system

Abstract

We provide the system. [Solution] A means for collecting data from past promotional programs and sales performance data, A means of analyzing collected data to extract basic information and effective selling points of products for sale, A means for generating a production script based on the analyzed information, A means of generating multilingual and multicultural video content based on a script, A means of optimizing and distributing the generated video content in a format suitable for each viewing region, A system that includes this.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, the method including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, 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

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the modern global market, television shopping programs are still effective sales channels, but it is costly and time-consuming to produce programs corresponding to different languages and cultures. As a result, it is difficult to efficiently provide programs for specific regions or languages. Also, there is a need for a method to quickly and accurately provide product information and sales points that meet the needs of the market. Therefore, it is an issue to provide low-cost and quickly multilingual and multicultural television shopping programs and realize a product presentation that is easy for consumers to understand.

Means for Solving the Problems

[0005] To solve this problem, the present invention provides a system having the following configuration: means for collecting data from past promotional programs and sales performance data; means for analyzing the collected data and extracting basic information and effective selling points of the products being sold; means for generating production scripts based on the analyzed information; means for generating multilingual and multicultural video content based on the scripts; and means for optimizing the generated video content into a format suitable for each viewing region and distributing it. By using machine learning models for data analysis and generating scripts using natural language generation algorithms, this system can efficiently produce multilingual and multicultural television shopping programs and distribute them quickly and at low cost.

[0006] A "promotional program" is a television program broadcast with the purpose of selling or promoting a product or service.

[0007] "Data" refers to information expressed in numerical or textual form, which is processed and stored by computers and other devices.

[0008] "Sales performance data" refers to data that includes information such as actual sales, sales volume, and sales period for a specific product or service.

[0009] "Analysis" is a technical method for extracting and understanding information relevant to a specific purpose using collected data.

[0010] "Basic information" refers to fundamental information necessary for consumers to understand a product, such as its characteristics, usage instructions, and price.

[0011] A "selling point" is a feature or advantage of a product or service that is particularly appealing to customers.

[0012] A "production script" is a detailed description of the video and audio content, serving as a guideline for program production.

[0013] "Video content" refers to media files that include both video and audio, and which provide information and entertainment to viewers.

[0014] "Optimization" refers to adjusting or improving something so that it functions efficiently for a specific purpose (for example, adapting to a viewing environment).

[0015] A "machine learning model" is an algorithm that uses data to learn specific patterns or rules, and then automatically performs tasks such as prediction and classification.

[0016] A "natural language generation algorithm" is a technology that allows computers to understand human language and generate new texts based on that understanding.

[0017] "Multilingual and multicultural support" refers to providing content adapted to different regions and cultural spheres in multiple languages. [Brief explanation of the drawing]

[0018] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8]It is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] It shows an emotion map to which a plurality of emotions are mapped. [Figure 10] It shows an emotion map to which a plurality of emotions are mapped. [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Example 2 when the emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when the emotion engine is combined.

Mode for Carrying Out the Invention

[0019] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0020] First, the terms used in the following description will be explained.

[0021] In the following embodiments, the signed processor (hereinafter simply referred to as "processor") may be one arithmetic unit or a combination of a plurality of arithmetic units. Also, the processor may be one type of arithmetic unit or a combination of a plurality of types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0022] In the following embodiments, signed RAM (Random Access Memory) is a memory that temporarily stores information and is used as work memory by the processor.

[0023] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0024] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0025] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0026] [First Embodiment]

[0027] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0028] As shown in Figure 1, the 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.

[0029] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0030] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0031] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and 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.

[0032] 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 perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0033] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0034] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0035] As shown in Figure 2, in the data processing device 12, a specific processing 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" related to the technology of this 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 according to the specific processing program 56 executed on the RAM 30.

[0036] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0037] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0038] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0039] The system according to the present invention collects and analyzes data from past promotional programs and sales performance data, extracts basic information and effective selling points of products, generates production scripts based on this information, and ultimately generates and distributes multilingual and multicultural video content. A detailed embodiment of this system is described below.

[0040] Data collection

[0041] The server connects to databases of past promotional programs and sales performance data to collect necessary data. For example, it retrieves recorded data of past TV shopping programs for a specific product, as well as data such as sales volume, sales period, and viewer feedback.

[0042] Data analysis

[0043] The server analyzes the collected data. First, it cleans the data, filling in missing data and removing outliers. Next, it inputs this cleaned data into a machine learning model to identify factors related to sales performance. For example, the analysis might reveal that certain time periods are advantageous for sales, or that the characteristics of certain products contribute to high sales performance.

[0044] Extracting selling points

[0045] Based on the analysis results, the server automatically extracts basic information about each product (e.g., product characteristics, usage instructions, price) and effective selling points (e.g., features that result in high customer retention rates). This information is then used to generate subsequent scripts.

[0046] Script generation

[0047] The server generates a production script based on the extracted basic information and selling points. In this process, a natural language generation (NLG) algorithm is used to automatically generate text describing the characteristics and benefits of each product. For example, a script highlighting the features of a new cooking utensil might be generated.

[0048] Video content generation

[0049] The server provides the generated script to the video generation AI, which uses AI Cast to create multilingual and multicultural video content. For example, videos in multiple languages, such as English, Spanish, and French, are created simultaneously. AI Cast adds narration and captions corresponding to each language and synthesizes the video and audio.

[0050] Optimization and delivery

[0051] The server optimizes the generated video content into a format suitable for the environment of each viewing region and uploads it to the relevant television station or online distribution platform. For example, optimizations may be made to provide high-resolution video in one region and low-resolution video in another.

[0052] View and purchase

[0053] Users select and watch shopping programs available via television or the internet. For example, users can use a remote control or mouse to choose a program featuring products that interest them. If a user is interested in a product, they can purchase it through telephone orders or online shopping sites.

[0054] Specific example:

[0055] The server collects and analyzes TV shopping program data for a specific detergent over the past six months and discovers that sales are highest during a particular time slot. Based on this information, the server extracts appropriate selling points (e.g., "Powerful cleaning effect with a small amount") and generates a script that emphasizes the detergent's characteristics. Furthermore, based on this information, it generates three video content pieces in English, Japanese, and Chinese, and distributes them optimized for each market.

[0056] Through the steps described above, the system according to the present invention can efficiently produce multilingual and multicultural television shopping programs and distribute them quickly and at low cost.

[0057] The following describes the processing flow.

[0058] Step 1:

[0059] The server connects to databases of past promotional programs and sales performance data to collect necessary data. Specifically, it retrieves recordings of past TV shopping programs, sales figures, sales periods, viewer feedback, and more.

[0060] Step 2:

[0061] The server cleans the collected data. Specifically, it performs tasks such as filling in missing data, removing outliers, and normalizing the data. For example, it standardizes the format of sales performance data and converts it into a format suitable for analysis algorithms.

[0062] Step 3:

[0063] The server inputs the cleaned data into a machine learning model to identify factors associated with sales performance. Specifically, it uses algorithms such as regression analysis, clustering, and decision trees to analyze how specific time periods or product characteristics influence sales.

[0064] Step 4:

[0065] The server extracts basic information and effective selling points for each product based on the analysis results of a machine learning model. Specifically, it uses text mining techniques to extract keywords from product reviews and sales history to identify features that are effective for sales.

[0066] Step 5:

[0067] The server generates a production script based on the extracted basic information and selling points. Specifically, it uses a natural language generation (NLG) algorithm to generate text that explains the product's characteristics and benefits.

[0068] Step 6:

[0069] The server provides the generated script to the video generation AI, which then uses AI Cast to create multilingual and multicultural video content. Specifically, AI Cast adds narration and captions corresponding to each language and synthesizes the video and audio.

[0070] Step 7:

[0071] The server optimizes the generated video content for each viewing region. Specifically, this involves changing the video resolution, converting file formats, and adding subtitles.

[0072] Step 8:

[0073] The server uploads optimized video content to the relevant television stations and online distribution platforms. Specifically, it distributes the content using FTP or APIs.

[0074] Step 9:

[0075] Users select and watch shopping programs available via television or the internet. Specifically, they select programs using a remote control or mouse.

[0076] Step 10:

[0077] When a user finds a product they are interested in, they purchase it. Specifically, they place an order via telephone or online shopping site.

[0078] Through the steps described above, the system according to the present invention can effectively produce and distribute multilingual and multicultural television shopping programs.

[0079] (Example 1)

[0080] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0081] In today's consumer market, to improve the efficiency of sales promotion, it is necessary to effectively utilize past promotional data and sales performance data to provide rapid and effective marketing tools. However, conventional systems have struggled to automate all processes from data collection, cleaning, and analysis to script generation, video creation, and distribution, as well as to efficiently produce and distribute multilingual and multicultural video content.

[0082] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0083] In this invention, the server includes means for collecting data from past promotional programs and sales performance data; means for cleaning, supplementing, and analyzing the collected data; means for extracting basic information and effective selling points of products from the analyzed data; means for generating production scripts based on the extracted information; means for generating multilingual and multicultural video content based on the scripts; and means for optimizing the generated video content into a format suitable for each viewing region and distributing it. This enables the efficient production and distribution of multilingual and multicultural video content.

[0084] "Promotional program data" refers to recorded data and related information of programs that were previously broadcast for the purpose of sales promotion.

[0085] "Sales performance data" refers to data related to actual sales activities, such as the quantity of products sold, the sales period, and viewer feedback.

[0086] "Cleaning" refers to the process of filling in missing values ​​and removing outliers from data.

[0087] A "machine learning model" refers to an algorithm or statistical model used for data analysis, designed to learn patterns and regularities from data.

[0088] A "selling point" refers to a product's characteristics, advantages, or effective elements for sales promotion.

[0089] A "natural language generation algorithm" is an algorithm that generates sentences in natural language based on extracted information.

[0090] "Multilingual and multicultural video content" refers to content that can generate video material that supports multiple languages ​​and cultures.

[0091] "Optimization" refers to the process of adjusting generated video content to suit the environment and requirements of each viewing region.

[0092] "Distribution" refers to the act of delivering generated video content to viewers online or through television stations.

[0093] The system according to the present invention collects and analyzes data from past promotional programs and sales performance data, extracts basic information and effective selling points of products, generates production scripts based on this information, and ultimately generates and distributes multilingual and multicultural video content. A detailed embodiment of this system is described below.

[0094] Data collection

[0095] The server connects to databases of past promotional programs and sales performance data to collect necessary data about a specific product. This includes recording data, sales volume, sales period, and viewer feedback. Specifically, the server uses SQL queries to retrieve the information.

[0096] Data analysis

[0097] The server cleans the collected data, imputing missing data and removing outliers. This cleaned data is then fed into a machine learning model to identify factors associated with sales performance. The Scikit-learn library in Python is used as the machine learning model.

[0098] Extracting selling points

[0099] Based on the analysis results, the server automatically extracts basic information about each product (e.g., product characteristics, usage instructions, price) and effective selling points (e.g., "powerful cleaning effect with a small amount").

[0100] Script generation

[0101] The server generates a production script using a natural language generation (NLG) algorithm based on the extracted basic information and selling points. Specifically, it uses OpenAI's GPT-3 to automatically generate text that describes the product's characteristics and advantages.

[0102] Video content generation

[0103] The server provides the generated script to the video generation AI, which then uses AI casting to create multilingual and multicultural video content. For example, it adds narration and captions in multiple languages, such as English, Spanish, and Japanese, and synthesizes the video and audio.

[0104] Optimization and delivery

[0105] The server optimizes and delivers the generated video content to the environment of each viewing region. Specifically, optimizations are made such as providing high-resolution video to some regions and low-resolution video to others. The optimized video is then uploaded to the relevant television stations and online distribution platforms.

[0106] View and purchase

[0107] Users can select and watch shopping programs delivered via television or the internet. For example, they can use a remote control or mouse to choose a program featuring products they are interested in, and then purchase the products by phone or through an online shopping site after watching the program.

[0108] Specific example

[0109] The server collected and analyzed TV shopping program data for a specific detergent over the past six months, revealing the highest sales volume during a particular time slot. Based on this information, it extracted the selling point "powerful cleaning effect with a small amount" and generated a script that emphasized the detergent's characteristics. Furthermore, based on this information, it generated three video content pieces in English, Japanese, and Chinese, and distributed them optimized for each market.

[0110] Through the steps described above, the system according to the present invention can efficiently produce multilingual and multicultural television shopping programs and distribute them quickly and at low cost.

[0111] Example of a prompt

[0112] Analyze TV shopping program data for a specific detergent from the past six months and extract its selling points. Then, generate scripts in English, Japanese, and Chinese based on those selling points. Please emphasize the detergent's characteristics and effects.

[0113] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0114] Step 1:

[0115] Data collection

[0116] The server connects to a database of promotional programs and a database of sales performance, collecting data such as recordings of past promotional programs, sales figures, sales periods, and viewer feedback.

[0117] Input: Information of the target database

[0118] Data processing: Extracting data using SQL queries.

[0119] Output: Collected dataset

[0120] Specific operation: The server executes an SQL query to retrieve video recording data and sales performance data for a specific product for the past six months.

[0121] Step 2:

[0122] Data cleaning

[0123] The server cleans the collected data, fills in missing data, and removes outliers.

[0124] Input: Collected dataset

[0125] Data processing: Imputation of missing data (imputation with mean values), removal of outliers.

[0126] Output: Cleaned dataset

[0127] Specific operation: Run a Python script to remove abnormal sales quantity data from the dataset and impute missing values ​​with the mean.

[0128] Step 3:

[0129] Data analysis

[0130] The server inputs the cleaned data into a machine learning model to identify factors associated with sales performance.

[0131] Input: Cleaned dataset

[0132] Data processing: Input data into a machine learning model and extract important features.

[0133] Output: Sales performance and related factors (specific time periods and product characteristics)

[0134] Specific operation: The server uses the Python Scikit-learn library to analyze specific time periods and product characteristics that contribute to sales performance using a model.

[0135] Step 4:

[0136] Extracting selling points

[0137] Based on the analysis results, the server automatically extracts basic product information and effective selling points.

[0138] Input: Sales performance and related factors

[0139] Data processing: Extraction of basic information, identification of effective selling points.

[0140] Output: Basic product information and selling points

[0141] Specific operation: Based on the extracted information, the server automatically identifies and extracts selling points such as "powerful cleaning effect with a small amount."

[0142] Step 5:

[0143] Script generation

[0144] The server generates a production script using a natural language generation (NLG) algorithm based on the extracted basic information and selling points.

[0145] Input: Basic product information and selling points

[0146] Data processing: Text generation using NLG algorithm

[0147] Output: Completed script

[0148] Specific operation: The server uses OpenAI's GPT-3 to generate a script that says, "This detergent has a powerful cleaning effect even in small amounts."

[0149] Step 6:

[0150] Video content generation

[0151] The server provides the generated script to the video generation AI, which then uses AI casting to generate multilingual and multicultural video content.

[0152] Input: Completed script

[0153] Data processing: Video and audio synthesis using AI casting.

[0154] Output: Multilingual and multicultural video content

[0155] Specific operation: The server generates videos based on the English, Japanese, and Chinese scripts, adding narration and captions in each language.

[0156] Step 7:

[0157] Optimization and delivery

[0158] The server optimizes the generated video content into a format suitable for the environment of each viewing region and then delivers it.

[0159] Input: Multilingual and multicultural video content

[0160] Data processing: Optimized for resolution and format suitable for each viewing region.

[0161] Output: Optimized video content

[0162] Specific operation: The server optimizes high-resolution videos for North America and low-resolution videos for regions with different internet speeds, and then uploads them to the distribution platform.

[0163] Step 8:

[0164] View and purchase

[0165] Users select shopping programs via television or the internet and then purchase products using telephone orders or online shopping after watching them.

[0166] Input: Optimized video content

[0167] Data processing: None

[0168] Output: Purchased items

[0169] Specific operation: The user uses the remote control to select a program about "special detergents," and after watching it, places a phone order or purchases the product from an online shopping site.

[0170] (Application Example 1)

[0171] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0172] Current advertising campaigns lack effective ways to leverage historical data and optimize in real time. Furthermore, generating and distributing multilingual video ads tailored to different regions and cultures is time-consuming and requires significant resources and time. In addition, there's a lack of mechanisms to quickly incorporate user feedback into subsequent advertising campaigns. A system is needed to address these challenges and maximize advertising effectiveness.

[0173] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0174] In this invention, the server includes means for collecting data from past promotional programs and sales performance data; means for analyzing the collected data and extracting basic information and effective selling points of the products being sold; means for generating production scripts based on the analyzed information; means for generating multilingual and multicultural video content based on the scripts; means for optimizing the generated video content into a format suitable for each viewing region and distributing it; means for collecting data from past advertising campaigns via smart devices and analyzing it in real time; means for automatically extracting effective advertising elements to generate advertising scripts; means for automatically generating multilingual video advertisements based on the generated advertising scripts; and means for collecting user feedback and using it to optimize the next advertising campaign. This makes it possible to effectively utilize past advertising data, generate and distribute optimized multilingual video advertisements in real time, and quickly reflect feedback.

[0175] "Data from past promotional programs" refers to information such as video, audio, and text related to promotions that were broadcast in the past.

[0176] "Sales performance data" refers to data related to the actual sales performance of a product, such as sales volume, sales period, and sales figures.

[0177] A "smart device" refers to a portable device with internet connectivity, such as a smartphone, smart glasses, or tablet.

[0178] "Real-time analysis" means performing analysis immediately the moment data is collected.

[0179] A "production script" is a text document that provides instructions regarding the content and flow of a video.

[0180] "Multilingual and multicultural video content" refers to videos produced in formats adapted to multiple languages ​​and cultures.

[0181] An "ad script" is text used to explain the content and selling points of an advertising video.

[0182] "Past advertising campaign data" refers to information such as viewing history, responses, and click-through rates of advertisements that were previously run.

[0183] "Optimizing and distributing content in a format suitable for each viewing region" means converting video content to a quality and format appropriate for each region and uploading it to the distribution platform.

[0184] "Automatically extracting effective advertising elements" means that the system automatically identifies and extracts elements (e.g., catchphrases and colors) that maximize the effectiveness of an advertisement.

[0185] "Collecting user feedback and using it to optimize the next advertising campaign" means collecting user reactions and evaluations as data and incorporating them into the production of future advertisements.

[0186] The system according to the present invention aims to optimize advertising campaigns and is configured as follows.

[0187] Data collection

[0188] The server first collects data from past promotional programs and sales performance data. Simultaneously, it also collects data such as viewing history and click-through rates of past advertising campaigns via smart devices.

[0189] Data Analysis

[0190] The collected data is analyzed on the server. Machine learning models are used for data analysis to identify factors that maximize advertising effectiveness. Specifically, algorithms such as random forests are used to analyze factors related to sales performance.

[0191] Extracting selling points

[0192] Based on the analysis results, the server automatically extracts basic information and effective selling points for each product. During this process, effective advertising elements such as colors and catchphrases are identified.

[0193] Script generation

[0194] The server generates advertising scripts based on extracted basic information and selling points. Using natural language generation (NLG) algorithms, it generates text that effectively explains the characteristics and benefits of each product. These scripts are generated in multiple languages ​​for multilingual support.

[0195] Video content generation

[0196] Based on the generated ad script, the server automatically creates multilingual video ads. Video editing libraries such as moviepy are used for video generation. The generated videos are saved in a format adapted to each viewing region.

[0197] Optimization and delivery

[0198] The server optimizes the generated video content into a format suitable for the environment of each viewing region and uploads it to the video distribution platform. For example, it adjusts high-resolution and low-resolution videos for each region.

[0199] Gathering feedback and optimizing future ads

[0200] Users view advertising campaigns and provide feedback. The server collects this feedback in real time and uses it to optimize the next advertising campaign.

[0201] Specific example

[0202] For example, by analyzing data from TV shopping programs that showed high effectiveness during specific time slots, it was found that certain colors (e.g., red) had a high click-through rate. Based on this information, a script and video featuring product introductions with a predominantly red background were generated and distributed in multiple languages.

[0203] Example of a prompt

[0204] "Please explain in detail how to extract the most effective advertising selling points based on past advertising campaign data and sales performance data, and how to generate multilingual advertising content. In particular, please explain the specific analysis procedures and generation process, focusing on the differences in effectiveness based on color and time of day."

[0205] The system according to the present invention makes it possible to effectively utilize past advertising data, generate and deliver optimized multilingual video advertisements in real time, and quickly incorporate feedback.

[0206] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0207] Step 1:

[0208] The server collects data from past promotional programs and sales performance data. Specifically, it retrieves recorded data of TV shopping programs, sales volume, sales period, and viewer feedback from the database. The data collected also includes viewing history and click-through rates of past advertising campaigns. The input is data from the database, and the output is the collected data to be analyzed.

[0209] Step 2:

[0210] The server analyzes the collected data. The data is first cleaned, including the imputation of missing data and the removal of outliers. Next, this cleaned data is input into a machine learning model (e.g., a random forest) to identify factors associated with sales performance. The input is the cleaned data, and the output is the factors associated with sales performance.

[0211] Step 3:

[0212] The server automatically extracts basic information and effective selling points for each product based on the analysis results. Specifically, it uses a machine learning model's importance ranking to identify features with high impact. The input is the analysis results, and the output is the extracted basic information and selling points.

[0213] Step 4:

[0214] The server generates advertising scripts based on the extracted basic information and selling points. It automatically generates text describing the characteristics and benefits of each product using a natural language generation (NLG) algorithm. The input is basic information and selling points, and the output is the generated script.

[0215] Step 5:

[0216] The server generates multilingual video ads based on the generated script. Video editing libraries such as moviepy are used for video generation, and narration and captions are added based on the script. The input is the generated script, and the output is the video ad.

