database" is a data storage system that stores past buzz posting data and retrieves related information through queries.

[1484] "Natural language processing" is a technology for analyzing and processing language data, and is used for preprocessing and analyzing text.

[1485] A "machine learning model" is an algorithm that learns patterns from large amounts of data and makes predictions and classifications based on new data.

[1486] An "emotion engine" is a system that analyzes a user's facial expressions and voice to identify the user's current emotional state.

[1487] An "advertising copy" is a post that conveys information about a specific product or service on social media and aims to pique the interest of target users.

[1488] "Noise reduction" is the process of removing irrelevant or false information from the text data being analyzed.

[1489] "Stop word removal" is the process of removing words that are too general to contribute to the analysis (such as "no" and "wa") during text analysis.

[1490] "Normalization" is the process of correcting spelling variations and typos and standardizing data to maintain consistency in text data.

[1491] "Keywords" are important words that represent the main theme or topic of the post content, and increase search and spreading effectiveness on social media.

[1492] A "hashtag" is a linguistic element used on social media to tag posts related to a particular theme or topic, improving the discoverability of the posts.

[1493] The present invention relates to a system that enables efficient communication of content that you want to create buzz on social media to specific targets. By combining this system with an emotion engine that analyzes user emotions, more effective communication of information can be achieved. Specific means for realizing this system are described below.

[1494] This system is implemented primarily using the following hardware and software:

[1495] Hardware

[1496] Smartphone: Used by the user to input information and receive results from the system.

[1497] Server: The central computing resource that stores data, analyzes it, and processes the generated submissions.

[1498] software

[1499] Natural Language Processing (NLP) libraries: Libraries for preprocessing and parsing text (e.g., spaCy, NLTK).

[1500] Machine learning frameworks: Frameworks for learning from large amounts of data and making predictions or classifications (e.g., TensorFlow, PyTorch).

[1501] Emotion analysis engine: Software to analyze a user's facial expressions and voice to classify their emotional state (e.g., OpenCV, Google Cloud Speech-to-Text).

[1502] The server is implemented based on a system including the following means:

[1503] 1. Means of accepting input from users

[1504] Users access the system's interface using their smartphones, where they input the content they want to create buzz and target information (age group, gender, interests, etc.). This information is then sent from the smartphone to the server.

[1505] 2. How to obtain information from the database

[1506] The server receives the information sent by the user and then executes a query to retrieve posts related to the user's target information from a database of past buzz posts. The query includes filtering criteria based on the target information.

[1507] 3. Preprocessing and Natural Language Processing Methods

[1508] The server preprocesses the submitted data by removing noise, stop words, and normalizing the text. This step prepares the data for analysis.

[1509] 4. Analysis Methods Using Machine Learning Models

[1510] The server applies natural language processing (NLP) to the preprocessed post data, performing morphological analysis to extract important keywords, phrases, and sentiment (positive, negative, neutral).

[1511] 5. User sentiment analysis using a sentiment analysis engine

[1512] The server uses an emotion analysis engine to analyze the user's emotions. This analysis is performed using the smartphone's camera and microphone to capture the user's facial expressions and voice. Based on the analysis results, the user's emotions are classified as positive, negative, or neutral.

[1513] 6. How to generate and present recommendations to users

[1514] The server adjusts the predictions generated by the machine learning model based on the user's emotions recognized by the emotion analysis engine, and creates optimal post content. Specifically, this includes the keywords, hashtags, tone (emotion) of the post to be used, and specific post content templates. The generated post content is presented to the user on the smartphone screen.

[1515] Specific examples

[1516] For example, if a user wants to "introduce a new product" and specifies the target audience as "20-30 year old women interested in fashion," the system will operate as follows:

[1517] 1. Accepting input from the user

[1518] User entered information: "New product introduction", "Female, 20-30 years old, interested in fashion".

[1519] 2. Retrieving information from the database

[1520] A server receives the information submitted by the user and performs a database query to retrieve relevant past buzz posts.

[1521] 3. Preprocessing and Natural Language Processing Methods

[1522] The server removes noise from the acquired posting data, removes stop words, and normalizes the text.

[1523] 4. Analysis Methods Using Machine Learning Models

[1524] The server applies natural language processing to the preprocessed data to extract important keywords, phrases, and sentiment.

[1525] 5. User Emotion Analysis Method Using an Emotion Engine

[1526] The server analyzes facial expression and voice data sent from the user's device and recognizes that the user's emotions are positive.

[1527] 6. How to generate and present recommendations to users

[1528] The server generates posts that are appropriate for positive emotions based on the analysis results of the emotion engine:

[1529] Recommended keywords: "new products," "trends," "up your style"

[1530] Recommended hashtags: "New products," "Trends," "Fashion," "Style up"

[1531] Post template: "New fashion items have arrived! Don't miss out on these trendy items. Just a glimpse of them will definitely elevate your style! New Products Trends Fashion Style Up"

[1532] Prompt Sentence Examples

[1533] markdown

[1534] New Product Introduction

[1535] Generate positive posts showcasing fashion products targeted at women aged 20-30.

[1536] ---

[1537] Target Information:

[1538] Age range: 20-30 years old

[1539] Gender: Female

[1540] Interests: Fashion

[1541] User's emotional state:

[1542] positive

[1543] Generated content:

[1544] A new fashion item has arrived! This trendy item is a must-see. Just a glimpse of it will definitely enhance your style!

[1545] New products Trend fashion Style up

[1546] In this way, the present invention provides a system that generates optimal posting content based on the user's emotional state and target information, and can efficiently create buzz on SNS.

[1547] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1548] Step 1:

[1549] The content that the user wants to create buzz and target information are accepted.

[1550] Specific behavior:

[1551] Users launch the application on their smartphone and access the interface, where they input the content they want to create buzz about (e.g., "Introducing a new product") and target information (e.g., "Women aged 20-30 who are interested in fashion"). This information is then sent from the smartphone to the server.

[1552] Input: User-entered content and target information they want to create buzz about.

[1553] Output: Information data sent to the server.

[1554] Step 2:

[1555] Retrieve information from a database.

[1556] Specific behavior:

[1557] The server receives the information sent by the user and then executes a database query to retrieve posts related to the specified target information from a database of past buzz posts. The query, including the filtering criteria, is sent to the database.

[1558] Input: User-specified target information.

[1559] Output: Relevant buzz post data retrieved from the database.

[1560] Step 3:

[1561] Preprocess the acquired post data.

[1562] Specific behavior:

[1563] The server preprocesses the submitted data, removing noise, stop words, and normalizing the text to make it easier to analyze. This process uses natural language processing technology.

[1564] Input: Post data retrieved from the database.

[1565] Output: Preprocessed text data.

[1566] Step 4:

[1567] Perform text analysis on the preprocessed data.

[1568] Specific behavior:

[1569] The server applies natural language processing (NLP) to the preprocessed data, performing morphological analysis to extract key keywords, phrases, and sentiment (positive, negative, neutral). NLP libraries (e.g., spaCy, NLTK) are used.

[1570] Input: Preprocessed text data.

[1571] Output: Parsed keywords, phrases and sentiment information.

[1572] Step 5:

[1573] The analysis results are input into a machine learning model to generate optimal post content.

