system

Abstract

We provide the system. [Solution] A means for collecting past sales data and order data, Methods for preprocessing and normalizing the collected data, A means for training a generative artificial intelligence model using preprocessed data, A means of integrating newly collected data in real time into a generative artificial intelligence model, A method for predicting future demand using a generative artificial intelligence model and automatically calculating production quantities, A means of notifying the user of the calculated prediction results, 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, including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of the chatbot's character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the conventional method for calculating the production quantity of in-store promotional products, it only relies on past sales data and order data, and cannot appropriately reflect elements that change in real time, resulting in problems such as stock shortages and surpluses. Therefore, it has been difficult to efficiently produce promotional products and optimize resources.

Means for Solving the Problems

[0005] To solve this problem, the present invention provides the following means. First, it provides means for collecting past sales data and order data, and preprocessing and normalizing it. Next, it includes means for training a generative artificial intelligence model using this preprocessed data, and further means for integrating new data collected in real time into this model. Then, it provides means for predicting future demand using the generative artificial intelligence model and automatically calculating production quantities. It also includes means for notifying the user of the calculated prediction results, and the notification method can be either a dashboard display or email notification. This solves the conventional problem and enables highly accurate calculation of production quantities for promotional items.

[0006] "Sales data" refers to information such as the quantity, sales amount, and sales period of products sold in the past.

[0007] "Order data" refers to data related to orders, including the quantity of goods ordered, the order date, and information about the ordering party.

[0008] "Preprocessing" refers to processes that convert collected data into a format that is easy for generative artificial intelligence to learn from, such as deduplication, imputation of missing values, and formatting of the data.

[0009] "Normalization" refers to a data transformation method that makes it easier to compare different datasets by fitting the data into a specific range.

[0010] A "generative artificial intelligence model" refers to an algorithm that learns specific patterns for a given task based on collected, pre-processed, and normalized data, and then performs predictions and generation.

[0011] "Real-time data" refers to data that is instantly reflected from the moment it is collected, and always contains the latest information.

[0012] "Predicting future demand" refers to estimating the quantity of goods that will be needed in the future, based on historical and real-time data.

[0013] "Automatically calculating production quantities" refers to a process where the system automatically determines the required production quantity of promotional items without manual intervention from the user.

[0014] "Dashboard display" refers to a method of providing users with important data through a visually clear and easy-to-understand interface.

[0015] "Email notification" refers to a method of sending prediction results and notification information to users via email. [Brief explanation of the drawing]

[0016] [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] This 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] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11]It is a sequence diagram showing the processing flow of the data processing system in Embodiment 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 Embodiment 2 when combined with an emotion engine. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when combined with an emotion engine.

Mode for Carrying Out the Invention

[0017] 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.

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

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

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

[0021] 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.

[0022] 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).

[0023] 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."

[0024] [First Embodiment]

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

[0026] 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.

[0027] 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).

[0028] 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.

[0029] 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.

[0030] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form 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.

[0031] 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.

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

[0033] 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.

[0034] 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.

[0035] 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.

[0036] 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".

[0037] System Overview

[0038] This invention is a system for more precisely calculating the production quantity of in-store promotional items. Specifically, a server collects past sales and order data, preprocesses and normalizes this data, and trains a generative artificial intelligence model. Newly acquired real-time data is integrated into this model to predict future demand. By notifying the user of the prediction results, the system automatically calculates the appropriate production quantity of promotional items.

[0039] Program Processing Description

[0040] Data collection

[0041] The server accesses the "sales management system" and "order management system" to collect historical sales and order data. It also collects market trends and competitor information from external data sources as needed. This ensures the richness and accuracy of the data.

[0042] Data Integration and Preprocessing

[0043] The server removes duplicate data from the collected data and performs imputation on incomplete data. For example, if there are missing values ​​in order data, it will use the mean or median to fill them in. It also normalizes the data and converts it into a format that is easy for generative artificial intelligence models to learn from.

[0044] Learning of generative artificial intelligence models

[0045] The server uses preprocessed data to train generative artificial intelligence models (e.g., LSTM or Transformer). These models have the ability to extract patterns from historical data and predict future demand. After training is complete, the models are evaluated using validation datasets to check their accuracy and make any necessary adjustments.

[0046] Real-time data integration

[0047] The server collects daily acquisition data and new order status data in real time. This real-time data is integrated with existing datasets and undergoes further data cleaning and formatting. This ensures that the model always incorporates the latest information.

[0048] Prediction and calculation

[0049] The server uses a generative artificial intelligence model to input the latest data, including real-time data, and predicts the production quantity of future promotional items. This prediction result is stored in a database and can be updated and recalculated immediately if necessary.

[0050] Providing results

[0051] The server notifies the user's device of the prediction results. Notification methods include dashboard display and email notifications. Users can connect to the server from their device to check the prediction results. This allows users to plan and execute promotional item production based on appropriate production quantities.

[0052] Specific example

[0053] For example, when a new product is launched, the user enters product information into the "sales management system." The server collects necessary data based on past data of similar products and market trends, and makes predictions using a generative artificial intelligence model. As a result, the predicted production quantity is notified to the user. Based on that quantity, the user can proceed with the production of promotional materials for the new product. This entire process significantly reduces the risk of inventory shortages or surpluses, enabling the efficient production of promotional materials.

[0054] As described above, this system is a powerful tool for significantly improving the accuracy of predicting the production quantity of promotional items, streamlining the execution process, and optimizing risk management.

[0055] The following describes the processing flow.

[0056] Step 1:

[0057] The server connects to the "sales management system" and "order management system" to collect historical sales and order data. It also obtains market trends and competitor information from external data sources.

[0058] Step 2:

[0059] The server stores the collected data in a database and performs duplicate data removal and data interpolation. For example, it applies historical mean or median values ​​to missing items.

[0060] Step 3:

[0061] The server normalizes the pre-processed data and transforms it into a format that is easy for generative artificial intelligence models to learn from. In this process, numerical data is scaled and categorical data is encoded.

[0062] Step 4:

[0063] The server uses pre-processed data to train a generative artificial intelligence model (e.g., LSTM or Transformer). The model recognizes past patterns and adjusts the parameters necessary for forecasting future demand.

[0064] Step 5:

[0065] The server evaluates the trained generative artificial intelligence model on a validation dataset to verify the model's accuracy. It then adjusts hyperparameters as needed to build an optimal predictive model.

[0066] Step 6:

[0067] The server periodically acquires real-time data such as daily acquisition results and new order status, and integrates it with the existing database. This data is also preprocessed and normalized.

[0068] Step 7:

[0069] The server inputs the latest real-time data into a generative artificial intelligence model to predict future demand for promotional items. The prediction results include approximate production quantities and their range of variation.

[0070] Step 8:

[0071] The server saves the calculated prediction results to a database. Furthermore, it notifies the user's device of the results. This notification may be displayed on a dashboard or sent via email.

[0072] Step 9:

[0073] Users access the server from their terminal and view the predicted production quantity on the dashboard. If necessary, they adjust and determine the actual production quantity based on the prediction results.

[0074] Step 10:

[0075] The user enters the determined production quantity and sends it to the server. The server then saves this information back to the database and sends the necessary production instructions to the order management system.

[0076] (Example 1)

[0077] 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."

[0078] Traditional systems for calculating the production quantity of in-store promotional items often relied on simple predictions based solely on historical data, making it difficult to accurately reflect market fluctuations and real-time conditions. Furthermore, manual data processing and management were required, leading to increased time, effort, and costs. This resulted in frequent inventory shortages or surpluses, making it difficult to create efficient production plans for promotional items.

[0079] 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.

[0080] In this invention, the server includes means for collecting historical sales data and order data, means for eliminating duplicates and imputing missing values ​​in the collected data, and means for normalizing the data and converting it into a format that is easy for a generating AI model to learn. This enables efficient and highly accurate predictions.

[0081] "Past sales data" refers to information about sales records and performance at the time a transaction took place.

[0082] "Order data" refers to records and information related to the ordering and procurement of goods.

[0083] "Means of collection" refers to the functions and devices that allow a server to access multiple data sources and obtain the necessary data.

[0084] "Means of eliminating duplicates" refer to methods or functions for deleting or merging data with identical content from an acquired dataset.

[0085] "Methods for imputing missing values" refer to functions or techniques for filling in missing values ​​or information in a dataset using the mean or median.

[0086] "Methods for normalizing data" refer to methods and functions for preparing data in a way that makes it easier for generative AI models to learn, by transforming it into a certain range or format.

[0087] A "generative AI model" is a machine learning model that uses artificial intelligence technology to learn patterns from past data and predict future trends and demand.

[0088] "Real-time data" refers to the latest information obtained based on ongoing transactions and events.

[0089] "Predictive means" refers to methods and functions for predicting future demand and fluctuations using generative AI models and calculating the results.

[0090] "Means of notification" refer to methods and functions for communicating prediction results to the user, and can take the form of a dashboard display or email notification.

[0091] Modes for carrying out the invention

[0092] This invention is a system for more precisely calculating the production quantity of in-store promotional items, and its main components are a server, terminals, and users. This system is implemented in the following steps.

[0093] System Overview

[0094] This system collects historical sales and order data, preprocesses and normalizes it, and uses it to train a generating AI model. It then integrates newly acquired data in real time with this model to predict future demand. By notifying the user of the prediction results, it automatically calculates the appropriate quantity of promotional items to produce.

[0095] Hardware and software used

[0096] Server: Performs data collection, preprocessing, normalization, AI model training, and prediction.

[0097] Terminal: Provides an interface for the user to view the prediction results.

[0098] Software: Sales management systems, order management systems, cloud databases, access to external data sources, data processing libraries, generative AI model libraries (e.g., LSTM and Transformer).

[0099] Specific example

[0100] When a new product is launched, the user enters product information into the "sales management system." For example, they might enter "details of new product A that is being promoted." The server collects necessary data, including historical data on similar products and market trends, and preprocesses and normalizes it. Then, it uses a generative AI model to predict future demand and calculate the appropriate quantity of promotional items to produce. This result is then communicated to the user.

[0101] Examples of prompt statements

[0102] For example, enter the following as a prompt for a generative AI model:

[0103] When entering the new product "○○" into the sales management system, please predict the required quantity of promotional materials to be produced, taking into account past sales data and market trends.

[0104] The server generates and provides prediction results to the user based on the above procedure. This significantly reduces the risk of inventory shortages or surpluses, and enables efficient planning of promotional item production. Furthermore, continuous integration of real-time data allows for highly accurate predictions that reflect the latest market conditions.

[0105] This system can operate in conjunction with common data infrastructures, regardless of the type of machine learning model or algorithm used. Its flexible application of generative AI models makes it adaptable to a variety of business environments.

[0106] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0107] Step 1:

[0108] The server collects past sales and order data.

[0109] Specifically, the server accesses the "sales management system" and "order management system" every night to retrieve historical sales and order data. This also includes retrieving market trends and competitor information via external APIs as needed. The input is raw data from the sales management system and order management system, and the output is all the collected raw data.

[0110] Step 2:

[0111] The server removes duplicate data and imputes missing values.

[0112] In terms of specific operations, the server analyzes the acquired dataset and removes or merges duplicate data with the same date and product ID. Furthermore, missing values ​​within the dataset are imputed using the mean or median of past data. The input is the collected raw data, and the output is the data with duplicates removed and missing values ​​imputed.

[0113] Step 3:

[0114] The server normalizes the data and converts it into a format that is easy for the generating AI model to learn from.

[0115] In terms of specific operations, the server scales quantitative and price data to a particular range and converts categorical data into numerical data using techniques such as one-hot encoding. The input is data that has undergone deduplicating and imputation, and the output is normalized data that has been converted into a format that is easy for the AI ​​model to learn from.

[0116] Step 4:

[0117] The server trains the generated AI model.

[0118] In terms of operation, the server inputs normalized data into a generative AI model (e.g., LSTM or Transformer) to train the model. During the training process, patterns are extracted from past data, and parameters are adjusted to predict future demand. The input is normalized data, and the output is the trained generative AI model.

[0119] Step 5:

[0120] The server integrates real-time data.

[0121] In terms of specific operations, the server collects daily sales performance and new order data in real time and integrates it into an existing dataset. Data cleaning and formatting are performed again as needed. The input is newly acquired data in real time, and the output is an integrated dataset containing the latest information.

[0122] Step 6:

[0123] The server makes predictions and calculates the production quantity.

[0124] In practice, the server uses a generative AI model to input the latest data and predict the production quantity of future promotional items. This prediction is performed periodically, and the data is saved to the database or updated immediately as needed. The input is the integrated latest data, and the output is the predicted production quantity.

[0125] Step 7:

[0126] The server notifies the user's device of the prediction results.

[0127] In terms of specific operations, the server transmits the prediction results to the user's terminal. At this time, dashboard display and email notifications can be used. When new prediction data is generated, the server displays the update information on the user's dashboard via the web server. It also notifies the user of the prediction results via email, depending on the settings. The input is the predicted production quantity, and the output is the notification to the user.

[0128] (Application Example 1)

[0129] 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."

[0130] The objective of this invention is to provide a system for appropriately predicting the production quantity of promotional items. Conventional methods rely solely on past sales and order data for predictions, making it difficult to reflect new data obtained in real time. As a result, it is not possible to respond quickly to fluctuations in demand. Furthermore, the methods for notifying users of the prediction results are limited, making it difficult for users to easily check them. This leads to risks of inventory shortages or surpluses, making it difficult to produce promotional items efficiently.

[0131] 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.

[0132] In this invention, the server includes means for collecting past sales data and order data; means for preprocessing and normalizing the collected data; means for training a generative artificial intelligence model using the preprocessed data; means for integrating newly collected data in real time into the generative artificial intelligence model; means for predicting future demand using the generative artificial intelligence model and automatically calculating production quantities; means for notifying the user of the calculated prediction results; and means for displaying the notified prediction results on a smart device. This enables precise sales forecasting including real-time data, and allows users to quickly check the prediction results via a smart device and create an appropriate production plan for promotional items.

[0133] "Past sales data" refers to data that records the sales performance of a product over a certain past period.

[0134] "Order data" refers to data that includes information about orders for goods and services.

[0135] "Preprocessing" is the process of converting collected data into a format suitable for analysis and learning.

[0136] "Normalization" is a transformation process that brings data values ​​within a certain range.

[0137] A "generative artificial intelligence model" is an algorithm or structure that learns from large amounts of data and makes predictions or generates new data.

[0138] "New data collected in real time" refers to data that is being collected in real time and is immediately available for processing and analysis.

[0139] "Integration" is the process of combining and linking different datasets into one.

[0140] "Predicting future demand and automatically calculating production quantities" refers to the act of using generative artificial intelligence models to estimate future consumer demand and determine the corresponding production volume.

[0141] "Notification" refers to communicating prediction results or information to the user through specific means.

[0142] A "smart device" is an electronic device that has communication capabilities and computing power, and allows users to display and operate information.

[0143] A "dashboard display" is an interface that visually organizes and presents information.

[0144] "Email notification" refers to a method of sending information in the form of an email.

[0145] A "remote database" is a data storage system located remotely and accessible via a network.

[0146] "External data sources" refer to data obtained from data providers outside the company, such as public databases.

[0147] System Overview

[0148] This invention is a system that streamlines inventory management by precisely predicting the production quantity of promotional items for physical stores. The system collects historical sales and order data, preprocesses and normalizes this data, then uses a generative artificial intelligence model for training and integrates real-time data to predict future demand. The prediction results are then communicated to the user's smart device.

[0149] Program Processing Description

[0150] Data collection

[0151] The server accesses the "sales management system" and "order management system" to collect historical sales and order data. This data is obtained from remote databases and external data sources.

[0152] Data Integration and Preprocessing

[0153] The server removes duplicate data from the collected data and performs imputation on incomplete data. It also normalizes the data and converts it into a format that is easy for generative artificial intelligence models to learn. Python and general-purpose data analysis libraries (such as Pandas and NumPy) are used in this process.

[0154] Learning of generative artificial intelligence models

[0155] The server uses preprocessed data to train generative artificial intelligence models (for example, LSTM or Transformer models using Keras). These models extract patterns from past data and predict future demand.

[0156] Real-time data integration

[0157] The server collects daily sales performance and new order status data in real time. This data is obtained from POS systems and sensor devices. The real-time data is integrated with existing datasets and then cleaned and formatted again.

[0158] Prediction and calculation

[0159] The server uses a generative artificial intelligence model to input the latest data and predict the production quantity of future promotional items. This prediction is stored in a database and can be updated and recalculated instantly as needed.

[0160] Providing results

[0161] The server notifies the user of the prediction results on their smart device. Notification methods include a dashboard display on the smart device and email notifications. Users can check the prediction results using their smartphones or tablets.

[0162] Specific example

[0163] For example, when a new product is launched, the user enters product information into the "sales management system." The server collects necessary data based on past data of similar products and market trends, and uses a generative artificial intelligence model to predict future demand. As a result, the predicted production quantity is notified to the user's smartphone. Based on that quantity, the user can proceed with the production of promotional materials for the new product. This entire process significantly reduces the risk of inventory shortages or surpluses, enabling the efficient production of promotional materials.