[0217] Step 6:

[0218] The server optimizes the generated video content to a format suitable for the environment of each viewing region. For example, it might provide high-resolution video in one region and low-resolution video in another. The input is a video advertisement, and the output is the optimized video content.

[0219] Step 7:

[0220] The server uploads optimized video content to video distribution platforms. It optimizes the video to match the different specifications of each distribution platform and then uploads it. The input is the optimized video content, and the output is the completion of the upload to each platform.

[0221] Step 8:

[0222] Users view advertising campaigns and provide feedback. For example, they might watch an ad on a smart device and enter ratings and comments within the application. The input is the user's feedback, and the output is the collected feedback data.

[0223] Step 9:

[0224] The server analyzes user feedback and uses it to optimize the next advertising campaign. It analyzes feedback data in real time and incorporates it into the generation of future ad scripts and videos. The input is feedback data, and the output is ideas for improving the next ad campaign.

[0225] This series of processing steps allows for the effective use of past advertising data, the real-time generation and delivery of optimized, multilingual video ads, and the rapid incorporation of feedback.

[0226] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0227] The system according to the present invention collects and analyzes data from past promotional programs and sales performance data, extracts basic information and effective selling points of products, generates production scripts based on this information, and ultimately generates and distributes multilingual and multicultural video content. Furthermore, this system incorporates an emotion engine that recognizes user emotions and customizes content based on the recognized emotions. A detailed embodiment of this system is described below.

[0228] Data collection

[0229] The server connects to databases of past promotional programs and sales performance data to collect necessary data. For example, it retrieves recorded data of past TV shopping programs, as well as data such as sales volume, sales period, and viewer feedback.

[0230] Data analysis

[0231] The server cleans the collected data. Specifically, it imputes missing data, removes outliers, and normalizes the data. Next, this cleaned data is input into a machine learning model to identify factors related to sales performance. For example, the analysis may reveal that certain time periods are advantageous for sales, or that the characteristics of specific products influence sales performance.

[0232] Extracting selling points

[0233] Based on the analysis results, the server automatically extracts basic information about each product (e.g., product characteristics, usage instructions, price) and effective selling points (e.g., features that result in high customer retention rates). This information is then used to generate subsequent scripts.

[0234] Script generation

[0235] The server generates a production script based on the extracted basic information and selling points. In this process, a natural language generation (NLG) algorithm is used to automatically generate text describing the product's characteristics and benefits. For example, a script highlighting the features of a new cooking utensil might be generated.

[0236] Video content generation

[0237] The server provides the generated script to the video generation AI, which then uses AI Cast to create multilingual and multicultural video content. Specifically, videos in multiple languages, such as English, Spanish, and French, are created simultaneously. AI Cast adds narration and captions corresponding to each language and synthesizes the video and audio.

[0238] Optimization and delivery

[0239] The server optimizes the generated video content into a format suitable for the environment of each viewing region and uploads it to the relevant television station or online distribution platform. For example, optimizations may be made to provide high-resolution video in one region and low-resolution video in another.

[0240] Customization using an emotion engine

[0241] The server uses an emotion engine to recognize the emotions of users viewing the delivered content. Specifically, it analyzes the user's facial expressions, voice tone, and body language in real time to identify the user's emotional state. For example, it uses cameras and microphones to detect the user's smiles and reactions indicating interest.

[0242] Dynamic content adjustment

[0243] The server dynamically adjusts the product presentation method and script based on the recognized user's emotions. For example, if the user shows interest, it adds a detailed description of the product; conversely, if the user shows little interest, it switches to a different product.

[0244] View and purchase

[0245] Users select and watch shopping programs available via television or the internet. For example, users can use a remote control or mouse to choose a program featuring products that interest them. If a user is interested in a product, they can purchase it through telephone orders or online shopping sites.

[0246] Specific example:

[0247] Suppose a server collects and analyzes TV shopping program data for a specific detergent over the past six months and discovers that sales are highest during a particular time slot. Based on this information, the server extracts appropriate selling points (e.g., "Powerful cleaning effect with a small amount") and generates a script that emphasizes the detergent's characteristics. Furthermore, based on this information, it generates three video content pieces in English, Japanese, and Chinese, and distributes them optimized for each market. In doing so, it uses an emotion engine to recognize user sentiment, adding detailed explanations if the user shows interest, and switching to other products or approaches if interest is low, thereby achieving even more effective sales promotion.

[0248] Through the steps described above, the system according to the present invention can efficiently produce and distribute multilingual and multicultural television shopping programs and provide customized content based on the user's emotions.

[0249] The following describes the processing flow.

[0250] Step 1:

[0251] The server collects past promotional program data and sales performance data. Specifically, it connects to a database to retrieve recordings of past TV shopping programs, sales figures, sales periods, viewer feedback, and other information.

[0252] Step 2:

[0253] The server cleans the collected data. Specifically, it performs tasks such as filling in missing data, removing outliers, and normalizing the data. For example, it standardizes the format of sales performance data and converts it into a format suitable for analysis algorithms.

[0254] Step 3:

[0255] The server inputs the cleaned data into a machine learning model to identify factors associated with sales performance. Specifically, it uses algorithms such as regression analysis, clustering, and decision trees to analyze how specific time periods or product characteristics influence sales.

[0256] Step 4:

[0257] The server extracts basic information and effective selling points for each product based on the analysis results of a machine learning model. Specifically, it uses text mining techniques to extract keywords from product reviews and sales history to identify features that are effective for sales.

[0258] Step 5:

[0259] The server generates a production script based on the extracted basic information and selling points. In this process, a natural language generation (NLG) algorithm is used to generate text that explains the product's characteristics and benefits.

[0260] Step 6:

[0261] The server provides the generated script to the video generation AI, which then uses AI Cast to create multilingual and multicultural video content. Specifically, it creates videos in multiple languages, such as English, Spanish, and French. AI Cast adds narration and captions corresponding to each language and synthesizes the video and audio.

[0262] Step 7:

[0263] The server optimizes the generated video content to a format suitable for the environment of each viewing region. Specifically, this involves changing the video resolution, converting file formats, and adding subtitles.

[0264] Step 8:

[0265] The server uploads optimized video content to the relevant television stations and online distribution platforms. Specifically, it distributes the content using FTP or APIs.

[0266] Step 9:

[0267] Users select and watch shopping programs available via television or the internet. For example, users can use a remote control or mouse to choose a program that features products they are interested in.

[0268] Step 10:

[0269] The emotion engine recognizes the emotions of users watching the streamed content. Specifically, it analyzes the user's facial expressions, voice tone, and body language in real time to identify the user's emotional state. For example, it can analyze emotions based on the user's camera footage and microphone recordings.

[0270] Step 11:

[0271] The server dynamically adjusts the product presentation method and script based on the recognized user's emotions. For example, if the user shows interest, it adds a detailed description of the product; conversely, if the user shows little interest, it switches to a different product.

[0272] Step 12:

[0273] When a user finds a product they are interested in, they purchase it. Specifically, they place an order via telephone or online shopping site.

[0274] Specific example:

[0275] The server collected and analyzed TV shopping program data for a specific detergent over the past six months and discovered that sales were highest during a particular time slot. Based on this information, the server extracted appropriate selling points (e.g., "Powerful cleaning effect with a small amount") and generated a script that emphasized the detergent's characteristics. Furthermore, it generated three video content versions—English, Japanese, and Chinese—and distributed them optimized for each market. Using an emotion engine, the system analyzes the viewer's emotions in real time, adding detailed explanations of the detergent if they show interest, and switching to other products if they show little interest, thereby achieving more effective sales promotion.

[0276] (Example 2)

[0277] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0278] In today's world, promotional activities are becoming increasingly diverse and complex. In particular, the creation and distribution of multilingual and multicultural content is difficult to manage effectively and efficiently using traditional methods. Furthermore, while there is a need to recognize audience emotions in real time and customize content accordingly, there is a lack of concrete technological means to achieve this. As a result, there are challenges in fully achieving improved sales efficiency and optimized user experience.

[0279] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0280] In this invention, the server includes means for collecting data on past promotional programs and sales performance data, means for cleaning, analyzing, and extracting basic information on the sold products and effective sales points from the collected data, means for generating a production script based on the analyzed information, means for generating multi-language and multi-cultural video content based on the script, means for optimizing and distributing the generated video content in a format suitable for each viewing region, and means for recognizing the emotions of viewing users and customizing the content to be distributed based on them. As a result, it becomes possible to efficiently and effectively produce and distribute multi-language and multi-cultural video content, and it becomes possible to improve sales efficiency and optimize the user experience through flexible content customization based on the emotions of viewers.

[0281] "Data on promotional programs" refers to information such as videos, audios, and texts produced for the purpose of selling products or promoting services.

[0282] "Sales performance data" refers to historical information on the sales of specific products or services, including data such as sales volume, period, revenue, and customer feedback.

[0283] "Cleaning" is a process of complementing missing data, deleting outliers, and normalizing data from a dataset.

[0284] "Analysis" is a process of identifying factors related to sales performance using techniques such as machine learning models for the collected data.

[0285] "Basic information" refers to basic information necessary to understand a product, such as its characteristics, usage method, and price.

[0286] "Sales point" refers to specific features or advantages of a product, which are important elements that appear attractive to customers.

[0287] A "production script" is a document or scenario created to explain the characteristics and benefits of a product.

[0288] "Multilingual and multicultural video content" refers to videos produced to be compatible with multiple languages ​​and different cultures, and includes narration and captions in each language.

[0289] "Optimization" is the process of converting generated video content into a format suitable for the environment of each viewing region.

[0290] "Means of distribution" refer to platforms and devices used to provide optimized video content to viewers.

[0291] "Recognizing emotions" is the process of analyzing the viewer's facial expressions, tone of voice, and body language in real time to identify the user's emotional state.

[0292] "Customizing" is the process of dynamically adjusting the way a product is presented and the script based on the emotions of the perceived audience.

[0293] The system according to the present invention collects and analyzes data from past promotional programs and sales performance data, extracts basic information and effective selling points of products, generates production scripts based on this information, and ultimately generates and distributes multilingual and multicultural video content. Furthermore, this system incorporates an emotion engine that recognizes user emotions and customizes content based on the recognized emotions. A detailed embodiment of this system is described below.

[0294] Data collection

[0295] The server uses a database management system to collect data from past promotional programs and sales performance data. Specifically, it uses SQL or NoSQL databases to retrieve data such as recordings of past TV shopping programs, sales volume, sales period, and viewer feedback.

[0296] Data analysis

[0297] The server uses data preprocessing tools (e.g., Python's Pandas library) to clean the collected data. Specifically, it performs data imputation, removal of outliers, and data normalization. Next, this cleaned data is input into a machine learning model (e.g., TENSORFLOW® or Scikit-learn) to identify factors related to sales performance. For example, the analysis may reveal that certain time periods are advantageous for sales, or that the characteristics of certain products influence sales performance.

[0298] Extracting selling points

[0299] Based on the analysis results, the server automatically extracts basic information about each product (e.g., product characteristics, usage instructions, price) and effective selling points (e.g., features that result in high customer retention rates). This information is then used to generate subsequent scripts.

[0300] Script generation

[0301] The server generates a production script based on the extracted basic information and selling points. This process uses natural language generation (NLG) algorithms (such as GPT-3) to automatically generate text describing the product's characteristics and benefits. For example, a script highlighting the features of a new cooking utensil might be generated.

[0302] Video content generation

[0303] The server provides the generated script to the video generation AI, which uses AI casts to generate multilingual and multicultural video content. Specifically, videos corresponding to multiple languages such as English, Spanish, French, etc. are created simultaneously. The AI casts add narrations and captions corresponding to each language and synthesize video and audio.

[0304] Optimization and Delivery

[0305] The server optimizes the generated video content into a format suitable for the environment of each viewing region and uploads it to relevant broadcasting stations and online distribution platforms. For example, optimizations such as providing high-resolution videos in a specific region and low-resolution videos in another region are carried out.

[0306] Customization by Emotion Engine

[0307] The server uses the emotion engine to recognize the emotions of the users watching the distributed content. Specifically, it analyzes the users' expressions, voice tones, and body languages in real time to identify the users' emotional states. For example, it detects reactions indicating the users' smiles and interests using cameras and microphones.

[0308] Dynamic Content Adjustment

[0309] The server dynamically adjusts the presentation method and script of the product based on the recognized emotions of the users. For example, when the user shows interest, it adds a detailed description of the product, and conversely, when the interest is low, it makes adjustments such as switching to other products.

[0310] Viewing and Purchase

[0311] Users select and watch shopping programs available via television or the internet. For example, users can use a remote control or mouse to choose a program featuring products that interest them. If a user is interested in a product, they can purchase it through telephone orders or online shopping sites.

[0312] Specific example

[0313] Suppose a server collects and analyzes TV shopping program data for a specific detergent over the past six months and discovers that sales are highest during a particular time slot. Based on this information, the server extracts appropriate selling points (e.g., "Powerful cleaning effect with a small amount") and generates a script that emphasizes the detergent's characteristics. Furthermore, based on this information, it generates three video content pieces in English, Japanese, and Chinese, and distributes them optimized for each market. In doing so, it uses an emotion engine to recognize user sentiment, adding detailed explanations if the user shows interest, and switching to other products or approaches if interest is low, thereby achieving even more effective sales promotion.

[0314] Example of a prompt

[0315] Collect TV shopping program data for specialty detergents from the past six months and identify the time slots with the highest sales. Next, extract the selling points and generate video scripts in English, Japanese, and Chinese. Use an emotion engine to adjust the content based on user reactions.

[0316] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0317] Step 1: Data Collection

[0318] The server uses a database management system to collect data from past promotional programs and sales performance data. This includes data such as recordings of past TV shopping programs, sales volume, sales period, and viewer feedback, using SQL or NoSQL databases. The input is database connection information, and the output is the collected raw data.

[0319] Step 2: Data Cleaning

[0320] The server uses data preprocessing tools (e.g., Python's Pandas library) to clean the raw data. Specifically, it performs data imputation, removal of outliers, and data normalization. The input is the raw data collected in step 1, and the output is the cleaned data.

[0321] Step 3: Data Analysis

[0322] The server inputs the cleaned data into a machine learning model (e.g., TensorFlow or Scikit-learn) to identify factors associated with sales performance. For example, the analysis might reveal that certain time periods are favorable for sales or that the characteristics of specific products influence sales performance. The input is the cleaned data, and the output is the analysis results (e.g., factors associated with sales performance).

[0323] Step 4: Extracting the selling points

[0324] The server extracts basic information and effective selling points for each product based on the analysis results. Specifically, it targets product characteristics, usage methods, pricing, and features that contribute to high customer retention rates. The input is the analysis results, and the output is the extracted basic information and selling points.

[0325] Step 5: Generate the script

[0326] The server generates a production script based on the extracted basic information and selling points. In this process, a natural language generation (NLG) algorithm (e.g., GPT-3) is used to automatically generate text describing the product's characteristics and benefits. The input is the basic information and selling points, and the output is the generated script.

[0327] Step 6: Generate video content

[0328] The server provides the generated script to the video generation AI, which then uses AI casting to generate multilingual and multicultural video content. Specifically, videos in multiple languages, such as English, Spanish, and French, are created simultaneously. The input is the generated script, and the output is multilingual video content.

[0329] Step 7: Optimization and Delivery

[0330] The server optimizes the generated video content into a format suitable for the environment of each viewing region and uploads it to the relevant broadcasters and online distribution platforms. For example, it may optimize the content by providing high-resolution video in one region and low-resolution video in another. The input is the generated video content, and the output is the optimized video file.

[0331] Step 8: Customization with the emotion engine

[0332] The server uses an emotion engine to recognize the emotions of users viewing the delivered content. Specifically, it analyzes the user's facial expressions, voice tone, and body language in real time to identify the user's emotional state. The input is real-time data of the user during viewing, and the output is the recognized emotional state.

[0333] Step 9: Dynamic Content Adjustment

[0334] The server dynamically adjusts the product presentation method and script based on the recognized user's emotions. For example, if the user shows interest, it adds a detailed description of the product; conversely, if the user shows little interest, it switches to another product. The input is the recognized emotional state, and the output is the adjusted content.

[0335] Step 10: View and Purchase

[0336] Users select and watch shopping programs available via television or the internet. For example, users can use a remote control or mouse to choose a program featuring products they are interested in. If a user is interested, they can purchase the product through telephone ordering or an online shopping site. The input is the user's viewing selection and emotional reaction, and the output is the action of purchasing the product.

[0337] (Application Example 2)

[0338] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0339] Current advertising systems lack the functionality to dynamically adjust ad content in response to the emotions of target users. This makes it difficult to maximize advertising effectiveness and capture user interest. Furthermore, automatically generating multilingual and multicultural video content and distributing it in formats suitable for each viewing region is not easy. To address these challenges, a system is needed that can extract effective selling points based on collected data and customize ad content according to user emotions.

[0340] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0341] In this invention, the server includes means for collecting data from past promotional programs and sales performance data; means for analyzing the collected data and extracting basic information and effective selling points of the products being sold; means for generating production scripts based on the analyzed information; means for generating multilingual and multicultural video content based on the scripts; means for optimizing the generated video content into a format suitable for each viewing region and distributing it; means for analyzing user emotions on a smart device; and means for dynamically adjusting advertising content based on the analyzed user emotions. This enables customized advertising that responds to user emotions, thereby maximizing advertising effectiveness.

[0342] A "promotional program" is video content intended to promote the sale of a product or service.

[0343] "Sales performance data" refers to data that includes the quantity of products sold in the past, the sales period, and customer feedback.

[0344] "Data analysis" is the process of organizing collected data and extracting information that is useful for sales.

[0345] "Basic information" refers to fundamental information such as the product's characteristics, usage instructions, and price.

[0346] An "effective selling point" is a feature or characteristic that is particularly effective in promoting the sale of a product.

[0347] A "production script" is a script or scenario used to generate video or advertising content.

[0348] "Multilingual and multicultural support" refers to the function of generating and providing content that is compatible with different languages ​​and cultures.

[0349] "Video content" refers to digital content that includes audio and video.

[0350] A "smart device" is an internet-connected device such as a smartphone, tablet, or smart glasses.

[0351] "User emotion" refers to the psychological state determined from the user's facial expressions, tone of voice, body language, etc.

[0352] "Dynamic adjustment" means changing the content in real time.

[0353] "Advertising content" refers to media content created for the purpose of promoting products or services.

[0354] The system for carrying out this invention is primarily executed by a server. The embodiments for carrying out the invention are described in detail below.

[0355] 1. Data Collection

[0356] The server collects data on past promotional programs and sales performance. This includes recordings of promotional programs, sales volume, sales period, and customer feedback data. The database records numerous promotional programs and their corresponding sales performance.

[0357] 2. Data Analysis

[0358] The server analyzes the collected data. First, it cleans the data, filling in missing data and removing outliers. Next, it uses the cleaned data to analyze factors related to sales performance using a machine learning model. This reveals things like specific time slots or product characteristics that are advantageous for sales.

[0359] 3. Extracting selling points

[0360] The server extracts basic information and effective selling points based on the analysis results. Basic information includes product characteristics, usage, and price. Effective selling points are product features deemed particularly effective for sales promotion.

[0361] 4. Script generation

[0362] The server generates a production script based on the extracted basic information and selling points. Using a natural language generation algorithm, it automatically generates text to describe the product's characteristics and benefits. For example, a script highlighting the features of a new cooking utensil might be generated here.

[0363] 5. Video content generation

[0364] The server generates multilingual and multicultural video content using the generated script. A generation AI model is used to add narration and captions corresponding to each language, and to synthesize the video and audio. For example, the same content can be generated in English, Spanish, French, and other languages.

[0365] 6. Distribution and Optimization

[0366] The server optimizes the generated video content into a format suitable for each viewing region and uploads it to television stations and online distribution platforms. For example, it might provide high-resolution video in certain regions and low-resolution video in others.

[0367] 7. Sentiment Analysis and Dynamic Content Adjustment

[0368] The system analyzes user emotions using smart devices (smartphones and smart glasses). It employs an emotion recognition engine that analyzes the user's facial expressions, voice tone, and body language in real time. Based on the detected emotions, the advertising content is dynamically adjusted. For example, if the user shows interest, a detailed description of the product is added; if interest is low, the system switches to a different product.

[0369] 8. Specific Examples

[0370] The server collects and analyzes promotional program data for a specific detergent over the past six months, discovering that sales are highest during certain time slots. Based on this information, the server extracts "powerful cleaning effect with a small amount" as a selling point and generates a script that emphasizes the detergent's characteristics. Video content is generated in English, Japanese, and Chinese, and distributed optimized for each market. When a user watches this ad on their smartphone and shows interest, a more detailed explanation is added; otherwise, the ad switches to the next product.

[0371] 9. Example of a prompt message

[0372] The next ad you'll show is for a new detergent. Use the following selling points to create a 30-second ad script:

[0373] A small amount provides powerful cleaning.

[0374] Made with environmentally friendly ingredients.

[0375] Effective against all types of dirt.

[0376] Thus, the system for implementing the present invention provides dynamic advertisements that can maximize user interest.

[0377] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0378] Step 1:

[0379] The server collects data from past promotional programs and sales performance data. Inputs include recordings of promotional programs, sales volume, sales period, and customer feedback data. This data is retrieved from a database and prepared for data analysis. The output is a list of the retrieved data.

[0380] Step 2:

[0381] The server analyzes the collected data. It uses collected promotional program data and sales performance data as input. First, it performs data cleaning, including imputing missing data and removing outliers. Next, a machine learning model analyzes factors related to sales performance, identifying the impact of time of day and product characteristics on sales. The output is the analysis results, providing information on specific sales trends and effective selling points.

[0382] Step 3:

[0383] The server extracts basic information and effective selling points based on the analysis results. It uses the analyzed data as input. From the analyzed information, it identifies basic information such as product characteristics, usage, and price, as well as features advantageous for sales promotion. The output is a list of basic information and selling points.