[1574] Specific behavior:

[1575] The server inputs the data analyzed by natural language processing into a machine learning model. The machine learning model predicts the optimal posting pattern for the target based on the results learned from past buzz post data. A machine learning framework (e.g., TensorFlow, PyTorch) is used here.

[1576] Input: Parsed data (keywords, phrases, sentiment information).

[1577] Output: Generated post content, suggested keywords and hashtags.

[1578] Step 6:

[1579] Analyze user sentiment.

[1580] Specific behavior:

[1581] The server analyzes the user's emotions using an emotion engine. It collects and analyzes the user's facial expression and voice data using the smartphone's camera and microphone. The emotion engine classifies the user's emotional state (positive, negative, neutral).

[1582] Input: User's facial expression data, voice data.

[1583] Output: The user's emotional state (positive, negative, neutral).

[1584] Step 7:

[1585] Recommendations are generated and presented to the user.

[1586] Specific behavior:

[1587] The server generates and adjusts optimal post content based on the user's emotional state. It combines the analysis results from the emotion engine with the output of the machine learning model to create templates for keywords, hashtags, post tone (emotion), and specific post content to use. The generated post content is presented to the user on their smartphone screen.

[1588] Input: Sentiment engine analysis results, machine learning model output.

[1589] Output: The post content, suggested keywords, and hashtags presented to the user.

[1590] This allows users to upload content to social media that best suits their emotional state, making it possible to create buzz efficiently and effectively.

[1591] 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.

[1592] 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.

[1593] 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.

[1594] [Fourth embodiment]

[1595] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1596] 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.

[1597] 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).

[1598] 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.

[1599] 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.

[1600] 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).

[1601] 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.

[1602] 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.

[1603] 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.

[1604] 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.

[1605] 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.

[1606] 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.

[1607] 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."

[1608] This invention relates to a system for efficiently communicating content that you want to create buzz on social networking sites to specific targets. Below, we will explain the program processing of this system in natural language.

[1609] System configuration

[1610] The system mainly consists of the following means:

[1611] 1. Means of accepting input from users

[1612] 2. How to obtain information from the database

[1613] 3. Natural Language Processing and Text Analysis Methods

[1614] 4. How to apply machine learning models

[1615] 5. How to generate and present recommendations to users

[1616] Program processing

[1617] 1. Means of accepting input from users

[1618] Users access the system's interface using a device (such as a smartphone or PC). They input the content they want to create buzz about and target information (age group, gender, interests, etc.). This information is then sent from the device to the server.

[1619] 2. How to obtain information from the database

[1620] The server receives the information sent by the user and then executes a query to retrieve buzz posts related to the specified target information from a database that stores past buzz post data. Here, the query includes filtering conditions based on the target information.

[1621] 3. Natural Language Processing and Text Analysis Methods

[1622] The server preprocesses the acquired post data, including noise removal, stop word removal, and text normalization. It then uses natural language processing (NLP) techniques to perform morphological analysis of the text and extract important keywords, phrases, and sentiment (positive, negative, neutral).

[1623] 4. How to apply machine learning models

[1624] The server then inputs the data into a machine learning model based on the analysis results. This model learns from past success stories (buzzworthy posts) and predicts the best posting pattern for the target audience. The output generates templates for keywords, hashtags, post tone (emotion), and specific post content to use.

[1625] 5. How to generate and present recommendations to users

[1626] The server then compiles the final recommendations and creates a list of suggested posts for the user, which may include, for example:

[1627] Example keywords

[1628] Hashtag examples

[1629] Post Template

[1630] The device displays the received recommendations to the user, who can then review and edit them and post them directly to the social networking site.

[1631] Specific examples

[1632] For example, if a user wants to create buzz about the opening of a new cafe and specifies the target audience as "women in their 20s who live in urban areas and love cafes," the system will operate as follows:

[1633] 1. Accepting input from the user

[1634] User input: "New cafe opening", "Female in her 20s, urban resident, cafe lover"

[1635] 2. Retrieving information from the database

[1636] The server runs a database query to retrieve past trending posts, such as:

[1637] "Cafe hopping and new discoveries"

[1638] "I went to a new cafe! It's stylish and comfortable - perfect for Instagram."

[1639] "All the cafes in the city are great, but this one was especially great! I want to connect with other cafe lovers."

[1640] 3. Natural Language Processing and Text Analytics

[1641] The server extracts important keywords and phrases from the above submission data:

[1642] Keywords: "cafe," "new," "urban," "Instagrammable"

[1643] Emotion: Positive

[1644] 4. Applying machine learning models

[1645] The server feeds the extracted data into a machine learning model to generate recommendations for:

[1646] Recommended keywords: "new cafe," "Instagrammable," "urban area," "cafe hopping"

[1647] Recommended hashtags: "cafe hopping," "new discoveries," "Instagrammable," "urban areas"

[1648] Post template: "A new cafe has opened! It has a relaxed atmosphere and is sure to be Instagram-worthy! Cafe hopping, urban areas, and connecting with fellow cafe lovers."

[1649] 5. Generating and presenting recommendations to the user

[1650] The device displays the recommended content to the user, who then checks and edits it and posts it to the social networking site.

[1651] The above is a specific embodiment of the present invention, which makes it possible to efficiently and accurately generate and present buzzworthy posts aimed at a target audience.

[1652] The processing flow will be explained below.

[1653] Step 1:

[1654] Users access the system's interface using a device (smartphone or PC). They input the content they want to create buzz about and target information (age group, gender, interests, etc.). This information is then sent from the device to the server.

[1655] Step 2:

[1656] The server receives the information sent by the user and then executes a query from a database of past buzz posts to retrieve posts related to the target information specified by the user. The query includes filtering conditions based on the target information.

[1657] Step 3:

[1658] The server preprocesses the submitted data retrieved from the database by removing noise, removing stop words, and normalizing the text, making the data easier to analyze.

[1659] Step 4:

[1660] The server then applies natural language processing (NLP) to the preprocessed post data. This step involves morphological analysis to extract important keywords, phrases, and sentiment (positive, negative, neutral).

[1661] Step 5:

[1662] The server then inputs the extracted keywords, phrases, and sentiment data into a machine learning model, which learns from past success stories (buzzworthy posts) and predicts the optimal posting pattern for the target audience.

[1663] Step 6:

[1664] The server creates recommended posts for users based on the predictions generated by the machine learning model, including optimal keywords, hashtags, post tone (emotion), and specific post content templates.

[1665] Step 7:

[1666] The server sends the generated recommendations to the device, which then displays them to the user. The user can then review and edit the recommendations and post them directly to the social networking site.

[1667] Specific examples

[1668] Step 1:

[1669] A user enters the following information into the system: "A new cafe is opening" and "A woman in her 20s who loves cafes and lives in an urban area."

[1670] Step 2:

[1671] A server receives the information submitted by the user and performs a database query to retrieve relevant past buzz posts.

[1672] Step 3:

[1673] The server removes noise from the acquired posting data, removes stop words, and normalizes the text.

[1674] Step 4:

[1675] The server applies natural language processing to the preprocessed data to extract important keywords, phrases, and sentiment.