[0164] Examples of prompt sentences:

[0165] Collect new sales data and feed it into an AI model to predict future demand. We want to build an application that notifies us of the prediction results in real time. Historical sales data is located here: [sales_data.csv]. Order data is located here: [order_data.csv]. Real-time data is located here: [new_data.csv].

[0166] As described above, this system significantly improves the accuracy of predicting the production quantity of promotional items and is a powerful tool for creating efficient production plans for promotional items.

[0167] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0168] Step 1:

[0169] The server collects historical sales and order data from the "sales management system" and "order management system." This data includes information such as product name, sales quantity, sales date, order quantity, and order date. The collected data is stored in a database.

[0170] Input: Sales data, order data

[0171] Output: Raw data stored in the database

[0172] The server also accesses remote databases and external data sources to obtain market trends and competitive information.

[0173] Step 2:

[0174] The server preprocesses the collected data. It removes duplicate data and imputes incomplete data, for example, by filling in missing values ​​with the median. Furthermore, it normalizes the data and converts it into a format that is easy for AI models to learn from.

[0175] Input: Collected raw data

[0176] Output: Normalized preprocessed data

[0177] Specifically, we will use Python's Pandas and NumPy libraries to perform data cleaning and imputation.

[0178] Step 3:

[0179] The server uses preprocessed data to train a generative artificial intelligence model. This model consists of components such as LSTM (Long Short-Term Memory) and Transformers. The training process is executed using Keras and TENSORFLOW®.

[0180] Input: Normalized preprocessed data

[0181] Output: Trained generative artificial intelligence model

[0182] The server splits the dataset into a training set and a test set, and then trains and evaluates the model.

[0183] Step 4:

[0184] The server collects daily sales performance and new order status data in real time. This data is obtained from POS systems and sensor devices. The real-time data is integrated with existing datasets and then cleaned and formatted again.

[0185] Input: Real-time data

[0186] Output: Updated dataset

[0187] The server processes real-time data quickly and incorporates it into model updates.

[0188] Step 5:

[0189] The server uses a generative artificial intelligence model to predict the production quantities of future promotional items. The predicted data is stored in a database and can be updated and recalculated instantly.

[0190] Input: Updated dataset

[0191] Output: Predicted production quantity

[0192] The server runs a prediction algorithm to calculate production quantities based on future demand.

[0193] Step 6:

[0194] The server notifies the user of the prediction results on their smart device. Notification methods include a dashboard display on the smart device and email notifications. Users can check the prediction results using their smartphone or tablet.

[0195] Input: Predicted production quantity

[0196] Output: Notification to the user

[0197] The server either sends emails using the SMTP protocol or displays results on a web-based dashboard.

[0198] 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.

[0199] System Overview

[0200] This invention provides a system that combines an emotion engine with a system for precisely calculating the production quantity of in-store promotional items, thereby recognizing and reflecting user emotions to achieve even more accurate predictions. This system has means for collecting past sales data and order data, and preprocesses and normalizes the collected data. Furthermore, it uses the preprocessed data to train a generative artificial intelligence model, integrates newly collected data in real time, and predicts future demand. Finally, it incorporates an emotion engine that recognizes user emotions, and reflects user emotion data in the predictions.

[0201] Program Processing Description

[0202] Data collection

[0203] The server connects to the "Sales Management System" and "Order Management System" to collect historical sales and order data. Furthermore, it acquires market trends and competitor information from external data sources to ensure comprehensive data richness and accuracy.

[0204] Data Integration and Preprocessing

[0205] The server stores the collected data in a database and performs duplicate data removal and missing data completion. For example, it applies the historical mean or median to missing items. It also normalizes the data and converts it into a format that is easy for generative artificial intelligence models to learn from.

[0206] Learning of generative artificial intelligence models

[0207] The server uses pre-processed data to train a generative artificial intelligence model (e.g., LSTM or Transformer). The model recognizes past patterns and adjusts the parameters necessary for forecasting future demand.

[0208] Real-time data integration

[0209] The server periodically acquires real-time data such as daily acquisition results and new order status, and integrates it with the existing database. This real-time data is also properly preprocessed and normalized to maintain a consistent dataset.

[0210] Collection of user sentiment data

[0211] The user connects to the server via their device, and the emotion engine recognizes the user's emotions. The emotion engine uses facial recognition and voice analysis to understand the user's emotional state. The user's emotional data is sent to the server and stored in a database.

[0212] Analysis and integration of emotional data

[0213] The server analyzes the collected user sentiment data and incorporates the results into a generative artificial intelligence model. This enables highly accurate predictions that take into account the impact of sentiment data on demand forecasting.

[0214] Prediction and calculation

[0215] The server uses a generative artificial intelligence model to input the latest integrated data (sales data, order data, real-time data, and sentiment data) and predict the production quantity of future promotional items. The prediction results include an approximate production quantity and range of variation, taking sentiment data into account.

[0216] Providing results

[0217] The server saves the calculated prediction results to a database and notifies the user's device. Notification methods include dashboard display and email notifications. Users can check the prediction results through their device and make adjustments as needed.

[0218] Specific example

[0219] For example, when a new product is launched, the user enters product information into the "sales management system." The server collects necessary data based on past data of similar products, market trends, and user sentiment data, and makes predictions using a generative artificial intelligence model. For instance, if a user shows very positive sentiment, this is analyzed as an indication of increased demand, and the predicted production quantity will also increase. The results are notified to the user, who can then plan and execute a production plan for promotional items based on this information. This significantly reduces the risk of inventory shortages or surpluses, enabling the efficient production of promotional items.

[0220] As described above, this system significantly improves the accuracy of predicting the production quantity of promotional items, and by incorporating user sentiment data into the prediction process, it achieves more precise and adaptable predictions.

[0221] The following describes the processing flow.

[0222] Step 1:

[0223] The server connects to the sales management system and order management system, collecting historical sales and order data. Furthermore, it also obtains market trends and competitor information from external data sources.

[0224] Step 2:

[0225] The server stores the collected data in a database and performs duplicate removal and missing data insertion. For missing data, it applies the historical mean or median.

[0226] Step 3:

[0227] The server normalizes the pre-processed data and transforms it into a format that is easy for generative artificial intelligence models to learn. Numerical data is scaled, and categorical data is encoded.

[0228] Step 4:

[0229] The server uses pre-processed data to train generative artificial intelligence models (e.g., LSTM or Transformer). In the process, it extracts patterns from past data.

[0230] Step 5:

[0231] The server evaluates the trained generative artificial intelligence model on a validation dataset to verify the model's accuracy. It then adjusts hyperparameters as needed to build an optimal predictive model.

[0232] Step 6:

[0233] The server periodically acquires real-time data such as daily acquisition results and new order status, and this data is similarly pre-processed and normalized. The real-time data is then integrated into the database.

[0234] Step 7:

[0235] The user connects to the server via their device and activates the emotion engine. The emotion engine collects the user's emotional data through facial recognition and voice analysis, and sends this data to the server.

[0236] Step 8:

[0237] The server analyzes the collected user sentiment data and integrates the results into a generative artificial intelligence model. It analyzes the impact of the sentiment data on predictions and reflects this in the model.

[0238] Step 9:

[0239] The server uses a generative artificial intelligence model to input integrated, up-to-date data (sales data, order data, real-time data, and sentiment data) and predict the production quantity of future promotional items.

[0240] Step 10:

[0241] The server saves the prediction results to a database and notifies the user's device. Notifications are made through methods such as dashboard displays and email notifications.

[0242] Step 11:

[0243] Users access the server from their terminal and view the predicted production quantity on the dashboard. If necessary, they adjust and determine the actual production quantity based on the prediction results.

[0244] Step 12:

[0245] The user sends the determined production quantity from their terminal to the server. The server then saves this information back to the database and sends the necessary production instructions to the order management system.

[0246] Specific example

[0247] For example, when a new product launch is decided, the user enters the new product information into the sales management system. The server collects historical data on similar products and market information, as well as user sentiment data. If the user shows positive sentiment, it is analyzed that this positive sentiment will lead to increased demand. The server inputs this data into a generative artificial intelligence model to predict the production quantity of promotional items with high accuracy. The prediction results are notified to the user via a dashboard or email, and the user then decides on the appropriate production quantity based on that information.

[0248] (Example 2)

[0249] 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".

[0250] Currently, many companies predict the production quantity of promotional items based on past sales and order data, but the accuracy of this method is limited. Furthermore, even when utilizing real-time data, the prediction accuracy does not improve sufficiently because qualitative data such as user sentiment is not considered. Additionally, there is a lack of timely and effective methods for notifying users of the prediction results. To address these challenges, a more accurate and flexible prediction system is needed.

[0251] 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.

[0252] In this invention, the server includes means for collecting past sales data and order data; means for preprocessing and normalizing the collected data; means for training a generative artificial intelligence model using the preprocessed data; means for integrating newly collected data in real time into the generative artificial intelligence model; means for collecting user sentiment data; means for analyzing the collected user sentiment data and integrating it into the generative artificial intelligence model; means for predicting future demand using the generative artificial intelligence model and automatically calculating production quantities; and means for notifying the user of the calculated prediction results. This enables highly accurate demand forecasting that includes user sentiment data.

[0253] "Past sales data" refers to data that records the past sales performance of a particular product or service.

[0254] "Order data" refers to data related to orders placed by companies to receive the supply of products or services.

[0255] "Means of collection" refers to methods or devices for acquiring and storing specific information.

[0256] "Preprocessing" refers to the process of shaping, cleaning, and normalizing data, which is necessary for analysis and model training.

[0257] "Normalization" is the process of unifying the variability of data and bringing it within a specific range.

[0258] A "generative artificial intelligence model" is a type of AI algorithm that learns from collected data and uses it to predict future situations.

[0259] "Real-time data" refers to the latest information on ongoing events and situations.

[0260] "Emotional data" refers to data that records information about a user's emotional state.

[0261] "Analysis" is a scientific method for finding patterns and rules based on collected data.

[0262] "Prediction" is the process of estimating future situations or events based on collected data.

[0263] "Notification" refers to an action or mechanism for informing a user of specific information.

[0264] A "dashboard display" is an interface that provides information to users visually.

[0265] "Email notification" refers to a method of sending information to users via email.

[0266] A "cloud database" is an online database that can be accessed via the internet.

[0267] "External data sources" refer to data provided by external services or data providers, rather than data from within a company.

[0268] This invention provides a system for accurately calculating the production quantity of in-store promotional items. This system collects historical sales and order data, preprocesses and normalizes this data, and then trains a generative artificial intelligence model. Furthermore, it integrates newly collected data and user sentiment data in real time to achieve advanced demand forecasting and production quantity calculation.

[0269] Data collection

[0270] Sales and order data collection via server

[0271] The server periodically connects to the "sales management system" and "order management system" via specific APIs to automatically collect the necessary data. API calls, for example, save sales and order data for the past year to the database.

[0272] Server-based market trend and competitor information gathering

[0273] The server automatically retrieves market trends and competitor information from external data sources (e.g., Google Trends and social media analysis tools). This ensures the richness and accuracy of the data.

[0274] Data Integration and Preprocessing

[0275] Server-based data storage and cleaning

[0276] The server stores the collected data in a database and performs duplicate data removal and missing data imputation. Missing data is filled in using historical mean and median values.

[0277] Data normalization by the server

[0278] The server converts the data into a unified format and shapes it so that it can be easily trained by generative artificial intelligence models. Specifically, it scales sales quantities to a range of 0 to 1 and performs one-hot encoding on categorical data.

[0279] Learning of generative artificial intelligence model

[0280] AI model learning by server

[0281] The server uses the pre - processed data to train a generative artificial intelligence model (e.g., LSTM or Transformer). This learning process improves the prediction accuracy.

[0282] Integration of real - time data

[0283] Collection and integration of real - time data by server

[0284] The server acquires real - time data such as daily sales performance and order status, and integrates it with the existing database. The new data is appropriately pre - processed and merged into the existing dataset.

[0285] Collection of user's emotion data

[0286] Provision of emotion data by user

[0287] The user obtains the emotional state through the terminal, and the emotion engine uses facial recognition and voice analysis. The user provides emotion information to the emotion engine using, for example, a camera or a microphone.

[0288] Storage of emotion data by server

[0289] The server stores the acquired user's emotion data in the database and integrates it with other data.

[0290] Analysis and integration of emotion data

[0291] Analysis of emotion data by server

[0292] The server analyzes emotional data and incorporates it into a generative artificial intelligence model to improve prediction accuracy. For example, positive emotions may indicate increased demand, thus influencing the predicted values.

[0293] Prediction and calculation

[0294] Server-based future demand forecasting

[0295] The server uses a generative artificial intelligence model to calculate the future production quantity of promotional items based on integrated, up-to-date data. The prediction results include sales data, order data, real-time data, and sentiment data.

[0296] Providing results

[0297] Server-based notification of results

[0298] The server saves the calculated prediction results to a database and notifies the user's device. Notifications are made via a dashboard display or email.

[0299] User review and adjustment of results

[0300] Users can view the prediction results through their devices and make adjustments as needed. This allows for the creation of efficient promotional material production plans. For example, if a user expresses positive emotions, the predicted production quantity will also increase.

[0301] Specific example

[0302] For example, when a new product is launched, the user enters product information into the "sales management system." The server uses a generative artificial intelligence model to make predictions based on past data of similar products, market trends, and user sentiment data. For instance, if a user expresses very positive sentiment, an increase in demand is predicted, and the production quantity will also increase. The results are notified to the user, who can then plan and execute a promotional item production plan based on that information.

[0303] Example of prompt text

[0304] User: I have positive feelings about the new product A. Please provide a sales forecast.

[0305] Server: Based on past sales data, order data, market trend information, and user sentiment data, a generative AI model is used to predict the demand for new product A.

[0306] This system enables highly accurate demand prediction including user sentiment data, and realizes efficient sales promotion activities.

[0307] The flow of the specific process in Example 2 will be described using FIG. 13.

[0308] Step 1: Data collection

[0309] The server regularly connects to the "sales management system" and the "order management system" to collect past sales data and order data. As a specific operation, a data collection request is sent via an API, and data for the past year is saved in the database. The input is the API request, and the output is the acquired sales data and order data.

[0310] Step 2: Collection of market trends and competitive information

[0311] The server connects to external data sources (e.g., Google Trends and SNS analysis tools) to collect market trends and competitive information. As a specific operation, a request is sent via an API, and the collected data is saved in the database. The input is the information request from the external data source, and the output is the acquired market trends and competitive information.

[0312] Step 3: Data preprocessing and normalization

[0313] The server stores the collected data in a database and performs duplicate data removal and missing data imputation. Specifically, it removes duplicate data and imputes missing data with historical mean or median values. Furthermore, it converts the data into a unified format (e.g., numerical scaling, one-hot encoding of categorical data) to make it easier for AI models to learn. The input is the various data acquired, and the output is the pre-processed dataset.

[0314] Step 4: Training a generative artificial intelligence model

[0315] The server uses pre-processed data to train generative artificial intelligence models (e.g., LSTM and Transformer). Specifically, it inputs data into the AI ​​model and adjusts the model's parameters while training. The input is the pre-processed dataset, and the output is the optimized AI model.

[0316] Step 5: Collect and integrate real-time data

[0317] The server periodically retrieves daily real-time data (new sales figures and order status) and integrates it with the existing database. Specifically, it collects real-time data via an API and merges the new data into the existing dataset. The input is real-time data, and the output is the updated integrated dataset.

[0318] Step 6: Collecting user sentiment data

[0319] The user accesses the emotion engine through their device and acquires their emotional state using facial recognition and voice analysis. Specifically, the user uses the camera and microphone to provide emotional information to the emotion engine. The input is the user's facial expressions and voice data, and the output is the analyzed emotional data.

[0320] Step 7: Analysis and integration of emotional data

[0321] The server analyzes user sentiment data and incorporates it into a generative artificial intelligence model. Specifically, it quantifies sentiment data, cross-analyzes it with sales data, and inputs it into the AI ​​model to improve prediction accuracy. The input is sentiment data and an integrated dataset, and the output is the final dataset with the sentiment data integrated.

[0322] Step 8: Forecast future demand and calculate production quantities.

[0323] The server uses a generative artificial intelligence model based on the latest integrated data to predict the production quantity of future promotional items. Specifically, it inputs the integrated data into the AI ​​model and performs demand forecasting. As a result, a demand forecast value is output. The input is the integrated dataset, and the output is the predicted production quantity of promotional items.

[0324] Step 9: Notification of Results

[0325] The server saves the calculated prediction results to a database and notifies the user's terminal. Notifications are made via a dashboard display or email. Specifically, the server displays the prediction results on a dashboard or sends them to the user via email. The input is the prediction result data, and the output is the notification to the user.