[0384] Step 4:

[0385] The server generates a production script based on the extracted basic information and selling points. It uses a list of basic information and selling points as input. A natural language generation algorithm is used to automatically generate text describing the product's characteristics and benefits. The output is the generated script.

[0386] Step 5:

[0387] The server generates multilingual and multicultural video content using the generated script. It uses the generated script as input. Utilizing a generation AI model, it adds narration and captions corresponding to each language and synthesizes the video and audio. The output is video content compatible with multiple languages.

[0388] Step 6:

[0389] The server optimizes the generated video content into a format suitable for each viewing region and distributes it. It uses the generated video content as input. Technical adjustments, such as high or low resolution, are made to meet the needs of each viewing region before uploading it to television stations and online distribution platforms. The output is the optimized video content.

[0390] Step 7:

[0391] The device analyzes the user's emotions. It uses the user's camera video and audio data as input. Utilizing an emotion recognition engine, it performs facial expression and voice analysis to identify the user's emotional state. The output is the recognized user emotion information.

[0392] Step 8:

[0393] The server dynamically adjusts ad content based on the analyzed user's sentiment. It uses recognized user sentiment information and already generated ad content as input. For example, if the user shows interest, it adds a detailed description; if not, it switches to another product. The output is the adjusted ad content.

[0394] Step 9:

[0395] Users view tailored advertising content. Dynamically adjusted advertising content from the server is used as input, providing the user with the optimal advertising experience. The output is the user's viewing behavior and feedback.

[0396] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0397] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include those described above. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions shown by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0398] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0399] [Second Embodiment]

[0400] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0401] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0402] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0403] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0404] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0405] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0406] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0407] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0408] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0409] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0410] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0411] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. 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".

[0412] The system according to the present invention collects and analyzes data from past promotional programs and sales performance data, extracts basic information and effective selling points of products, generates production scripts based on this information, and ultimately generates and distributes multilingual and multicultural video content. A detailed embodiment of this system is described below.

[0413] Data collection

[0414] The server connects to databases of past promotional programs and sales performance data to collect necessary data. For example, it retrieves recorded data of past TV shopping programs for a specific product, as well as data such as sales volume, sales period, and viewer feedback.

[0415] Data analysis

[0416] The server analyzes the collected data. First, it cleans the data, filling in missing data and removing outliers. Next, it inputs this cleaned data into a machine learning model to identify factors related to sales performance. For example, the analysis might reveal that certain time periods are advantageous for sales, or that the characteristics of certain products contribute to high sales performance.

[0417] Extracting selling points

[0418] Based on the analysis results, the server automatically extracts basic information about each product (e.g., product characteristics, usage instructions, price) and effective selling points (e.g., features that result in high customer retention rates). This information is then used to generate subsequent scripts.

[0419] Script generation

[0420] The server generates a production script based on the extracted basic information and selling points. In this process, a natural language generation (NLG) algorithm is used to automatically generate text describing the characteristics and benefits of each product. For example, a script highlighting the features of a new cooking utensil might be generated.

[0421] Video content generation

[0422] The server provides the generated script to the video generation AI, which uses AI Cast to create multilingual and multicultural video content. For example, videos in multiple languages, such as English, Spanish, and French, are created simultaneously. AI Cast adds narration and captions corresponding to each language and synthesizes the video and audio.

[0423] Optimization and delivery

[0424] The server optimizes the generated video content into a format suitable for the environment of each viewing region and uploads it to the relevant television station or online distribution platform. For example, optimizations may be made to provide high-resolution video in one region and low-resolution video in another.

[0425] View and purchase

[0426] Users select and watch shopping programs available via television or the internet. For example, users can use a remote control or mouse to choose a program featuring products that interest them. If a user is interested in a product, they can purchase it through telephone orders or online shopping sites.

[0427] Specific example:

[0428] The server collects and analyzes TV shopping program data for a specific detergent over the past six months and discovers that sales are highest during a particular time slot. Based on this information, the server extracts appropriate selling points (e.g., "Powerful cleaning effect with a small amount") and generates a script that emphasizes the detergent's characteristics. Furthermore, based on this information, it generates three video content pieces in English, Japanese, and Chinese, and distributes them optimized for each market.

[0429] Through the steps described above, the system according to the present invention can efficiently produce multilingual and multicultural television shopping programs and distribute them quickly and at low cost.

[0430] The following describes the processing flow.

[0431] Step 1:

[0432] The server connects to databases of past promotional programs and sales performance data to collect necessary data. Specifically, it retrieves recordings of past TV shopping programs, sales figures, sales periods, viewer feedback, and more.

[0433] Step 2:

[0434] The server cleans the collected data. Specifically, it performs tasks such as filling in missing data, removing outliers, and normalizing the data. For example, it standardizes the format of sales performance data and converts it into a format suitable for analysis algorithms.

[0435] Step 3:

[0436] The server inputs the cleaned data into a machine learning model to identify factors associated with sales performance. Specifically, it uses algorithms such as regression analysis, clustering, and decision trees to analyze how specific time periods or product characteristics influence sales.

[0437] Step 4:

[0438] The server extracts basic information and effective selling points for each product based on the analysis results of a machine learning model. Specifically, it uses text mining techniques to extract keywords from product reviews and sales history to identify features that are effective for sales.

[0439] Step 5:

[0440] The server generates a production script based on the extracted basic information and selling points. Specifically, it uses a natural language generation (NLG) algorithm to generate text that explains the product's characteristics and benefits.

[0441] Step 6:

[0442] The server provides the generated script to the video generation AI, which then uses AI Cast to create multilingual and multicultural video content. Specifically, AI Cast adds narration and captions corresponding to each language and synthesizes the video and audio.

[0443] Step 7:

[0444] The server optimizes the generated video content for each viewing region. Specifically, this involves changing the video resolution, converting file formats, and adding subtitles.

[0445] Step 8:

[0446] The server uploads optimized video content to the relevant television stations and online distribution platforms. Specifically, it distributes the content using FTP or APIs.

[0447] Step 9:

[0448] Users select and watch shopping programs available via television or the internet. Specifically, they select programs using a remote control or mouse.

[0449] Step 10:

[0450] When a user finds a product they are interested in, they purchase it. Specifically, they place an order via telephone or online shopping site.

[0451] Through the steps described above, the system according to the present invention can effectively produce and distribute multilingual and multicultural television shopping programs.

[0452] (Example 1)

[0453] Next, we will describe Example 1. 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."

[0454] In today's consumer market, to improve the efficiency of sales promotion, it is necessary to effectively utilize past promotional data and sales performance data to provide rapid and effective marketing tools. However, conventional systems have struggled to automate all processes from data collection, cleaning, and analysis to script generation, video creation, and distribution, as well as to efficiently produce and distribute multilingual and multicultural video content.

[0455] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0456] In this invention, the server includes means for collecting data from past promotional programs and sales performance data; means for cleaning, supplementing, and analyzing the collected data; means for extracting basic information and effective selling points of products from the analyzed data; means for generating production scripts based on the extracted information; means for generating multilingual and multicultural video content based on the scripts; and means for optimizing the generated video content into a format suitable for each viewing region and distributing it. This enables the efficient production and distribution of multilingual and multicultural video content.

[0457] "Promotional program data" refers to recorded data and related information of programs that were previously broadcast for the purpose of sales promotion.

[0458] "Sales performance data" refers to data related to actual sales activities, such as the quantity of products sold, the sales period, and viewer feedback.

[0459] "Cleaning" refers to the process of filling in missing values ​​and removing outliers from data.

[0460] A "machine learning model" refers to an algorithm or statistical model used for data analysis, designed to learn patterns and regularities from data.

[0461] A "selling point" refers to a product's characteristics, advantages, or effective elements for sales promotion.

[0462] A "natural language generation algorithm" is an algorithm that generates sentences in natural language based on extracted information.

[0463] "Multilingual and multicultural video content" refers to content that can generate video material that supports multiple languages ​​and cultures.

[0464] "Optimization" refers to the process of adjusting generated video content to suit the environment and requirements of each viewing region.

[0465] "Distribution" refers to the act of delivering generated video content to viewers online or through television stations.

[0466] The system according to the present invention collects and analyzes data from past promotional programs and sales performance data, extracts basic information and effective selling points of products, generates production scripts based on this information, and ultimately generates and distributes multilingual and multicultural video content. A detailed embodiment of this system is described below.

[0467] Data collection

[0468] The server connects to databases of past promotional programs and sales performance data to collect necessary data about a specific product. This includes recording data, sales volume, sales period, and viewer feedback. Specifically, the server uses SQL queries to retrieve the information.

[0469] Data analysis

[0470] The server cleans the collected data, imputing missing data and removing outliers. This cleaned data is then fed into a machine learning model to identify factors associated with sales performance. The Scikit-learn library in Python is used as the machine learning model.

[0471] Extracting selling points

[0472] Based on the analysis results, the server automatically extracts basic information about each product (e.g., product characteristics, usage instructions, price) and effective selling points (e.g., "powerful cleaning effect with a small amount").

[0473] Script generation

[0474] The server generates a production script using a natural language generation (NLG) algorithm based on the extracted basic information and selling points. Specifically, it uses OpenAI's GPT-3 to automatically generate text that describes the product's characteristics and advantages.

[0475] Video content generation

[0476] The server provides the generated script to the video generation AI, which then uses AI casting to create multilingual and multicultural video content. For example, it adds narration and captions in multiple languages, such as English, Spanish, and Japanese, and synthesizes the video and audio.

[0477] Optimization and delivery

[0478] The server optimizes and delivers the generated video content to the environment of each viewing region. Specifically, optimizations are made such as providing high-resolution video to some regions and low-resolution video to others. The optimized video is then uploaded to the relevant television stations and online distribution platforms.

[0479] View and purchase

[0480] Users can select and watch shopping programs delivered via television or the internet. For example, they can use a remote control or mouse to choose a program featuring products they are interested in, and then purchase the products by phone or through an online shopping site after watching the program.

[0481] Specific example

[0482] The server collected and analyzed TV shopping program data for a specific detergent over the past six months, revealing the highest sales volume during a particular time slot. Based on this information, it extracted the selling point "powerful cleaning effect with a small amount" and generated a script that emphasized the detergent's characteristics. Furthermore, based on this information, it generated three video content pieces in English, Japanese, and Chinese, and distributed them optimized for each market.

[0483] Through the steps described above, the system according to the present invention can efficiently produce multilingual and multicultural television shopping programs and distribute them quickly and at low cost.

[0484] Example of a prompt

[0485] Analyze TV shopping program data for a specific detergent from the past six months and extract its selling points. Then, generate scripts in English, Japanese, and Chinese based on those selling points. Please emphasize the detergent's characteristics and effects.

[0486] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0487] Step 1:

[0488] Data collection

[0489] The server connects to a database of promotional programs and a database of sales performance, collecting data such as recordings of past promotional programs, sales figures, sales periods, and viewer feedback.

[0490] Input: Information of the target database

[0491] Data processing: Extracting data using SQL queries.

[0492] Output: Collected dataset

[0493] Specific operation: The server executes an SQL query to retrieve video recording data and sales performance data for a specific product for the past six months.

[0494] Step 2:

[0495] Data cleaning

[0496] The server cleans the collected data, fills in missing data, and removes outliers.

[0497] Input: Collected dataset

[0498] Data processing: Imputation of missing data (imputation with mean values), removal of outliers.

[0499] Output: Cleaned dataset

[0500] Specific operation: Run a Python script to remove abnormal sales quantity data from the dataset and impute missing values ​​with the mean.

[0501] Step 3:

[0502] Data analysis

[0503] The server inputs the cleaned data into a machine learning model to identify factors associated with sales performance.

[0504] Input: Cleaned dataset

[0505] Data processing: Input data into a machine learning model and extract important features.

[0506] Output: Sales performance and related factors (specific time periods and product characteristics)

[0507] Specific operation: The server uses the Python Scikit-learn library to analyze specific time periods and product characteristics that contribute to sales performance using a model.

[0508] Step 4:

[0509] Extracting selling points

[0510] Based on the analysis results, the server automatically extracts basic product information and effective selling points.

[0511] Input: Sales performance and related factors

[0512] Data processing: Extraction of basic information, identification of effective selling points.

[0513] Output: Basic product information and selling points

[0514] Specific operation: Based on the extracted information, the server automatically identifies and extracts selling points such as "powerful cleaning effect with a small amount."

[0515] Step 5:

[0516] Script generation

[0517] The server generates a production script using a natural language generation (NLG) algorithm based on the extracted basic information and selling points.

[0518] Input: Basic product information and selling points

[0519] Data processing: Text generation using NLG algorithm

[0520] Output: Completed script

[0521] Specific operation: The server uses OpenAI's GPT-3 to generate a script that says, "This detergent has a powerful cleaning effect even in small amounts."

[0522] Step 6:

[0523] Video content generation

[0524] The server provides the generated script to the video generation AI, which then uses AI casting to generate multilingual and multicultural video content.

[0525] Input: Completed script

[0526] Data processing: Video and audio synthesis using AI casting.

[0527] Output: Multilingual and multicultural video content

[0528] Specific operation: The server generates videos based on the English, Japanese, and Chinese scripts, adding narration and captions in each language.

[0529] Step 7:

[0530] Optimization and delivery

[0531] The server optimizes the generated video content into a format suitable for the environment of each viewing region and then delivers it.

[0532] Input: Multilingual and multicultural video content

[0533] Data processing: Optimized for resolution and format suitable for each viewing region.

[0534] Output: Optimized video content

[0535] Specific operation: The server optimizes high-resolution videos for North America and low-resolution videos for regions with different internet speeds, and then uploads them to the distribution platform.

[0536] Step 8:

[0537] View and purchase

[0538] Users select shopping programs via television or the internet and then purchase products using telephone orders or online shopping after watching them.

[0539] Input: Optimized video content

[0540] Data processing: None

[0541] Output: Purchased items

[0542] Specific operation: The user uses the remote control to select a program about "special detergents," and after watching it, places a phone order or purchases the product from an online shopping site.

[0543] (Application Example 1)

[0544] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0545] Current advertising campaigns lack effective ways to leverage historical data and optimize in real time. Furthermore, generating and distributing multilingual video ads tailored to different regions and cultures is time-consuming and requires significant resources and time. In addition, there's a lack of mechanisms to quickly incorporate user feedback into subsequent advertising campaigns. A system is needed to address these challenges and maximize advertising effectiveness.

[0546] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0547] In this invention, the server includes means for collecting data from past promotional programs and sales performance data; means for analyzing the collected data and extracting basic information and effective selling points of the products being sold; means for generating production scripts based on the analyzed information; means for generating multilingual and multicultural video content based on the scripts; means for optimizing the generated video content into a format suitable for each viewing region and distributing it; means for collecting data from past advertising campaigns via smart devices and analyzing it in real time; means for automatically extracting effective advertising elements to generate advertising scripts; means for automatically generating multilingual video advertisements based on the generated advertising scripts; and means for collecting user feedback and using it to optimize the next advertising campaign. This makes it possible to effectively utilize past advertising data, generate and distribute optimized multilingual video advertisements in real time, and quickly reflect feedback.

[0548] "Data from past promotional programs" refers to information such as video, audio, and text related to promotions that were broadcast in the past.

[0549] "Sales performance data" refers to data related to the actual sales performance of a product, such as sales volume, sales period, and sales figures.

[0550] A "smart device" refers to a portable device with internet connectivity, such as a smartphone, smart glasses, or tablet.

[0551] "Real-time analysis" means performing analysis immediately the moment data is collected.

[0552] A "production script" is a text document that provides instructions regarding the content and flow of a video.

[0553] "Multilingual and multicultural video content" refers to videos produced in formats adapted to multiple languages ​​and cultures.

[0554] An "ad script" is text used to explain the content and selling points of an advertising video.

[0555] "Past advertising campaign data" refers to information such as viewing history, responses, and click-through rates of advertisements that were previously run.

[0556] "Optimizing and distributing content in a format suitable for each viewing region" means converting video content to a quality and format appropriate for each region and uploading it to the distribution platform.

[0557] "Automatically extracting effective advertising elements" means that the system automatically identifies and extracts elements (e.g., catchphrases and colors) that maximize the effectiveness of an advertisement.

[0558] "Collecting user feedback and using it to optimize the next advertising campaign" means collecting user reactions and evaluations as data and incorporating them into the production of future advertisements.

[0559] The system according to the present invention aims to optimize advertising campaigns and is configured as follows.

[0560] Data collection

[0561] The server first collects data from past promotional programs and sales performance data. Simultaneously, it also collects data such as viewing history and click-through rates of past advertising campaigns via smart devices.

[0562] Data Analysis

[0563] The collected data is analyzed on the server. Machine learning models are used for data analysis to identify factors that maximize advertising effectiveness. Specifically, algorithms such as random forests are used to analyze factors related to sales performance.

[0564] Extracting selling points

[0565] Based on the analysis results, the server automatically extracts basic information and effective selling points for each product. During this process, effective advertising elements such as colors and catchphrases are identified.

[0566] Script generation

[0567] The server generates advertising scripts based on extracted basic information and selling points. Using natural language generation (NLG) algorithms, it generates text that effectively explains the characteristics and benefits of each product. These scripts are generated in multiple languages ​​for multilingual support.

[0568] Video content generation

[0569] Based on the generated ad script, the server automatically creates multilingual video ads. Video editing libraries such as moviepy are used for video generation. The generated videos are saved in a format adapted to each viewing region.

[0570] Optimization and delivery

[0571] The server optimizes the generated video content into a format suitable for the environment of each viewing region and uploads it to the video distribution platform. For example, it adjusts high-resolution and low-resolution videos for each region.

[0572] Gathering feedback and optimizing future ads

[0573] Users view advertising campaigns and provide feedback. The server collects this feedback in real time and uses it to optimize the next advertising campaign.

[0574] Specific example

[0575] For example, by analyzing data from TV shopping programs that showed high effectiveness during specific time slots, it was found that certain colors (e.g., red) had a high click-through rate. Based on this information, a script and video featuring product introductions with a predominantly red background were generated and distributed in multiple languages.

[0576] Example of a prompt

[0577] "Please explain in detail how to extract the most effective advertising selling points based on past advertising campaign data and sales performance data, and how to generate multilingual advertising content. In particular, please explain the specific analysis procedures and generation process, focusing on the differences in effectiveness based on color and time of day."

[0578] The system according to the present invention makes it possible to effectively utilize past advertising data, generate and deliver optimized multilingual video advertisements in real time, and quickly incorporate feedback.

[0579] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0580] Step 1:

[0581] The server collects data from past promotional programs and sales performance data. Specifically, it retrieves recorded data of TV shopping programs, sales volume, sales period, and viewer feedback from the database. The data collected also includes viewing history and click-through rates of past advertising campaigns. The input is data from the database, and the output is the collected data to be analyzed.

[0582] Step 2:

[0583] The server analyzes the collected data. The data is first cleaned, including the imputation of missing data and the removal of outliers. Next, this cleaned data is input into a machine learning model (e.g., a random forest) to identify factors associated with sales performance. The input is the cleaned data, and the output is the factors associated with sales performance.

[0584] Step 3:

[0585] The server automatically extracts basic information and effective selling points for each product based on the analysis results. Specifically, it uses a machine learning model's importance ranking to identify features with high impact. The input is the analysis results, and the output is the extracted basic information and selling points.

[0586] Step 4:

[0587] The server generates advertising scripts based on the extracted basic information and selling points. It automatically generates text describing the characteristics and benefits of each product using a natural language generation (NLG) algorithm. The input is basic information and selling points, and the output is the generated script.

[0588] Step 5:

[0589] The server generates multilingual video ads based on the generated script. Video editing libraries such as moviepy are used for video generation, and narration and captions are added based on the script. The input is the generated script, and the output is the video ad.

[0590] Step 6:

[0591] The server optimizes the generated video content to a format suitable for the environment of each viewing region. For example, it might provide high-resolution video in one region and low-resolution video in another. The input is a video advertisement, and the output is the optimized video content.

[0592] Step 7:

[0593] The server uploads optimized video content to video distribution platforms. It optimizes the video to match the different specifications of each distribution platform and then uploads it. The input is the optimized video content, and the output is the completion of the upload to each platform.

[0594] Step 8:

[0595] Users view advertising campaigns and provide feedback. For example, they might watch an ad on a smart device and enter ratings and comments within the application. The input is the user's feedback, and the output is the collected feedback data.

[0596] Step 9:

[0597] The server analyzes user feedback and uses it to optimize the next advertising campaign. It analyzes feedback data in real time and incorporates it into the generation of future ad scripts and videos. The input is feedback data, and the output is ideas for improving the next ad campaign.

[0598] This series of processing steps allows for the effective use of past advertising data, the real-time generation and delivery of optimized, multilingual video ads, and the rapid incorporation of feedback.

[0599] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0600] The system according to the present invention collects and analyzes data from past promotional programs and sales performance data, extracts basic information and effective selling points of products, generates production scripts based on this information, and ultimately generates and distributes multilingual and multicultural video content. Furthermore, this system incorporates an emotion engine that recognizes user emotions and customizes content based on the recognized emotions. A detailed embodiment of this system is described below.

[0601] Data collection

[0602] The server connects to databases of past promotional programs and sales performance data to collect necessary data. For example, it retrieves recorded data of past TV shopping programs, as well as data such as sales volume, sales period, and viewer feedback.

[0603] Data analysis

[0604] The server cleans the collected data. Specifically, it imputes missing data, removes outliers, and normalizes the data. Next, this cleaned data is input into a machine learning model to identify factors related to sales performance. For example, the analysis may reveal that certain time periods are advantageous for sales, or that the characteristics of specific products influence sales performance.