[1676] Step 5:

[1677] The server inputs the extracted data into a machine learning model to predict the optimal posting patterns for the target audience.

[1678] Step 6:

[1679] The server generates recommended posts based on predictions from the machine learning model.

[1680] Step 7:

[1681] The server sends the generated recommendations to the device, which displays them to the user, who then checks and edits them and posts them to the SNS.

[1682] Through these processing steps, the system can effectively generate and present buzz posts to the user's target audience.

[1683] Example 1

[1684] 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."

[1685] Currently, many users are using social networking services (SNS) to spread information, but it is not easy to efficiently create buzz for a specific target audience. Existing methods make it difficult to predict what content and format will create buzz, and require a great deal of time and effort. Furthermore, selecting appropriate keywords and hashtags is difficult, so generating optimal post content requires specialized knowledge. Therefore, there is a need for a system that can efficiently and accurately generate and present buzzworthy posts for specific targets.

[1686] 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.

[1687] In this invention, the server includes means for receiving content and target information from a user that the user wants to create a buzz, means for retrieving related posts from a database storing past buzz post data, means for preprocessing the retrieved post data and performing text analysis using natural language processing, means for inputting the analysis results into a machine learning model to generate optimal post content, means for presenting the generated post content and recommended keywords and hashtags to the user, and means for the user to edit and confirm the generated post content and then post it to an SNS. This allows users to effectively create buzz posts aimed at a specific target demographic.

[1688] "Means for accepting content to be buzzed and target information from users" refers to an interface or function that allows users to input the content they want to buzz and information about the target person, and send it to the server.

[1689] "Means for retrieving related posts from a database storing data on past buzz posts" refers to a function in which the server queries the database, retrieves data on past buzz posts, and retrieves posts related to the target information specified by the user.

[1690] "Means for preprocessing acquired posted data and performing text analysis using natural language processing" refers to a function that performs preprocessing on acquired posted data, such as noise removal, stop word removal, and text normalization, and then analyzes the text using natural language processing technology.

[1691] "A means of inputting the analysis results into a machine learning model to generate optimal post content" refers to a function that inputs data obtained from text analysis into a machine learning model and uses the model, which has learned from past success stories, to generate post content that is most suitable for the target demographic.

[1692] "Means of presenting generated post content and recommended keywords and hashtags to users" refers to the function of organizing post content, keywords, hashtags, etc. generated by machine learning models and displaying and suggesting them to users.

[1693] "Means for users to post generated posts to SNS after editing and reviewing them" refers to a function that allows users to review and edit the suggested content and post it directly to the SNS platform.

[1694] "Preprocessing of posted data retrieved from the database" refers to the process of removing unnecessary information (noise) from the posted data, deleting commonly used words (stop words) that are irrelevant to a specific analysis, and normalizing the strings to maintain consistency.

[1695] "Natural language processing" is a technology that enables computers to understand, interpret, and generate human language, and in the present invention refers to the technology used for morphological analysis of text data and extraction of important keywords and emotions.

[1696] A "machine learning model" refers to an algorithm that learns patterns from given data and makes predictions and classifications for new data. In the context of this invention, it is used to learn from past buzz posts and generate post content that is optimal for the target demographic.

[1697] "Target information" refers to the attribute information (e.g., age group, gender, interests, etc.) of the intended recipients of the post that the user wants to make go viral, and the content of the post is optimized based on that information.

[1698] This invention relates to a system for efficiently communicating content that users want to create buzz on social media to specific targets. This system accepts input from users, acquires and analyzes relevant data from past buzz posts, and generates optimal post content using a machine learning model, which is then presented to the user. The configuration and operation of this system are described in detail below.

[1699] System configuration

[1700] The system mainly consists of the following means:

[1701] 1. Means of accepting input from users

[1702] 2. How to obtain information from the database

[1703] 3. Natural Language Processing and Text Analysis Methods

[1704] 4. How to apply machine learning models

[1705] 5. How to generate and present recommendations to users

[1706] 6. Means of posting to social media

[1707] Processing Details

[1708] 1. Accepting input from the user

[1709] Users access the system's interface using a smartphone or PC. They input the content they want to create buzz and target information (age group, gender, interests, etc.), and this information is sent from the device to the server. The input is in a specific data format, and the server receives it and begins processing.

[1710] 2. Retrieving information from the database

[1711] After receiving the information sent by the user, the server executes a database query based on that information. The database stores past buzz post data and retrieves relevant post data based on specific filtering criteria. For example, a query is executed to retrieve information about "women in their 20s, urban areas, cafe lovers."

[1712] 3. Natural Language Processing and Text Analytics

[1713] The server preprocesses the acquired post data. This preprocessing includes noise removal (removal of special symbols and unnecessary spaces), stopword removal (frequent phrases that are generally ignored), and text normalization (standardization of characters). Then, using natural language processing technology (e.g., the morphological analysis tool MeCab), it performs morphological analysis of the text data and extracts important keywords, phrases, and sentiment (positive, negative, neutral).

[1714] 4. Applying machine learning models

[1715] Based on the analysis results, the server inputs the data into a machine learning model. The machine learning model learns from past successful posting data and predicts the optimal posting pattern for the target demographic. For example, the model predicts the keywords, hashtags, and tone (emotion) to be used in the post, and generates a specific post content template.

[1716] 5. Generating and presenting recommendations to the user

[1717] Finally, the server presents recommendations to the user, including example keywords, example hashtags, and post templates, which the user can review and edit to select the post they deem most appropriate.

[1718] 6. How to post to social media

[1719] The final post content that users have considered can be posted directly to the SNS platform, and users can easily publish the generated post content to the SNS through their device.

[1720] Specific examples

[1721] For example, if a user wants to create buzz about the opening of a new cafe and specifies the target audience as "women in their 20s who live in urban areas and love cafes," the system will operate as follows:

[1722] 1. Accepting input from the user

[1723] User input: "New cafe opening", "Female in her 20s, urban resident, cafe lover"

[1724] 2. Retrieving information from the database

[1725] The server runs a database query to retrieve past viral posts, such as:

[1726] "Cafe hopping and new discoveries"

[1727] "I went to a new cafe! It's stylish and comfortable - perfect for Instagram."

[1728] "All the cafes in the city are great, but this one was especially great! I want to connect with other cafe lovers."

[1729] 3. Natural Language Processing and Text Analytics

[1730] The server extracts important keywords and phrases from the above submission data:

[1731] Keywords: "cafe," "new," "urban," "Instagrammable"

[1732] Emotion: Positive

[1733] 4. Applying machine learning models

[1734] The server feeds the extracted data into a machine learning model to generate the following recommendations:

[1735] Recommended keywords: "new cafe," "Instagrammable," "urban area," "cafe hopping"

[1736] Recommended hashtags: "cafe hopping," "new discoveries," "Instagrammable," "urban areas"

[1737] Post template: "A new cafe has opened! It has a relaxed atmosphere and is sure to be Instagram-worthy! Cafe hopping, urban areas, and connecting with fellow cafe lovers."

[1738] 5. Generating and presenting recommendations to the user

[1739] The device displays the recommended content to the user, who can then review it, edit it if necessary, and post it to social media.