[0326] Step 10: Review and adjust results

[0327] Users can view prediction results via their devices and make adjustments as needed. Specifically, users view prediction results on a dashboard and then create a production plan for promotional items based on those results. The input is the prediction result, and the output is the adjusted production plan.

[0328] This completes the entire processing flow, enabling highly accurate demand forecasting and production quantity calculation for promotional items.

[0329] (Application Example 2)

[0330] 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".

[0331] Traditional demand forecasting systems make predictions based on sales and order data, but they often lack accuracy because they don't take into account customer emotions and reactions at the actual point of sale. As a result, they can either overestimate or underestimate the quantity of promotional items to be produced. This can lead to inefficient inventory management and potentially lost business opportunities.

[0332] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for collecting past sales data and order data, means for preprocessing and normalizing the collected data, means for training a generative artificial intelligence model using the preprocessed data, means for integrating newly collected data in real time into the generative artificial intelligence model, means for predicting future demand using the generative artificial intelligence model and automatically calculating the production quantity, means for notifying the user of the calculated prediction results, and means for collecting user sentiment data and integrating it into the generative artificial intelligence model. As a result, customer sentiment data can be incorporated into demand forecasting, improving forecasting accuracy and enabling more accurate estimation of the production quantity of promotional items.

[0333] "Past sales data" refers to data that shows the history of sales transactions that have taken place in the past, and includes details such as quantity, date, and product name.

[0334] "Order data" refers to data related to orders for goods and services, including information such as the order date, order quantity, and ordering source information.

[0335] "Collection means" refers to methods or devices for collecting specific data, and includes means of acquiring information using sensors, APIs, databases, etc.

[0336] "Preprocessing means" refers to methods and techniques for converting collected data into an appropriate format, making it easy to use for analysis and model training.

[0337] "Methods of normalization" refer to methods and techniques for unifying data variability and bringing it within a certain range, and include techniques such as standardization and scaling.

[0338] "Generative artificial intelligence models" refer to machine learning algorithms and deep learning models that learn patterns from large amounts of data and predict future outcomes.

[0339] "Real-time data" refers to data about events currently in progress or data generated at that moment, including data that is collected and processed immediately.

[0340] "Integration methods" refer to methods and techniques for combining information obtained from multiple data sources to create a unified dataset.

[0341] "Predictive means" refers to methods and techniques that use generative artificial intelligence models to estimate future demand and determine the required production quantity.

[0342] "Notification means" refers to methods and technologies for informing users of the calculated prediction results, and includes means such as dashboards, smart device displays, and email notifications.

[0343] "Emotional data" refers to information that indicates a user's emotional state, and includes data obtained from facial expressions, voice, behavioral patterns, etc.

[0344] System Overview

[0345] The present invention is a system for predicting the production quantity of promotional items using user sentiment data, and mainly includes the following elements: means for collecting past sales data and order data, preprocessing means, normalization means, real-time data integration means, generative artificial intelligence model learning means, demand forecasting means, notification means, and sentiment data collection means.

[0346] Program Embodiment

[0347] 1. Data Collection

[0348] The server connects to the "sales management system" and "order management system" to collect historical sales and order data. Furthermore, it acquires market trends and competitor information from external data sources to ensure comprehensive data richness and accuracy. Using smartphones and smart glasses, it collects user emotion data in real time from customers' facial expressions and voices in stores.

[0349] 2. Data Integration and Preprocessing

[0350] The server stores the collected data in a database and uses Pandas and NumPy to remove duplicate data and impute missing data. For example, it applies the historical mean or median to missing items. It also normalizes the data and converts it into a format that is easy for generative artificial intelligence models to learn.

[0351] 3. Training of generative artificial intelligence models

[0352] The server uses preprocessed data to train generative artificial intelligence models (e.g., LSTM or Transformer) using TensorFlow or Keras. The models recognize past patterns and adjust the parameters necessary for future demand forecasting.

[0353] 4. Integration of real-time data

[0354] The server periodically acquires real-time data such as daily acquisition results and new order status, and integrates it with existing databases using Apache® Kafka or RabbitMQ. This real-time data is also preprocessed and normalized to maintain a consistent dataset.

[0355] 5. Collection of emotional data

[0356] The user's smartphone or smart glasses use camera APIs and microphone APIs to recognize the user's emotional state from their facial expressions and voice via an emotion engine, and then send that data to the server.

[0357] 6. Analysis and Integration of Emotional Data

[0358] The server analyzes the collected user sentiment data using Scikit-learn and TensorFlow, and incorporates the results into a generative artificial intelligence model. This enables highly accurate predictions that take into account the impact of sentiment data on demand forecasting.

[0359] 7. Prediction and Calculation

[0360] The server inputs the latest integrated data (sales data, order data, real-time data, sentiment data) into a generative artificial intelligence model to predict the production quantity of future promotional items. For example, if a user expresses very positive sentiment, this is analyzed as an indication of increased demand, and the predicted production quantity will also increase.

[0361] 8. Providing the results

[0362] The server notifies users of the calculated prediction results through a smartphone app using React Native or a smart glasses app using ARKit / ARCore. Based on this, users can plan and execute promotional product production plans.

[0363] Specific example

[0364] For example, imagine a store selling a new product, with staff wearing smart glasses. When a customer shows interest in the product, emotional data is collected. This data is sent to a server in real time and integrated with past sales data and market trends. A generative artificial intelligence model analyzes this data to predict demand and calculate the quantity of promotional items to produce for the new product. Staff can instantly check the information through the smart glasses' HUD and take appropriate action.

[0365] Example of a prompt

[0366] Prompt: "To accurately predict the quantity of promotional items to be produced at stores, collect historical sales data, order data, and user sentiment data in real time, and use a generative artificial intelligence model to make predictions."

[0367] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0368] Step 1:

[0369] Data collection

[0370] The server connects to the "sales management system" and "order management system" to collect historical sales and order data. Specifically, it retrieves data from the database using a REST API and also collects market trends and competitor information via external APIs. In addition, smart devices (smartphones and smart glasses) collect customer facial expressions and voices in real time. This data is sent to the server as raw data and stored in the database.

[0371] Inputs: Sales data, order data, market trend data, competitor information, sentiment data

[0372] Output: Collected raw dataset

[0373] Step 2:

[0374] Data Integration and Preprocessing

[0375] The server processes the collected raw data using Pandas and NumPy. Specifically, it removes duplicate data and imputes missing data with the mean or median. Next, it normalizes the data and converts it into a format suitable for training the model. For example, it scales numerical data and one-hot encodes categorical data.

[0376] Input: Collected raw dataset

[0377] Output: Preprocessed normalized dataset

[0378] Step 3:

[0379] Learning of generative artificial intelligence models

[0380] The server uses TensorFlow and Keras to train generative artificial intelligence models (such as LSTM and Transformer) on preprocessed datasets. Specifically, it splits the data into a training set and a test set, trains the model using the training set, and evaluates it using the test set. It adjusts parameters and sets the number of epochs to ensure the model learns properly.

[0381] Input: Preprocessed normalized dataset

[0382] Output: Trained generative artificial intelligence model

[0383] Step 4:

[0384] Real-time data integration

[0385] The server uses Apache Kafka and RabbitMQ to collect new sales data, order status, and other information in real time and integrate it with existing databases. This data is also preprocessed and normalized. Real-time data is sent to the server as soon as it is collected, ensuring that it is always up-to-date.

[0386] Input: Real-time dataset

[0387] Output: Integrated modern database

[0388] Step 5:

[0389] Collection of emotional data

[0390] The user's smart device uses camera and microphone APIs to collect customer facial expressions and voice data, and analyzes the user's emotional state through an emotion engine. This emotional data is sent to a server via WebSocket or REST API and stored in a database.

[0391] Input: Facial expression data, audio data

[0392] Output: Analyzed sentiment data

[0393] Step 6:

[0394] Analysis and integration of emotional data

[0395] The server uses Scikit-learn and TensorFlow to analyze the collected sentiment data. Specifically, it analyzes the sentiment data and extracts factors that influence demand forecasting. Then, this sentiment data is integrated into a generative artificial intelligence model to improve the model's accuracy.

[0396] Input: Analyzed sentiment data

[0397] Output: Integrated sentiment dataset

[0398] Step 7:

[0399] Prediction and calculation

[0400] The server inputs the integrated dataset into a generative artificial intelligence model to predict future demand. Specifically, it uses LSTM and Transformer models to calculate predicted sales and production quantities. For example, if there are many positive user sentiments, it analyzes this as an indication of increased demand and increases the predicted production quantity accordingly.

[0401] Input: Integrated dataset

[0402] Output: Estimated production quantity of promotional items

[0403] Step 8:

[0404] Providing results

[0405] The server notifies users of prediction results through a smartphone app using React Native or a smart glasses app using ARKit / ARCore. Specifically, the prediction results are displayed on a dashboard, allowing users to immediately check and take action.

[0406] Input: Estimated production quantity of promotional items

[0407] Output: Notifications to the user (dashboard display, smart device display)

[0408] Example of a prompt

[0409] Prompt: "To accurately predict the quantity of promotional items to be produced at stores, collect historical sales data, order data, and user sentiment data in real time, and use a generative artificial intelligence model to make predictions."

[0410] 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.

[0411] 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 the following. 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 indicated 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.

[0412] 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.

[0413] [Second Embodiment]

[0414] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0415] 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.

[0416] 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).

[0417] 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.

[0418] 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.

[0419] 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).

[0420] 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.

[0421] 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.

[0422] 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.

[0423] 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.

[0424] 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.

[0425] 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".

[0426] System Overview

[0427] This invention is a system for more precisely calculating the production quantity of in-store promotional items. Specifically, a server collects past sales and order data, preprocesses and normalizes this data, and trains a generative artificial intelligence model. Newly acquired real-time data is integrated into this model to predict future demand. By notifying the user of the prediction results, the system automatically calculates the appropriate production quantity of promotional items.

[0428] Program Processing Description

[0429] Data collection

[0430] The server accesses the "sales management system" and "order management system" to collect historical sales and order data. It also collects market trends and competitor information from external data sources as needed. This ensures the richness and accuracy of the data.

[0431] Data Integration and Preprocessing

[0432] The server removes duplicate data from the collected data and performs imputation on incomplete data. For example, if there are missing values ​​in order data, it will use the mean or median to fill them in. It also normalizes the data and converts it into a format that is easy for generative artificial intelligence models to learn from.

[0433] Learning of generative artificial intelligence models

[0434] The server uses preprocessed data to train generative artificial intelligence models (e.g., LSTM or Transformer). These models have the ability to extract patterns from historical data and predict future demand. After training is complete, the models are evaluated using validation datasets to check their accuracy and make any necessary adjustments.

[0435] Real-time data integration

[0436] The server collects daily acquisition data and new order status data in real time. This real-time data is integrated with existing datasets and undergoes further data cleaning and formatting. This ensures that the model always incorporates the latest information.

[0437] Prediction and calculation

[0438] The server uses a generative artificial intelligence model to input the latest data, including real-time data, and predicts the production quantity of future promotional items. This prediction result is stored in a database and can be updated and recalculated immediately if necessary.

[0439] Providing results

[0440] The server notifies the user's device of the prediction results. Notification methods include dashboard display and email notifications. Users can connect to the server from their device to check the prediction results. This allows users to plan and execute promotional item production based on appropriate production quantities.

[0441] Specific example

[0442] For example, when a new product is launched, the user enters product information into the "sales management system." The server collects necessary data based on past data of similar products and market trends, and makes predictions using a generative artificial intelligence model. As a result, the predicted production quantity is notified to the user. Based on that quantity, the user can proceed with the production of promotional materials for the new product. This entire process significantly reduces the risk of inventory shortages or surpluses, enabling the efficient production of promotional materials.

[0443] As described above, this system is a powerful tool for significantly improving the accuracy of predicting the production quantity of promotional items, streamlining the execution process, and optimizing risk management.

[0444] The following describes the processing flow.

[0445] Step 1:

[0446] The server connects to the "sales management system" and "order management system" to collect historical sales and order data. It also obtains market trends and competitor information from external data sources.

[0447] Step 2:

[0448] The server stores the collected data in a database and performs duplicate data removal and data interpolation. For example, it applies historical mean or median values ​​to missing items.

[0449] Step 3:

[0450] The server normalizes the pre-processed data and transforms it into a format that is easy for generative artificial intelligence models to learn from. In this process, numerical data is scaled and categorical data is encoded.

[0451] Step 4:

[0452] The server uses pre-processed data to train a generative artificial intelligence model (e.g., LSTM or Transformer). The model recognizes past patterns and adjusts the parameters necessary for forecasting future demand.

[0453] Step 5:

[0454] The server evaluates the trained generative artificial intelligence model on a validation dataset to verify the model's accuracy. It then adjusts hyperparameters as needed to build an optimal predictive model.

[0455] Step 6:

[0456] The server periodically acquires real-time data such as daily acquisition results and new order status, and integrates it with the existing database. This data is also preprocessed and normalized.

[0457] Step 7:

[0458] The server inputs the latest real-time data into a generative artificial intelligence model to predict future demand for promotional items. The prediction results include approximate production quantities and their range of variation.

[0459] Step 8:

[0460] The server saves the calculated prediction results to a database. Furthermore, it notifies the user's device of the results. This notification may be displayed on a dashboard or sent via email.

[0461] Step 9:

[0462] Users access the server from their terminal and view the predicted production quantity on the dashboard. If necessary, they adjust and determine the actual production quantity based on the prediction results.

[0463] Step 10:

[0464] The user enters the determined production quantity and sends it to the server. The server then saves this information back to the database and sends the necessary production instructions to the order management system.

[0465] (Example 1)

[0466] 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."

[0467] Traditional systems for calculating the production quantity of in-store promotional items often relied on simple predictions based solely on historical data, making it difficult to accurately reflect market fluctuations and real-time conditions. Furthermore, manual data processing and management were required, leading to increased time, effort, and costs. This resulted in frequent inventory shortages or surpluses, making it difficult to create efficient production plans for promotional items.

[0468] 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.

[0469] In this invention, the server includes means for collecting historical sales data and order data, means for eliminating duplicates and imputing missing values ​​in the collected data, and means for normalizing the data and converting it into a format that is easy for a generating AI model to learn. This enables efficient and highly accurate predictions.

[0470] "Past sales data" refers to information about sales records and performance at the time a transaction took place.

[0471] "Order data" refers to records and information related to the ordering and procurement of goods.

[0472] "Means of collection" refers to the functions and devices that allow a server to access multiple data sources and obtain the necessary data.

[0473] "Means of eliminating duplicates" refer to methods or functions for deleting or merging data with identical content from an acquired dataset.

[0474] "Methods for imputing missing values" refer to functions or techniques for filling in missing values ​​or information in a dataset using the mean or median.

[0475] "Methods for normalizing data" refer to methods and functions for preparing data in a way that makes it easier for generative AI models to learn, by transforming it into a certain range or format.

[0476] A "generative AI model" is a machine learning model that uses artificial intelligence technology to learn patterns from past data and predict future trends and demand.

[0477] "Real-time data" refers to the latest information obtained based on ongoing transactions and events.

[0478] "Predictive means" refers to methods and functions for predicting future demand and fluctuations using generative AI models and calculating the results.

[0479] "Means of notification" refer to methods and functions for communicating prediction results to the user, and can take the form of a dashboard display or email notification.

[0480] Modes for carrying out the invention

[0481] This invention is a system for more precisely calculating the production quantity of in-store promotional items, and its main components are a server, terminals, and users. This system is implemented in the following steps.

[0482] System Overview

[0483] This system collects historical sales and order data, preprocesses and normalizes it, and uses it to train a generating AI model. It then integrates newly acquired data in real time with this model to predict future demand. By notifying the user of the prediction results, it automatically calculates the appropriate quantity of promotional items to produce.

[0484] Hardware and software used

[0485] Server: Performs data collection, preprocessing, normalization, AI model training, and prediction.

[0486] Terminal: Provides an interface for the user to view the prediction results.

[0487] Software: Sales management systems, order management systems, cloud databases, access to external data sources, data processing libraries, generative AI model libraries (e.g., LSTM and Transformer).

[0488] Specific example

[0489] When a new product is launched, the user enters product information into the "sales management system." For example, they might enter "details of new product A that is being promoted." The server collects necessary data, including historical data on similar products and market trends, and preprocesses and normalizes it. Then, it uses a generative AI model to predict future demand and calculate the appropriate quantity of promotional items to produce. This result is then communicated to the user.

[0490] Examples of prompt statements

[0491] For example, enter the following as a prompt for a generative AI model:

[0492] When entering the new product "○○" into the sales management system, please predict the required quantity of promotional materials to be produced, taking into account past sales data and market trends.

[0493] The server generates and provides prediction results to the user based on the above procedure. This significantly reduces the risk of inventory shortages or surpluses, and enables efficient planning of promotional item production. Furthermore, continuous integration of real-time data allows for highly accurate predictions that reflect the latest market conditions.