[0605] Extracting selling points

[0606] Based on the analysis results, the server automatically extracts basic information about each product (e.g., product characteristics, usage instructions, price) and effective selling points (e.g., features that result in high customer retention rates). This information is then used to generate subsequent scripts.

[0607] Script generation

[0608] The server generates a production script based on the extracted basic information and selling points. In this process, a natural language generation (NLG) algorithm is used to automatically generate text describing the product's characteristics and benefits. For example, a script highlighting the features of a new cooking utensil might be generated.

[0609] Video content generation

[0610] The server provides the generated script to the video generation AI, which then uses AI Cast to create multilingual and multicultural video content. Specifically, videos in multiple languages, such as English, Spanish, and French, are created simultaneously. AI Cast adds narration and captions corresponding to each language and synthesizes the video and audio.

[0611] Optimization and delivery

[0612] The server optimizes the generated video content into a format suitable for the environment of each viewing region and uploads it to the relevant television station or online distribution platform. For example, optimizations may be made to provide high-resolution video in one region and low-resolution video in another.

[0613] Customization using an emotion engine

[0614] The server uses an emotion engine to recognize the emotions of users viewing the delivered content. Specifically, it analyzes the user's facial expressions, voice tone, and body language in real time to identify the user's emotional state. For example, it uses cameras and microphones to detect the user's smiles and reactions indicating interest.

[0615] Dynamic content adjustment

[0616] The server dynamically adjusts the product presentation method and script based on the recognized user's emotions. For example, if the user shows interest, it adds a detailed description of the product; conversely, if the user shows little interest, it switches to a different product.

[0617] View and purchase

[0618] Users select and watch shopping programs available via television or the internet. For example, users can use a remote control or mouse to choose a program featuring products that interest them. If a user is interested in a product, they can purchase it through telephone orders or online shopping sites.

[0619] Specific example:

[0620] Suppose a server collects and analyzes TV shopping program data for a specific detergent over the past six months and discovers that sales are highest during a particular time slot. Based on this information, the server extracts appropriate selling points (e.g., "Powerful cleaning effect with a small amount") and generates a script that emphasizes the detergent's characteristics. Furthermore, based on this information, it generates three video content pieces in English, Japanese, and Chinese, and distributes them optimized for each market. In doing so, it uses an emotion engine to recognize user sentiment, adding detailed explanations if the user shows interest, and switching to other products or approaches if interest is low, thereby achieving even more effective sales promotion.

[0621] Through the steps described above, the system according to the present invention can efficiently produce and distribute multilingual and multicultural television shopping programs and provide customized content based on the user's emotions.

[0622] The following describes the processing flow.

[0623] Step 1:

[0624] The server collects past promotional program data and sales performance data. Specifically, it connects to a database to retrieve recordings of past TV shopping programs, sales figures, sales periods, viewer feedback, and other information.

[0625] Step 2:

[0626] The server cleans the collected data. Specifically, it performs tasks such as filling in missing data, removing outliers, and normalizing the data. For example, it standardizes the format of sales performance data and converts it into a format suitable for analysis algorithms.

[0627] Step 3:

[0628] The server inputs the cleaned data into a machine learning model to identify factors associated with sales performance. Specifically, it uses algorithms such as regression analysis, clustering, and decision trees to analyze how specific time periods or product characteristics influence sales.

[0629] Step 4:

[0630] The server extracts basic information and effective selling points for each product based on the analysis results of a machine learning model. Specifically, it uses text mining techniques to extract keywords from product reviews and sales history to identify features that are effective for sales.

[0631] Step 5:

[0632] The server generates a production script based on the extracted basic information and selling points. In this process, a natural language generation (NLG) algorithm is used to generate text that explains the product's characteristics and benefits.

[0633] Step 6:

[0634] The server provides the generated script to the video generation AI, which then uses AI Cast to create multilingual and multicultural video content. Specifically, it creates videos in multiple languages, such as English, Spanish, and French. AI Cast adds narration and captions corresponding to each language and synthesizes the video and audio.

[0635] Step 7:

[0636] The server optimizes the generated video content to a format suitable for the environment of each viewing region. Specifically, this involves changing the video resolution, converting file formats, and adding subtitles.

[0637] Step 8:

[0638] The server uploads optimized video content to the relevant television stations and online distribution platforms. Specifically, it distributes the content using FTP or APIs.

[0639] Step 9:

[0640] Users select and watch shopping programs available via television or the internet. For example, users can use a remote control or mouse to choose a program that features products they are interested in.

[0641] Step 10:

[0642] The emotion engine recognizes the emotions of users watching the streamed content. Specifically, it analyzes the user's facial expressions, voice tone, and body language in real time to identify the user's emotional state. For example, it can analyze emotions based on the user's camera footage and microphone recordings.

[0643] Step 11:

[0644] The server dynamically adjusts the product presentation method and script based on the recognized user's emotions. For example, if the user shows interest, it adds a detailed description of the product; conversely, if the user shows little interest, it switches to a different product.

[0645] Step 12:

[0646] When a user finds a product they are interested in, they purchase it. Specifically, they place an order via telephone or online shopping site.

[0647] Specific example:

[0648] The server collected and analyzed TV shopping program data for a specific detergent over the past six months and discovered that sales were highest during a particular time slot. Based on this information, the server extracted appropriate selling points (e.g., "Powerful cleaning effect with a small amount") and generated a script that emphasized the detergent's characteristics. Furthermore, it generated three video content versions—English, Japanese, and Chinese—and distributed them optimized for each market. Using an emotion engine, the system analyzes the viewer's emotions in real time, adding detailed explanations of the detergent if they show interest, and switching to other products if they show little interest, thereby achieving more effective sales promotion.

[0649] (Example 2)

[0650] Next, we will describe Example 2. 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".

[0651] In today's world, promotional activities are becoming increasingly diverse and complex. In particular, the creation and distribution of multilingual and multicultural content is difficult to manage effectively and efficiently using traditional methods. Furthermore, while there is a need to recognize audience emotions in real time and customize content accordingly, there is a lack of concrete technological means to achieve this. As a result, there are challenges in fully achieving improved sales efficiency and optimized user experience.

[0652] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0653] In this invention, the server includes means for collecting data from past promotional programs and sales performance data; means for cleaning and analyzing the collected data and extracting basic information and effective selling points of the products being sold; means for generating production scripts based on the analyzed information; means for generating multilingual and multicultural video content based on the scripts; means for optimizing and distributing the generated video content in a format suitable for each viewing region; and means for recognizing the emotions of the viewing users and customizing the content to be distributed accordingly. This enables the effective and efficient production and distribution of multilingual and multicultural video content, and improves sales efficiency and optimizes the user experience through flexible content customization based on the emotions of the viewers.

[0654] "Promotional program data" refers to information such as videos, audio, and text produced for the purpose of advertising products or services.

[0655] "Sales performance data" refers to historical information about the sales of a specific product or service, including sales volume, period, revenue, and customer feedback.

[0656] "Cleaning" is the process of filling in missing data, removing outliers, and normalizing data from a dataset.

[0657] "Analysis" is the process of identifying factors related to sales performance using methods such as machine learning models on collected data.

[0658] "Basic information" refers to fundamental information necessary to understand a product, such as its characteristics, usage instructions, and price.

[0659] A "selling point" is a specific feature or advantage of a product, and it is an important element that makes it appealing to customers.

[0660] A "production script" is a document or scenario created to explain the characteristics and benefits of a product.

[0661] "Multilingual and multicultural video content" refers to videos produced to be compatible with multiple languages ​​and different cultures, and includes narration and captions in each language.

[0662] "Optimization" is the process of converting generated video content into a format suitable for the environment of each viewing region.

[0663] "Means of distribution" refer to platforms and devices used to provide optimized video content to viewers.

[0664] "Recognizing emotions" is the process of analyzing the viewer's facial expressions, tone of voice, and body language in real time to identify the user's emotional state.

[0665] "Customizing" is the process of dynamically adjusting the way a product is presented and the script based on the emotions of the perceived audience.

[0666] The system according to the present invention collects and analyzes data from past promotional programs and sales performance data, extracts basic information and effective selling points of products, generates production scripts based on this information, and ultimately generates and distributes multilingual and multicultural video content. Furthermore, this system incorporates an emotion engine that recognizes user emotions and customizes content based on the recognized emotions. A detailed embodiment of this system is described below.

[0667] Data collection

[0668] The server uses a database management system to collect data from past promotional programs and sales performance data. Specifically, it uses SQL or NoSQL databases to retrieve data such as recordings of past TV shopping programs, sales volume, sales period, and viewer feedback.

[0669] Data analysis

[0670] The server uses data preprocessing tools (e.g., Python's Pandas library) to clean the collected data. Specifically, it performs data imputation, removal of outliers, and data normalization. Next, this cleaned data is input into a machine learning model (e.g., TensorFlow or Scikit-learn) to identify factors related to sales performance. For example, the analysis may reveal that certain time periods are advantageous for sales, or that the characteristics of certain products influence sales performance.

[0671] Extracting selling points

[0672] Based on the analysis results, the server automatically extracts basic information about each product (e.g., product characteristics, usage instructions, price) and effective selling points (e.g., features that result in high customer retention rates). This information is then used to generate subsequent scripts.

[0673] Script generation

[0674] The server generates a production script based on the extracted basic information and selling points. This process uses natural language generation (NLG) algorithms (such as GPT-3) to automatically generate text describing the product's characteristics and benefits. For example, a script highlighting the features of a new cooking utensil might be generated.

[0675] Video content generation

[0676] The server provides the generated script to the video generation AI, which then uses AI Cast to create multilingual and multicultural video content. Specifically, videos in multiple languages, such as English, Spanish, and French, are created simultaneously. AI Cast adds narration and captions corresponding to each language and synthesizes the video and audio.

[0677] Optimization and delivery

[0678] The server optimizes the generated video content into a format suitable for the environment of each viewing region and uploads it to the relevant broadcasters and online distribution platforms. For example, optimizations may be made to provide high-resolution video in certain regions and low-resolution video in other regions.

[0679] Customization using an emotion engine

[0680] The server uses an emotion engine to recognize the emotions of users viewing the delivered content. Specifically, it analyzes the user's facial expressions, voice tone, and body language in real time to identify the user's emotional state. For example, it uses cameras and microphones to detect the user's smiles and reactions indicating interest.

[0681] Dynamic content adjustment

[0682] The server dynamically adjusts the product presentation method and script based on the recognized user's emotions. For example, if the user shows interest, it adds a detailed description of the product; conversely, if the user shows little interest, it switches to a different product.

[0683] View and purchase

[0684] Users select and watch shopping programs available via television or the internet. For example, users can use a remote control or mouse to choose a program featuring products that interest them. If a user is interested in a product, they can purchase it through telephone orders or online shopping sites.

[0685] Specific example

[0686] Suppose a server collects and analyzes TV shopping program data for a specific detergent over the past six months and discovers that sales are highest during a particular time slot. Based on this information, the server extracts appropriate selling points (e.g., "Powerful cleaning effect with a small amount") and generates a script that emphasizes the detergent's characteristics. Furthermore, based on this information, it generates three video content pieces in English, Japanese, and Chinese, and distributes them optimized for each market. In doing so, it uses an emotion engine to recognize user sentiment, adding detailed explanations if the user shows interest, and switching to other products or approaches if interest is low, thereby achieving even more effective sales promotion.

[0687] Example of a prompt

[0688] Collect TV shopping program data for specialty detergents from the past six months and identify the time slots with the highest sales. Next, extract the selling points and generate video scripts in English, Japanese, and Chinese. Use an emotion engine to adjust the content based on user reactions.

[0689] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0690] Step 1: Data Collection

[0691] The server uses a database management system to collect data from past promotional programs and sales performance data. This includes data such as recordings of past TV shopping programs, sales volume, sales period, and viewer feedback, using SQL or NoSQL databases. The input is database connection information, and the output is the collected raw data.

[0692] Step 2: Data Cleaning

[0693] The server uses data preprocessing tools (e.g., Python's Pandas library) to clean the raw data. Specifically, it performs data imputation, removal of outliers, and data normalization. The input is the raw data collected in step 1, and the output is the cleaned data.

[0694] Step 3: Data Analysis

[0695] The server inputs the cleaned data into a machine learning model (e.g., TensorFlow or Scikit-learn) to identify factors associated with sales performance. For example, the analysis might reveal that certain time periods are favorable for sales or that the characteristics of specific products influence sales performance. The input is the cleaned data, and the output is the analysis results (e.g., factors associated with sales performance).

[0696] Step 4: Extracting the selling points

[0697] The server extracts basic information and effective selling points for each product based on the analysis results. Specifically, it targets product characteristics, usage methods, pricing, and features that contribute to high customer retention rates. The input is the analysis results, and the output is the extracted basic information and selling points.

[0698] Step 5: Generate the script

[0699] The server generates a production script based on the extracted basic information and selling points. In this process, a natural language generation (NLG) algorithm (e.g., GPT-3) is used to automatically generate text describing the product's characteristics and benefits. The input is the basic information and selling points, and the output is the generated script.

[0700] Step 6: Generate video content

[0701] The server provides the generated script to the video generation AI, which then uses AI casting to generate multilingual and multicultural video content. Specifically, videos in multiple languages, such as English, Spanish, and French, are created simultaneously. The input is the generated script, and the output is multilingual video content.

[0702] Step 7: Optimization and Delivery

[0703] The server optimizes the generated video content into a format suitable for the environment of each viewing region and uploads it to the relevant broadcasters and online distribution platforms. For example, it may optimize the content by providing high-resolution video in one region and low-resolution video in another. The input is the generated video content, and the output is the optimized video file.

[0704] Step 8: Customization with the emotion engine

[0705] The server uses an emotion engine to recognize the emotions of users viewing the delivered content. Specifically, it analyzes the user's facial expressions, voice tone, and body language in real time to identify the user's emotional state. The input is real-time data of the user during viewing, and the output is the recognized emotional state.

[0706] Step 9: Dynamic Content Adjustment

[0707] The server dynamically adjusts the product presentation method and script based on the recognized user's emotions. For example, if the user shows interest, it adds a detailed description of the product; conversely, if the user shows little interest, it switches to another product. The input is the recognized emotional state, and the output is the adjusted content.

[0708] Step 10: View and Purchase

[0709] Users select and watch shopping programs available via television or the internet. For example, users can use a remote control or mouse to choose a program featuring products they are interested in. If a user is interested, they can purchase the product through telephone ordering or an online shopping site. The input is the user's viewing selection and emotional reaction, and the output is the action of purchasing the product.

[0710] (Application Example 2)

[0711] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0712] Current advertising systems lack the functionality to dynamically adjust ad content in response to the emotions of target users. This makes it difficult to maximize advertising effectiveness and capture user interest. Furthermore, automatically generating multilingual and multicultural video content and distributing it in formats suitable for each viewing region is not easy. To address these challenges, a system is needed that can extract effective selling points based on collected data and customize ad content according to user emotions.

[0713] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0714] In this invention, the server includes means for collecting data from past promotional programs and sales performance data; means for analyzing the collected data and extracting basic information and effective selling points of the products being sold; means for generating production scripts based on the analyzed information; means for generating multilingual and multicultural video content based on the scripts; means for optimizing the generated video content into a format suitable for each viewing region and distributing it; means for analyzing user emotions on a smart device; and means for dynamically adjusting advertising content based on the analyzed user emotions. This enables customized advertising that responds to user emotions, thereby maximizing advertising effectiveness.

[0715] A "promotional program" is video content intended to promote the sale of a product or service.

[0716] "Sales performance data" refers to data that includes the quantity of products sold in the past, the sales period, and customer feedback.

[0717] "Data analysis" is the process of organizing collected data and extracting information that is useful for sales.

[0718] "Basic information" refers to fundamental information such as the product's characteristics, usage instructions, and price.

[0719] An "effective selling point" is a feature or characteristic that is particularly effective in promoting the sale of a product.

[0720] A "production script" is a script or scenario used to generate video or advertising content.

[0721] "Multilingual and multicultural support" refers to the function of generating and providing content that is compatible with different languages ​​and cultures.

[0722] "Video content" refers to digital content that includes audio and video.

[0723] A "smart device" is an internet-connected device such as a smartphone, tablet, or smart glasses.

[0724] "User emotion" refers to the psychological state determined from the user's facial expressions, tone of voice, body language, etc.

[0725] "Dynamic adjustment" means changing the content in real time.

[0726] "Advertising content" refers to media content created for the purpose of promoting products or services.

[0727] The system for carrying out this invention is primarily executed by a server. The embodiments for carrying out the invention are described in detail below.

[0728] 1. Data Collection

[0729] The server collects data on past promotional programs and sales performance. This includes recordings of promotional programs, sales volume, sales period, and customer feedback data. The database records numerous promotional programs and their corresponding sales performance.

[0730] 2. Data Analysis

[0731] The server analyzes the collected data. First, it cleans the data, filling in missing data and removing outliers. Next, it uses the cleaned data to analyze factors related to sales performance using a machine learning model. This reveals things like specific time slots or product characteristics that are advantageous for sales.

[0732] 3. Extracting selling points

[0733] The server extracts basic information and effective selling points based on the analysis results. Basic information includes product characteristics, usage, and price. Effective selling points are product features deemed particularly effective for sales promotion.

[0734] 4. Script generation

[0735] The server generates a production script based on the extracted basic information and selling points. Using a natural language generation algorithm, it automatically generates text to describe the product's characteristics and benefits. For example, a script highlighting the features of a new cooking utensil might be generated here.

[0736] 5. Video content generation

[0737] The server generates multilingual and multicultural video content using the generated script. A generation AI model is used to add narration and captions corresponding to each language, and to synthesize the video and audio. For example, the same content can be generated in English, Spanish, French, and other languages.

[0738] 6. Distribution and Optimization

[0739] The server optimizes the generated video content into a format suitable for each viewing region and uploads it to television stations and online distribution platforms. For example, it might provide high-resolution video in certain regions and low-resolution video in others.

[0740] 7. Sentiment Analysis and Dynamic Content Adjustment

[0741] The system analyzes user emotions using smart devices (smartphones and smart glasses). It employs an emotion recognition engine that analyzes the user's facial expressions, voice tone, and body language in real time. Based on the detected emotions, the advertising content is dynamically adjusted. For example, if the user shows interest, a detailed description of the product is added; if interest is low, the system switches to a different product.

[0742] 8. Specific Examples

[0743] The server collects and analyzes promotional program data for a specific detergent over the past six months, discovering that sales are highest during certain time slots. Based on this information, the server extracts "powerful cleaning effect with a small amount" as a selling point and generates a script that emphasizes the detergent's characteristics. Video content is generated in English, Japanese, and Chinese, and distributed optimized for each market. When a user watches this ad on their smartphone and shows interest, a more detailed explanation is added; otherwise, the ad switches to the next product.

[0744] 9. Example of a prompt message

[0745] The next ad you'll show is for a new detergent. Use the following selling points to create a 30-second ad script:

[0746] A small amount provides powerful cleaning.

[0747] Made with environmentally friendly ingredients.

[0748] Effective against all types of dirt.

[0749] Thus, the system for implementing the present invention provides dynamic advertisements that can maximize user interest.

[0750] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0751] Step 1:

[0752] The server collects data from past promotional programs and sales performance data. Inputs include recordings of promotional programs, sales volume, sales period, and customer feedback data. This data is retrieved from a database and prepared for data analysis. The output is a list of the retrieved data.

[0753] Step 2:

[0754] The server analyzes the collected data. It uses collected promotional program data and sales performance data as input. First, it performs data cleaning, including imputing missing data and removing outliers. Next, a machine learning model analyzes factors related to sales performance, identifying the impact of time of day and product characteristics on sales. The output is the analysis results, providing information on specific sales trends and effective selling points.

[0755] Step 3:

[0756] The server extracts basic information and effective selling points based on the analysis results. It uses the analyzed data as input. From the analyzed information, it identifies basic information such as product characteristics, usage, and price, as well as features advantageous for sales promotion. The output is a list of basic information and selling points.

[0757] Step 4:

[0758] The server generates a production script based on the extracted basic information and selling points. It uses a list of basic information and selling points as input. A natural language generation algorithm is used to automatically generate text describing the product's characteristics and benefits. The output is the generated script.

[0759] Step 5:

[0760] The server generates multilingual and multicultural video content using the generated script. It uses the generated script as input. Utilizing a generation AI model, it adds narration and captions corresponding to each language and synthesizes the video and audio. The output is video content compatible with multiple languages.

[0761] Step 6:

[0762] The server optimizes the generated video content into a format suitable for each viewing region and distributes it. It uses the generated video content as input. Technical adjustments, such as high or low resolution, are made to meet the needs of each viewing region before uploading it to television stations and online distribution platforms. The output is the optimized video content.

[0763] Step 7:

[0764] The device analyzes the user's emotions. It uses the user's camera video and audio data as input. Utilizing an emotion recognition engine, it performs facial expression and voice analysis to identify the user's emotional state. The output is the recognized user emotion information.

[0765] Step 8:

[0766] The server dynamically adjusts ad content based on the analyzed user's sentiment. It uses recognized user sentiment information and already generated ad content as input. For example, if the user shows interest, it adds a detailed description; if not, it switches to another product. The output is the adjusted ad content.

[0767] Step 9:

[0768] Users view tailored advertising content. Dynamically adjusted advertising content from the server is used as input, providing the user with the optimal advertising experience. The output is the user's viewing behavior and feedback.

[0769] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0770] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include those described above. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions shown by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0771] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0772] [Third Embodiment]

[0773] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0774] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0775] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0776] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0777] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0778] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0779] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0780] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0781] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0782] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0783] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0784] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0785] The system according to the present invention collects and analyzes data from past promotional programs and sales performance data, extracts basic information and effective selling points of products, generates production scripts based on this information, and ultimately generates and distributes multilingual and multicultural video content. A detailed embodiment of this system is described below.