[1740] Examples of prompt statements

[1741] You might enter the following prompts for the system's generated AI model:

[1742] "I want to create a buzz about the opening of a new cafe among women in their 20s who love cafes and live in urban areas. Based on past data on buzzworthy posts, could you recommend some keywords, hashtags, and a post template?"

[1743] The above is a specific embodiment of the present invention, which makes it possible to efficiently and accurately generate and present buzzworthy posts aimed at a target audience.

[1744] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1745] Step 1: Accepting input from the user

[1746] Users access the system's interface using a smartphone or PC. They input the content they want to create buzz about (e.g., "A new cafe has opened") and target information (e.g., "Women in their 20s who live in urban areas and love cafes"). The device receives this information and sends it to the server. The server temporarily stores the input information before proceeding to the next processing step.

[1747] Step 2: Retrieving information from the database

[1748] The server receives target information sent by the user. Based on this information, the server sends a query to the database to retrieve related posts from past buzz post data. For example, it searches the database for posts related to "women in their 20s" and "cafes" and extracts related post data. Input: Target information. Output: Related past buzz post data.

[1749] Step 3: Natural Language Processing and Text Analytics

[1750] The server preprocesses the acquired post data. Preprocessing includes removing noise (removing special symbols and unnecessary spaces), deleting stop words (frequent words that are generally ignored), and normalizing the text (standardizing characters). Natural language processing techniques (such as morphological analysis tools) are then used to perform morphological analysis of the text data, extract important keywords and phrases, and analyze sentiment (positive, negative, neutral). Input: Past buzz post data. Output: Preprocessed text data and analysis results.

[1751] Step 4: Applying the machine learning model

[1752] The server inputs the preprocessed text data and analysis results into a machine learning model. The machine learning model learns from past posting data of successful cases and predicts the optimal posting pattern for the target demographic. This prediction generates templates for the keywords, hashtags, tone (emotion) of the post, and specific post content to be used. Input: Analysis results. Output: Optimal post content, recommended keywords, recommended hashtags.

[1753] Step 5: Generate recommendations and present them to the user

[1754] The server organizes the final recommendations and creates a list of post content to suggest to the user. The list includes example keywords, example hashtags, and post templates. The device displays the received recommendations to the user. The user can review and edit these recommendations and select the most suitable post content. Input: Best post content, recommended keywords, recommended hashtags. Output: List of posts presented to the user.

[1755] Step 6: How to post to social media

[1756] The user reviews the suggested content and edits it as necessary. Through the device interface, the user can post the final post directly to social media. This allows for efficient spreading of content that is targeted to a specific demographic. Input: Final post content after user editing. Output: Post to social media.

[1757] (Application example 1)

[1758] 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."

[1759] In order to efficiently create buzz on social media for specific content, it is necessary to generate posts that are optimal for the target audience. However, conventional systems that utilize artificial intelligence and machine learning require the manual collection and analysis of data and the generation of post content, which is time-consuming and labor-intensive. In addition, there is uncertainty as to whether the generated post content will actually appeal to the target, so a reliable method is needed to increase the probability of buzz.

[1760] 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.

[1761] In this invention, the server includes means for receiving content and target information from a user that the user wants to create a buzz, means for retrieving related posts from a database storing past buzz post data, means for preprocessing the retrieved post data and performing text analysis using natural language processing, means for inputting the analysis results into a machine learning model to generate optimal post content, means for presenting the generated post content and recommended keywords and hashtags to the user, and means for posting the generated optimal post content to an SNS via an application on a smart device. This allows users to avoid complex manual data collection and analysis work, quickly and easily generate post content that is most effective for their target, and significantly increases the chances of creating a buzz on an SNS.

[1762] "User" refers to any individual or entity that uses the System.

[1763] "Buzzworthy content" refers to information or messages posted on social media with the aim of attracting a lot of attention.

[1764] "Target information" is data that indicates the attributes and interests of the target audience for a particular post, including, for example, age group, gender, place of residence, and interests.

[1765] "Database" refers to a collection of information that stores collected data on past buzz posts and makes it searchable and retrievalable.

[1766] "Preprocessing" refers to the process of processing the data by removing noise, deleting stop words, normalizing text, etc. from the acquired submission data.

[1767] "Natural language processing" refers to the technology that enables computers to understand and analyze human language.

[1768] "Text analysis" refers to the process of using natural language processing techniques to interpret the meaning and structure of a document and extract important information.

[1769] A "machine learning model" refers to a statistical method for learning patterns and rules from past data and making predictions and classifications for future data.

[1770] "Keywords" are particularly important words or phrases in your post that will grab your target's attention.

[1771] A "hashtag" is a type of tagging notation used in social media posts, and refers to a keyword that makes it easier to group posts related to a particular topic or theme.

[1772] "Smart device" refers to a portable electronic device that can connect to a network, such as a smartphone or tablet.

[1773] "Application" refers to a software program that runs on a smart device and provides specific functionality.

[1774] "SNS" is an abbreviation for social networking service, which refers to a platform where users can share information and interact with each other online.

[1775] This invention relates to a system for efficiently communicating content that you want to create buzz on SNS to designated targets, and includes the following means and processes.

[1776] Program processing explanation

[1777] The server receives the content and target information from the user and performs the following processing based on this: First, it retrieves data related to the specified target information from a database of past buzz posts, and then performs preprocessing to remove noise, stop words, and normalize the data.

[1778] The server then performs text analysis using natural language processing (NLP) techniques, including morphological analysis, keyword extraction, and sentiment analysis of the acquired data, using software such as NLTK (Natural Language Toolkit) and Scikit-learn.

[1779] The server then inputs the analyzed data into a machine learning model, which uses classification techniques such as logistic regression to predict the best post content, keywords, and hashtags for the target audience based on past success stories.

[1780] Finally, the generated optimal post content, recommended keywords, and hashtags are presented to the user, who posts it to a social networking site via an application on their smart device.

[1781] Specific examples

[1782] For example, if a user wants to create buzz about the opening of a new cafe and specifies "women in their 20s who love cafes and live in urban areas" as their target, the system will operate as follows.

[1783] User Input

[1784] The user wants to create buzz: "New cafe opening"

[1785] Target information: "Women in their 20s who love cafes and live in urban areas"

[1786] Server Processing

[1787] Get past trending posts related to your target information from the database. Example:

[1788] "Cafe hopping and new discoveries"

[1789] "I went to a new cafe! It's stylish and comfortable - perfect for Instagram."

[1790] "All the cafes in the city are great, but this one was especially great! I want to connect with other cafe lovers."

[1791] Keyword extraction from analysis results:

[1792] Keywords: "cafe," "new," "urban," "Instagrammable"

[1793] Emotion: Positive

[1794] Generate suggested posts

[1795] Recommended keywords: "new cafe," "Instagrammable," "urban area," "cafe hopping"

[1796] Recommended hashtags: "cafe hopping," "new discoveries," "Instagrammable," "urban areas"

[1797] Post template: "A new cafe has opened! It has a relaxed atmosphere and is sure to be Instagram-worthy! Cafe hopping, urban areas, and connecting with fellow cafe lovers."

[1798] Prompt Sentence Examples

[1799] The following sentences can be used as prompts for users to input to a generative AI model:

[1800] "Generate the best social media post content to promote the opening of a new cafe. The target audience is women in their 20s who love cafes and live in urban areas. Past related post data is shown below. Please include keywords and hashtags that should be used in the post."