[0494] This system can operate in conjunction with common data infrastructures, regardless of the type of machine learning model or algorithm used. Its flexible application of generative AI models makes it adaptable to a variety of business environments.

[0495] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0496] Step 1:

[0497] The server collects past sales and order data.

[0498] Specifically, the server accesses the "sales management system" and "order management system" every night to retrieve historical sales and order data. This also includes retrieving market trends and competitor information via external APIs as needed. The input is raw data from the sales management system and order management system, and the output is all the collected raw data.

[0499] Step 2:

[0500] The server removes duplicate data and imputes missing values.

[0501] In terms of specific operations, the server analyzes the acquired dataset and removes or merges duplicate data with the same date and product ID. Furthermore, missing values ​​within the dataset are imputed using the mean or median of past data. The input is the collected raw data, and the output is the data with duplicates removed and missing values ​​imputed.

[0502] Step 3:

[0503] The server normalizes the data and converts it into a format that is easy for the generating AI model to learn from.

[0504] In terms of specific operations, the server scales quantitative and price data to a particular range and converts categorical data into numerical data using techniques such as one-hot encoding. The input is data that has undergone deduplicating and imputation, and the output is normalized data that has been converted into a format that is easy for the AI ​​model to learn from.

[0505] Step 4:

[0506] The server trains the generated AI model.

[0507] In terms of operation, the server inputs normalized data into a generative AI model (e.g., LSTM or Transformer) to train the model. During the training process, patterns are extracted from past data, and parameters are adjusted to predict future demand. The input is normalized data, and the output is the trained generative AI model.

[0508] Step 5:

[0509] The server integrates real-time data.

[0510] In terms of specific operations, the server collects daily sales performance and new order data in real time and integrates it into an existing dataset. Data cleaning and formatting are performed again as needed. The input is newly acquired data in real time, and the output is an integrated dataset containing the latest information.

[0511] Step 6:

[0512] The server makes predictions and calculates the production quantity.

[0513] In practice, the server uses a generative AI model to input the latest data and predict the production quantity of future promotional items. This prediction is performed periodically, and the data is saved to the database or updated immediately as needed. The input is the integrated latest data, and the output is the predicted production quantity.

[0514] Step 7:

[0515] The server notifies the user's device of the prediction results.

[0516] In terms of specific operations, the server transmits the prediction results to the user's terminal. At this time, dashboard display and email notifications can be used. When new prediction data is generated, the server displays the update information on the user's dashboard via the web server. It also notifies the user of the prediction results via email, depending on the settings. The input is the predicted production quantity, and the output is the notification to the user.

[0517] (Application Example 1)

[0518] 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."

[0519] The objective of this invention is to provide a system for appropriately predicting the production quantity of promotional items. Conventional methods rely solely on past sales and order data for predictions, making it difficult to reflect new data obtained in real time. As a result, it is not possible to respond quickly to fluctuations in demand. Furthermore, the methods for notifying users of the prediction results are limited, making it difficult for users to easily check them. This leads to risks of inventory shortages or surpluses, making it difficult to produce promotional items efficiently.

[0520] 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.

[0521] In this invention, the server includes means for collecting past sales data and order data; means for preprocessing and normalizing the collected data; means for training a generative artificial intelligence model using the preprocessed data; means for integrating newly collected data in real time into the generative artificial intelligence model; means for predicting future demand using the generative artificial intelligence model and automatically calculating production quantities; means for notifying the user of the calculated prediction results; and means for displaying the notified prediction results on a smart device. This enables precise sales forecasting including real-time data, and allows users to quickly check the prediction results via a smart device and create an appropriate production plan for promotional items.

[0522] "Past sales data" refers to data that records the sales performance of a product over a certain past period.

[0523] "Order data" refers to data that includes information about orders for goods and services.

[0524] "Preprocessing" is the process of converting collected data into a format suitable for analysis and learning.

[0525] "Normalization" is a transformation process that brings data values ​​within a certain range.

[0526] A "generative artificial intelligence model" is an algorithm or structure that learns from large amounts of data and makes predictions or generates new data.

[0527] "New data collected in real time" refers to data that is being collected in real time and is immediately available for processing and analysis.

[0528] "Integration" is the process of combining and linking different datasets into one.

[0529] "Predicting future demand and automatically calculating production quantities" refers to the act of using generative artificial intelligence models to estimate future consumer demand and determine the corresponding production volume.

[0530] "Notification" refers to communicating prediction results or information to the user through specific means.

[0531] A "smart device" is an electronic device that has communication capabilities and computing power, and allows users to display and operate information.

[0532] A "dashboard display" is an interface that visually organizes and presents information.

[0533] "Email notification" refers to a method of sending information in the form of an email.

[0534] A "remote database" is a data storage system located remotely and accessible via a network.

[0535] "External data sources" refer to data obtained from data providers outside the company, such as public databases.

[0536] System Overview

[0537] This invention is a system that streamlines inventory management by precisely predicting the production quantity of promotional items for physical stores. The system collects historical sales and order data, preprocesses and normalizes this data, then uses a generative artificial intelligence model for training and integrates real-time data to predict future demand. The prediction results are then communicated to the user's smart device.

[0538] Program Processing Description

[0539] Data collection

[0540] The server accesses the "sales management system" and "order management system" to collect historical sales and order data. This data is obtained from remote databases and external data sources.

[0541] Data Integration and Preprocessing

[0542] The server removes duplicate data from the collected data and performs imputation on incomplete data. It also normalizes the data and converts it into a format that is easy for generative artificial intelligence models to learn. Python and general-purpose data analysis libraries (such as Pandas and NumPy) are used in this process.

[0543] Learning of generative artificial intelligence models

[0544] The server uses preprocessed data to train generative artificial intelligence models (for example, LSTM or Transformer models using Keras). These models extract patterns from past data and predict future demand.

[0545] Real-time data integration

[0546] The server collects daily sales performance and new order status data in real time. This data is obtained from POS systems and sensor devices. The real-time data is integrated with existing datasets and then cleaned and formatted again.

[0547] Prediction and calculation

[0548] The server uses a generative artificial intelligence model to input the latest data and predict the production quantity of future promotional items. This prediction is stored in a database and can be updated and recalculated instantly as needed.

[0549] Providing results

[0550] The server notifies the user of the prediction results on their smart device. Notification methods include a dashboard display on the smart device and email notifications. Users can check the prediction results using their smartphones or tablets.

[0551] Specific example

[0552] For example, when a new product is launched, the user enters product information into the "sales management system." The server collects necessary data based on past data of similar products and market trends, and uses a generative artificial intelligence model to predict future demand. As a result, the predicted production quantity is notified to the user's smartphone. Based on that quantity, the user can proceed with the production of promotional materials for the new product. This entire process significantly reduces the risk of inventory shortages or surpluses, enabling the efficient production of promotional materials.

[0553] Examples of prompt sentences:

[0554] Collect new sales data and feed it into an AI model to predict future demand. We want to build an application that notifies us of the prediction results in real time. Historical sales data is located here: [sales_data.csv]. Order data is located here: [order_data.csv]. Real-time data is located here: [new_data.csv].

[0555] As described above, this system significantly improves the accuracy of predicting the production quantity of promotional items and is a powerful tool for creating efficient production plans for promotional items.

[0556] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0557] Step 1:

[0558] The server collects historical sales and order data from the "sales management system" and "order management system." This data includes information such as product name, sales quantity, sales date, order quantity, and order date. The collected data is stored in a database.

[0559] Input: Sales data, order data

[0560] Output: Raw data stored in the database

[0561] The server also accesses remote databases and external data sources to obtain market trends and competitive information.

[0562] Step 2:

[0563] The server preprocesses the collected data. It removes duplicate data and imputes incomplete data, for example, by filling in missing values ​​with the median. Furthermore, it normalizes the data and converts it into a format that is easy for AI models to learn from.

[0564] Input: Collected raw data

[0565] Output: Normalized preprocessed data

[0566] Specifically, we will use Python's Pandas and NumPy libraries to perform data cleaning and imputation.

[0567] Step 3:

[0568] The server uses preprocessed data to train a generative artificial intelligence model. This model consists of components such as LSTM (Long Short-Term Memory) and Transformer. The training process is executed using Keras and TensorFlow.

[0569] Input: Normalized preprocessed data

[0570] Output: Trained generative artificial intelligence model

[0571] The server splits the dataset into a training set and a test set, and then trains and evaluates the model.

[0572] Step 4:

[0573] The server collects daily sales performance and new order status data in real time. This data is obtained from POS systems and sensor devices. The real-time data is integrated with existing datasets and then cleaned and formatted again.

[0574] Input: Real-time data

[0575] Output: Updated dataset

[0576] The server processes real-time data quickly and incorporates it into model updates.

[0577] Step 5:

[0578] The server uses a generative artificial intelligence model to predict the production quantities of future promotional items. The predicted data is stored in a database and can be updated and recalculated instantly.

[0579] Input: Updated dataset

[0580] Output: Predicted production quantity

[0581] The server runs a prediction algorithm to calculate production quantities based on future demand.

[0582] Step 6:

[0583] The server notifies the user of the prediction results on their smart device. Notification methods include a dashboard display on the smart device and email notifications. Users can check the prediction results using their smartphone or tablet.

[0584] Input: Predicted production quantity

[0585] Output: Notification to the user

[0586] The server either sends emails using the SMTP protocol or displays results on a web-based dashboard.

[0587] 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.

[0588] System Overview

[0589] This invention provides a system that combines an emotion engine with a system for precisely calculating the production quantity of in-store promotional items, thereby recognizing and reflecting user emotions to achieve even more accurate predictions. This system has means for collecting past sales data and order data, and preprocesses and normalizes the collected data. Furthermore, it uses the preprocessed data to train a generative artificial intelligence model, integrates newly collected data in real time, and predicts future demand. Finally, it incorporates an emotion engine that recognizes user emotions, and reflects user emotion data in the predictions.

[0590] Program Processing Description

[0591] Data collection

[0592] The server connects to the "Sales Management System" and "Order Management System" to collect historical sales and order data. Furthermore, it acquires market trends and competitor information from external data sources to ensure comprehensive data richness and accuracy.

[0593] Data Integration and Preprocessing

[0594] The server stores the collected data in a database and performs duplicate data removal and missing data completion. For example, it applies the historical mean or median to missing items. It also normalizes the data and converts it into a format that is easy for generative artificial intelligence models to learn from.

[0595] Learning of generative artificial intelligence models

[0596] The server uses pre-processed data to train a generative artificial intelligence model (e.g., LSTM or Transformer). The model recognizes past patterns and adjusts the parameters necessary for forecasting future demand.

[0597] Real-time data integration

[0598] The server periodically acquires real-time data such as daily acquisition results and new order status, and integrates it with the existing database. This real-time data is also properly preprocessed and normalized to maintain a consistent dataset.

[0599] Collection of user sentiment data

[0600] The user connects to the server via their device, and the emotion engine recognizes the user's emotions. The emotion engine uses facial recognition and voice analysis to understand the user's emotional state. The user's emotional data is sent to the server and stored in a database.

[0601] Analysis and integration of emotional data

[0602] The server analyzes the collected user sentiment data and incorporates the results into a generative artificial intelligence model. This enables highly accurate predictions that take into account the impact of sentiment data on demand forecasting.

[0603] Prediction and calculation

[0604] The server uses a generative artificial intelligence model to input the latest integrated data (sales data, order data, real-time data, and sentiment data) and predict the production quantity of future promotional items. The prediction results include an approximate production quantity and range of variation, taking sentiment data into account.

[0605] Providing results

[0606] The server saves the calculated prediction results to a database and notifies the user's device. Notification methods include dashboard display and email notifications. Users can check the prediction results through their device and make adjustments as needed.

[0607] Specific example

[0608] For example, when a new product is launched, the user enters product information into the "sales management system." The server collects necessary data based on past data of similar products, market trends, and user sentiment data, and makes predictions using a generative artificial intelligence model. For instance, if a user shows very positive sentiment, this is analyzed as an indication of increased demand, and the predicted production quantity will also increase. The results are notified to the user, who can then plan and execute a production plan for promotional items based on this information. This significantly reduces the risk of inventory shortages or surpluses, enabling the efficient production of promotional items.

[0609] As described above, this system significantly improves the accuracy of predicting the production quantity of promotional items, and by incorporating user sentiment data into the prediction process, it achieves more precise and adaptable predictions.

[0610] The following describes the processing flow.

[0611] Step 1:

[0612] The server connects to the sales management system and order management system, collecting historical sales and order data. Furthermore, it also obtains market trends and competitor information from external data sources.

[0613] Step 2:

[0614] The server stores the collected data in a database and performs duplicate removal and missing data insertion. For missing data, it applies the historical mean or median.

[0615] Step 3:

[0616] The server normalizes the pre-processed data and transforms it into a format that is easy for generative artificial intelligence models to learn. Numerical data is scaled, and categorical data is encoded.

[0617] Step 4:

[0618] The server uses pre-processed data to train generative artificial intelligence models (e.g., LSTM or Transformer). In the process, it extracts patterns from past data.

[0619] Step 5:

[0620] The server evaluates the trained generative artificial intelligence model on a validation dataset to verify the model's accuracy. It then adjusts hyperparameters as needed to build an optimal predictive model.

[0621] Step 6:

[0622] The server periodically acquires real-time data such as daily acquisition results and new order status, and this data is similarly pre-processed and normalized. The real-time data is then integrated into the database.

[0623] Step 7:

[0624] The user connects to the server via their device and activates the emotion engine. The emotion engine collects the user's emotional data through facial recognition and voice analysis, and sends this data to the server.

[0625] Step 8:

[0626] The server analyzes the collected user sentiment data and integrates the results into a generative artificial intelligence model. It analyzes the impact of the sentiment data on predictions and reflects this in the model.

[0627] Step 9:

[0628] The server uses a generative artificial intelligence model to input integrated, up-to-date data (sales data, order data, real-time data, and sentiment data) and predict the production quantity of future promotional items.

[0629] Step 10:

[0630] The server saves the prediction results to a database and notifies the user's device. Notifications are made through methods such as dashboard displays and email notifications.

[0631] Step 11:

[0632] Users access the server from their terminal and view the predicted production quantity on the dashboard. If necessary, they adjust and determine the actual production quantity based on the prediction results.

[0633] Step 12:

[0634] The user sends the determined production quantity from their terminal to the server. The server then saves this information back to the database and sends the necessary production instructions to the order management system.

[0635] Specific example

[0636] For example, when a new product launch is decided, the user enters the new product information into the sales management system. The server collects historical data on similar products and market information, as well as user sentiment data. If the user shows positive sentiment, it is analyzed that this positive sentiment will lead to increased demand. The server inputs this data into a generative artificial intelligence model to predict the production quantity of promotional items with high accuracy. The prediction results are notified to the user via a dashboard or email, and the user then decides on the appropriate production quantity based on that information.

[0637] (Example 2)

[0638] 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".

[0639] Currently, many companies predict the production quantity of promotional items based on past sales and order data, but the accuracy of this method is limited. Furthermore, even when utilizing real-time data, the prediction accuracy does not improve sufficiently because qualitative data such as user sentiment is not considered. Additionally, there is a lack of timely and effective methods for notifying users of the prediction results. To address these challenges, a more accurate and flexible prediction system is needed.

[0640] 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.

[0641] In this invention, the server includes means for collecting past sales data and order data; means for preprocessing and normalizing the collected data; means for training a generative artificial intelligence model using the preprocessed data; means for integrating newly collected data in real time into the generative artificial intelligence model; means for collecting user sentiment data; means for analyzing the collected user sentiment data and integrating it into the generative artificial intelligence model; means for predicting future demand using the generative artificial intelligence model and automatically calculating production quantities; and means for notifying the user of the calculated prediction results. This enables highly accurate demand forecasting that includes user sentiment data.

[0642] "Past sales data" refers to data that records the past sales performance of a particular product or service.

[0643] "Order data" refers to data related to orders placed by companies to receive the supply of products or services.

[0644] "Means of collection" refers to methods or devices for acquiring and storing specific information.

[0645] "Preprocessing" refers to the process of shaping, cleaning, and normalizing data, which is necessary for analysis and model training.

[0646] "Normalization" is the process of unifying the variability of data and bringing it within a specific range.

[0647] A "generative artificial intelligence model" is a type of AI algorithm that learns from collected data and uses it to predict future situations.

[0648] "Real-time data" refers to the latest information on ongoing events and situations.

[0649] "Emotional data" refers to data that records information about a user's emotional state.

[0650] "Analysis" is a scientific method for finding patterns and rules based on collected data.

[0651] "Prediction" is the process of estimating future situations or events based on collected data.

[0652] "Notification" refers to an action or mechanism for informing a user of specific information.

[0653] A "dashboard display" is an interface that provides information to users visually.

[0654] "Email notification" refers to a method of sending information to users via email.

[0655] A "cloud database" is an online database that can be accessed via the internet.

[0656] "External data sources" refer to data provided by external services or data providers, rather than data from within a company.