[0786] Data collection

[0787] The server connects to databases of past promotional programs and sales performance data to collect necessary data. For example, it retrieves recorded data of past TV shopping programs for a specific product, as well as data such as sales volume, sales period, and viewer feedback.

[0788] Data analysis

[0789] The server analyzes the collected data. First, it cleans the data, filling in missing data and removing outliers. Next, it inputs this cleaned data into a machine learning model to identify factors related to sales performance. For example, the analysis might reveal that certain time periods are advantageous for sales, or that the characteristics of certain products contribute to high sales performance.

[0790] Extracting selling points

[0791] Based on the analysis results, the server automatically extracts basic information about each product (e.g., product characteristics, usage instructions, price) and effective selling points (e.g., features that result in high customer retention rates). This information is then used to generate subsequent scripts.

[0792] Script generation

[0793] The server generates a production script based on the extracted basic information and selling points. In this process, a natural language generation (NLG) algorithm is used to automatically generate text describing the characteristics and benefits of each product. For example, a script highlighting the features of a new cooking utensil might be generated.

[0794] Video content generation

[0795] The server provides the generated script to the video generation AI, which uses AI Cast to create multilingual and multicultural video content. For example, videos in multiple languages, such as English, Spanish, and French, are created simultaneously. AI Cast adds narration and captions corresponding to each language and synthesizes the video and audio.

[0796] Optimization and delivery

[0797] The server optimizes the generated video content into a format suitable for the environment of each viewing region and uploads it to the relevant television station or online distribution platform. For example, optimizations may be made to provide high-resolution video in one region and low-resolution video in another.

[0798] View and purchase

[0799] Users select and watch shopping programs available via television or the internet. For example, users can use a remote control or mouse to choose a program featuring products that interest them. If a user is interested in a product, they can purchase it through telephone orders or online shopping sites.

[0800] Specific example:

[0801] The server collects and analyzes TV shopping program data for a specific detergent over the past six months and discovers that sales are highest during a particular time slot. Based on this information, the server extracts appropriate selling points (e.g., "Powerful cleaning effect with a small amount") and generates a script that emphasizes the detergent's characteristics. Furthermore, based on this information, it generates three video content pieces in English, Japanese, and Chinese, and distributes them optimized for each market.

[0802] Through the steps described above, the system according to the present invention can efficiently produce multilingual and multicultural television shopping programs and distribute them quickly and at low cost.

[0803] The following describes the processing flow.

[0804] Step 1:

[0805] The server connects to databases of past promotional programs and sales performance data to collect necessary data. Specifically, it retrieves recordings of past TV shopping programs, sales figures, sales periods, viewer feedback, and more.

[0806] Step 2:

[0807] The server cleans the collected data. Specifically, it performs tasks such as filling in missing data, removing outliers, and normalizing the data. For example, it standardizes the format of sales performance data and converts it into a format suitable for analysis algorithms.

[0808] Step 3:

[0809] The server inputs the cleaned data into a machine learning model to identify factors associated with sales performance. Specifically, it uses algorithms such as regression analysis, clustering, and decision trees to analyze how specific time periods or product characteristics influence sales.

[0810] Step 4:

[0811] The server extracts basic information and effective selling points for each product based on the analysis results of a machine learning model. Specifically, it uses text mining techniques to extract keywords from product reviews and sales history to identify features that are effective for sales.

[0812] Step 5:

[0813] The server generates a production script based on the extracted basic information and selling points. Specifically, it uses a natural language generation (NLG) algorithm to generate text that explains the product's characteristics and benefits.

[0814] Step 6:

[0815] The server provides the generated script to the video generation AI, which then uses AI Cast to create multilingual and multicultural video content. Specifically, AI Cast adds narration and captions corresponding to each language and synthesizes the video and audio.

[0816] Step 7:

[0817] The server optimizes the generated video content for each viewing region. Specifically, this involves changing the video resolution, converting file formats, and adding subtitles.

[0818] Step 8:

[0819] The server uploads optimized video content to the relevant television stations and online distribution platforms. Specifically, it distributes the content using FTP or APIs.

[0820] Step 9:

[0821] Users select and watch shopping programs available via television or the internet. Specifically, they select programs using a remote control or mouse.

[0822] Step 10:

[0823] When a user finds a product they are interested in, they purchase it. Specifically, they place an order via telephone or online shopping site.

[0824] Through the steps described above, the system according to the present invention can effectively produce and distribute multilingual and multicultural television shopping programs.

[0825] (Example 1)

[0826] Next, we will describe Example 1. 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."

[0827] In today's consumer market, to improve the efficiency of sales promotion, it is necessary to effectively utilize past promotional data and sales performance data to provide rapid and effective marketing tools. However, conventional systems have struggled to automate all processes from data collection, cleaning, and analysis to script generation, video creation, and distribution, as well as to efficiently produce and distribute multilingual and multicultural video content.

[0828] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0829] In this invention, the server includes means for collecting data from past promotional programs and sales performance data; means for cleaning, supplementing, and analyzing the collected data; means for extracting basic information and effective selling points of products from the analyzed data; means for generating production scripts based on the extracted information; means for generating multilingual and multicultural video content based on the scripts; and means for optimizing the generated video content into a format suitable for each viewing region and distributing it. This enables the efficient production and distribution of multilingual and multicultural video content.

[0830] "Promotional program data" refers to recorded data and related information of programs that were previously broadcast for the purpose of sales promotion.

[0831] "Sales performance data" refers to data related to actual sales activities, such as the quantity of products sold, the sales period, and viewer feedback.

[0832] "Cleaning" refers to the process of filling in missing values ​​and removing outliers from data.

[0833] A "machine learning model" refers to an algorithm or statistical model used for data analysis, designed to learn patterns and regularities from data.

[0834] A "selling point" refers to a product's characteristics, advantages, or effective elements for sales promotion.

[0835] A "natural language generation algorithm" is an algorithm that generates sentences in natural language based on extracted information.

[0836] "Multilingual and multicultural video content" refers to content that can generate video material that supports multiple languages ​​and cultures.

[0837] "Optimization" refers to the process of adjusting generated video content to suit the environment and requirements of each viewing region.

[0838] "Distribution" refers to the act of delivering generated video content to viewers online or through television stations.

[0839] The system according to the present invention collects and analyzes data from past promotional programs and sales performance data, extracts basic information and effective selling points of products, generates production scripts based on this information, and ultimately generates and distributes multilingual and multicultural video content. A detailed embodiment of this system is described below.

[0840] Data collection

[0841] The server connects to databases of past promotional programs and sales performance data to collect necessary data about a specific product. This includes recording data, sales volume, sales period, and viewer feedback. Specifically, the server uses SQL queries to retrieve the information.

[0842] Data analysis

[0843] The server cleans the collected data, imputing missing data and removing outliers. This cleaned data is then fed into a machine learning model to identify factors associated with sales performance. The Scikit-learn library in Python is used as the machine learning model.

[0844] Extracting selling points

[0845] Based on the analysis results, the server automatically extracts basic information about each product (e.g., product characteristics, usage instructions, price) and effective selling points (e.g., "powerful cleaning effect with a small amount").

[0846] Script generation

[0847] The server generates a production script using a natural language generation (NLG) algorithm based on the extracted basic information and selling points. Specifically, it uses OpenAI's GPT-3 to automatically generate text that describes the product's characteristics and advantages.

[0848] Video content generation

[0849] The server provides the generated script to the video generation AI, which then uses AI casting to create multilingual and multicultural video content. For example, it adds narration and captions in multiple languages, such as English, Spanish, and Japanese, and synthesizes the video and audio.

[0850] Optimization and delivery

[0851] The server optimizes and delivers the generated video content to the environment of each viewing region. Specifically, optimizations are made such as providing high-resolution video to some regions and low-resolution video to others. The optimized video is then uploaded to the relevant television stations and online distribution platforms.

[0852] View and purchase

[0853] Users can select and watch shopping programs delivered via television or the internet. For example, they can use a remote control or mouse to choose a program featuring products they are interested in, and then purchase the products by phone or through an online shopping site after watching the program.

[0854] Specific example

[0855] The server collected and analyzed TV shopping program data for a specific detergent over the past six months, revealing the highest sales volume during a particular time slot. Based on this information, it extracted the selling point "powerful cleaning effect with a small amount" and generated a script that emphasized the detergent's characteristics. Furthermore, based on this information, it generated three video content pieces in English, Japanese, and Chinese, and distributed them optimized for each market.

[0856] Through the steps described above, the system according to the present invention can efficiently produce multilingual and multicultural television shopping programs and distribute them quickly and at low cost.

[0857] Example of a prompt

[0858] Analyze TV shopping program data for a specific detergent from the past six months and extract its selling points. Then, generate scripts in English, Japanese, and Chinese based on those selling points. Please emphasize the detergent's characteristics and effects.

[0859] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0860] Step 1:

[0861] Data collection

[0862] The server connects to a database of promotional programs and a database of sales performance, collecting data such as recordings of past promotional programs, sales figures, sales periods, and viewer feedback.

[0863] Input: Information of the target database

[0864] Data processing: Extracting data using SQL queries.

[0865] Output: Collected dataset

[0866] Specific operation: The server executes an SQL query to retrieve video recording data and sales performance data for a specific product for the past six months.

[0867] Step 2:

[0868] Data cleaning

[0869] The server cleans the collected data, fills in missing data, and removes outliers.

[0870] Input: Collected dataset

[0871] Data processing: Imputation of missing data (imputation with mean values), removal of outliers.

[0872] Output: Cleaned dataset

[0873] Specific operation: Run a Python script to remove abnormal sales quantity data from the dataset and impute missing values ​​with the mean.

[0874] Step 3:

[0875] Data analysis

[0876] The server inputs the cleaned data into a machine learning model to identify factors associated with sales performance.

[0877] Input: Cleaned dataset

[0878] Data processing: Input data into a machine learning model and extract important features.

[0879] Output: Sales performance and related factors (specific time periods and product characteristics)

[0880] Specific operation: The server uses the Python Scikit-learn library to analyze specific time periods and product characteristics that contribute to sales performance using a model.

[0881] Step 4:

[0882] Extracting selling points

[0883] Based on the analysis results, the server automatically extracts basic product information and effective selling points.

[0884] Input: Sales performance and related factors

[0885] Data processing: Extraction of basic information, identification of effective selling points.

[0886] Output: Basic product information and selling points

[0887] Specific operation: Based on the extracted information, the server automatically identifies and extracts selling points such as "powerful cleaning effect with a small amount."

[0888] Step 5:

[0889] Script generation

[0890] The server generates a production script using a natural language generation (NLG) algorithm based on the extracted basic information and selling points.

[0891] Input: Basic product information and selling points

[0892] Data processing: Text generation using NLG algorithm

[0893] Output: Completed script

[0894] Specific operation: The server uses OpenAI's GPT-3 to generate a script that says, "This detergent has a powerful cleaning effect even in small amounts."

[0895] Step 6:

[0896] Video content generation

[0897] The server provides the generated script to the video generation AI, which then uses AI casting to generate multilingual and multicultural video content.

[0898] Input: Completed script

[0899] Data processing: Video and audio synthesis using AI casting.

[0900] Output: Multilingual and multicultural video content

[0901] Specific operation: The server generates videos based on the English, Japanese, and Chinese scripts, adding narration and captions in each language.

[0902] Step 7:

[0903] Optimization and delivery

[0904] The server optimizes the generated video content into a format suitable for the environment of each viewing region and then delivers it.

[0905] Input: Multilingual and multicultural video content

[0906] Data processing: Optimized for resolution and format suitable for each viewing region.

[0907] Output: Optimized video content

[0908] Specific operation: The server optimizes high-resolution videos for North America and low-resolution videos for regions with different internet speeds, and then uploads them to the distribution platform.

[0909] Step 8:

[0910] View and purchase

[0911] Users select shopping programs via television or the internet and then purchase products using telephone orders or online shopping after watching them.

[0912] Input: Optimized video content

[0913] Data processing: None

[0914] Output: Purchased items

[0915] Specific operation: The user uses the remote control to select a program about "special detergents," and after watching it, places a phone order or purchases the product from an online shopping site.

[0916] (Application Example 1)

[0917] Next, we will explain Application Example 1. In the following explanation, 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."

[0918] Current advertising campaigns lack effective ways to leverage historical data and optimize in real time. Furthermore, generating and distributing multilingual video ads tailored to different regions and cultures is time-consuming and requires significant resources and time. In addition, there's a lack of mechanisms to quickly incorporate user feedback into subsequent advertising campaigns. A system is needed to address these challenges and maximize advertising effectiveness.

[0919] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0920] In this invention, the server includes means for collecting data from past promotional programs and sales performance data; means for analyzing the collected data and extracting basic information and effective selling points of the products being sold; means for generating production scripts based on the analyzed information; means for generating multilingual and multicultural video content based on the scripts; means for optimizing the generated video content into a format suitable for each viewing region and distributing it; means for collecting data from past advertising campaigns via smart devices and analyzing it in real time; means for automatically extracting effective advertising elements to generate advertising scripts; means for automatically generating multilingual video advertisements based on the generated advertising scripts; and means for collecting user feedback and using it to optimize the next advertising campaign. This makes it possible to effectively utilize past advertising data, generate and distribute optimized multilingual video advertisements in real time, and quickly reflect feedback.

[0921] "Data from past promotional programs" refers to information such as video, audio, and text related to promotions that were broadcast in the past.

[0922] "Sales performance data" refers to data related to the actual sales performance of a product, such as sales volume, sales period, and sales figures.

[0923] A "smart device" refers to a portable device with internet connectivity, such as a smartphone, smart glasses, or tablet.

[0924] "Real-time analysis" means performing analysis immediately the moment data is collected.

[0925] A "production script" is a text document that provides instructions regarding the content and flow of a video.

[0926] "Multilingual and multicultural video content" refers to videos produced in formats adapted to multiple languages ​​and cultures.

[0927] An "ad script" is text used to explain the content and selling points of an advertising video.

[0928] "Past advertising campaign data" refers to information such as viewing history, responses, and click-through rates of advertisements that were previously run.

[0929] "Optimizing and distributing content in a format suitable for each viewing region" means converting video content to a quality and format appropriate for each region and uploading it to the distribution platform.

[0930] "Automatically extracting effective advertising elements" means that the system automatically identifies and extracts elements (e.g., catchphrases and colors) that maximize the effectiveness of an advertisement.

[0931] "Collecting user feedback and using it to optimize the next advertising campaign" means collecting user reactions and evaluations as data and incorporating them into the production of future advertisements.

[0932] The system according to the present invention aims to optimize advertising campaigns and is configured as follows.

[0933] Data collection

[0934] The server first collects data from past promotional programs and sales performance data. Simultaneously, it also collects data such as viewing history and click-through rates of past advertising campaigns via smart devices.

[0935] Data Analysis

[0936] The collected data is analyzed on the server. Machine learning models are used for data analysis to identify factors that maximize advertising effectiveness. Specifically, algorithms such as random forests are used to analyze factors related to sales performance.

[0937] Extracting selling points

[0938] Based on the analysis results, the server automatically extracts basic information and effective selling points for each product. During this process, effective advertising elements such as colors and catchphrases are identified.

[0939] Script generation

[0940] The server generates advertising scripts based on extracted basic information and selling points. Using natural language generation (NLG) algorithms, it generates text that effectively explains the characteristics and benefits of each product. These scripts are generated in multiple languages ​​for multilingual support.

[0941] Video content generation

[0942] Based on the generated ad script, the server automatically creates multilingual video ads. Video editing libraries such as moviepy are used for video generation. The generated videos are saved in a format adapted to each viewing region.

[0943] Optimization and delivery

[0944] The server optimizes the generated video content into a format suitable for the environment of each viewing region and uploads it to the video distribution platform. For example, it adjusts high-resolution and low-resolution videos for each region.

[0945] Gathering feedback and optimizing future ads

[0946] Users view advertising campaigns and provide feedback. The server collects this feedback in real time and uses it to optimize the next advertising campaign.

[0947] Specific example

[0948] For example, by analyzing data from TV shopping programs that showed high effectiveness during specific time slots, it was found that certain colors (e.g., red) had a high click-through rate. Based on this information, a script and video featuring product introductions with a predominantly red background were generated and distributed in multiple languages.

[0949] Example of a prompt

[0950] "Please explain in detail how to extract the most effective advertising selling points based on past advertising campaign data and sales performance data, and how to generate multilingual advertising content. In particular, please explain the specific analysis procedures and generation process, focusing on the differences in effectiveness based on color and time of day."

[0951] The system according to the present invention makes it possible to effectively utilize past advertising data, generate and deliver optimized multilingual video advertisements in real time, and quickly incorporate feedback.

[0952] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0953] Step 1:

[0954] The server collects data from past promotional programs and sales performance data. Specifically, it retrieves recorded data of TV shopping programs, sales volume, sales period, and viewer feedback from the database. The data collected also includes viewing history and click-through rates of past advertising campaigns. The input is data from the database, and the output is the collected data to be analyzed.

[0955] Step 2:

[0956] The server analyzes the collected data. The data is first cleaned, including the imputation of missing data and the removal of outliers. Next, this cleaned data is input into a machine learning model (e.g., a random forest) to identify factors associated with sales performance. The input is the cleaned data, and the output is the factors associated with sales performance.

[0957] Step 3:

[0958] The server automatically extracts basic information and effective selling points for each product based on the analysis results. Specifically, it uses a machine learning model's importance ranking to identify features with high impact. The input is the analysis results, and the output is the extracted basic information and selling points.

[0959] Step 4:

[0960] The server generates advertising scripts based on the extracted basic information and selling points. It automatically generates text describing the characteristics and benefits of each product using a natural language generation (NLG) algorithm. The input is basic information and selling points, and the output is the generated script.

[0961] Step 5:

[0962] The server generates multilingual video ads based on the generated script. Video editing libraries such as moviepy are used for video generation, and narration and captions are added based on the script. The input is the generated script, and the output is the video ad.

[0963] Step 6:

[0964] The server optimizes the generated video content to a format suitable for the environment of each viewing region. For example, it might provide high-resolution video in one region and low-resolution video in another. The input is a video advertisement, and the output is the optimized video content.

[0965] Step 7:

[0966] The server uploads optimized video content to video distribution platforms. It optimizes the video to match the different specifications of each distribution platform and then uploads it. The input is the optimized video content, and the output is the completion of the upload to each platform.

[0967] Step 8:

[0968] Users view advertising campaigns and provide feedback. For example, they might watch an ad on a smart device and enter ratings and comments within the application. The input is the user's feedback, and the output is the collected feedback data.

[0969] Step 9:

[0970] The server analyzes user feedback and uses it to optimize the next advertising campaign. It analyzes feedback data in real time and incorporates it into the generation of future ad scripts and videos. The input is feedback data, and the output is ideas for improving the next ad campaign.

[0971] This series of processing steps allows for the effective use of past advertising data, the real-time generation and delivery of optimized, multilingual video ads, and the rapid incorporation of feedback.

[0972] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0973] The system according to the present invention collects and analyzes data from past promotional programs and sales performance data, extracts basic information and effective selling points of products, generates production scripts based on this information, and ultimately generates and distributes multilingual and multicultural video content. Furthermore, this system incorporates an emotion engine that recognizes user emotions and customizes content based on the recognized emotions. A detailed embodiment of this system is described below.

[0974] Data collection

[0975] The server connects to databases of past promotional programs and sales performance data to collect necessary data. For example, it retrieves recorded data of past TV shopping programs, as well as data such as sales volume, sales period, and viewer feedback.

[0976] Data analysis

[0977] The server cleans the collected data. Specifically, it imputes missing data, removes outliers, and normalizes the data. Next, this cleaned data is input into a machine learning model to identify factors related to sales performance. For example, the analysis may reveal that certain time periods are advantageous for sales, or that the characteristics of specific products influence sales performance.

[0978] Extracting selling points

[0979] Based on the analysis results, the server automatically extracts basic information about each product (e.g., product characteristics, usage instructions, price) and effective selling points (e.g., features that result in high customer retention rates). This information is then used to generate subsequent scripts.

[0980] Script generation

[0981] The server generates a production script based on the extracted basic information and selling points. In this process, a natural language generation (NLG) algorithm is used to automatically generate text describing the product's characteristics and benefits. For example, a script highlighting the features of a new cooking utensil might be generated.

[0982] Video content generation

[0983] The server provides the generated script to the video generation AI, which then uses AI Cast to create multilingual and multicultural video content. Specifically, videos in multiple languages, such as English, Spanish, and French, are created simultaneously. AI Cast adds narration and captions corresponding to each language and synthesizes the video and audio.

[0984] Optimization and delivery

[0985] The server optimizes the generated video content into a format suitable for the environment of each viewing region and uploads it to the relevant television station or online distribution platform. For example, optimizations may be made to provide high-resolution video in one region and low-resolution video in another.

[0986] Customization using an emotion engine

[0987] The server uses an emotion engine to recognize the emotions of users viewing the delivered content. Specifically, it analyzes the user's facial expressions, voice tone, and body language in real time to identify the user's emotional state. For example, it uses cameras and microphones to detect the user's smiles and reactions indicating interest.

[0988] Dynamic content adjustment

[0989] The server dynamically adjusts the product presentation method and script based on the recognized user's emotions. For example, if the user shows interest, it adds a detailed description of the product; conversely, if the user shows little interest, it switches to a different product.

[0990] View and purchase

[0991] Users select and watch shopping programs available via television or the internet. For example, users can use a remote control or mouse to choose a program featuring products that interest them. If a user is interested in a product, they can purchase it through telephone orders or online shopping sites.