[1801] This allows users to quickly generate posts that are suitable for their target audience and are likely to create buzz, and then easily post them on social media.

[1802] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1803] Step 1:

[1804] Users enter the content they want to create buzz on social media and target information.

[1805] Input: What you want to create a buzz about (e.g., the opening of a new cafe), target information (e.g., women in their 20s who love cafes and live in urban areas)

[1806] What happens: A user provides input via an application on their smart device, which is sent to the server.

[1807] Step 2:

[1808] The server retrieves post data related to the target information from a database of past buzz posts.

[1809] Input: Target information sent by the user

[1810] Data processing: Perform database queries based on target information and filter relevant post data

[1811] Output: Data of past buzz posts (e.g., "Cafe hopping, new discoveries")

[1812] What happens: The server uses SQL to run a search query against the database to retrieve the relevant data.

[1813] Step 3:

[1814] Preprocess the submitted data received by the server.

[1815] Input: Previous buzz post data obtained

[1816] Data processing: noise removal, stopword removal, text normalization

[1817] Output: Clean post data after preprocessing

[1818] Specific operation: The server uses a natural language processing library such as NLTK to preprocess the text data.

[1819] Step 4:

[1820] The server performs text analysis using natural language processing technology.

[1821] Input: Clean post data after preprocessing

[1822] Data processing: morphological analysis of text, keyword extraction, sentiment analysis

[1823] Output: Key keywords, phrases, and emotional state (positive, negative, neutral)

[1824] Specific operation: The server uses Scikit-learn or similar tools to apply models for morphological analysis and keyword extraction.

[1825] Step 5:

[1826] The server inputs the analysis results into a machine learning model to generate optimal post content.

[1827] Input: Keywords, phrases, and emotional states obtained through natural language processing

[1828] Data Computing: Prediction and Content Generation with Machine Learning Models

[1829] Output: Best post content, recommended keywords, recommended hashtags, post templates

[1830] What it does: The server uses machine learning models such as logistic regression to generate targeted posts.

[1831] Step 6:

[1832] The server presents the generated post content and recommended keywords and hashtags to the user.

[1833] Input: Best post content, suggested keywords, suggested hashtags, post template

[1834] Output: What the user sees on their smart device

[1835] Specific behavior: The server sends the generated content to the user interface so that the user can view it on their smart device.

[1836] Step 7:

[1837] The optimal post content that the user has checked and edited is posted to SNS.

[1838] Input: Post content provided by the server, edits made by the user

[1839] Output: Posts published on social media

[1840] Specific behavior: The user publishes a post using the interface that allows them to post directly to the social networking site from the application.

[1841] 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.

[1842] This invention relates to a system that enables efficient communication of content that you want to create buzz on social media to specific targets. This system achieves even more effective communication by combining it with an emotion engine that recognizes and analyzes user emotions. Below, we will explain the program processing of this system in natural language.

[1843] System configuration

[1844] The system mainly consists of the following means:

[1845] 1. Means of accepting input from users

[1846] 2. How to obtain information from the database

[1847] 3. Natural Language Processing and Text Analysis Methods

[1848] 4. How to apply machine learning models

[1849] 5. User Emotion Analysis Method Using Emotion Engine

[1850] 6. How to generate and present recommendations to users

[1851] Program processing

[1852] 1. Means of accepting input from users

[1853] Users access the system's interface using a device (such as a smartphone or PC). They input the content they want to create buzz about and target information (age group, gender, interests, etc.). This information is then sent from the device to the server.

[1854] 2. How to obtain information from the database

[1855] The server receives the information sent by the user and then executes a query from a database of past buzz posts to retrieve posts related to the target information specified by the user. The query includes filtering conditions based on the target information.

[1856] 3. Natural Language Processing and Text Analysis Methods

[1857] The server preprocesses the submitted data by removing noise, removing stop words, and normalizing the text, making the data easier to analyze.

[1858] 4. How to apply machine learning models

[1859] The server then applies natural language processing (NLP) to the preprocessed post data. This step involves morphological analysis to extract important keywords, phrases, and sentiment (positive, negative, neutral).

[1860] 5. User Emotion Analysis Method Using Emotion Engine

[1861] The server uses an emotion engine to analyze the user's emotions based on the user's facial expressions, tone of voice, or text input. Based on the analysis results, the user's emotions are classified as positive, negative, or neutral.

[1862] 6. How to generate and present recommendations to users

[1863] The server adjusts the predictions generated by the machine learning model based on the user's emotions recognized by the emotion engine to create optimal post content, including keywords, hashtags, post tone (emotion), and specific post content templates to use.

[1864] For example, if the user's emotions are positive, the system generates posts with a bright and cheerful tone. If the emotions are negative, the system emphasizes an encouraging and comforting tone.

[1865] Specific examples

[1866] For example, if a user wants to create buzz about the opening of a new cafe and specifies the target audience as "women in their 20s who live in urban areas and love cafes," the system will operate as follows:

[1867] 1. Accepting input from the user

[1868] User input: "New cafe opening", "Female in her 20s, urban resident, cafe lover"

[1869] 2. Retrieving information from the database

[1870] A server receives the information submitted by the user and performs a database query to retrieve relevant past buzz posts.

[1871] 3. Natural Language Processing and Text Analytics

[1872] The server removes noise from the acquired posting data, removes stop words, and normalizes the text.

[1873] 4. Applying machine learning models

[1874] The server applies natural language processing to the preprocessed data to extract important keywords, phrases, and sentiment.

[1875] 5. User Emotion Analysis Using an Emotion Engine

[1876] The server analyzes facial expression data, voice data, or input text sent from the user's device and recognizes that the user's emotions are positive.

[1877] 6. Generating and Presenting Recommendations to the User

[1878] The server generates posts that are appropriate for positive emotions based on the analysis results of the emotion engine:

[1879] Recommended keywords: "new cafe," "Instagrammable," "urban area," "cafe hopping"

[1880] Recommended hashtags: "cafe hopping," "new discoveries," "Instagrammable," "urban areas"

[1881] Post template: "A new cafe has opened! It has a relaxed atmosphere and is sure to be Instagram-worthy! Cafe hopping, urban areas, and connecting with fellow cafe lovers."

[1882] Through the above processing steps, the system can effectively generate and present buzz posts to the user's target demographic, and provide optimal content based on the user's emotions.

[1883] The processing flow will be explained below.

[1884] Step 1:

[1885] Users access the system's interface using a device (smartphone or PC). They input the content they want to create buzz about (e.g., "A new cafe has opened") and target information (e.g., "Women in their 20s who live in urban areas and love cafes"). This information is then sent from the device to the server.

[1886] Step 2:

[1887] The server receives the input information sent by the user. It then executes a query to retrieve posts related to the specified target information from a database of past buzz posts. For example, it filters past posts related to "women in their 20s" and "cafes" based on the user's target information.

[1888] Step 3:

[1889] The server preprocesses the submitted data retrieved from the database by removing noise, removing stop words, and normalizing the text, making the data easier to analyze.