[0657] This invention provides a system for accurately calculating the production quantity of in-store promotional items. This system collects historical sales and order data, preprocesses and normalizes this data, and then trains a generative artificial intelligence model. Furthermore, it integrates newly collected data and user sentiment data in real time to achieve advanced demand forecasting and production quantity calculation.

[0658] Data collection

[0659] Sales and order data collection via server

[0660] The server periodically connects to the "sales management system" and "order management system" via specific APIs to automatically collect the necessary data. API calls, for example, save sales and order data for the past year to the database.

[0661] Server-based market trend and competitor information gathering

[0662] The server automatically retrieves market trends and competitor information from external data sources (e.g., Google Trends and social media analysis tools). This ensures the richness and accuracy of the data.

[0663] Data Integration and Preprocessing

[0664] Server-based data storage and cleaning

[0665] The server stores the collected data in a database and performs duplicate data removal and missing data imputation. Missing data is filled in using historical mean and median values.

[0666] Data normalization by the server

[0667] The server converts the data into a unified format and shapes it so that it can be easily trained by generative artificial intelligence models. Specifically, it scales sales quantities to a range of 0 to 1 and performs one-hot encoding on categorical data.

[0668] Learning of generative artificial intelligence models

[0669] Server-based AI model training

[0670] The server uses the pre-processed data to train generative artificial intelligence models (e.g., LSTM or Transformer). This training process improves the accuracy of predictions.

[0671] Real-time data integration

[0672] Real-time data collection and integration by the server.

[0673] The server acquires real-time data such as daily sales performance and order status, and integrates it with the existing database. The new data is appropriately preprocessed and merged into the existing dataset.

[0674] Collection of user sentiment data

[0675] User provision of sentiment data

[0676] The user, through their device, allows the emotion engine to acquire their emotional state using facial recognition and voice analysis. The user provides emotional information to the emotion engine, for example, using the camera or microphone.

[0677] Storage of emotional data by a server

[0678] The server stores the acquired user sentiment data in a database and integrates it with other data.

[0679] Analysis and integration of emotional data

[0680] Server-based analysis of emotional data

[0681] The server analyzes emotional data and incorporates it into a generative artificial intelligence model to improve prediction accuracy. For example, positive emotions may indicate increased demand, thus influencing the predicted values.

[0682] Prediction and calculation

[0683] Server-based future demand forecasting

[0684] The server uses a generative artificial intelligence model to calculate the future production quantity of promotional items based on integrated, up-to-date data. The prediction results include sales data, order data, real-time data, and sentiment data.

[0685] Providing results

[0686] Server-based notification of results

[0687] The server saves the calculated prediction results to a database and notifies the user's device. Notifications are made via a dashboard display or email.

[0688] User review and adjustment of results

[0689] Users can view the prediction results through their devices and make adjustments as needed. This allows for the creation of efficient promotional material production plans. For example, if a user expresses positive emotions, the predicted production quantity will also increase.

[0690] Specific example

[0691] For example, when a new product is launched, the user enters product information into the "sales management system." The server uses a generative artificial intelligence model to make predictions based on past data of similar products, market trends, and user sentiment data. For instance, if a user expresses very positive sentiment, an increase in demand is predicted, and the production quantity will also increase. The results are notified to the user, who can then plan and execute a promotional item production plan based on that information.

[0692] Example of a prompt

[0693] User: I have positive feelings about the new product A. Please provide a sales forecast.

[0694] Server: Based on past sales data, order data, market trend information, and user sentiment data, a generative AI model predicts the demand for new product A.

[0695] This system enables highly accurate demand forecasting, including user sentiment data, leading to more efficient promotional activities.

[0696] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0697] Step 1: Data Collection

[0698] The server periodically connects to the "Sales Management System" and "Order Management System" to collect historical sales and order data. Specifically, it sends data collection requests via API and stores the past year's worth of data in the database. The input is the API request, and the output is the retrieved sales and order data.

[0699] Step 2: Gather market trends and competitive information

[0700] The server connects to external data sources (e.g., Google Trends and social media analysis tools) to collect market trends and competitor information. Specifically, it sends requests via APIs and stores the collected data in a database. The input is information requests from external data sources, and the output is the retrieved market trends and competitor information.

[0701] Step 3: Data preprocessing and normalization

[0702] The server stores the collected data in a database and performs duplicate data removal and missing data imputation. Specifically, it removes duplicate data and imputes missing data with historical mean or median values. Furthermore, it converts the data into a unified format (e.g., numerical scaling, one-hot encoding of categorical data) to make it easier for AI models to learn. The input is the various data acquired, and the output is the pre-processed dataset.

[0703] Step 4: Training a generative artificial intelligence model

[0704] The server uses pre-processed data to train generative artificial intelligence models (e.g., LSTM and Transformer). Specifically, it inputs data into the AI ​​model and adjusts the model's parameters while training. The input is the pre-processed dataset, and the output is the optimized AI model.

[0705] Step 5: Collect and integrate real-time data

[0706] The server periodically retrieves daily real-time data (new sales figures and order status) and integrates it with the existing database. Specifically, it collects real-time data via an API and merges the new data into the existing dataset. The input is real-time data, and the output is the updated integrated dataset.

[0707] Step 6: Collecting user sentiment data

[0708] The user accesses the emotion engine through their device and acquires their emotional state using facial recognition and voice analysis. Specifically, the user uses the camera and microphone to provide emotional information to the emotion engine. The input is the user's facial expressions and voice data, and the output is the analyzed emotional data.

[0709] Step 7: Analysis and integration of emotional data

[0710] The server analyzes user sentiment data and incorporates it into a generative artificial intelligence model. Specifically, it quantifies sentiment data, cross-analyzes it with sales data, and inputs it into the AI ​​model to improve prediction accuracy. The input is sentiment data and an integrated dataset, and the output is the final dataset with the sentiment data integrated.

[0711] Step 8: Forecast future demand and calculate production quantities.

[0712] The server uses a generative artificial intelligence model based on the latest integrated data to predict the production quantity of future promotional items. Specifically, it inputs the integrated data into the AI ​​model and performs demand forecasting. As a result, a demand forecast value is output. The input is the integrated dataset, and the output is the predicted production quantity of promotional items.

[0713] Step 9: Notification of Results

[0714] The server saves the calculated prediction results to a database and notifies the user's terminal. Notifications are made via a dashboard display or email. Specifically, the server displays the prediction results on a dashboard or sends them to the user via email. The input is the prediction result data, and the output is the notification to the user.

[0715] Step 10: Review and adjust results

[0716] Users can view prediction results via their devices and make adjustments as needed. Specifically, users view prediction results on a dashboard and then create a production plan for promotional items based on those results. The input is the prediction result, and the output is the adjusted production plan.

[0717] This completes the entire processing flow, enabling highly accurate demand forecasting and production quantity calculation for promotional items.

[0718] (Application Example 2)

[0719] 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."

[0720] Traditional demand forecasting systems make predictions based on sales and order data, but they often lack accuracy because they don't take into account customer emotions and reactions at the actual point of sale. As a result, they can either overestimate or underestimate the quantity of promotional items to be produced. This can lead to inefficient inventory management and potentially lost business opportunities.

[0721] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for collecting past sales data and order data, means for preprocessing and normalizing the collected data, means for training a generative artificial intelligence model using the preprocessed data, means for integrating newly collected data in real time into the generative artificial intelligence model, means for predicting future demand using the generative artificial intelligence model and automatically calculating the production quantity, means for notifying the user of the calculated prediction results, and means for collecting user sentiment data and integrating it into the generative artificial intelligence model. As a result, customer sentiment data can be incorporated into demand forecasting, improving forecasting accuracy and enabling more accurate estimation of the production quantity of promotional items.

[0722] "Past sales data" refers to data that shows the history of sales transactions that have taken place in the past, and includes details such as quantity, date, and product name.

[0723] "Order data" refers to data related to orders for goods and services, including information such as the order date, order quantity, and ordering source information.

[0724] "Collection means" refers to methods or devices for collecting specific data, and includes means of acquiring information using sensors, APIs, databases, etc.

[0725] "Preprocessing means" refers to methods and techniques for converting collected data into an appropriate format, making it easy to use for analysis and model training.

[0726] "Methods of normalization" refer to methods and techniques for unifying data variability and bringing it within a certain range, and include techniques such as standardization and scaling.

[0727] "Generative artificial intelligence models" refer to machine learning algorithms and deep learning models that learn patterns from large amounts of data and predict future outcomes.

[0728] "Real-time data" refers to data about events currently in progress or data generated at that moment, including data that is collected and processed immediately.

[0729] "Integration methods" refer to methods and techniques for combining information obtained from multiple data sources to create a unified dataset.

[0730] "Predictive means" refers to methods and techniques that use generative artificial intelligence models to estimate future demand and determine the required production quantity.

[0731] "Notification means" refers to methods and technologies for informing users of the calculated prediction results, and includes means such as dashboards, smart device displays, and email notifications.

[0732] "Emotional data" refers to information that indicates a user's emotional state, and includes data obtained from facial expressions, voice, behavioral patterns, etc.

[0733] System Overview

[0734] The present invention is a system for predicting the production quantity of promotional items using user sentiment data, and mainly includes the following elements: means for collecting past sales data and order data, preprocessing means, normalization means, real-time data integration means, generative artificial intelligence model learning means, demand forecasting means, notification means, and sentiment data collection means.

[0735] Program Embodiment

[0736] 1. Data Collection

[0737] The server connects to the "sales management system" and "order management system" to collect historical sales and order data. Furthermore, it acquires market trends and competitor information from external data sources to ensure comprehensive data richness and accuracy. Using smartphones and smart glasses, it collects user emotion data in real time from customers' facial expressions and voices in stores.

[0738] 2. Data Integration and Preprocessing

[0739] The server stores the collected data in a database and uses Pandas and NumPy to remove duplicate data and impute missing data. For example, it applies the historical mean or median to missing items. It also normalizes the data and converts it into a format that is easy for generative artificial intelligence models to learn.

[0740] 3. Training of generative artificial intelligence models

[0741] The server uses preprocessed data to train generative artificial intelligence models (e.g., LSTM or Transformer) using TensorFlow or Keras. The models recognize past patterns and adjust the parameters necessary for future demand forecasting.

[0742] 4. Integration of real-time data

[0743] The server periodically acquires real-time data such as daily acquisition results and new order status, and integrates it with existing databases using Apache Kafka or RabbitMQ. This real-time data is also preprocessed and normalized to maintain a consistent dataset.

[0744] 5. Collection of emotional data

[0745] The user's smartphone or smart glasses use camera APIs and microphone APIs to recognize the user's emotional state from their facial expressions and voice via an emotion engine, and then send that data to the server.

[0746] 6. Analysis and Integration of Emotional Data

[0747] The server analyzes the collected user sentiment data using Scikit-learn and TensorFlow, and incorporates the results into a generative artificial intelligence model. This enables highly accurate predictions that take into account the impact of sentiment data on demand forecasting.

[0748] 7. Prediction and Calculation

[0749] The server inputs the latest integrated data (sales data, order data, real-time data, sentiment data) into a generative artificial intelligence model to predict the production quantity of future promotional items. For example, if a user expresses very positive sentiment, this is analyzed as an indication of increased demand, and the predicted production quantity will also increase.

[0750] 8. Providing the results

[0751] The server notifies users of the calculated prediction results through a smartphone app using React Native or a smart glasses app using ARKit / ARCore. Based on this, users can plan and execute promotional product production plans.

[0752] Specific example

[0753] For example, imagine a store selling a new product, with staff wearing smart glasses. When a customer shows interest in the product, emotional data is collected. This data is sent to a server in real time and integrated with past sales data and market trends. A generative artificial intelligence model analyzes this data to predict demand and calculate the quantity of promotional items to produce for the new product. Staff can instantly check the information through the smart glasses' HUD and take appropriate action.

[0754] Example of a prompt

[0755] Prompt: "To accurately predict the quantity of promotional items to be produced at stores, collect historical sales data, order data, and user sentiment data in real time, and use a generative artificial intelligence model to make predictions."

[0756] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0757] Step 1:

[0758] Data collection

[0759] The server connects to the "sales management system" and "order management system" to collect historical sales and order data. Specifically, it retrieves data from the database using a REST API and also collects market trends and competitor information via external APIs. In addition, smart devices (smartphones and smart glasses) collect customer facial expressions and voices in real time. This data is sent to the server as raw data and stored in the database.

[0760] Inputs: Sales data, order data, market trend data, competitor information, sentiment data

[0761] Output: Collected raw dataset

[0762] Step 2:

[0763] Data Integration and Preprocessing

[0764] The server processes the collected raw data using Pandas and NumPy. Specifically, it removes duplicate data and imputes missing data with the mean or median. Next, it normalizes the data and converts it into a format suitable for training the model. For example, it scales numerical data and one-hot encodes categorical data.

[0765] Input: Collected raw dataset

[0766] Output: Preprocessed normalized dataset

[0767] Step 3:

[0768] Learning of generative artificial intelligence models

[0769] The server uses TensorFlow and Keras to train generative artificial intelligence models (such as LSTM and Transformer) on preprocessed datasets. Specifically, it splits the data into a training set and a test set, trains the model using the training set, and evaluates it using the test set. It adjusts parameters and sets the number of epochs to ensure the model learns properly.

[0770] Input: Preprocessed normalized dataset

[0771] Output: Trained generative artificial intelligence model

[0772] Step 4:

[0773] Real-time data integration

[0774] The server uses Apache Kafka and RabbitMQ to collect new sales data, order status, and other information in real time and integrate it with existing databases. This data is also preprocessed and normalized. Real-time data is sent to the server as soon as it is collected, ensuring that it is always up-to-date.

[0775] Input: Real-time dataset

[0776] Output: Integrated modern database

[0777] Step 5:

[0778] Collection of emotional data

[0779] The user's smart device uses camera and microphone APIs to collect customer facial expressions and voice data, and analyzes the user's emotional state through an emotion engine. This emotional data is sent to a server via WebSocket or REST API and stored in a database.

[0780] Input: Facial expression data, audio data

[0781] Output: Analyzed sentiment data

[0782] Step 6:

[0783] Analysis and integration of emotional data

[0784] The server uses Scikit-learn and TensorFlow to analyze the collected sentiment data. Specifically, it analyzes the sentiment data and extracts factors that influence demand forecasting. Then, this sentiment data is integrated into a generative artificial intelligence model to improve the model's accuracy.

[0785] Input: Analyzed sentiment data

[0786] Output: Integrated sentiment dataset

[0787] Step 7:

[0788] Prediction and calculation

[0789] The server inputs the integrated dataset into a generative artificial intelligence model to predict future demand. Specifically, it uses LSTM and Transformer models to calculate predicted sales and production quantities. For example, if there are many positive user sentiments, it analyzes this as an indication of increased demand and increases the predicted production quantity accordingly.

[0790] Input: Integrated dataset

[0791] Output: Estimated production quantity of promotional items

[0792] Step 8:

[0793] Providing results

[0794] The server notifies users of prediction results through a smartphone app using React Native or a smart glasses app using ARKit / ARCore. Specifically, the prediction results are displayed on a dashboard, allowing users to immediately check and take action.

[0795] Input: Estimated production quantity of promotional items

[0796] Output: Notifications to the user (dashboard display, smart device display)

[0797] Example of a prompt

[0798] Prompt: "To accurately predict the quantity of promotional items to be produced at stores, collect historical sales data, order data, and user sentiment data in real time, and use a generative artificial intelligence model to make predictions."

[0799] 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.

[0800] 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 the following. 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 indicated 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.

[0801] 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.

[0802] [Third Embodiment]

[0803] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0804] 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.

[0805] 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).

[0806] 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.

[0807] 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.

[0808] 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).

[0809] 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.

[0810] 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.

[0811] 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.

[0812] 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.

[0813] 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.

[0814] 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".

[0815] System Overview

[0816] This invention is a system for more precisely calculating the production quantity of in-store promotional items. Specifically, a server collects past sales and order data, preprocesses and normalizes this data, and trains a generative artificial intelligence model. Newly acquired real-time data is integrated into this model to predict future demand. By notifying the user of the prediction results, the system automatically calculates the appropriate production quantity of promotional items.

[0817] Program Processing Description

[0818] Data collection

[0819] The server accesses the "sales management system" and "order management system" to collect historical sales and order data. It also collects market trends and competitor information from external data sources as needed. This ensures the richness and accuracy of the data.

[0820] Data Integration and Preprocessing

[0821] The server removes duplicate data from the collected data and performs imputation on incomplete data. For example, if there are missing values ​​in order data, it will use the mean or median to fill them in. It also normalizes the data and converts it into a format that is easy for generative artificial intelligence models to learn from.

[0822] Learning of generative artificial intelligence models

[0823] The server uses preprocessed data to train generative artificial intelligence models (e.g., LSTM or Transformer). These models have the ability to extract patterns from historical data and predict future demand. After training is complete, the models are evaluated using validation datasets to check their accuracy and make any necessary adjustments.