[0992] Specific example:

[0993] Suppose a server collects and analyzes TV shopping program data for a specific detergent over the past six months and discovers that sales are highest during a particular time slot. Based on this information, the server extracts appropriate selling points (e.g., "Powerful cleaning effect with a small amount") and generates a script that emphasizes the detergent's characteristics. Furthermore, based on this information, it generates three video content pieces in English, Japanese, and Chinese, and distributes them optimized for each market. In doing so, it uses an emotion engine to recognize user sentiment, adding detailed explanations if the user shows interest, and switching to other products or approaches if interest is low, thereby achieving even more effective sales promotion.

[0994] Through the steps described above, the system according to the present invention can efficiently produce and distribute multilingual and multicultural television shopping programs and provide customized content based on the user's emotions.

[0995] The following describes the processing flow.

[0996] Step 1:

[0997] The server collects past promotional program data and sales performance data. Specifically, it connects to a database to retrieve recordings of past TV shopping programs, sales figures, sales periods, viewer feedback, and other information.

[0998] Step 2:

[0999] The server cleans the collected data. Specifically, it performs tasks such as filling in missing data, removing outliers, and normalizing the data. For example, it standardizes the format of sales performance data and converts it into a format suitable for analysis algorithms.

[1000] Step 3:

[1001] The server inputs the cleaned data into a machine learning model to identify factors associated with sales performance. Specifically, it uses algorithms such as regression analysis, clustering, and decision trees to analyze how specific time periods or product characteristics influence sales.

[1002] Step 4:

[1003] The server extracts basic information and effective selling points for each product based on the analysis results of a machine learning model. Specifically, it uses text mining techniques to extract keywords from product reviews and sales history to identify features that are effective for sales.

[1004] Step 5:

[1005] The server generates a production script based on the extracted basic information and selling points. In this process, a natural language generation (NLG) algorithm is used to generate text that explains the product's characteristics and benefits.

[1006] Step 6:

[1007] The server provides the generated script to the video generation AI, which then uses AI Cast to create multilingual and multicultural video content. Specifically, it creates videos in multiple languages, such as English, Spanish, and French. AI Cast adds narration and captions corresponding to each language and synthesizes the video and audio.

[1008] Step 7:

[1009] The server optimizes the generated video content to a format suitable for the environment of each viewing region. Specifically, this involves changing the video resolution, converting file formats, and adding subtitles.

[1010] Step 8:

[1011] The server uploads optimized video content to the relevant television stations and online distribution platforms. Specifically, it distributes the content using FTP or APIs.

[1012] Step 9:

[1013] Users select and watch shopping programs available via television or the internet. For example, users can use a remote control or mouse to choose a program that features products they are interested in.

[1014] Step 10:

[1015] The emotion engine recognizes the emotions of users watching the streamed content. Specifically, it analyzes the user's facial expressions, voice tone, and body language in real time to identify the user's emotional state. For example, it can analyze emotions based on the user's camera footage and microphone recordings.

[1016] Step 11:

[1017] The server dynamically adjusts the product presentation method and script based on the recognized user's emotions. For example, if the user shows interest, it adds a detailed description of the product; conversely, if the user shows little interest, it switches to a different product.

[1018] Step 12:

[1019] When a user finds a product they are interested in, they purchase it. Specifically, they place an order via telephone or online shopping site.

[1020] Specific example:

[1021] The server collected and analyzed TV shopping program data for a specific detergent over the past six months and discovered that sales were highest during a particular time slot. Based on this information, the server extracted appropriate selling points (e.g., "Powerful cleaning effect with a small amount") and generated a script that emphasized the detergent's characteristics. Furthermore, it generated three video content versions—English, Japanese, and Chinese—and distributed them optimized for each market. Using an emotion engine, the system analyzes the viewer's emotions in real time, adding detailed explanations of the detergent if they show interest, and switching to other products if they show little interest, thereby achieving more effective sales promotion.

[1022] (Example 2)

[1023] Next, we will describe Example 2. 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."

[1024] In today's world, promotional activities are becoming increasingly diverse and complex. In particular, the creation and distribution of multilingual and multicultural content is difficult to manage effectively and efficiently using traditional methods. Furthermore, while there is a need to recognize audience emotions in real time and customize content accordingly, there is a lack of concrete technological means to achieve this. As a result, there are challenges in fully achieving improved sales efficiency and optimized user experience.

[1025] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1026] In this invention, the server includes means for collecting data from past promotional programs and sales performance data; means for cleaning and analyzing the collected data and extracting basic information and effective selling points of the products being sold; means for generating production scripts based on the analyzed information; means for generating multilingual and multicultural video content based on the scripts; means for optimizing and distributing the generated video content in a format suitable for each viewing region; and means for recognizing the emotions of the viewing users and customizing the content to be distributed accordingly. This enables the effective and efficient production and distribution of multilingual and multicultural video content, and improves sales efficiency and optimizes the user experience through flexible content customization based on the emotions of the viewers.

[1027] "Promotional program data" refers to information such as videos, audio, and text produced for the purpose of advertising products or services.

[1028] "Sales performance data" refers to historical information about the sales of a specific product or service, including sales volume, period, revenue, and customer feedback.

[1029] "Cleaning" is the process of filling in missing data, removing outliers, and normalizing data from a dataset.

[1030] "Analysis" is the process of identifying factors related to sales performance using methods such as machine learning models on collected data.

[1031] "Basic information" refers to fundamental information necessary to understand a product, such as its characteristics, usage instructions, and price.

[1032] A "selling point" is a specific feature or advantage of a product, and it is an important element that makes it appealing to customers.

[1033] A "production script" is a document or scenario created to explain the characteristics and benefits of a product.

[1034] "Multilingual and multicultural video content" refers to videos produced to be compatible with multiple languages ​​and different cultures, and includes narration and captions in each language.

[1035] "Optimization" is the process of converting generated video content into a format suitable for the environment of each viewing region.

[1036] "Means of distribution" refer to platforms and devices used to provide optimized video content to viewers.

[1037] "Recognizing emotions" is the process of analyzing the viewer's facial expressions, tone of voice, and body language in real time to identify the user's emotional state.

[1038] "Customizing" is the process of dynamically adjusting the way a product is presented and the script based on the emotions of the perceived audience.

[1039] The system according to the present invention collects and analyzes data from past promotional programs and sales performance data, extracts basic information and effective selling points of products, generates production scripts based on this information, and ultimately generates and distributes multilingual and multicultural video content. Furthermore, this system incorporates an emotion engine that recognizes user emotions and customizes content based on the recognized emotions. A detailed embodiment of this system is described below.

[1040] Data collection

[1041] The server uses a database management system to collect data from past promotional programs and sales performance data. Specifically, it uses SQL or NoSQL databases to retrieve data such as recordings of past TV shopping programs, sales volume, sales period, and viewer feedback.

[1042] Data analysis

[1043] The server uses data preprocessing tools (e.g., Python's Pandas library) to clean the collected data. Specifically, it performs data imputation, removal of outliers, and data normalization. Next, this cleaned data is input into a machine learning model (e.g., TensorFlow or Scikit-learn) to identify factors related to sales performance. For example, the analysis may reveal that certain time periods are advantageous for sales, or that the characteristics of certain products influence sales performance.

[1044] Extracting selling points

[1045] Based on the analysis results, the server automatically extracts basic information about each product (e.g., product characteristics, usage instructions, price) and effective selling points (e.g., features that result in high customer retention rates). This information is then used to generate subsequent scripts.

[1046] Script generation

[1047] The server generates a production script based on the extracted basic information and selling points. This process uses natural language generation (NLG) algorithms (such as GPT-3) to automatically generate text describing the product's characteristics and benefits. For example, a script highlighting the features of a new cooking utensil might be generated.

[1048] Video content generation

[1049] The server provides the generated script to the video generation AI, which then uses AI Cast to create multilingual and multicultural video content. Specifically, videos in multiple languages, such as English, Spanish, and French, are created simultaneously. AI Cast adds narration and captions corresponding to each language and synthesizes the video and audio.

[1050] Optimization and delivery

[1051] The server optimizes the generated video content into a format suitable for the environment of each viewing region and uploads it to the relevant broadcasters and online distribution platforms. For example, optimizations may be made to provide high-resolution video in certain regions and low-resolution video in other regions.

[1052] Customization using an emotion engine

[1053] The server uses an emotion engine to recognize the emotions of users viewing the delivered content. Specifically, it analyzes the user's facial expressions, voice tone, and body language in real time to identify the user's emotional state. For example, it uses cameras and microphones to detect the user's smiles and reactions indicating interest.

[1054] Dynamic content adjustment

[1055] The server dynamically adjusts the product presentation method and script based on the recognized user's emotions. For example, if the user shows interest, it adds a detailed description of the product; conversely, if the user shows little interest, it switches to a different product.

[1056] View and purchase

[1057] Users select and watch shopping programs available via television or the internet. For example, users can use a remote control or mouse to choose a program featuring products that interest them. If a user is interested in a product, they can purchase it through telephone orders or online shopping sites.

[1058] Specific example

[1059] Suppose a server collects and analyzes TV shopping program data for a specific detergent over the past six months and discovers that sales are highest during a particular time slot. Based on this information, the server extracts appropriate selling points (e.g., "Powerful cleaning effect with a small amount") and generates a script that emphasizes the detergent's characteristics. Furthermore, based on this information, it generates three video content pieces in English, Japanese, and Chinese, and distributes them optimized for each market. In doing so, it uses an emotion engine to recognize user sentiment, adding detailed explanations if the user shows interest, and switching to other products or approaches if interest is low, thereby achieving even more effective sales promotion.

[1060] Example of a prompt

[1061] Collect TV shopping program data for specialty detergents from the past six months and identify the time slots with the highest sales. Next, extract the selling points and generate video scripts in English, Japanese, and Chinese. Use an emotion engine to adjust the content based on user reactions.

[1062] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1063] Step 1: Data Collection

[1064] The server uses a database management system to collect data from past promotional programs and sales performance data. This includes data such as recordings of past TV shopping programs, sales volume, sales period, and viewer feedback, using SQL or NoSQL databases. The input is database connection information, and the output is the collected raw data.

[1065] Step 2: Data Cleaning

[1066] The server uses data preprocessing tools (e.g., Python's Pandas library) to clean the raw data. Specifically, it performs data imputation, removal of outliers, and data normalization. The input is the raw data collected in step 1, and the output is the cleaned data.

[1067] Step 3: Data Analysis

[1068] The server inputs the cleaned data into a machine learning model (e.g., TensorFlow or Scikit-learn) to identify factors associated with sales performance. For example, the analysis might reveal that certain time periods are favorable for sales or that the characteristics of specific products influence sales performance. The input is the cleaned data, and the output is the analysis results (e.g., factors associated with sales performance).

[1069] Step 4: Extracting the selling points

[1070] The server extracts basic information and effective selling points for each product based on the analysis results. Specifically, it targets product characteristics, usage methods, pricing, and features that contribute to high customer retention rates. The input is the analysis results, and the output is the extracted basic information and selling points.

[1071] Step 5: Generate the script

[1072] The server generates a production script based on the extracted basic information and selling points. In this process, a natural language generation (NLG) algorithm (e.g., GPT-3) is used to automatically generate text describing the product's characteristics and benefits. The input is the basic information and selling points, and the output is the generated script.

[1073] Step 6: Generate video content

[1074] The server provides the generated script to the video generation AI, which then uses AI casting to generate multilingual and multicultural video content. Specifically, videos in multiple languages, such as English, Spanish, and French, are created simultaneously. The input is the generated script, and the output is multilingual video content.

[1075] Step 7: Optimization and Delivery

[1076] The server optimizes the generated video content into a format suitable for the environment of each viewing region and uploads it to the relevant broadcasters and online distribution platforms. For example, it may optimize the content by providing high-resolution video in one region and low-resolution video in another. The input is the generated video content, and the output is the optimized video file.

[1077] Step 8: Customization with the emotion engine

[1078] The server uses an emotion engine to recognize the emotions of users viewing the delivered content. Specifically, it analyzes the user's facial expressions, voice tone, and body language in real time to identify the user's emotional state. The input is real-time data of the user during viewing, and the output is the recognized emotional state.

[1079] Step 9: Dynamic Content Adjustment

[1080] The server dynamically adjusts the product presentation method and script based on the recognized user's emotions. For example, if the user shows interest, it adds a detailed description of the product; conversely, if the user shows little interest, it switches to another product. The input is the recognized emotional state, and the output is the adjusted content.

[1081] Step 10: View and Purchase

[1082] Users select and watch shopping programs available via television or the internet. For example, users can use a remote control or mouse to choose a program featuring products they are interested in. If a user is interested, they can purchase the product through telephone ordering or an online shopping site. The input is the user's viewing selection and emotional reaction, and the output is the action of purchasing the product.

[1083] (Application Example 2)

[1084] Next, we will explain application example 2. In the following explanation, 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."

[1085] Current advertising systems lack the functionality to dynamically adjust ad content in response to the emotions of target users. This makes it difficult to maximize advertising effectiveness and capture user interest. Furthermore, automatically generating multilingual and multicultural video content and distributing it in formats suitable for each viewing region is not easy. To address these challenges, a system is needed that can extract effective selling points based on collected data and customize ad content according to user emotions.

[1086] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1087] In this invention, the server includes means for collecting data from past promotional programs and sales performance data; means for analyzing the collected data and extracting basic information and effective selling points of the products being sold; means for generating production scripts based on the analyzed information; means for generating multilingual and multicultural video content based on the scripts; means for optimizing the generated video content into a format suitable for each viewing region and distributing it; means for analyzing user emotions on a smart device; and means for dynamically adjusting advertising content based on the analyzed user emotions. This enables customized advertising that responds to user emotions, thereby maximizing advertising effectiveness.

[1088] A "promotional program" is video content intended to promote the sale of a product or service.

[1089] "Sales performance data" refers to data that includes the quantity of products sold in the past, the sales period, and customer feedback.

[1090] "Data analysis" is the process of organizing collected data and extracting information that is useful for sales.

[1091] "Basic information" refers to fundamental information such as the product's characteristics, usage instructions, and price.

[1092] An "effective selling point" is a feature or characteristic that is particularly effective in promoting the sale of a product.

[1093] A "production script" is a script or scenario used to generate video or advertising content.

[1094] "Multilingual and multicultural support" refers to the function of generating and providing content that is compatible with different languages ​​and cultures.

[1095] "Video content" refers to digital content that includes audio and video.

[1096] A "smart device" is an internet-connected device such as a smartphone, tablet, or smart glasses.

[1097] "User emotion" refers to the psychological state determined from the user's facial expressions, tone of voice, body language, etc.

[1098] "Dynamic adjustment" means changing the content in real time.

[1099] "Advertising content" refers to media content created for the purpose of promoting products or services.

[1100] The system for carrying out this invention is primarily executed by a server. The embodiments for carrying out the invention are described in detail below.

[1101] 1. Data Collection

[1102] The server collects data on past promotional programs and sales performance. This includes recordings of promotional programs, sales volume, sales period, and customer feedback data. The database records numerous promotional programs and their corresponding sales performance.

[1103] 2. Data Analysis

[1104] The server analyzes the collected data. First, it cleans the data, filling in missing data and removing outliers. Next, it uses the cleaned data to analyze factors related to sales performance using a machine learning model. This reveals things like specific time slots or product characteristics that are advantageous for sales.

[1105] 3. Extracting selling points

[1106] The server extracts basic information and effective selling points based on the analysis results. Basic information includes product characteristics, usage, and price. Effective selling points are product features deemed particularly effective for sales promotion.

[1107] 4. Script generation

[1108] The server generates a production script based on the extracted basic information and selling points. Using a natural language generation algorithm, it automatically generates text to describe the product's characteristics and benefits. For example, a script highlighting the features of a new cooking utensil might be generated here.

[1109] 5. Video content generation

[1110] The server generates multilingual and multicultural video content using the generated script. A generation AI model is used to add narration and captions corresponding to each language, and to synthesize the video and audio. For example, the same content can be generated in English, Spanish, French, and other languages.

[1111] 6. Distribution and Optimization

[1112] The server optimizes the generated video content into a format suitable for each viewing region and uploads it to television stations and online distribution platforms. For example, it might provide high-resolution video in certain regions and low-resolution video in others.

[1113] 7. Sentiment Analysis and Dynamic Content Adjustment

[1114] The system analyzes user emotions using smart devices (smartphones and smart glasses). It employs an emotion recognition engine that analyzes the user's facial expressions, voice tone, and body language in real time. Based on the detected emotions, the advertising content is dynamically adjusted. For example, if the user shows interest, a detailed description of the product is added; if interest is low, the system switches to a different product.

[1115] 8. Specific Examples

[1116] The server collects and analyzes promotional program data for a specific detergent over the past six months, discovering that sales are highest during certain time slots. Based on this information, the server extracts "powerful cleaning effect with a small amount" as a selling point and generates a script that emphasizes the detergent's characteristics. Video content is generated in English, Japanese, and Chinese, and distributed optimized for each market. When a user watches this ad on their smartphone and shows interest, a more detailed explanation is added; otherwise, the ad switches to the next product.

[1117] 9. Example of a prompt message

[1118] The next ad you'll show is for a new detergent. Use the following selling points to create a 30-second ad script:

[1119] A small amount provides powerful cleaning.

[1120] Made with environmentally friendly ingredients.

[1121] Effective against all types of dirt.

[1122] Thus, the system for implementing the present invention provides dynamic advertisements that can maximize user interest.

[1123] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1124] Step 1:

[1125] The server collects data from past promotional programs and sales performance data. Inputs include recordings of promotional programs, sales volume, sales period, and customer feedback data. This data is retrieved from a database and prepared for data analysis. The output is a list of the retrieved data.

[1126] Step 2:

[1127] The server analyzes the collected data. It uses collected promotional program data and sales performance data as input. First, it performs data cleaning, including imputing missing data and removing outliers. Next, a machine learning model analyzes factors related to sales performance, identifying the impact of time of day and product characteristics on sales. The output is the analysis results, providing information on specific sales trends and effective selling points.

[1128] Step 3:

[1129] The server extracts basic information and effective selling points based on the analysis results. It uses the analyzed data as input. From the analyzed information, it identifies basic information such as product characteristics, usage, and price, as well as features advantageous for sales promotion. The output is a list of basic information and selling points.

[1130] Step 4:

[1131] The server generates a production script based on the extracted basic information and selling points. It uses a list of basic information and selling points as input. A natural language generation algorithm is used to automatically generate text describing the product's characteristics and benefits. The output is the generated script.

[1132] Step 5:

[1133] The server generates multilingual and multicultural video content using the generated script. It uses the generated script as input. Utilizing a generation AI model, it adds narration and captions corresponding to each language and synthesizes the video and audio. The output is video content compatible with multiple languages.

[1134] Step 6:

[1135] The server optimizes the generated video content into a format suitable for each viewing region and distributes it. It uses the generated video content as input. Technical adjustments, such as high or low resolution, are made to meet the needs of each viewing region before uploading it to television stations and online distribution platforms. The output is the optimized video content.

[1136] Step 7:

[1137] The device analyzes the user's emotions. It uses the user's camera video and audio data as input. Utilizing an emotion recognition engine, it performs facial expression and voice analysis to identify the user's emotional state. The output is the recognized user emotion information.

[1138] Step 8:

[1139] The server dynamically adjusts ad content based on the analyzed user's sentiment. It uses recognized user sentiment information and already generated ad content as input. For example, if the user shows interest, it adds a detailed description; if not, it switches to another product. The output is the adjusted ad content.

[1140] Step 9:

[1141] Users view tailored advertising content. Dynamically adjusted advertising content from the server is used as input, providing the user with the optimal advertising experience. The output is the user's viewing behavior and feedback.

[1142] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1143] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include those described above. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions shown by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1144] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[1145] [Fourth Embodiment]

[1146] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1147] As shown in Figure 7, the 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.

[1148] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1149] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[1150] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1151] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1152] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1153] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive 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 robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[1154] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1155] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[1156] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1157] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1158] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1159] The system according to the present invention collects and analyzes data from past promotional programs and sales performance data, extracts basic information and effective selling points of products, generates production scripts based on this information, and ultimately generates and distributes multilingual and multicultural video content. A detailed embodiment of this system is described below.

[1160] Data collection

[1161] The server connects to databases of past promotional programs and sales performance data to collect necessary data. For example, it retrieves recorded data of past TV shopping programs for a specific product, as well as data such as sales volume, sales period, and viewer feedback.

[1162] Data analysis

[1163] The server analyzes the collected data. First, it cleans the data, filling in missing data and removing outliers. Next, it inputs this cleaned data into a machine learning model to identify factors related to sales performance. For example, the analysis might reveal that certain time periods are advantageous for sales, or that the characteristics of certain products contribute to high sales performance.

[1164] Extracting selling points

[1165] Based on the analysis results, the server automatically extracts basic information about each product (e.g., product characteristics, usage instructions, price) and effective selling points (e.g., features that result in high customer retention rates). This information is then used to generate subsequent scripts.

[1166] Script generation

[1167] The server generates a production script based on the extracted basic information and selling points. In this process, a natural language generation (NLG) algorithm is used to automatically generate text describing the characteristics and benefits of each product. For example, a script highlighting the features of a new cooking utensil might be generated.

[1168] Video content generation

[1169] The server provides the generated script to the video generation AI, which uses AI Cast to create multilingual and multicultural video content. For example, videos in multiple languages, such as English, Spanish, and French, are created simultaneously. AI Cast adds narration and captions corresponding to each language and synthesizes the video and audio.

[1170] Optimization and delivery

[1171] The server optimizes the generated video content into a format suitable for the environment of each viewing region and uploads it to the relevant television station or online distribution platform. For example, optimizations may be made to provide high-resolution video in one region and low-resolution video in another.

[1172] View and purchase

[1173] Users select and watch shopping programs available via television or the internet. For example, users can use a remote control or mouse to choose a program featuring products that interest them. If a user is interested in a product, they can purchase it through telephone orders or online shopping sites.