[1890] Step 4:

[1891] The server then applies natural language processing (NLP) to the preprocessed post data. This step involves morphological analysis to extract important keywords (e.g., "new cafe"), phrases, and sentiment (positive, negative, neutral).

[1892] Step 5:

[1893] The server inputs data into a machine learning model based on keywords and sentiment extracted through natural language processing. This model learns from past success stories (buzzworthy posts) and predicts the best posting pattern for the target audience. For example, it generates recommended hashtags such as "Instagrammable" and "cafe hopping."

[1894] Step 6:

[1895] The server analyzes the user's emotions using an emotion engine. This analysis is based on the user's facial expression recognition, voice analysis, or text input. For example, if the user has a positive expression while typing "new cafe," this emotion is recognized as positive.

[1896] Step 7:

[1897] The server adjusts the predictions generated by the machine learning model based on the user's emotions recognized by the emotion engine. For example, if the user's emotions are positive, it generates posts with a bright tone. On the other hand, if the emotions are negative, it generates posts that contain encouragement and comfort.

[1898] Step 8:

[1899] The server then compiles the final recommendations and creates a list of suggested posts for the user, including the best keywords, hashtags, and post templates, for example:

[1900] Recommended keywords: "new cafe," "Instagrammable," "urban area," "cafe hopping"

[1901] Recommended hashtags: "cafe hopping," "new discoveries," "Instagrammable," "urban areas"

[1902] Post template: "A new cafe has opened! It has a relaxed atmosphere and is sure to be Instagram-worthy! Cafe hopping, urban areas, and connecting with fellow cafe lovers."

[1903] Step 9:

[1904] The server sends the generated recommendations to the device. The device displays the received recommendations to the user. The user can then review and edit the recommendations and post them directly to the SNS.

[1905] Through the above processing steps, the system can effectively generate buzz posts for the user's target demographic and provide optimal content according to the user's emotions.

[1906] Example 2

[1907] 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."

[1908] With conventional SNS posting systems, it was difficult for users to effectively communicate information to their target audience. Also, because only uniform post content could be generated without considering the user's emotions, the effectiveness of the posts was limited. As a result, it was difficult for the posts to go viral, and it was difficult for the posts to reach the intended audience.

[1909] 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.

[1910] In this invention, the server includes means for receiving content to be made buzzworthy and target information from a user, means for retrieving related posts from a database storing past buzz posting data, means for preprocessing the retrieved posting data and performing text analysis using natural language processing, means for recognizing and analyzing user emotions using a sentiment analysis engine, means for inputting the analysis results into a machine learning algorithm to generate optimal posting content, and means for presenting the generated posting content and recommended keywords and hashtags to the user. This makes it possible to generate posting content that is effective for the target demographic and in line with their emotions.

[1911] "Means for receiving content to be buzzed and target information from a user" refers to an interface or process for receiving content to be buzzed and target information input by a user.

[1912] "Means for retrieving related posts from a database storing past buzz posting data" refers to a function for searching and retrieving information related to the user's requirements from stored past buzz posting data.

[1913] "Means for preprocessing acquired posted data and performing text analysis using natural language processing" refers to the process of removing noise, deleting stop words, and normalizing acquired posted data, and then analyzing the content using natural language processing.

[1914] "Means for recognizing and analyzing user emotions using an emotion analysis engine" refers to the function of analyzing emotions from data such as text, voice, and facial expressions entered by the user and classifying them as positive, negative, neutral, etc.

[1915] "Method of inputting analysis results into a machine learning algorithm to generate optimal post content" refers to the process of generating appropriate post content through a machine learning model based on the results of natural language processing and sentiment analysis.

[1916] "Means for presenting generated post content and recommended keywords and hashtags to users" refers to functions and interfaces for providing users with post content generated by the system and effective keyword and hashtag suggestions.

[1917] A specific embodiment for carrying out the present invention will be described below. The system of the present invention realizes effective information transmission by combining users, terminals, and servers.

[1918] System configuration and hardware / software used

[1919] 1. Means of accepting input from users

[1920] Users access the system interface via a web browser on their device (e.g., smartphone, PC). They then access the system and enter the content they want to create buzz about and target information (age group, gender, interests, etc.). This information is sent to the server as an HTTP request.

[1921] 2. How to obtain information from the database

[1922] The server receives the information sent by the user, then executes an SQL query against a database of past buzz posts (e.g., MySQL or PostgreSQL) to search for and retrieve posts related to the target information specified by the user.

[1923] 3. Natural Language Processing and Text Analysis Methods

[1924] The server uses a Python NLP library (e.g., NLTK, spaCy) to preprocess the submitted data. Specifically, it performs the following steps:

[1925] Noise removal: Remove unnecessary special characters and HTML tags.

[1926] Stopword removal: Removal of common words that have no meaning (e.g. "wa", "o", "ni").

[1927] Text normalization: Convert all text to lower case for consistency.

[1928] 4. How to apply machine learning models

[1929] The server then performs natural language processing (NLP) on the preprocessed text data, including morphological analysis and sentiment analysis, using TensorFlow and PyTorch to extract key keywords, phrases, and sentiment (positive, negative, neutral).

[1930] 5. User Emotion Analysis Method Using Emotion Engine

[1931] The server uses an emotion engine (e.g., IBM Watson's Tone Analyzer) to analyze emotions from the user's input data (e.g., text, voice, facial expressions, etc.). Based on the analysis results, the user's emotions are classified as positive, negative, or neutral.

[1932] 6. How to generate and present recommendations to users

[1933] The server combines the results of the machine learning model and the emotion engine to generate optimal posts for the target users. Specifically, it generates the following elements:

[1934] Recommended Keywords

[1935] Recommended hashtags

[1936] The tone (emotion) of the post

[1937] Specific post content template

[1938] The user can review the generated post and make any necessary adjustments.

[1939] Specific examples

[1940] For example, if a user wants to create buzz about the opening of a new cafe and specifies the target audience as "women in their 20s who live in urban areas and love cafes," the system will operate as follows:

[1941] 1. Accepting input from the user

[1942] The user fills in the form with "a new cafe opening" and "a woman in her 20s who loves cafes and lives in an urban area" and submits it.

[1943] 2. Retrieving information from the database

[1944] Based on the input information received by the server, a database query is executed to retrieve related past buzz posts.

[1945] 3. Natural Language Processing and Text Analytics

[1946] The server removes noise from the acquired submission data, removes stop words, and normalizes the text.

[1947] 4. Applying machine learning models

[1948] The server applies NLP to the pre-processed data to extract important keywords, phrases, and sentiment.

[1949] 5. User Emotion Analysis Using an Emotion Engine

[1950] The server analyzes the user's text input and recognizes it as a positive emotion.

[1951] 6. Generating and Presenting Recommendations to the User

[1952] Based on the sentiment and NLP results, the server generates the following post with a positive tone:

[1953] Recommended keywords: "new cafe," "Instagrammable," "urban area," "cafe hopping"

[1954] Recommended hashtags: "cafe hopping," "new discoveries," "Instagrammable," "urban areas"

[1955] Post template: "A new cafe has opened! It has a relaxed atmosphere and is sure to be Instagram-worthy! Cafe hopping, urban areas, and I want to connect with fellow cafe lovers."