[0824] Real-time data integration

[0825] The server collects daily acquisition data and new order status data in real time. This real-time data is integrated with existing datasets and undergoes further data cleaning and formatting. This ensures that the model always incorporates the latest information.

[0826] Prediction and calculation

[0827] The server uses a generative artificial intelligence model to input the latest data, including real-time data, and predicts the production quantity of future promotional items. This prediction result is stored in a database and can be updated and recalculated immediately if necessary.

[0828] Providing results

[0829] The server notifies the user's device of the prediction results. Notification methods include dashboard display and email notifications. Users can connect to the server from their device to check the prediction results. This allows users to plan and execute promotional item production based on appropriate production quantities.

[0830] Specific example

[0831] For example, when a new product is launched, the user enters product information into the "sales management system." The server collects necessary data based on past data of similar products and market trends, and makes predictions using a generative artificial intelligence model. As a result, the predicted production quantity is notified to the user. Based on that quantity, the user can proceed with the production of promotional materials for the new product. This entire process significantly reduces the risk of inventory shortages or surpluses, enabling the efficient production of promotional materials.

[0832] As described above, this system is a powerful tool for significantly improving the accuracy of predicting the production quantity of promotional items, streamlining the execution process, and optimizing risk management.

[0833] The following describes the processing flow.

[0834] Step 1:

[0835] The server connects to the "sales management system" and "order management system" to collect historical sales and order data. It also obtains market trends and competitor information from external data sources.

[0836] Step 2:

[0837] The server stores the collected data in a database and performs duplicate data removal and data interpolation. For example, it applies historical mean or median values ​​to missing items.

[0838] Step 3:

[0839] The server normalizes the pre-processed data and transforms it into a format that is easy for generative artificial intelligence models to learn from. In this process, numerical data is scaled and categorical data is encoded.

[0840] Step 4:

[0841] The server uses pre-processed data to train a generative artificial intelligence model (e.g., LSTM or Transformer). The model recognizes past patterns and adjusts the parameters necessary for forecasting future demand.

[0842] Step 5:

[0843] The server evaluates the trained generative artificial intelligence model on a validation dataset to verify the model's accuracy. It then adjusts hyperparameters as needed to build an optimal predictive model.

[0844] Step 6:

[0845] The server periodically acquires real-time data such as daily acquisition results and new order status, and integrates it with the existing database. This data is also preprocessed and normalized.

[0846] Step 7:

[0847] The server inputs the latest real-time data into a generative artificial intelligence model to predict future demand for promotional items. The prediction results include approximate production quantities and their range of variation.

[0848] Step 8:

[0849] The server saves the calculated prediction results to a database. Furthermore, it notifies the user's device of the results. This notification may be displayed on a dashboard or sent via email.

[0850] Step 9:

[0851] Users access the server from their terminal and view the predicted production quantity on the dashboard. If necessary, they adjust and determine the actual production quantity based on the prediction results.

[0852] Step 10:

[0853] The user enters the determined production quantity and sends it to the server. The server then saves this information back to the database and sends the necessary production instructions to the order management system.

[0854] (Example 1)

[0855] 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."

[0856] Traditional systems for calculating the production quantity of in-store promotional items often relied on simple predictions based solely on historical data, making it difficult to accurately reflect market fluctuations and real-time conditions. Furthermore, manual data processing and management were required, leading to increased time, effort, and costs. This resulted in frequent inventory shortages or surpluses, making it difficult to create efficient production plans for promotional items.

[0857] 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.

[0858] In this invention, the server includes means for collecting historical sales data and order data, means for eliminating duplicates and imputing missing values ​​in the collected data, and means for normalizing the data and converting it into a format that is easy for a generating AI model to learn. This enables efficient and highly accurate predictions.

[0859] "Past sales data" refers to information about sales records and performance at the time a transaction took place.

[0860] "Order data" refers to records and information related to the ordering and procurement of goods.

[0861] "Means of collection" refers to the functions and devices that allow a server to access multiple data sources and obtain the necessary data.

[0862] "Means of eliminating duplicates" refer to methods or functions for deleting or merging data with identical content from an acquired dataset.

[0863] "Methods for imputing missing values" refer to functions or techniques for filling in missing values ​​or information in a dataset using the mean or median.

[0864] "Methods for normalizing data" refer to methods and functions for preparing data in a way that makes it easier for generative AI models to learn, by transforming it into a certain range or format.

[0865] A "generative AI model" is a machine learning model that uses artificial intelligence technology to learn patterns from past data and predict future trends and demand.

[0866] "Real-time data" refers to the latest information obtained based on ongoing transactions and events.

[0867] "Predictive means" refers to methods and functions for predicting future demand and fluctuations using generative AI models and calculating the results.

[0868] "Means of notification" refer to methods and functions for communicating prediction results to the user, and can take the form of a dashboard display or email notification.

[0869] Modes for carrying out the invention

[0870] This invention is a system for more precisely calculating the production quantity of in-store promotional items, and its main components are a server, terminals, and users. This system is implemented in the following steps.

[0871] System Overview

[0872] This system collects historical sales and order data, preprocesses and normalizes it, and uses it to train a generating AI model. It then integrates newly acquired data in real time with this model to predict future demand. By notifying the user of the prediction results, it automatically calculates the appropriate quantity of promotional items to produce.

[0873] Hardware and software used

[0874] Server: Performs data collection, preprocessing, normalization, AI model training, and prediction.

[0875] Terminal: Provides an interface for the user to view the prediction results.

[0876] Software: Sales management systems, order management systems, cloud databases, access to external data sources, data processing libraries, generative AI model libraries (e.g., LSTM and Transformer).

[0877] Specific example

[0878] When a new product is launched, the user enters product information into the "sales management system." For example, they might enter "details of new product A that is being promoted." The server collects necessary data, including historical data on similar products and market trends, and preprocesses and normalizes it. Then, it uses a generative AI model to predict future demand and calculate the appropriate quantity of promotional items to produce. This result is then communicated to the user.

[0879] Examples of prompt statements

[0880] For example, enter the following as a prompt for a generative AI model:

[0881] When entering the new product "○○" into the sales management system, please predict the required quantity of promotional materials to be produced, taking into account past sales data and market trends.

[0882] The server generates and provides prediction results to the user based on the above procedure. This significantly reduces the risk of inventory shortages or surpluses, and enables efficient planning of promotional item production. Furthermore, continuous integration of real-time data allows for highly accurate predictions that reflect the latest market conditions.

[0883] This system can operate in conjunction with common data infrastructures, regardless of the type of machine learning model or algorithm used. Its flexible application of generative AI models makes it adaptable to a variety of business environments.

[0884] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0885] Step 1:

[0886] The server collects past sales and order data.

[0887] Specifically, the server accesses the "sales management system" and "order management system" every night to retrieve historical sales and order data. This also includes retrieving market trends and competitor information via external APIs as needed. The input is raw data from the sales management system and order management system, and the output is all the collected raw data.

[0888] Step 2:

[0889] The server removes duplicate data and imputes missing values.

[0890] In terms of specific operations, the server analyzes the acquired dataset and removes or merges duplicate data with the same date and product ID. Furthermore, missing values ​​within the dataset are imputed using the mean or median of past data. The input is the collected raw data, and the output is the data with duplicates removed and missing values ​​imputed.

[0891] Step 3:

[0892] The server normalizes the data and converts it into a format that is easy for the generating AI model to learn from.

[0893] In terms of specific operations, the server scales quantitative and price data to a particular range and converts categorical data into numerical data using techniques such as one-hot encoding. The input is data that has undergone deduplicating and imputation, and the output is normalized data that has been converted into a format that is easy for the AI ​​model to learn from.

[0894] Step 4:

[0895] The server trains the generated AI model.

[0896] In terms of operation, the server inputs normalized data into a generative AI model (e.g., LSTM or Transformer) to train the model. During the training process, patterns are extracted from past data, and parameters are adjusted to predict future demand. The input is normalized data, and the output is the trained generative AI model.

[0897] Step 5:

[0898] The server integrates real-time data.

[0899] In terms of specific operations, the server collects daily sales performance and new order data in real time and integrates it into an existing dataset. Data cleaning and formatting are performed again as needed. The input is newly acquired data in real time, and the output is an integrated dataset containing the latest information.

[0900] Step 6:

[0901] The server makes predictions and calculates the production quantity.

[0902] In practice, the server uses a generative AI model to input the latest data and predict the production quantity of future promotional items. This prediction is performed periodically, and the data is saved to the database or updated immediately as needed. The input is the integrated latest data, and the output is the predicted production quantity.

[0903] Step 7:

[0904] The server notifies the user's device of the prediction results.

[0905] In terms of specific operations, the server transmits the prediction results to the user's terminal. At this time, dashboard display and email notifications can be used. When new prediction data is generated, the server displays the update information on the user's dashboard via the web server. It also notifies the user of the prediction results via email, depending on the settings. The input is the predicted production quantity, and the output is the notification to the user.

[0906] (Application Example 1)

[0907] 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."

[0908] The objective of this invention is to provide a system for appropriately predicting the production quantity of promotional items. Conventional methods rely solely on past sales and order data for predictions, making it difficult to reflect new data obtained in real time. As a result, it is not possible to respond quickly to fluctuations in demand. Furthermore, the methods for notifying users of the prediction results are limited, making it difficult for users to easily check them. This leads to risks of inventory shortages or surpluses, making it difficult to produce promotional items efficiently.

[0909] 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.

[0910] In this invention, the server includes means for collecting past sales data and order data; means for preprocessing and normalizing the collected data; means for training a generative artificial intelligence model using the preprocessed data; means for integrating newly collected data in real time into the generative artificial intelligence model; means for predicting future demand using the generative artificial intelligence model and automatically calculating production quantities; means for notifying the user of the calculated prediction results; and means for displaying the notified prediction results on a smart device. This enables precise sales forecasting including real-time data, and allows users to quickly check the prediction results via a smart device and create an appropriate production plan for promotional items.

[0911] "Past sales data" refers to data that records the sales performance of a product over a certain past period.

[0912] "Order data" refers to data that includes information about orders for goods and services.

[0913] "Preprocessing" is the process of converting collected data into a format suitable for analysis and learning.

[0914] "Normalization" is a transformation process that brings data values ​​within a certain range.

[0915] A "generative artificial intelligence model" is an algorithm or structure that learns from large amounts of data and makes predictions or generates new data.

[0916] "New data collected in real time" refers to data that is being collected in real time and is immediately available for processing and analysis.

[0917] "Integration" is the process of combining and linking different datasets into one.

[0918] "Predicting future demand and automatically calculating production quantities" refers to the act of using generative artificial intelligence models to estimate future consumer demand and determine the corresponding production volume.

[0919] "Notification" refers to communicating prediction results or information to the user through specific means.

[0920] A "smart device" is an electronic device that has communication capabilities and computing power, and allows users to display and operate information.

[0921] A "dashboard display" is an interface that visually organizes and presents information.

[0922] "Email notification" refers to a method of sending information in the form of an email.

[0923] A "remote database" is a data storage system located remotely and accessible via a network.

[0924] "External data sources" refer to data obtained from data providers outside the company, such as public databases.

[0925] System Overview

[0926] This invention is a system that streamlines inventory management by precisely predicting the production quantity of promotional items for physical stores. The system collects historical sales and order data, preprocesses and normalizes this data, then uses a generative artificial intelligence model for training and integrates real-time data to predict future demand. The prediction results are then communicated to the user's smart device.

[0927] Program Processing Description

[0928] Data collection

[0929] The server accesses the "sales management system" and "order management system" to collect historical sales and order data. This data is obtained from remote databases and external data sources.

[0930] Data Integration and Preprocessing

[0931] The server removes duplicate data from the collected data and performs imputation on incomplete data. It also normalizes the data and converts it into a format that is easy for generative artificial intelligence models to learn. Python and general-purpose data analysis libraries (such as Pandas and NumPy) are used in this process.

[0932] Learning of generative artificial intelligence models

[0933] The server uses preprocessed data to train generative artificial intelligence models (for example, LSTM or Transformer models using Keras). These models extract patterns from past data and predict future demand.

[0934] Real-time data integration

[0935] The server collects daily sales performance and new order status data in real time. This data is obtained from POS systems and sensor devices. The real-time data is integrated with existing datasets and then cleaned and formatted again.

[0936] Prediction and calculation

[0937] The server uses a generative artificial intelligence model to input the latest data and predict the production quantity of future promotional items. This prediction is stored in a database and can be updated and recalculated instantly as needed.

[0938] Providing results

[0939] The server notifies the user of the prediction results on their smart device. Notification methods include a dashboard display on the smart device and email notifications. Users can check the prediction results using their smartphones or tablets.

[0940] Specific example

[0941] For example, when a new product is launched, the user enters product information into the "sales management system." The server collects necessary data based on past data of similar products and market trends, and uses a generative artificial intelligence model to predict future demand. As a result, the predicted production quantity is notified to the user's smartphone. Based on that quantity, the user can proceed with the production of promotional materials for the new product. This entire process significantly reduces the risk of inventory shortages or surpluses, enabling the efficient production of promotional materials.

[0942] Examples of prompt sentences:

[0943] Collect new sales data and feed it into an AI model to predict future demand. We want to build an application that notifies us of the prediction results in real time. Historical sales data is located here: [sales_data.csv]. Order data is located here: [order_data.csv]. Real-time data is located here: [new_data.csv].

[0944] As described above, this system significantly improves the accuracy of predicting the production quantity of promotional items and is a powerful tool for creating efficient production plans for promotional items.

[0945] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0946] Step 1:

[0947] The server collects historical sales and order data from the "sales management system" and "order management system." This data includes information such as product name, sales quantity, sales date, order quantity, and order date. The collected data is stored in a database.

[0948] Input: Sales data, order data

[0949] Output: Raw data stored in the database

[0950] The server also accesses remote databases and external data sources to obtain market trends and competitive information.

[0951] Step 2:

[0952] The server preprocesses the collected data. It removes duplicate data and imputes incomplete data, for example, by filling in missing values ​​with the median. Furthermore, it normalizes the data and converts it into a format that is easy for AI models to learn from.

[0953] Input: Collected raw data

[0954] Output: Normalized preprocessed data

[0955] Specifically, we will use Python's Pandas and NumPy libraries to perform data cleaning and imputation.

[0956] Step 3:

[0957] The server uses preprocessed data to train a generative artificial intelligence model. This model consists of components such as LSTM (Long Short-Term Memory) and Transformer. The training process is executed using Keras and TensorFlow.

[0958] Input: Normalized preprocessed data

[0959] Output: Trained generative artificial intelligence model

[0960] The server splits the dataset into a training set and a test set, and then trains and evaluates the model.

[0961] Step 4:

[0962] The server collects daily sales performance and new order status data in real time. This data is obtained from POS systems and sensor devices. The real-time data is integrated with existing datasets and then cleaned and formatted again.

[0963] Input: Real-time data

[0964] Output: Updated dataset

[0965] The server processes real-time data quickly and incorporates it into model updates.

[0966] Step 5:

[0967] The server uses a generative artificial intelligence model to predict the production quantities of future promotional items. The predicted data is stored in a database and can be updated and recalculated instantly.

[0968] Input: Updated dataset

[0969] Output: Predicted production quantity

[0970] The server runs a prediction algorithm to calculate production quantities based on future demand.

[0971] Step 6:

[0972] The server notifies the user of the prediction results on their smart device. Notification methods include a dashboard display on the smart device and email notifications. Users can check the prediction results using their smartphone or tablet.

[0973] Input: Predicted production quantity

[0974] Output: Notification to the user

[0975] The server either sends emails using the SMTP protocol or displays results on a web-based dashboard.

[0976] 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.

[0977] System Overview

[0978] This invention provides a system that combines an emotion engine with a system for precisely calculating the production quantity of in-store promotional items, thereby recognizing and reflecting user emotions to achieve even more accurate predictions. This system has means for collecting past sales data and order data, and preprocesses and normalizes the collected data. Furthermore, it uses the preprocessed data to train a generative artificial intelligence model, integrates newly collected data in real time, and predicts future demand. Finally, it incorporates an emotion engine that recognizes user emotions, and reflects user emotion data in the predictions.

[0979] Program Processing Description

[0980] Data collection

[0981] The server connects to the "Sales Management System" and "Order Management System" to collect historical sales and order data. Furthermore, it acquires market trends and competitor information from external data sources to ensure comprehensive data richness and accuracy.

[0982] Data Integration and Preprocessing

[0983] The server stores the collected data in a database and performs duplicate data removal and missing data completion. For example, it applies the historical mean or median to missing items. It also normalizes the data and converts it into a format that is easy for generative artificial intelligence models to learn from.

[0984] Learning of generative artificial intelligence models

[0985] The server uses pre-processed data to train a generative artificial intelligence model (e.g., LSTM or Transformer). The model recognizes past patterns and adjusts the parameters necessary for forecasting future demand.

[0986] Real-time data integration

[0987] The server periodically acquires real-time data such as daily acquisition results and new order status, and integrates it with the existing database. This real-time data is also properly preprocessed and normalized to maintain a consistent dataset.