[1174] Specific example:

[1175] The server collects and analyzes TV shopping program data for a specific detergent over the past six months and discovers that sales are highest during a particular time slot. Based on this information, the server extracts appropriate selling points (e.g., "Powerful cleaning effect with a small amount") and generates a script that emphasizes the detergent's characteristics. Furthermore, based on this information, it generates three video content pieces in English, Japanese, and Chinese, and distributes them optimized for each market.

[1176] Through the steps described above, the system according to the present invention can efficiently produce multilingual and multicultural television shopping programs and distribute them quickly and at low cost.

[1177] The following describes the processing flow.

[1178] Step 1:

[1179] The server connects to databases of past promotional programs and sales performance data to collect necessary data. Specifically, it retrieves recordings of past TV shopping programs, sales figures, sales periods, viewer feedback, and more.

[1180] Step 2:

[1181] The server cleans the collected data. Specifically, it performs tasks such as filling in missing data, removing outliers, and normalizing the data. For example, it standardizes the format of sales performance data and converts it into a format suitable for analysis algorithms.

[1182] Step 3:

[1183] The server inputs the cleaned data into a machine learning model to identify factors associated with sales performance. Specifically, it uses algorithms such as regression analysis, clustering, and decision trees to analyze how specific time periods or product characteristics influence sales.

[1184] Step 4:

[1185] The server extracts basic information and effective selling points for each product based on the analysis results of a machine learning model. Specifically, it uses text mining techniques to extract keywords from product reviews and sales history to identify features that are effective for sales.

[1186] Step 5:

[1187] The server generates a production script based on the extracted basic information and selling points. Specifically, it uses a natural language generation (NLG) algorithm to generate text that explains the product's characteristics and benefits.

[1188] Step 6:

[1189] The server provides the generated script to the video generation AI, which then uses AI Cast to create multilingual and multicultural video content. Specifically, AI Cast adds narration and captions corresponding to each language and synthesizes the video and audio.

[1190] Step 7:

[1191] The server optimizes the generated video content for each viewing region. Specifically, this involves changing the video resolution, converting file formats, and adding subtitles.

[1192] Step 8:

[1193] The server uploads optimized video content to the relevant television stations and online distribution platforms. Specifically, it distributes the content using FTP or APIs.

[1194] Step 9:

[1195] Users select and watch shopping programs available via television or the internet. Specifically, they select programs using a remote control or mouse.

[1196] Step 10:

[1197] When a user finds a product they are interested in, they purchase it. Specifically, they place an order via telephone or online shopping site.

[1198] Through the steps described above, the system according to the present invention can effectively produce and distribute multilingual and multicultural television shopping programs.

[1199] (Example 1)

[1200] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1201] In today's consumer market, to improve the efficiency of sales promotion, it is necessary to effectively utilize past promotional data and sales performance data to provide rapid and effective marketing tools. However, conventional systems have struggled to automate all processes from data collection, cleaning, and analysis to script generation, video creation, and distribution, as well as to efficiently produce and distribute multilingual and multicultural video content.

[1202] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1203] In this invention, the server includes means for collecting data from past promotional programs and sales performance data; means for cleaning, supplementing, and analyzing the collected data; means for extracting basic information and effective selling points of products from the analyzed data; means for generating production scripts based on the extracted information; means for generating multilingual and multicultural video content based on the scripts; and means for optimizing the generated video content into a format suitable for each viewing region and distributing it. This enables the efficient production and distribution of multilingual and multicultural video content.

[1204] "Promotional program data" refers to recorded data and related information of programs that were previously broadcast for the purpose of sales promotion.

[1205] "Sales performance data" refers to data related to actual sales activities, such as the quantity of products sold, the sales period, and viewer feedback.

[1206] "Cleaning" refers to the process of filling in missing values ​​and removing outliers from data.

[1207] A "machine learning model" refers to an algorithm or statistical model used for data analysis, designed to learn patterns and regularities from data.

[1208] A "selling point" refers to a product's characteristics, advantages, or effective elements for sales promotion.

[1209] A "natural language generation algorithm" is an algorithm that generates sentences in natural language based on extracted information.

[1210] "Multilingual and multicultural video content" refers to content that can generate video material that supports multiple languages ​​and cultures.

[1211] "Optimization" refers to the process of adjusting generated video content to suit the environment and requirements of each viewing region.

[1212] "Distribution" refers to the act of delivering generated video content to viewers online or through television stations.

[1213] The system according to the present invention collects and analyzes data from past promotional programs and sales performance data, extracts basic information and effective selling points of products, generates production scripts based on this information, and ultimately generates and distributes multilingual and multicultural video content. A detailed embodiment of this system is described below.

[1214] Data collection

[1215] The server connects to databases of past promotional programs and sales performance data to collect necessary data about a specific product. This includes recording data, sales volume, sales period, and viewer feedback. Specifically, the server uses SQL queries to retrieve the information.

[1216] Data analysis

[1217] The server cleans the collected data, imputing missing data and removing outliers. This cleaned data is then fed into a machine learning model to identify factors associated with sales performance. The Scikit-learn library in Python is used as the machine learning model.

[1218] Extracting selling points

[1219] Based on the analysis results, the server automatically extracts basic information about each product (e.g., product characteristics, usage instructions, price) and effective selling points (e.g., "powerful cleaning effect with a small amount").

[1220] Script generation

[1221] The server generates a production script using a natural language generation (NLG) algorithm based on the extracted basic information and selling points. Specifically, it uses OpenAI's GPT-3 to automatically generate text that describes the product's characteristics and advantages.

[1222] Video content generation

[1223] The server provides the generated script to the video generation AI, which then uses AI casting to create multilingual and multicultural video content. For example, it adds narration and captions in multiple languages, such as English, Spanish, and Japanese, and synthesizes the video and audio.

[1224] Optimization and delivery

[1225] The server optimizes and delivers the generated video content to the environment of each viewing region. Specifically, optimizations are made such as providing high-resolution video to some regions and low-resolution video to others. The optimized video is then uploaded to the relevant television stations and online distribution platforms.

[1226] View and purchase

[1227] Users can select and watch shopping programs delivered via television or the internet. For example, they can use a remote control or mouse to choose a program featuring products they are interested in, and then purchase the products by phone or through an online shopping site after watching the program.

[1228] Specific example

[1229] The server collected and analyzed TV shopping program data for a specific detergent over the past six months, revealing the highest sales volume during a particular time slot. Based on this information, it extracted the selling point "powerful cleaning effect with a small amount" and generated a script that emphasized the detergent's characteristics. Furthermore, based on this information, it generated three video content pieces in English, Japanese, and Chinese, and distributed them optimized for each market.

[1230] Through the steps described above, the system according to the present invention can efficiently produce multilingual and multicultural television shopping programs and distribute them quickly and at low cost.

[1231] Example of a prompt

[1232] Analyze TV shopping program data for a specific detergent from the past six months and extract its selling points. Then, generate scripts in English, Japanese, and Chinese based on those selling points. Please emphasize the detergent's characteristics and effects.

[1233] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1234] Step 1:

[1235] Data collection

[1236] The server connects to a database of promotional programs and a database of sales performance, collecting data such as recordings of past promotional programs, sales figures, sales periods, and viewer feedback.

[1237] Input: Information of the target database

[1238] Data processing: Extracting data using SQL queries.

[1239] Output: Collected dataset

[1240] Specific operation: The server executes an SQL query to retrieve video recording data and sales performance data for a specific product for the past six months.

[1241] Step 2:

[1242] Data cleaning

[1243] The server cleans the collected data, fills in missing data, and removes outliers.

[1244] Input: Collected dataset

[1245] Data processing: Imputation of missing data (imputation with mean values), removal of outliers.

[1246] Output: Cleaned dataset

[1247] Specific operation: Run a Python script to remove abnormal sales quantity data from the dataset and impute missing values ​​with the mean.

[1248] Step 3:

[1249] Data analysis

[1250] The server inputs the cleaned data into a machine learning model to identify factors associated with sales performance.

[1251] Input: Cleaned dataset

[1252] Data processing: Input data into a machine learning model and extract important features.

[1253] Output: Sales performance and related factors (specific time periods and product characteristics)

[1254] Specific operation: The server uses the Python Scikit-learn library to analyze specific time periods and product characteristics that contribute to sales performance using a model.

[1255] Step 4:

[1256] Extracting selling points

[1257] Based on the analysis results, the server automatically extracts basic product information and effective selling points.

[1258] Input: Sales performance and related factors

[1259] Data processing: Extraction of basic information, identification of effective selling points.

[1260] Output: Basic product information and selling points

[1261] Specific operation: Based on the extracted information, the server automatically identifies and extracts selling points such as "powerful cleaning effect with a small amount."

[1262] Step 5:

[1263] Script generation

[1264] The server generates a production script using a natural language generation (NLG) algorithm based on the extracted basic information and selling points.

[1265] Input: Basic product information and selling points

[1266] Data processing: Text generation using NLG algorithm

[1267] Output: Completed script

[1268] Specific operation: The server uses OpenAI's GPT-3 to generate a script that says, "This detergent has a powerful cleaning effect even in small amounts."

[1269] Step 6:

[1270] Video content generation

[1271] The server provides the generated script to the video generation AI, which then uses AI casting to generate multilingual and multicultural video content.

[1272] Input: Completed script

[1273] Data processing: Video and audio synthesis using AI casting.

[1274] Output: Multilingual and multicultural video content

[1275] Specific operation: The server generates videos based on the English, Japanese, and Chinese scripts, adding narration and captions in each language.

[1276] Step 7:

[1277] Optimization and delivery

[1278] The server optimizes the generated video content into a format suitable for the environment of each viewing region and then delivers it.

[1279] Input: Multilingual and multicultural video content

[1280] Data processing: Optimized for resolution and format suitable for each viewing region.

[1281] Output: Optimized video content

[1282] Specific operation: The server optimizes high-resolution videos for North America and low-resolution videos for regions with different internet speeds, and then uploads them to the distribution platform.

[1283] Step 8:

[1284] View and purchase

[1285] Users select shopping programs via television or the internet and then purchase products using telephone orders or online shopping after watching them.

[1286] Input: Optimized video content

[1287] Data processing: None

[1288] Output: Purchased items

[1289] Specific operation: The user uses the remote control to select a program about "special detergents," and after watching it, places a phone order or purchases the product from an online shopping site.

[1290] (Application Example 1)

[1291] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1292] Current advertising campaigns lack effective ways to leverage historical data and optimize in real time. Furthermore, generating and distributing multilingual video ads tailored to different regions and cultures is time-consuming and requires significant resources and time. In addition, there's a lack of mechanisms to quickly incorporate user feedback into subsequent advertising campaigns. A system is needed to address these challenges and maximize advertising effectiveness.

[1293] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1294] In this invention, the server includes means for collecting data from past promotional programs and sales performance data; means for analyzing the collected data and extracting basic information and effective selling points of the products being sold; means for generating production scripts based on the analyzed information; means for generating multilingual and multicultural video content based on the scripts; means for optimizing the generated video content into a format suitable for each viewing region and distributing it; means for collecting data from past advertising campaigns via smart devices and analyzing it in real time; means for automatically extracting effective advertising elements to generate advertising scripts; means for automatically generating multilingual video advertisements based on the generated advertising scripts; and means for collecting user feedback and using it to optimize the next advertising campaign. This makes it possible to effectively utilize past advertising data, generate and distribute optimized multilingual video advertisements in real time, and quickly reflect feedback.

[1295] "Data from past promotional programs" refers to information such as video, audio, and text related to promotions that were broadcast in the past.

[1296] "Sales performance data" refers to data related to the actual sales performance of a product, such as sales volume, sales period, and sales figures.

[1297] A "smart device" refers to a portable device with internet connectivity, such as a smartphone, smart glasses, or tablet.

[1298] "Real-time analysis" means performing analysis immediately the moment data is collected.

[1299] A "production script" is a text document that provides instructions regarding the content and flow of a video.

[1300] "Multilingual and multicultural video content" refers to videos produced in formats adapted to multiple languages ​​and cultures.

[1301] An "ad script" is text used to explain the content and selling points of an advertising video.

[1302] "Past advertising campaign data" refers to information such as viewing history, responses, and click-through rates of advertisements that were previously run.

[1303] "Optimizing and distributing content in a format suitable for each viewing region" means converting video content to a quality and format appropriate for each region and uploading it to the distribution platform.

[1304] "Automatically extracting effective advertising elements" means that the system automatically identifies and extracts elements (e.g., catchphrases and colors) that maximize the effectiveness of an advertisement.

[1305] "Collecting user feedback and using it to optimize the next advertising campaign" means collecting user reactions and evaluations as data and incorporating them into the production of future advertisements.

[1306] The system according to the present invention aims to optimize advertising campaigns and is configured as follows.

[1307] Data collection

[1308] The server first collects data from past promotional programs and sales performance data. Simultaneously, it also collects data such as viewing history and click-through rates of past advertising campaigns via smart devices.

[1309] Data Analysis

[1310] The collected data is analyzed on the server. Machine learning models are used for data analysis to identify factors that maximize advertising effectiveness. Specifically, algorithms such as random forests are used to analyze factors related to sales performance.

[1311] Extracting selling points

[1312] Based on the analysis results, the server automatically extracts basic information and effective selling points for each product. During this process, effective advertising elements such as colors and catchphrases are identified.

[1313] Script generation

[1314] The server generates advertising scripts based on extracted basic information and selling points. Using natural language generation (NLG) algorithms, it generates text that effectively explains the characteristics and benefits of each product. These scripts are generated in multiple languages ​​for multilingual support.

[1315] Video content generation

[1316] Based on the generated ad script, the server automatically creates multilingual video ads. Video editing libraries such as moviepy are used for video generation. The generated videos are saved in a format adapted to each viewing region.

[1317] Optimization and delivery

[1318] The server optimizes the generated video content into a format suitable for the environment of each viewing region and uploads it to the video distribution platform. For example, it adjusts high-resolution and low-resolution videos for each region.

[1319] Gathering feedback and optimizing future ads

[1320] Users view advertising campaigns and provide feedback. The server collects this feedback in real time and uses it to optimize the next advertising campaign.

[1321] Specific example

[1322] For example, by analyzing data from TV shopping programs that showed high effectiveness during specific time slots, it was found that certain colors (e.g., red) had a high click-through rate. Based on this information, a script and video featuring product introductions with a predominantly red background were generated and distributed in multiple languages.

[1323] Example of a prompt

[1324] "Please explain in detail how to extract the most effective advertising selling points based on past advertising campaign data and sales performance data, and how to generate multilingual advertising content. In particular, please explain the specific analysis procedures and generation process, focusing on the differences in effectiveness based on color and time of day."

[1325] The system according to the present invention makes it possible to effectively utilize past advertising data, generate and deliver optimized multilingual video advertisements in real time, and quickly incorporate feedback.

[1326] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1327] Step 1:

[1328] The server collects data from past promotional programs and sales performance data. Specifically, it retrieves recorded data of TV shopping programs, sales volume, sales period, and viewer feedback from the database. The data collected also includes viewing history and click-through rates of past advertising campaigns. The input is data from the database, and the output is the collected data to be analyzed.

[1329] Step 2:

[1330] The server analyzes the collected data. The data is first cleaned, including the imputation of missing data and the removal of outliers. Next, this cleaned data is input into a machine learning model (e.g., a random forest) to identify factors associated with sales performance. The input is the cleaned data, and the output is the factors associated with sales performance.

[1331] Step 3:

[1332] The server automatically extracts basic information and effective selling points for each product based on the analysis results. Specifically, it uses a machine learning model's importance ranking to identify features with high impact. The input is the analysis results, and the output is the extracted basic information and selling points.

[1333] Step 4:

[1334] The server generates advertising scripts based on the extracted basic information and selling points. It automatically generates text describing the characteristics and benefits of each product using a natural language generation (NLG) algorithm. The input is basic information and selling points, and the output is the generated script.

[1335] Step 5:

[1336] The server generates multilingual video ads based on the generated script. Video editing libraries such as moviepy are used for video generation, and narration and captions are added based on the script. The input is the generated script, and the output is the video ad.

[1337] Step 6:

[1338] The server optimizes the generated video content to a format suitable for the environment of each viewing region. For example, it might provide high-resolution video in one region and low-resolution video in another. The input is a video advertisement, and the output is the optimized video content.

[1339] Step 7:

[1340] The server uploads optimized video content to video distribution platforms. It optimizes the video to match the different specifications of each distribution platform and then uploads it. The input is the optimized video content, and the output is the completion of the upload to each platform.

[1341] Step 8:

[1342] Users view advertising campaigns and provide feedback. For example, they might watch an ad on a smart device and enter ratings and comments within the application. The input is the user's feedback, and the output is the collected feedback data.

[1343] Step 9:

[1344] The server analyzes user feedback and uses it to optimize the next advertising campaign. It analyzes feedback data in real time and incorporates it into the generation of future ad scripts and videos. The input is feedback data, and the output is ideas for improving the next ad campaign.

[1345] This series of processing steps allows for the effective use of past advertising data, the real-time generation and delivery of optimized, multilingual video ads, and the rapid incorporation of feedback.

[1346] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1347] The system according to the present invention collects and analyzes data from past promotional programs and sales performance data, extracts basic information and effective selling points of products, generates production scripts based on this information, and ultimately generates and distributes multilingual and multicultural video content. Furthermore, this system incorporates an emotion engine that recognizes user emotions and customizes content based on the recognized emotions. A detailed embodiment of this system is described below.

[1348] Data collection

[1349] The server connects to databases of past promotional programs and sales performance data to collect necessary data. For example, it retrieves recorded data of past TV shopping programs, as well as data such as sales volume, sales period, and viewer feedback.

[1350] Data analysis

[1351] The server cleans the collected data. Specifically, it imputes missing data, removes outliers, and normalizes the data. Next, this cleaned data is input into a machine learning model to identify factors related to sales performance. For example, the analysis may reveal that certain time periods are advantageous for sales, or that the characteristics of specific products influence sales performance.

[1352] Extracting selling points

[1353] Based on the analysis results, the server automatically extracts basic information about each product (e.g., product characteristics, usage instructions, price) and effective selling points (e.g., features that result in high customer retention rates). This information is then used to generate subsequent scripts.

[1354] Script generation

[1355] The server generates a production script based on the extracted basic information and selling points. In this process, a natural language generation (NLG) algorithm is used to automatically generate text describing the product's characteristics and benefits. For example, a script highlighting the features of a new cooking utensil might be generated.

[1356] Video content generation

[1357] The server provides the generated script to the video generation AI, which then uses AI Cast to create multilingual and multicultural video content. Specifically, videos in multiple languages, such as English, Spanish, and French, are created simultaneously. AI Cast adds narration and captions corresponding to each language and synthesizes the video and audio.

[1358] Optimization and delivery

[1359] The server optimizes the generated video content into a format suitable for the environment of each viewing region and uploads it to the relevant television station or online distribution platform. For example, optimizations may be made to provide high-resolution video in one region and low-resolution video in another.

[1360] Customization using an emotion engine

[1361] The server uses an emotion engine to recognize the emotions of users viewing the delivered content. Specifically, it analyzes the user's facial expressions, voice tone, and body language in real time to identify the user's emotional state. For example, it uses cameras and microphones to detect the user's smiles and reactions indicating interest.

[1362] Dynamic content adjustment

[1363] The server dynamically adjusts the product presentation method and script based on the recognized user's emotions. For example, if the user shows interest, it adds a detailed description of the product; conversely, if the user shows little interest, it switches to a different product.

[1364] View and purchase

[1365] Users select and watch shopping programs available via television or the internet. For example, users can use a remote control or mouse to choose a program featuring products that interest them. If a user is interested in a product, they can purchase it through telephone orders or online shopping sites.

[1366] Specific example:

[1367] Suppose a server collects and analyzes TV shopping program data for a specific detergent over the past six months and discovers that sales are highest during a particular time slot. Based on this information, the server extracts appropriate selling points (e.g., "Powerful cleaning effect with a small amount") and generates a script that emphasizes the detergent's characteristics. Furthermore, based on this information, it generates three video content pieces in English, Japanese, and Chinese, and distributes them optimized for each market. In doing so, it uses an emotion engine to recognize user sentiment, adding detailed explanations if the user shows interest, and switching to other products or approaches if interest is low, thereby achieving even more effective sales promotion.

[1368] Through the steps described above, the system according to the present invention can efficiently produce and distribute multilingual and multicultural television shopping programs and provide customized content based on the user's emotions.

[1369] The following describes the processing flow.

[1370] Step 1:

[1371] The server collects past promotional program data and sales performance data. Specifically, it connects to a database to retrieve recordings of past TV shopping programs, sales figures, sales periods, viewer feedback, and other information.

[1372] Step 2:

[1373] The server cleans the collected data. Specifically, it performs tasks such as filling in missing data, removing outliers, and normalizing the data. For example, it standardizes the format of sales performance data and converts it into a format suitable for analysis algorithms.

[1374] Step 3:

[1375] The server inputs the cleaned data into a machine learning model to identify factors associated with sales performance. Specifically, it uses algorithms such as regression analysis, clustering, and decision trees to analyze how specific time periods or product characteristics influence sales.

[1376] Step 4:

[1377] The server extracts basic information and effective selling points for each product based on the analysis results of a machine learning model. Specifically, it uses text mining techniques to extract keywords from product reviews and sales history to identify features that are effective for sales.

[1378] Step 5:

[1379] The server generates a production script based on the extracted basic information and selling points. In this process, a natural language generation (NLG) algorithm is used to generate text that explains the product's characteristics and benefits.