[1956] Prompt Sentence Examples

[1957] Example prompts to input to a generative AI model:

[1958] We want to create buzz on social media about the opening of a new cafe. Our target audience is women in their 20s who love cafes and live in urban areas. We will generate effective posts based on data on past buzz posts. Since many users have positive feelings, we recommend creating posts with a bright and fun tone.

[1959] As a result, this system can not only effectively generate and present buzz posts to the user's target demographic, but also provide optimal content that reflects the user's emotions.

[1960] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1961] Divide the processing flow of the system program into processing steps

[1962] Step 1:

[1963] Step 2:

[1964] Step 3:

[1965] Step 4:

[1966] Step 5:

[1967] Step 6:

[1968] Specific explanation of each processing step

[1969] Step 1: Accepting input from the user

[1970] Input: A user accesses the system interface using a terminal and enters the content they want to buzz and target information into a form.

[1971] Processing: The terminal sends the input information to the server in the form of an HTTP request.

[1972] Output: The server temporarily stores the received content and target information in memory.

[1973] What happens: A user opens a browser, fills in a web form with the words "New cafe opening" and "Female, 20s, urban resident, cafe lover," and clicks the submit button.

[1974] Step 2: How to retrieve information from the database

[1975] Input: The server generates a database query based on the buzz content and target information received from the user.

[1976] Processing: The server executes an SQL query against a database of past buzz posts (e.g., MySQL, PostgreSQL) to search and retrieve relevant past posts.

[1977] Output: A dataset containing the retrieved submission data.

[1978] Specific operation: The server queries the database using the target information of "women in their 20s" and "urban areas," and extracts related past buzz posts using an SQL query.

[1979] Step 3: Natural language processing and text analysis tools

[1980] Input: The raw submission data obtained.

[1981] Processing: The server uses Python NLP libraries (e.g., NLTK, spaCy) to preprocess the data, specifically denoising, removing stop words, and normalizing the text.

[1982] Output: Preprocessed text data.

[1983] Specific operations: Noise removal removes HTML tags and special characters, stop word removal removes common words such as "は", "を", and "に", and text normalization standardizes case and removes unnecessary whitespace.

[1984] Step 4: How to apply the machine learning model

[1985] Input: Preprocessed text data.

[1986] Processing: The server performs natural language processing on the preprocessed data, performing morphological and sentiment analysis, specifically extracting key keywords, phrases, and sentiment (positive, negative, neutral) using TensorFlow and PyTorch.

[1987] Output: Extracted keywords, phrases, and sentiment data.

[1988] Specific operation: The server divides the text data into tokens, calculates and classifies the sentiment score of each token, and passes the result on to the next process.

[1989] Step 5: User emotion analysis method using emotion engine

[1990] Input: User text input, voice data, or facial expression data.

[1991] Processing: The server uses an emotion engine (e.g., Tone Analyzer) to analyze the user's emotion and determine whether it is classified as positive, negative, or neutral.

[1992] Output: User sentiment classification result.

[1993] Specific operation: Text and audio files entered by the user are passed to the emotion engine, emotion analysis is performed on each piece of data, and the analysis results are obtained.

[1994] Step 6: How to generate and present recommendations to the user

[1995] Input: Output data from the machine learning model and the sentiment engine.

[1996] Processing: The server integrates this data and generates posts that are optimally tailored to the user's target audience, including recommended keywords, hashtags, post tone, and specific post content templates.

[1997] Output: Generated post content, suggested keywords, suggested hashtags.

[1998] Specific behavior: Keywords to use: "New cafe," "Instagrammable," "Urban area," "Cafe hopping," recommended hashtags: "Cafe hopping," "New discovery," "Instagrammable," "Urban area," post template: "A new cafe has opened! It has a relaxed atmosphere and is sure to be Instagrammable! Cafe hopping, urban area, I want to connect with cafe lovers," and the following will be presented to the user.

[1999] As a result, this system can effectively generate and present buzz posts to the user's target demographic, and can provide optimal content that reflects the user's emotions.

[2000] (Application example 2)

[2001] 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."

[2002] Conventional social media buzz-generating posting systems perform analysis to maximize the impact on the target, but rarely take into account the user's own emotional state. As a result, the content of the post may not match the user's intended emotion or tone, resulting in an ineffective buzz. Another issue is that generating optimal post content to efficiently generate buzz among a specific target takes time and effort.

[2003] 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.

[2004] In this invention, the server includes means for receiving content to be made buzzworthy and target information from a user, means for retrieving related posts from a database storing past buzzworthy post data, means for preprocessing the retrieved post data and performing text analysis using natural language processing, means for inputting the analysis results into a machine learning model to generate optimal post content, means for presenting the generated post content and recommended keywords and hashtags to the user, and means for generating advertising copy based on the user's emotional state using an emotion engine that analyzes the user's emotions. This makes it possible to generate optimal post content according to the user's emotional state and effectively create buzz among the target demographic.

[2005] "Buzzworthy content" refers to content posted on social media that is intended to attract the attention of many users and be spread widely.

[2006] "Target information" refers to information about the target user demographic for buzz posts, such as a specific age group, gender, or interests.

[2007] The "database" is a data storage system that stores past buzz posting data and retrieves related information through queries.

[2008] "Natural language processing" is a technology for analyzing and processing language data, and is used for preprocessing and analyzing text.

[2009] A "machine learning model" is an algorithm that learns patterns from large amounts of data and makes predictions and classifications based on new data.

[2010] An "emotion engine" is a system that analyzes a user's facial expressions and voice to identify the user's current emotional state.

[2011] An "advertising copy" is a post that conveys information about a specific product or service on social media and aims to pique the interest of target users.

[2012] "Noise reduction" is the process of removing irrelevant or false information from the text data being analyzed.

[2013] "Stop word removal" is the process of removing words that are too general to contribute to the analysis (such as "no" and "wa") during text analysis.

[2014] "Normalization" is the process of correcting spelling variations and typos and standardizing data to maintain consistency in text data.

[2015] "Keywords" are important words that represent the main theme or topic of the post content, and increase search and spreading effectiveness on social media.

[2016] A "hashtag" is a linguistic element used on social media to tag posts related to a particular theme or topic, improving the discoverability of the posts.

[2017] The present invention relates to a system that enables efficient communication of content that you want to create buzz on social media to specific targets. By combining this system with an emotion engine that analyzes user emotions, more effective communication of information can be achieved. Specific means for realizing this system are described below.

[2018] This system is implemented primarily using the following hardware and software:

[2019] Hardware

[2020] Smartphone: Used by the user to input information and receive results from the system.

[2021] Server: The central computing resource that stores data, analyzes it, and processes the generated submissions.

[2022] software

[2023] Natural Language Processing (NLP) libraries: Libraries for preprocessing and parsing text (e.g., spaCy, NLTK).

[2024] Machine learning frameworks: Frameworks for learning from large amounts of data and making predictions or classifications (e.g., TensorFlow, PyTorch).

[2025] Emotion analysis engine: Software to analyze a user's facial expressions and voice to classify their emotional state (e.g., OpenCV, Google Cloud Speech-to-Text).