[0988] Collection of user sentiment data

[0989] The user connects to the server via their device, and the emotion engine recognizes the user's emotions. The emotion engine uses facial recognition and voice analysis to understand the user's emotional state. The user's emotional data is sent to the server and stored in a database.

[0990] Analysis and integration of emotional data

[0991] The server analyzes the collected user sentiment data and incorporates the results into a generative artificial intelligence model. This enables highly accurate predictions that take into account the impact of sentiment data on demand forecasting.

[0992] Prediction and calculation

[0993] The server uses a generative artificial intelligence model to input the latest integrated data (sales data, order data, real-time data, and sentiment data) and predict the production quantity of future promotional items. The prediction results include an approximate production quantity and range of variation, taking sentiment data into account.

[0994] Providing results

[0995] The server saves the calculated prediction results to a database and notifies the user's device. Notification methods include dashboard display and email notifications. Users can check the prediction results through their device and make adjustments as needed.

[0996] Specific example

[0997] For example, when a new product is launched, the user enters product information into the "sales management system." The server collects necessary data based on past data of similar products, market trends, and user sentiment data, and makes predictions using a generative artificial intelligence model. For instance, if a user shows very positive sentiment, this is analyzed as an indication of increased demand, and the predicted production quantity will also increase. The results are notified to the user, who can then plan and execute a production plan for promotional items based on this information. This significantly reduces the risk of inventory shortages or surpluses, enabling the efficient production of promotional items.

[0998] As described above, this system significantly improves the accuracy of predicting the production quantity of promotional items, and by incorporating user sentiment data into the prediction process, it achieves more precise and adaptable predictions.

[0999] The following describes the processing flow.

[1000] Step 1:

[1001] The server connects to the sales management system and order management system, collecting historical sales and order data. Furthermore, it also obtains market trends and competitor information from external data sources.

[1002] Step 2:

[1003] The server stores the collected data in a database and performs duplicate removal and missing data insertion. For missing data, it applies the historical mean or median.

[1004] Step 3:

[1005] The server normalizes the pre-processed data and transforms it into a format that is easy for generative artificial intelligence models to learn. Numerical data is scaled, and categorical data is encoded.

[1006] Step 4:

[1007] The server uses pre-processed data to train generative artificial intelligence models (e.g., LSTM or Transformer). In the process, it extracts patterns from past data.

[1008] Step 5:

[1009] The server evaluates the trained generative artificial intelligence model on a validation dataset to verify the model's accuracy. It then adjusts hyperparameters as needed to build an optimal predictive model.

[1010] Step 6:

[1011] The server periodically acquires real-time data such as daily acquisition results and new order status, and this data is similarly pre-processed and normalized. The real-time data is then integrated into the database.

[1012] Step 7:

[1013] The user connects to the server via their device and activates the emotion engine. The emotion engine collects the user's emotional data through facial recognition and voice analysis, and sends this data to the server.

[1014] Step 8:

[1015] The server analyzes the collected user sentiment data and integrates the results into a generative artificial intelligence model. It analyzes the impact of the sentiment data on predictions and reflects this in the model.

[1016] Step 9:

[1017] The server uses a generative artificial intelligence model to input integrated, up-to-date data (sales data, order data, real-time data, and sentiment data) and predict the production quantity of future promotional items.

[1018] Step 10:

[1019] The server saves the prediction results to a database and notifies the user's device. Notifications are made through methods such as dashboard displays and email notifications.

[1020] Step 11:

[1021] Users access the server from their terminal and view the predicted production quantity on the dashboard. If necessary, they adjust and determine the actual production quantity based on the prediction results.

[1022] Step 12:

[1023] The user sends the determined production quantity from their terminal to the server. The server then saves this information back to the database and sends the necessary production instructions to the order management system.

[1024] Specific example

[1025] For example, when a new product launch is decided, the user enters the new product information into the sales management system. The server collects historical data on similar products and market information, as well as user sentiment data. If the user shows positive sentiment, it is analyzed that this positive sentiment will lead to increased demand. The server inputs this data into a generative artificial intelligence model to predict the production quantity of promotional items with high accuracy. The prediction results are notified to the user via a dashboard or email, and the user then decides on the appropriate production quantity based on that information.

[1026] (Example 2)

[1027] 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."

[1028] Currently, many companies predict the production quantity of promotional items based on past sales and order data, but the accuracy of this method is limited. Furthermore, even when utilizing real-time data, the prediction accuracy does not improve sufficiently because qualitative data such as user sentiment is not considered. Additionally, there is a lack of timely and effective methods for notifying users of the prediction results. To address these challenges, a more accurate and flexible prediction system is needed.

[1029] 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.

[1030] In this invention, the server includes means for collecting past sales data and order data; means for preprocessing and normalizing the collected data; means for training a generative artificial intelligence model using the preprocessed data; means for integrating newly collected data in real time into the generative artificial intelligence model; means for collecting user sentiment data; means for analyzing the collected user sentiment data and integrating it into the generative artificial intelligence model; means for predicting future demand using the generative artificial intelligence model and automatically calculating production quantities; and means for notifying the user of the calculated prediction results. This enables highly accurate demand forecasting that includes user sentiment data.

[1031] "Past sales data" refers to data that records the past sales performance of a particular product or service.

[1032] "Order data" refers to data related to orders placed by companies to receive the supply of products or services.

[1033] "Means of collection" refers to methods or devices for acquiring and storing specific information.

[1034] "Preprocessing" refers to the process of shaping, cleaning, and normalizing data, which is necessary for analysis and model training.

[1035] "Normalization" is the process of unifying the variability of data and bringing it within a specific range.

[1036] A "generative artificial intelligence model" is a type of AI algorithm that learns from collected data and uses it to predict future situations.

[1037] "Real-time data" refers to the latest information on ongoing events and situations.

[1038] "Emotional data" refers to data that records information about a user's emotional state.

[1039] "Analysis" is a scientific method for finding patterns and rules based on collected data.

[1040] "Prediction" is the process of estimating future situations or events based on collected data.

[1041] "Notification" refers to an action or mechanism for informing a user of specific information.

[1042] A "dashboard display" is an interface that provides information to users visually.

[1043] "Email notification" refers to a method of sending information to users via email.

[1044] A "cloud database" is an online database that can be accessed via the internet.

[1045] "External data sources" refer to data provided by external services or data providers, rather than data from within a company.

[1046] This invention provides a system for accurately calculating the production quantity of in-store promotional items. This system collects historical sales and order data, preprocesses and normalizes this data, and then trains a generative artificial intelligence model. Furthermore, it integrates newly collected data and user sentiment data in real time to achieve advanced demand forecasting and production quantity calculation.

[1047] Data collection

[1048] Sales and order data collection via server

[1049] The server periodically connects to the "sales management system" and "order management system" via specific APIs to automatically collect the necessary data. API calls, for example, save sales and order data for the past year to the database.

[1050] Server-based market trend and competitor information gathering

[1051] The server automatically retrieves market trends and competitor information from external data sources (e.g., Google Trends and social media analysis tools). This ensures the richness and accuracy of the data.

[1052] Data Integration and Preprocessing

[1053] Server-based data storage and cleaning

[1054] The server stores the collected data in a database and performs duplicate data removal and missing data imputation. Missing data is filled in using historical mean and median values.

[1055] Data normalization by the server

[1056] The server converts the data into a unified format and shapes it so that it can be easily trained by generative artificial intelligence models. Specifically, it scales sales quantities to a range of 0 to 1 and performs one-hot encoding on categorical data.

[1057] Learning of generative artificial intelligence models

[1058] Server-based AI model training

[1059] The server uses the pre-processed data to train generative artificial intelligence models (e.g., LSTM or Transformer). This training process improves the accuracy of predictions.

[1060] Real-time data integration

[1061] Real-time data collection and integration by the server.

[1062] The server acquires real-time data such as daily sales performance and order status, and integrates it with the existing database. The new data is appropriately preprocessed and merged into the existing dataset.

[1063] Collection of user sentiment data

[1064] User provision of sentiment data

[1065] The user, through their device, allows the emotion engine to acquire their emotional state using facial recognition and voice analysis. The user provides emotional information to the emotion engine, for example, using the camera or microphone.

[1066] Storage of emotional data by a server

[1067] The server stores the acquired user sentiment data in a database and integrates it with other data.

[1068] Analysis and integration of emotional data

[1069] Server-based analysis of emotional data

[1070] The server analyzes emotional data and incorporates it into a generative artificial intelligence model to improve prediction accuracy. For example, positive emotions may indicate increased demand, thus influencing the predicted values.

[1071] Prediction and calculation

[1072] Server-based future demand forecasting

[1073] The server uses a generative artificial intelligence model to calculate the future production quantity of promotional items based on integrated, up-to-date data. The prediction results include sales data, order data, real-time data, and sentiment data.

[1074] Providing results

[1075] Server-based notification of results

[1076] The server saves the calculated prediction results to a database and notifies the user's device. Notifications are made via a dashboard display or email.

[1077] User review and adjustment of results

[1078] Users can view the prediction results through their devices and make adjustments as needed. This allows for the creation of efficient promotional material production plans. For example, if a user expresses positive emotions, the predicted production quantity will also increase.

[1079] Specific example

[1080] For example, when a new product is launched, the user enters product information into the "sales management system." The server uses a generative artificial intelligence model to make predictions based on past data of similar products, market trends, and user sentiment data. For instance, if a user expresses very positive sentiment, an increase in demand is predicted, and the production quantity will also increase. The results are notified to the user, who can then plan and execute a promotional item production plan based on that information.

[1081] Example of a prompt

[1082] User: I have positive feelings about the new product A. Please provide a sales forecast.

[1083] Server: Based on past sales data, order data, market trend information, and user sentiment data, a generative AI model predicts the demand for new product A.

[1084] This system enables highly accurate demand forecasting, including user sentiment data, leading to more efficient promotional activities.

[1085] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1086] Step 1: Data Collection

[1087] The server periodically connects to the "Sales Management System" and "Order Management System" to collect historical sales and order data. Specifically, it sends data collection requests via API and stores the past year's worth of data in the database. The input is the API request, and the output is the retrieved sales and order data.

[1088] Step 2: Gather market trends and competitive information

[1089] The server connects to external data sources (e.g., Google Trends and social media analysis tools) to collect market trends and competitor information. Specifically, it sends requests via APIs and stores the collected data in a database. The input is information requests from external data sources, and the output is the retrieved market trends and competitor information.

[1090] Step 3: Data preprocessing and normalization

[1091] The server stores the collected data in a database and performs duplicate data removal and missing data imputation. Specifically, it removes duplicate data and imputes missing data with historical mean or median values. Furthermore, it converts the data into a unified format (e.g., numerical scaling, one-hot encoding of categorical data) to make it easier for AI models to learn. The input is the various data acquired, and the output is the pre-processed dataset.

[1092] Step 4: Training a generative artificial intelligence model

[1093] The server uses pre-processed data to train generative artificial intelligence models (e.g., LSTM and Transformer). Specifically, it inputs data into the AI ​​model and adjusts the model's parameters while training. The input is the pre-processed dataset, and the output is the optimized AI model.

[1094] Step 5: Collect and integrate real-time data

[1095] The server periodically retrieves daily real-time data (new sales figures and order status) and integrates it with the existing database. Specifically, it collects real-time data via an API and merges the new data into the existing dataset. The input is real-time data, and the output is the updated integrated dataset.

[1096] Step 6: Collecting user sentiment data

[1097] The user accesses the emotion engine through their device and acquires their emotional state using facial recognition and voice analysis. Specifically, the user uses the camera and microphone to provide emotional information to the emotion engine. The input is the user's facial expressions and voice data, and the output is the analyzed emotional data.

[1098] Step 7: Analysis and integration of emotional data

[1099] The server analyzes user sentiment data and incorporates it into a generative artificial intelligence model. Specifically, it quantifies sentiment data, cross-analyzes it with sales data, and inputs it into the AI ​​model to improve prediction accuracy. The input is sentiment data and an integrated dataset, and the output is the final dataset with the sentiment data integrated.

[1100] Step 8: Forecast future demand and calculate production quantities.

[1101] The server uses a generative artificial intelligence model based on the latest integrated data to predict the production quantity of future promotional items. Specifically, it inputs the integrated data into the AI ​​model and performs demand forecasting. As a result, a demand forecast value is output. The input is the integrated dataset, and the output is the predicted production quantity of promotional items.

[1102] Step 9: Notification of Results

[1103] The server saves the calculated prediction results to a database and notifies the user's terminal. Notifications are made via a dashboard display or email. Specifically, the server displays the prediction results on a dashboard or sends them to the user via email. The input is the prediction result data, and the output is the notification to the user.

[1104] Step 10: Review and adjust results

[1105] Users can view prediction results via their devices and make adjustments as needed. Specifically, users view prediction results on a dashboard and then create a production plan for promotional items based on those results. The input is the prediction result, and the output is the adjusted production plan.

[1106] This completes the entire processing flow, enabling highly accurate demand forecasting and production quantity calculation for promotional items.

[1107] (Application Example 2)

[1108] 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."

[1109] Traditional demand forecasting systems make predictions based on sales and order data, but they often lack accuracy because they don't take into account customer emotions and reactions at the actual point of sale. As a result, they can either overestimate or underestimate the quantity of promotional items to be produced. This can lead to inefficient inventory management and potentially lost business opportunities.

[1110] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for collecting past sales data and order data, means for preprocessing and normalizing the collected data, means for training a generative artificial intelligence model using the preprocessed data, means for integrating newly collected data in real time into the generative artificial intelligence model, means for predicting future demand using the generative artificial intelligence model and automatically calculating the production quantity, means for notifying the user of the calculated prediction results, and means for collecting user sentiment data and integrating it into the generative artificial intelligence model. As a result, customer sentiment data can be incorporated into demand forecasting, improving forecasting accuracy and enabling more accurate estimation of the production quantity of promotional items.

[1111] "Past sales data" refers to data that shows the history of sales transactions that have taken place in the past, and includes details such as quantity, date, and product name.

[1112] "Order data" refers to data related to orders for goods and services, including information such as the order date, order quantity, and ordering source information.

[1113] "Collection means" refers to methods or devices for collecting specific data, and includes means of acquiring information using sensors, APIs, databases, etc.

[1114] "Preprocessing means" refers to methods and techniques for converting collected data into an appropriate format, making it easy to use for analysis and model training.

[1115] "Methods of normalization" refer to methods and techniques for unifying data variability and bringing it within a certain range, and include techniques such as standardization and scaling.

[1116] "Generative artificial intelligence models" refer to machine learning algorithms and deep learning models that learn patterns from large amounts of data and predict future outcomes.

[1117] "Real-time data" refers to data about events currently in progress or data generated at that moment, including data that is collected and processed immediately.

[1118] "Integration methods" refer to methods and techniques for combining information obtained from multiple data sources to create a unified dataset.

[1119] "Predictive means" refers to methods and techniques that use generative artificial intelligence models to estimate future demand and determine the required production quantity.

[1120] "Notification means" refers to methods and technologies for informing users of the calculated prediction results, and includes means such as dashboards, smart device displays, and email notifications.

[1121] "Emotional data" refers to information that indicates a user's emotional state, and includes data obtained from facial expressions, voice, behavioral patterns, etc.

[1122] System Overview

[1123] The present invention is a system for predicting the production quantity of promotional items using user sentiment data, and mainly includes the following elements: means for collecting past sales data and order data, preprocessing means, normalization means, real-time data integration means, generative artificial intelligence model learning means, demand forecasting means, notification means, and sentiment data collection means.

[1124] Program Embodiment

[1125] 1. Data Collection

[1126] The server connects to the "sales management system" and "order management system" to collect historical sales and order data. Furthermore, it acquires market trends and competitor information from external data sources to ensure comprehensive data richness and accuracy. Using smartphones and smart glasses, it collects user emotion data in real time from customers' facial expressions and voices in stores.

[1127] 2. Data Integration and Preprocessing

[1128] The server stores the collected data in a database and uses Pandas and NumPy to remove duplicate data and impute missing data. For example, it applies the historical mean or median to missing items. It also normalizes the data and converts it into a format that is easy for generative artificial intelligence models to learn.

[1129] 3. Training of generative artificial intelligence models

[1130] The server uses preprocessed data to train generative artificial intelligence models (e.g., LSTM or Transformer) using TensorFlow or Keras. The models recognize past patterns and adjust the parameters necessary for future demand forecasting.

[1131] 4. Integration of real-time data

[1132] The server periodically acquires real-time data such as daily acquisition results and new order status, and integrates it with existing databases using Apache Kafka or RabbitMQ. This real-time data is also preprocessed and normalized to maintain a consistent dataset.

[1133] 5. Collection of emotional data

[1134] The user's smartphone or smart glasses use camera APIs and microphone APIs to recognize the user's emotional state from their facial expressions and voice via an emotion engine, and then send that data to the server.