[1380] Step 6:

[1381] The server provides the generated script to the video generation AI, which then uses AI Cast to create multilingual and multicultural video content. Specifically, it creates videos in multiple languages, such as English, Spanish, and French. AI Cast adds narration and captions corresponding to each language and synthesizes the video and audio.

[1382] Step 7:

[1383] The server optimizes the generated video content to a format suitable for the environment of each viewing region. Specifically, this involves changing the video resolution, converting file formats, and adding subtitles.

[1384] Step 8:

[1385] The server uploads optimized video content to the relevant television stations and online distribution platforms. Specifically, it distributes the content using FTP or APIs.

[1386] Step 9:

[1387] Users select and watch shopping programs available via television or the internet. For example, users can use a remote control or mouse to choose a program that features products they are interested in.

[1388] Step 10:

[1389] The emotion engine recognizes the emotions of users watching the streamed content. Specifically, it analyzes the user's facial expressions, voice tone, and body language in real time to identify the user's emotional state. For example, it can analyze emotions based on the user's camera footage and microphone recordings.

[1390] Step 11:

[1391] The server dynamically adjusts the product presentation method and script based on the recognized user's emotions. For example, if the user shows interest, it adds a detailed description of the product; conversely, if the user shows little interest, it switches to a different product.

[1392] Step 12:

[1393] When a user finds a product they are interested in, they purchase it. Specifically, they place an order via telephone or online shopping site.

[1394] Specific example:

[1395] The server collected and analyzed TV shopping program data for a specific detergent over the past six months and discovered that sales were highest during a particular time slot. Based on this information, the server extracted appropriate selling points (e.g., "Powerful cleaning effect with a small amount") and generated a script that emphasized the detergent's characteristics. Furthermore, it generated three video content versions—English, Japanese, and Chinese—and distributed them optimized for each market. Using an emotion engine, the system analyzes the viewer's emotions in real time, adding detailed explanations of the detergent if they show interest, and switching to other products if they show little interest, thereby achieving more effective sales promotion.

[1396] (Example 2)

[1397] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1398] In today's world, promotional activities are becoming increasingly diverse and complex. In particular, the creation and distribution of multilingual and multicultural content is difficult to manage effectively and efficiently using traditional methods. Furthermore, while there is a need to recognize audience emotions in real time and customize content accordingly, there is a lack of concrete technological means to achieve this. As a result, there are challenges in fully achieving improved sales efficiency and optimized user experience.

[1399] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1400] In this invention, the server includes means for collecting data from past promotional programs and sales performance data; means for cleaning and analyzing the collected data and extracting basic information and effective selling points of the products being sold; means for generating production scripts based on the analyzed information; means for generating multilingual and multicultural video content based on the scripts; means for optimizing and distributing the generated video content in a format suitable for each viewing region; and means for recognizing the emotions of the viewing users and customizing the content to be distributed accordingly. This enables the effective and efficient production and distribution of multilingual and multicultural video content, and improves sales efficiency and optimizes the user experience through flexible content customization based on the emotions of the viewers.

[1401] "Promotional program data" refers to information such as videos, audio, and text produced for the purpose of advertising products or services.

[1402] "Sales performance data" refers to historical information about the sales of a specific product or service, including sales volume, period, revenue, and customer feedback.

[1403] "Cleaning" is the process of filling in missing data, removing outliers, and normalizing data from a dataset.

[1404] "Analysis" is the process of identifying factors related to sales performance using methods such as machine learning models on collected data.

[1405] "Basic information" refers to fundamental information necessary to understand a product, such as its characteristics, usage instructions, and price.

[1406] A "selling point" is a specific feature or advantage of a product, and it is an important element that makes it appealing to customers.

[1407] A "production script" is a document or scenario created to explain the characteristics and benefits of a product.

[1408] "Multilingual and multicultural video content" refers to videos produced to be compatible with multiple languages ​​and different cultures, and includes narration and captions in each language.

[1409] "Optimization" is the process of converting generated video content into a format suitable for the environment of each viewing region.

[1410] "Means of distribution" refer to platforms and devices used to provide optimized video content to viewers.

[1411] "Recognizing emotions" is the process of analyzing the viewer's facial expressions, tone of voice, and body language in real time to identify the user's emotional state.

[1412] "Customizing" is the process of dynamically adjusting the way a product is presented and the script based on the emotions of the perceived audience.

[1413] The system according to the present invention collects and analyzes data from past promotional programs and sales performance data, extracts basic information and effective selling points of products, generates production scripts based on this information, and ultimately generates and distributes multilingual and multicultural video content. Furthermore, this system incorporates an emotion engine that recognizes user emotions and customizes content based on the recognized emotions. A detailed embodiment of this system is described below.

[1414] Data collection

[1415] The server uses a database management system to collect data from past promotional programs and sales performance data. Specifically, it uses SQL or NoSQL databases to retrieve data such as recordings of past TV shopping programs, sales volume, sales period, and viewer feedback.

[1416] Data analysis

[1417] The server uses data preprocessing tools (e.g., Python's Pandas library) to clean the collected data. Specifically, it performs data imputation, removal of outliers, and data normalization. Next, this cleaned data is input into a machine learning model (e.g., TensorFlow or Scikit-learn) to identify factors related to sales performance. For example, the analysis may reveal that certain time periods are advantageous for sales, or that the characteristics of certain products influence sales performance.

[1418] Extracting selling points

[1419] Based on the analysis results, the server automatically extracts basic information about each product (e.g., product characteristics, usage instructions, price) and effective selling points (e.g., features that result in high customer retention rates). This information is then used to generate subsequent scripts.

[1420] Script generation

[1421] The server generates a production script based on the extracted basic information and selling points. This process uses natural language generation (NLG) algorithms (such as GPT-3) to automatically generate text describing the product's characteristics and benefits. For example, a script highlighting the features of a new cooking utensil might be generated.

[1422] Video content generation

[1423] The server provides the generated script to the video generation AI, which then uses AI Cast to create multilingual and multicultural video content. Specifically, videos in multiple languages, such as English, Spanish, and French, are created simultaneously. AI Cast adds narration and captions corresponding to each language and synthesizes the video and audio.

[1424] Optimization and delivery

[1425] The server optimizes the generated video content into a format suitable for the environment of each viewing region and uploads it to the relevant broadcasters and online distribution platforms. For example, optimizations may be made to provide high-resolution video in certain regions and low-resolution video in other regions.

[1426] Customization using an emotion engine

[1427] The server uses an emotion engine to recognize the emotions of users viewing the delivered content. Specifically, it analyzes the user's facial expressions, voice tone, and body language in real time to identify the user's emotional state. For example, it uses cameras and microphones to detect the user's smiles and reactions indicating interest.

[1428] Dynamic content adjustment

[1429] The server dynamically adjusts the product presentation method and script based on the recognized user's emotions. For example, if the user shows interest, it adds a detailed description of the product; conversely, if the user shows little interest, it switches to a different product.

[1430] View and purchase

[1431] Users select and watch shopping programs available via television or the internet. For example, users can use a remote control or mouse to choose a program featuring products that interest them. If a user is interested in a product, they can purchase it through telephone orders or online shopping sites.

[1432] Specific example

[1433] Suppose a server collects and analyzes TV shopping program data for a specific detergent over the past six months and discovers that sales are highest during a particular time slot. Based on this information, the server extracts appropriate selling points (e.g., "Powerful cleaning effect with a small amount") and generates a script that emphasizes the detergent's characteristics. Furthermore, based on this information, it generates three video content pieces in English, Japanese, and Chinese, and distributes them optimized for each market. In doing so, it uses an emotion engine to recognize user sentiment, adding detailed explanations if the user shows interest, and switching to other products or approaches if interest is low, thereby achieving even more effective sales promotion.

[1434] Example of a prompt

[1435] Collect TV shopping program data for specialty detergents from the past six months and identify the time slots with the highest sales. Next, extract the selling points and generate video scripts in English, Japanese, and Chinese. Use an emotion engine to adjust the content based on user reactions.

[1436] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1437] Step 1: Data Collection

[1438] The server uses a database management system to collect data from past promotional programs and sales performance data. This includes data such as recordings of past TV shopping programs, sales volume, sales period, and viewer feedback, using SQL or NoSQL databases. The input is database connection information, and the output is the collected raw data.

[1439] Step 2: Data Cleaning

[1440] The server uses data preprocessing tools (e.g., Python's Pandas library) to clean the raw data. Specifically, it performs data imputation, removal of outliers, and data normalization. The input is the raw data collected in step 1, and the output is the cleaned data.

[1441] Step 3: Data Analysis

[1442] The server inputs the cleaned data into a machine learning model (e.g., TensorFlow or Scikit-learn) to identify factors associated with sales performance. For example, the analysis might reveal that certain time periods are favorable for sales or that the characteristics of specific products influence sales performance. The input is the cleaned data, and the output is the analysis results (e.g., factors associated with sales performance).

[1443] Step 4: Extracting the selling points

[1444] The server extracts basic information and effective selling points for each product based on the analysis results. Specifically, it targets product characteristics, usage methods, pricing, and features that contribute to high customer retention rates. The input is the analysis results, and the output is the extracted basic information and selling points.

[1445] Step 5: Generate the script

[1446] The server generates a production script based on the extracted basic information and selling points. In this process, a natural language generation (NLG) algorithm (e.g., GPT-3) is used to automatically generate text describing the product's characteristics and benefits. The input is the basic information and selling points, and the output is the generated script.

[1447] Step 6: Generate video content

[1448] The server provides the generated script to the video generation AI, which then uses AI casting to generate multilingual and multicultural video content. Specifically, videos in multiple languages, such as English, Spanish, and French, are created simultaneously. The input is the generated script, and the output is multilingual video content.

[1449] Step 7: Optimization and Delivery

[1450] The server optimizes the generated video content into a format suitable for the environment of each viewing region and uploads it to the relevant broadcasters and online distribution platforms. For example, it may optimize the content by providing high-resolution video in one region and low-resolution video in another. The input is the generated video content, and the output is the optimized video file.

[1451] Step 8: Customization with the emotion engine

[1452] The server uses an emotion engine to recognize the emotions of users viewing the delivered content. Specifically, it analyzes the user's facial expressions, voice tone, and body language in real time to identify the user's emotional state. The input is real-time data of the user during viewing, and the output is the recognized emotional state.

[1453] Step 9: Dynamic Content Adjustment

[1454] The server dynamically adjusts the product presentation method and script based on the recognized user's emotions. For example, if the user shows interest, it adds a detailed description of the product; conversely, if the user shows little interest, it switches to another product. The input is the recognized emotional state, and the output is the adjusted content.

[1455] Step 10: View and Purchase

[1456] Users select and watch shopping programs available via television or the internet. For example, users can use a remote control or mouse to choose a program featuring products they are interested in. If a user is interested, they can purchase the product through telephone ordering or an online shopping site. The input is the user's viewing selection and emotional reaction, and the output is the action of purchasing the product.

[1457] (Application Example 2)

[1458] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1459] Current advertising systems lack the functionality to dynamically adjust ad content in response to the emotions of target users. This makes it difficult to maximize advertising effectiveness and capture user interest. Furthermore, automatically generating multilingual and multicultural video content and distributing it in formats suitable for each viewing region is not easy. To address these challenges, a system is needed that can extract effective selling points based on collected data and customize ad content according to user emotions.

[1460] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1461] In this invention, the server includes means for collecting data from past promotional programs and sales performance data; means for analyzing the collected data and extracting basic information and effective selling points of the products being sold; means for generating production scripts based on the analyzed information; means for generating multilingual and multicultural video content based on the scripts; means for optimizing the generated video content into a format suitable for each viewing region and distributing it; means for analyzing user emotions on a smart device; and means for dynamically adjusting advertising content based on the analyzed user emotions. This enables customized advertising that responds to user emotions, thereby maximizing advertising effectiveness.

[1462] A "promotional program" is video content intended to promote the sale of a product or service.

[1463] "Sales performance data" refers to data that includes the quantity of products sold in the past, the sales period, and customer feedback.

[1464] "Data analysis" is the process of organizing collected data and extracting information that is useful for sales.

[1465] "Basic information" refers to fundamental information such as the product's characteristics, usage instructions, and price.

[1466] An "effective selling point" is a feature or characteristic that is particularly effective in promoting the sale of a product.

[1467] A "production script" is a script or scenario used to generate video or advertising content.

[1468] "Multilingual and multicultural support" refers to the function of generating and providing content that is compatible with different languages ​​and cultures.

[1469] "Video content" refers to digital content that includes audio and video.

[1470] A "smart device" is an internet-connected device such as a smartphone, tablet, or smart glasses.

[1471] "User emotion" refers to the psychological state determined from the user's facial expressions, tone of voice, body language, etc.

[1472] "Dynamic adjustment" means changing the content in real time.

[1473] "Advertising content" refers to media content created for the purpose of promoting products or services.

[1474] The system for carrying out this invention is primarily executed by a server. The embodiments for carrying out the invention are described in detail below.

[1475] 1. Data Collection

[1476] The server collects data on past promotional programs and sales performance. This includes recordings of promotional programs, sales volume, sales period, and customer feedback data. The database records numerous promotional programs and their corresponding sales performance.

[1477] 2. Data Analysis

[1478] The server analyzes the collected data. First, it cleans the data, filling in missing data and removing outliers. Next, it uses the cleaned data to analyze factors related to sales performance using a machine learning model. This reveals things like specific time slots or product characteristics that are advantageous for sales.

[1479] 3. Extracting selling points

[1480] The server extracts basic information and effective selling points based on the analysis results. Basic information includes product characteristics, usage, and price. Effective selling points are product features deemed particularly effective for sales promotion.

[1481] 4. Script generation

[1482] The server generates a production script based on the extracted basic information and selling points. Using a natural language generation algorithm, it automatically generates text to describe the product's characteristics and benefits. For example, a script highlighting the features of a new cooking utensil might be generated here.

[1483] 5. Video content generation

[1484] The server generates multilingual and multicultural video content using the generated script. A generation AI model is used to add narration and captions corresponding to each language, and to synthesize the video and audio. For example, the same content can be generated in English, Spanish, French, and other languages.

[1485] 6. Distribution and Optimization

[1486] The server optimizes the generated video content into a format suitable for each viewing region and uploads it to television stations and online distribution platforms. For example, it might provide high-resolution video in certain regions and low-resolution video in others.

[1487] 7. Sentiment Analysis and Dynamic Content Adjustment

[1488] The system analyzes user emotions using smart devices (smartphones and smart glasses). It employs an emotion recognition engine that analyzes the user's facial expressions, voice tone, and body language in real time. Based on the detected emotions, the advertising content is dynamically adjusted. For example, if the user shows interest, a detailed description of the product is added; if interest is low, the system switches to a different product.

[1489] 8. Specific Examples

[1490] The server collects and analyzes promotional program data for a specific detergent over the past six months, discovering that sales are highest during certain time slots. Based on this information, the server extracts "powerful cleaning effect with a small amount" as a selling point and generates a script that emphasizes the detergent's characteristics. Video content is generated in English, Japanese, and Chinese, and distributed optimized for each market. When a user watches this ad on their smartphone and shows interest, a more detailed explanation is added; otherwise, the ad switches to the next product.

[1491] 9. Example of a prompt message

[1492] The next ad you'll show is for a new detergent. Use the following selling points to create a 30-second ad script:

[1493] A small amount provides powerful cleaning.

[1494] Made with environmentally friendly ingredients.

[1495] Effective against all types of dirt.

[1496] Thus, the system for implementing the present invention provides dynamic advertisements that can maximize user interest.

[1497] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1498] Step 1:

[1499] The server collects data from past promotional programs and sales performance data. Inputs include recordings of promotional programs, sales volume, sales period, and customer feedback data. This data is retrieved from a database and prepared for data analysis. The output is a list of the retrieved data.

[1500] Step 2:

[1501] The server analyzes the collected data. It uses collected promotional program data and sales performance data as input. First, it performs data cleaning, including imputing missing data and removing outliers. Next, a machine learning model analyzes factors related to sales performance, identifying the impact of time of day and product characteristics on sales. The output is the analysis results, providing information on specific sales trends and effective selling points.

[1502] Step 3:

[1503] The server extracts basic information and effective selling points based on the analysis results. It uses the analyzed data as input. From the analyzed information, it identifies basic information such as product characteristics, usage, and price, as well as features advantageous for sales promotion. The output is a list of basic information and selling points.

[1504] Step 4:

[1505] The server generates a production script based on the extracted basic information and selling points. It uses a list of basic information and selling points as input. A natural language generation algorithm is used to automatically generate text describing the product's characteristics and benefits. The output is the generated script.

[1506] Step 5:

[1507] The server generates multilingual and multicultural video content using the generated script. It uses the generated script as input. Utilizing a generation AI model, it adds narration and captions corresponding to each language and synthesizes the video and audio. The output is video content compatible with multiple languages.

[1508] Step 6:

[1509] The server optimizes the generated video content into a format suitable for each viewing region and distributes it. It uses the generated video content as input. Technical adjustments, such as high or low resolution, are made to meet the needs of each viewing region before uploading it to television stations and online distribution platforms. The output is the optimized video content.

[1510] Step 7:

[1511] The device analyzes the user's emotions. It uses the user's camera video and audio data as input. Utilizing an emotion recognition engine, it performs facial expression and voice analysis to identify the user's emotional state. The output is the recognized user emotion information.

[1512] Step 8:

[1513] The server dynamically adjusts ad content based on the analyzed user's sentiment. It uses recognized user sentiment information and already generated ad content as input. For example, if the user shows interest, it adds a detailed description; if not, it switches to another product. The output is the adjusted ad content.

[1514] Step 9:

[1515] Users view tailored advertising content. Dynamically adjusted advertising content from the server is used as input, providing the user with the optimal advertising experience. The output is the user's viewing behavior and feedback.

[1516] 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 controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1517] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include those described above. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions shown by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1518] 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 this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[1519] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1520] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[1521] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[1522] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[1523] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated based, for example, on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[1524] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[1525] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[1526] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[1527] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[1528] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[1529] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1530] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[1531] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[1532] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[1533] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[1534] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[1535] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[1536] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[1537] The following is further disclosed regarding the embodiments described above.

[1538] (Claim 1)

[1539] A means of collecting data from past promotional programs and sales performance data,

[1540] A means of analyzing collected data to extract basic information and effective selling points of products for sale,

[1541] A means for generating a production script based on the analyzed information,

[1542] A means of generating multilingual and multicultural video content based on a script,

[1543] A means of optimizing and distributing the generated video content in a format suitable for each viewing region,

[1544] A system that includes this.

[1545] (Claim 2)

[1546] The system according to claim 1, wherein a machine learning model is used for data analysis.

[1547] (Claim 3)

[1548] The system according to claim 1, wherein a natural language generation algorithm is used for script generation.

[1549] "Example 1"

[1550] (Claim 1)

[1551] A means of collecting data from past promotional programs and sales performance data,

[1552] Means for cleaning, supplementing, and analyzing the collected data,

[1553] A means for extracting basic information and effective selling points of products from analyzed data,

[1554] A means for generating a production script based on extracted information,

[1555] A means of generating multilingual and multicultural video content based on a script,

[1556] A means of optimizing and distributing the generated video content in a format suitable for each viewing region,

[1557] A system that includes this.

[1558] (Claim 2)

[1559] The system according to claim 1, wherein a machine learning model is used for data analysis.

[1560] (Claim 3)

[1561] The system according to claim 1, wherein a natural language generation algorithm is used for script generation.

[1562] "Application Example 1"

[1563] (Claim 1)

[1564] A means of collecting data from past promotional programs and sales performance data,

[1565] A means of analyzing collected data to extract basic information and effective selling points of products for sale,

[1566] A means for generating a production script based on the analyzed information,

[1567] A means of generating multilingual and multicultural video content based on a script,

[1568] A means of optimizing and distributing the generated video content in a format suitable for each viewing region,

[1569] A means of collecting data from past advertising campaigns via smart devices and analyzing it in real time,

[1570] A means of automatically extracting effective ad elements to generate ad scripts,

[1571] A means for automatically generating multilingual video ads based on the generated ad script,

[1572] A means of collecting user feedback and using it to optimize the next advertising campaign,

[1573] A system that includes this.

[1574] (Claim 2)

[1575] The system according to claim 1, wherein a machine learning model is used for data analysis.

[1576] (Claim 3)

[1577] The system according to claim 1, wherein a natural language generation algorithm is used for script generation.

[1578] "Example 2 of combining an emotion engine"

[1579] (Claim 1)

[1580] A means of collecting data from past promotional programs and sales performance data,

[1581] A means for cleaning and analyzing collected data and extracting basic information and effective selling points of products for sale,

[1582] A means for generating a production script based on the analyzed information,

[1583] A means of generating multilingual and multicultural video content based on a script,

[1584] A means of optimizing and distributing the generated video content in a format suitable for each viewing region,

[1585] A means of recognizing the emotions of viewers and customizing the content delivered based on those emotions,

[1586] A system that includes this.

[1587] (Claim 2)

[1588] The system according to claim 1, wherein a machine learning model is used for data analysis.

[1589] (Claim 3)

[1590] The system according to claim 1, wherein a natural language generation algorithm is used for script generation.

[1591] "Application example 2 when combining with an emotional engine"

[1592] (Claim 1)

[1593] A means of collecting data from past promotional programs and sales performance data,

[1594] A means of analyzing collected data to extract basic information and effective selling points of p...

Claims

1. A means of collecting data from past promotional programs and sales performance data, A means of analyzing collected data to extract basic information and effective selling points of products for sale, A means for generating a production script based on the analyzed information, A means of generating multilingual and multicultural video content based on a script, A means of optimizing and distributing the generated video content in a format suitable for each viewing region, A system that includes this.

2. The system according to claim 1, wherein a machine learning model is used for data analysis.

3. The system according to claim 1, wherein a natural language generation algorithm is used for script generation.

Images