[2026] The server is implemented based on a system including the following means:

[2027] 1. Means of accepting input from users

[2028] Users access the system's interface using their smartphones, where they input the content they want to create buzz and target information (age group, gender, interests, etc.). This information is then sent from the smartphone to the server.

[2029] 2. How to obtain information from the database

[2030] The server receives the information sent by the user and then executes a query to retrieve posts related to the user's target information from a database of past buzz posts. The query includes filtering criteria based on the target information.

[2031] 3. Preprocessing and Natural Language Processing Methods

[2032] The server preprocesses the submitted data by removing noise, stop words, and normalizing the text. This step prepares the data for analysis.

[2033] 4. Analysis Methods Using Machine Learning Models

[2034] The server applies natural language processing (NLP) to the preprocessed post data, performing morphological analysis to extract important keywords, phrases, and sentiment (positive, negative, neutral).

[2035] 5. User sentiment analysis using a sentiment analysis engine

[2036] The server uses an emotion analysis engine to analyze the user's emotions. This analysis is performed using the smartphone's camera and microphone to capture the user's facial expressions and voice. Based on the analysis results, the user's emotions are classified as positive, negative, or neutral.

[2037] 6. How to generate and present recommendations to users

[2038] The server adjusts the predictions generated by the machine learning model based on the user's emotions recognized by the emotion analysis engine, and creates optimal post content. Specifically, this includes the keywords, hashtags, tone (emotion) of the post to be used, and specific post content templates. The generated post content is presented to the user on the smartphone screen.

[2039] Specific examples

[2040] For example, if a user wants to "introduce a new product" and specifies the target audience as "20-30 year old women interested in fashion," the system will operate as follows:

[2041] 1. Accepting input from the user

[2042] User entered information: "New product introduction", "Female, 20-30 years old, interested in fashion".

[2043] 2. Retrieving information from the database

[2044] A server receives the information submitted by the user and performs a database query to retrieve relevant past buzz posts.

[2045] 3. Preprocessing and Natural Language Processing Methods

[2046] The server removes noise from the acquired posting data, removes stop words, and normalizes the text.

[2047] 4. Analysis Methods Using Machine Learning Models

[2048] The server applies natural language processing to the preprocessed data to extract important keywords, phrases, and sentiment.

[2049] 5. User Emotion Analysis Method Using an Emotion Engine

[2050] The server analyzes facial expression and voice data sent from the user's device and recognizes that the user's emotions are positive.

[2051] 6. How to generate and present recommendations to users

[2052] The server generates posts that are appropriate for positive emotions based on the analysis results of the emotion engine:

[2053] Recommended keywords: "new products," "trends," "up your style"

[2054] Recommended hashtags: "New products," "Trends," "Fashion," "Style up"

[2055] Post template: "New fashion items have arrived! Don't miss out on these trendy items. Just a glimpse of them will definitely elevate your style! New Products Trends Fashion Style Up"

[2056] Prompt Sentence Examples

[2057] markdown

[2058] New Product Introduction

[2059] Generate positive posts showcasing fashion products targeted at women aged 20-30.

[2060] ---

[2061] Target Information:

[2062] Age range: 20-30 years old

[2063] Gender: Female

[2064] Interests: Fashion

[2065] User's emotional state:

[2066] positive

[2067] Generated content:

[2068] A new fashion item has arrived! This trendy item is a must-see. Just a glimpse of it will definitely enhance your style!

[2069] New products Trend fashion Style up

[2070] In this way, the present invention provides a system that generates optimal posting content based on the user's emotional state and target information, and can efficiently create buzz on SNS.

[2071] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[2072] Step 1:

[2073] The content that the user wants to create buzz and target information are accepted.

[2074] Specific behavior:

[2075] Users launch the application on their smartphone and access the interface, where they input the content they want to create buzz about (e.g., "Introducing a new product") and target information (e.g., "Women aged 20-30 who are interested in fashion"). This information is then sent from the smartphone to the server.

[2076] Input: User-entered content and target information they want to create buzz about.

[2077] Output: Information data sent to the server.

[2078] Step 2:

[2079] Retrieve information from a database.

[2080] Specific behavior:

[2081] The server receives the information sent by the user and then executes a database query to retrieve posts related to the specified target information from a database of past buzz posts. The query, including the filtering criteria, is sent to the database.

[2082] Input: User-specified target information.

[2083] Output: Relevant buzz post data retrieved from the database.

[2084] Step 3:

[2085] Preprocess the acquired post data.

[2086] Specific behavior:

[2087] The server preprocesses the submitted data, removing noise, stop words, and normalizing the text to make it easier to analyze. This process uses natural language processing technology.

[2088] Input: Post data retrieved from the database.

[2089] Output: Preprocessed text data.

[2090] Step 4:

[2091] Perform text analysis on the preprocessed data.

[2092] Specific behavior:

[2093] The server applies natural language processing (NLP) to the preprocessed data, performing morphological analysis to extract key keywords, phrases, and sentiment (positive, negative, neutral). NLP libraries (e.g., spaCy, NLTK) are used.

[2094] Input: Preprocessed text data.

[2095] Output: Parsed keywords, phrases and sentiment information.

[2096] Step 5:

[2097] The analysis results are input into a machine learning model to generate optimal post content.

[2098] Specific behavior:

[2099] The server inputs the data analyzed by natural language processing into a machine learning model. The machine learning model predicts the optimal posting pattern for the target based on the results learned from past buzz post data. A machine learning framework (e.g., TensorFlow, PyTorch) is used here.

[2100] Input: Parsed data (keywords, phrases, sentiment information).

[2101] Output: Generated post content, suggested keywords and hashtags.

[2102] Step 6:

[2103] Analyze user sentiment.

[2104] Specific behavior:

[2105] The server analyzes the user's emotions using an emotion engine. It collects and analyzes the user's facial expression and voice data using the smartphone's camera and microphone. The emotion engine classifies the user's emotional state (positive, negative, neutral).

[2106] Input: User's facial expression data, voice data.

[2107] Output: The user's emotional state (positive, negative, neutral).

[2108] Step 7:

[2109] Recommendations are generated and presented to the user.

[2110] Specific behavior:

[2111] The server generates and adjusts optimal post content based on the user's emotional state. It combines the analysis results from the emotion engine with the output of the machine learning model to create templates for keywords, hashtags, post tone (emotion), and specific post content to use. The generated post content is presented to the user on their smartphone screen.

[2112] Input: Sentiment engine analysis results, machine learning model output.

[2113] Output: The post content, suggested keywords, and hashtags presented to the user.

[2114] This allows users to upload content to social media that best suits their emotional state, making it possible to create buzz efficiently and effectively.

[2115] 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.

[2116] 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,...

Claims

1. A means for receiving content and target information from a user that the user wants to create a buzz about; A means for retrieving related posts from a database that stores data on past buzz posts; A means for preprocessing the acquired post data and performing text analysis using natural language processing; A means to input the analysis results into a machine learning model to generate optimal post content, A means for presenting the generated post content and recommended keywords and hashtags to the user; A system including:

2. 2. The system according to claim 1, wherein preprocessing of the posted data acquired from the database includes noise removal, stop word removal, and normalization.

3. The system of claim 1, wherein the machine learning model learns from past buzz posts and predicts optimal posting patterns for a target.

Images