[1135] 6. Analysis and Integration of Emotional Data

[1136] The server analyzes the collected user sentiment data using Scikit-learn and TensorFlow, and incorporates the results into a generative artificial intelligence model. This enables highly accurate predictions that take into account the impact of sentiment data on demand forecasting.

[1137] 7. Prediction and Calculation

[1138] The server inputs the latest integrated data (sales data, order data, real-time data, sentiment data) into a generative artificial intelligence model to predict the production quantity of future promotional items. For example, if a user expresses very positive sentiment, this is analyzed as an indication of increased demand, and the predicted production quantity will also increase.

[1139] 8. Providing the results

[1140] The server notifies users of the calculated prediction results through a smartphone app using React Native or a smart glasses app using ARKit / ARCore. Based on this, users can plan and execute promotional product production plans.

[1141] Specific example

[1142] For example, imagine a store selling a new product, with staff wearing smart glasses. When a customer shows interest in the product, emotional data is collected. This data is sent to a server in real time and integrated with past sales data and market trends. A generative artificial intelligence model analyzes this data to predict demand and calculate the quantity of promotional items to produce for the new product. Staff can instantly check the information through the smart glasses' HUD and take appropriate action.

[1143] Example of a prompt

[1144] Prompt: "To accurately predict the quantity of promotional items to be produced at stores, collect historical sales data, order data, and user sentiment data in real time, and use a generative artificial intelligence model to make predictions."

[1145] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1146] Step 1:

[1147] Data collection

[1148] The server connects to the "sales management system" and "order management system" to collect historical sales and order data. Specifically, it retrieves data from the database using a REST API and also collects market trends and competitor information via external APIs. In addition, smart devices (smartphones and smart glasses) collect customer facial expressions and voices in real time. This data is sent to the server as raw data and stored in the database.

[1149] Inputs: Sales data, order data, market trend data, competitor information, sentiment data

[1150] Output: Collected raw dataset

[1151] Step 2:

[1152] Data Integration and Preprocessing

[1153] The server processes the collected raw data using Pandas and NumPy. Specifically, it removes duplicate data and imputes missing data with the mean or median. Next, it normalizes the data and converts it into a format suitable for training the model. For example, it scales numerical data and one-hot encodes categorical data.

[1154] Input: Collected raw dataset

[1155] Output: Preprocessed normalized dataset

[1156] Step 3:

[1157] Learning of generative artificial intelligence models

[1158] The server uses TensorFlow and Keras to train generative artificial intelligence models (such as LSTM and Transformer) on preprocessed datasets. Specifically, it splits the data into a training set and a test set, trains the model using the training set, and evaluates it using the test set. It adjusts parameters and sets the number of epochs to ensure the model learns properly.

[1159] Input: Preprocessed normalized dataset

[1160] Output: Trained generative artificial intelligence model

[1161] Step 4:

[1162] Real-time data integration

[1163] The server uses Apache Kafka and RabbitMQ to collect new sales data, order status, and other information in real time and integrate it with existing databases. This data is also preprocessed and normalized. Real-time data is sent to the server as soon as it is collected, ensuring that it is always up-to-date.

[1164] Input: Real-time dataset

[1165] Output: Integrated modern database

[1166] Step 5:

[1167] Collection of emotional data

[1168] The user's smart device uses camera and microphone APIs to collect customer facial expressions and voice data, and analyzes the user's emotional state through an emotion engine. This emotional data is sent to a server via WebSocket or REST API and stored in a database.

[1169] Input: Facial expression data, audio data

[1170] Output: Analyzed sentiment data

[1171] Step 6:

[1172] Analysis and integration of emotional data

[1173] The server uses Scikit-learn and TensorFlow to analyze the collected sentiment data. Specifically, it analyzes the sentiment data and extracts factors that influence demand forecasting. Then, this sentiment data is integrated into a generative artificial intelligence model to improve the model's accuracy.

[1174] Input: Analyzed sentiment data

[1175] Output: Integrated sentiment dataset

[1176] Step 7:

[1177] Prediction and calculation

[1178] The server inputs the integrated dataset into a generative artificial intelligence model to predict future demand. Specifically, it uses LSTM and Transformer models to calculate predicted sales and production quantities. For example, if there are many positive user sentiments, it analyzes this as an indication of increased demand and increases the predicted production quantity accordingly.

[1179] Input: Integrated dataset

[1180] Output: Estimated production quantity of promotional items

[1181] Step 8:

[1182] Providing results

[1183] The server notifies users of prediction results through a smartphone app using React Native or a smart glasses app using ARKit / ARCore. Specifically, the prediction results are displayed on a dashboard, allowing users to immediately check and take action.

[1184] Input: Estimated production quantity of promotional items

[1185] Output: Notifications to the user (dashboard display, smart device display)

[1186] Example of a prompt

[1187] Prompt: "To accurately predict the quantity of promotional items to be produced at stores, collect historical sales data, order data, and user sentiment data in real time, and use a generative artificial intelligence model to make predictions."

[1188] 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.

[1189] 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 the following. 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 indicated 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.

[1190] 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.

[1191] [Fourth Embodiment]

[1192] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1193] 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.

[1194] 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).

[1195] 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.

[1196] 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.

[1197] 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).

[1198] 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.

[1199] 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.

[1200] 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.

[1201] 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.

[1202] 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.

[1203] 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.

[1204] 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".

[1205] System Overview

[1206] This invention is a system for more precisely calculating the production quantity of in-store promotional items. Specifically, a server collects past sales and order data, preprocesses and normalizes this data, and trains a generative artificial intelligence model. Newly acquired real-time data is integrated into this model to predict future demand. By notifying the user of the prediction results, the system automatically calculates the appropriate production quantity of promotional items.

[1207] Program Processing Description

[1208] Data collection

[1209] The server accesses the "sales management system" and "order management system" to collect historical sales and order data. It also collects market trends and competitor information from external data sources as needed. This ensures the richness and accuracy of the data.

[1210] Data Integration and Preprocessing

[1211] The server removes duplicate data from the collected data and performs imputation on incomplete data. For example, if there are missing values ​​in order data, it will use the mean or median to fill them in. It also normalizes the data and converts it into a format that is easy for generative artificial intelligence models to learn from.

[1212] Learning of generative artificial intelligence models

[1213] The server uses preprocessed data to train generative artificial intelligence models (e.g., LSTM or Transformer). These models have the ability to extract patterns from historical data and predict future demand. After training is complete, the models are evaluated using validation datasets to check their accuracy and make any necessary adjustments.

[1214] Real-time data integration

[1215] The server collects daily acquisition data and new order status data in real time. This real-time data is integrated with existing datasets and undergoes further data cleaning and formatting. This ensures that the model always incorporates the latest information.

[1216] Prediction and calculation

[1217] The server uses a generative artificial intelligence model to input the latest data, including real-time data, and predicts the production quantity of future promotional items. This prediction result is stored in a database and can be updated and recalculated immediately if necessary.

[1218] Providing results

[1219] The server notifies the user's device of the prediction results. Notification methods include dashboard display and email notifications. Users can connect to the server from their device to check the prediction results. This allows users to plan and execute promotional item production based on appropriate production quantities.

[1220] Specific example

[1221] For example, when a new product is launched, the user enters product information into the "sales management system." The server collects necessary data based on past data of similar products and market trends, and makes predictions using a generative artificial intelligence model. As a result, the predicted production quantity is notified to the user. Based on that quantity, the user can proceed with the production of promotional materials for the new product. This entire process significantly reduces the risk of inventory shortages or surpluses, enabling the efficient production of promotional materials.

[1222] As described above, this system is a powerful tool for significantly improving the accuracy of predicting the production quantity of promotional items, streamlining the execution process, and optimizing risk management.

[1223] The following describes the processing flow.

[1224] Step 1:

[1225] The server connects to the "sales management system" and "order management system" to collect historical sales and order data. It also obtains market trends and competitor information from external data sources.

[1226] Step 2:

[1227] The server stores the collected data in a database and performs duplicate data removal and data interpolation. For example, it applies historical mean or median values ​​to missing items.

[1228] Step 3:

[1229] The server normalizes the pre-processed data and transforms it into a format that is easy for generative artificial intelligence models to learn from. In this process, numerical data is scaled and categorical data is encoded.

[1230] Step 4:

[1231] The server uses pre-processed data to train a generative artificial intelligence model (e.g., LSTM or Transformer). The model recognizes past patterns and adjusts the parameters necessary for forecasting future demand.

[1232] Step 5:

[1233] The server evaluates the trained generative artificial intelligence model on a validation dataset to verify the model's accuracy. It then adjusts hyperparameters as needed to build an optimal predictive model.

[1234] Step 6:

[1235] The server periodically acquires real-time data such as daily acquisition results and new order status, and integrates it with the existing database. This data is also preprocessed and normalized.

[1236] Step 7:

[1237] The server inputs the latest real-time data into a generative artificial intelligence model to predict future demand for promotional items. The prediction results include approximate production quantities and their range of variation.

[1238] Step 8:

[1239] The server saves the calculated prediction results to a database. Furthermore, it notifies the user's device of the results. This notification may be displayed on a dashboard or sent via email.

[1240] Step 9:

[1241] Users access the server from their terminal and view the predicted production quantity on the dashboard. If necessary, they adjust and determine the actual production quantity based on the prediction results.

[1242] Step 10:

[1243] The user enters the determined production quantity and sends it to the server. The server then saves this information back to the database and sends the necessary production instructions to the order management system.

[1244] (Example 1)

[1245] 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".

[1246] Traditional systems for calculating the production quantity of in-store promotional items often relied on simple predictions based solely on historical data, making it difficult to accurately reflect market fluctuations and real-time conditions. Furthermore, manual data processing and management were required, leading to increased time, effort, and costs. This resulted in frequent inventory shortages or surpluses, making it difficult to create efficient production plans for promotional items.

[1247] 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.

[1248] In this invention, the server includes means for collecting historical sales data and order data, means for eliminating duplicates and imputing missing values ​​in the collected data, and means for normalizing the data and converting it into a format that is easy for a generating AI model to learn. This enables efficient and highly accurate predictions.

[1249] "Past sales data" refers to information about sales records and performance at the time a transaction took place.

[1250] "Order data" refers to records and information related to the ordering and procurement of goods.

[1251] "Means of collection" refers to the functions and devices that allow a server to access multiple data sources and obtain the necessary data.

[1252] "Means of eliminating duplicates" refer to methods or functions for deleting or merging data with identical content from an acquired dataset.

[1253] "Methods for imputing missing values" refer to functions or techniques for filling in missing values ​​or information in a dataset using the mean or median.

[1254] "Methods for normalizing data" refer to methods and functions for preparing data in a way that makes it easier for generative AI models to learn, by transforming it into a certain range or format.

[1255] A "generative AI model" is a machine learning model that uses artificial intelligence technology to learn patterns from past data and predict future trends and demand.

[1256] "Real-time data" refers to the latest information obtained based on ongoing transactions and events.

[1257] "Predictive means" refers to methods and functions for predicting future demand and fluctuations using generative AI models and calculating the results.

[1258] "Means of notification" refer to methods and functions for communicating prediction results to the user, and can take the form of a dashboard display or email notification.

[1259] Modes for carrying out the invention

[1260] This invention is a system for more precisely calculating the production quantity of in-store promotional items, and its main components are a server, terminals, and users. This system is implemented in the following steps.

[1261] System Overview

[1262] This system collects historical sales and order data, preprocesses and normalizes it, and uses it to train a generating AI model. It then integrates newly acquired data in real time with this model to predict future demand. By notifying the user of the prediction results, it automatically calculates the appropriate quantity of promotional items to produce.

[1263] Hardware and software used

[1264] Server: Performs data collection, preprocessing, normalization, AI model training, and prediction.

[1265] Terminal: Provides an interface for the user to view the prediction results.

[1266] Software: Sales management systems, order management systems, cloud databases, access to external data sources, data processing libraries, generative AI model libraries (e.g., LSTM and Transformer).

[1267] Specific example

[1268] When a new product is launched, the user enters product information into the "sales management system." For example, they might enter "details of new product A that is being promoted." The server collects necessary data, including historical data on similar products and market trends, and preprocesses and normalizes it. Then, it uses a generative AI model to predict future demand and calculate the appropriate quantity of promotional items to produce. This result is then communicated to the user.

[1269] Examples of prompt statements

[1270] For example, enter the following as a prompt for a generative AI model:

[1271] When entering the new product "○○" into the sales management system, please predict the required quantity of promotional materials to be produced, taking into account past sales data and market trends.

[1272] The server generates and provides prediction results to the user based on the above procedure. This significantly reduces the risk of inventory shortages or surpluses, and enables efficient planning of promotional item production. Furthermore, continuous integration of real-time data allows for highly accurate predictions that reflect the latest market conditions.

[1273] This system can operate in conjunction with common data infrastructures, regardless of the type of machine learning model or algorithm used. Its flexible application of generative AI models makes it adaptable to a variety of business environments.

[1274] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1275] Step 1:

[1276] The server collects past sales and order data.

[1277] Specifically, the server accesses the "sales management system" and "order management system" every night to retrieve historical sales and order data. This also includes retrieving market trends and competitor information via external APIs as needed. The input is raw data from the sales management system and order management system, and the output is all the collected raw data.

[1278] Step 2:

[1279] The server removes duplicate data and imputes missing values.

[1280] In terms of specific operations, the server analyzes the acquired dataset and removes or merges duplicate data with the same date and product ID. Furthermore, missing values ​​within the dataset are imputed using the mean or median of past data. The input is the collected raw data, and the output is the data with duplicates removed and missing values ​​imputed.

[1281] Step 3:

[1282] The server normalizes the data and converts it into a format that is easy for the generating AI model to learn from.

[1283] In terms of specific operations, the server scales quantitative and price data to a particular range and converts categorical data into numerical data using techniques such as one-hot encoding. The input is data that has undergone deduplicating and imputation, and the output is normalized data that has been converted into a format that is easy for the AI ​​model to learn from.

[1284] Step 4:

[1285] The server trains the generated AI model.

[1286] In terms of operation, the server inputs normalized data into a generative AI model (e.g., LSTM or Transformer) to train the model. During the training process, patterns are extracted from past data, and parameters are adjusted to predict future demand. The input is normalized data, and the output is the trained generative AI model.

[1287] Step 5:

[1288] The server integrates real-time data.

[1289] In terms of specific operations, the server collects daily sales performance and new order data in real time and integrates it into an existing dataset. Data cleaning and formatting are performed again as needed. The input is newly acquired data in real time, and the output is an integrated dataset containing the latest information.

[1290] Step 6:

[1291] The server makes predictions and calculates the production quantity.

[1292] In practice, the server uses a generative AI model to input the latest data and predict the production quantity of future promotional items. This prediction is performed periodically, and the data is saved to the database or updated immediately as needed. The input is the integrated latest data, and the output is the predicted production quantity.

[1293] Step 7:

[1294] The server notifies the user's device of the prediction results.

[1295] In terms of specific operations, the server transmits the prediction results to the user's terminal. At this time, dashboard display and email notifications can be used. When new prediction data is generated, the server displays the update information on the user's dashboard via the web server. It also notifies the user of the prediction results via email, depending on the settings. The input is the predicted production quantity, and the output is the notification to the user.

[1296] (Application Example 1)

[1297] 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".

[1298] The objective of this invention is to provide a system for appropriately predicting the production quantity of promotional items. Conventional methods rely solely on past sales and order data for predictions, making it difficult to reflect new data obtained in real time. As a result, it is not possible to respond quickly to fluctuations in demand. Furthermore, the methods for notifying users of the prediction results are limited, making it difficult for users to easily check them. This leads to risks of inventory shortages or surpluses, making it difficult to produce promotional items efficiently.

[1299] 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.

[1300] In this invention, the server includes means for collecting past sales data and order data; means for preprocessing and normalizing the collected data; means for training a generative artificial intelligence model using the preprocessed data; means for integrating newly collected data in real time into the generative artificial intelligence model; means for predicting future demand using the generative artificial intelligence model and automatically calculating production quantities; means for notifying the user of the calculated prediction results; and means for displaying the notified prediction results on a smart device. This enables precise sales forecasting including real-time data, and allows users to quickly check the prediction results via a smart device and create an appropriate production plan for promotional items.

[1301] "Past sales data" refers to data that records the sales performance of a product over a certain past period.

[1302] "Order data" refers to data that includes information about orders for goods and services.

[1303] "Preprocessing" is the process of converting collected data into a format suitable for analysis and learning.

[1304] "Normalization" is a transformation process that brings data values ​​within a certain range.

[1305] A "generative artificial intelligence model" is an algorithm or structure that learns from large amounts of data and makes predictions or generates new data.

[1306] "New data collected in real time" refers to data that is being collected in real time and is immediately available for processing and analysis.

[1307] "Integration" is the process of combining and linking different datasets into one.

[1308] "Predicting future demand and automatically calculating production quantities" refers to the act of using generative artificial intelligence models to estimate future consumer demand and determine the corresponding production volume.

[1309] "Notification" refers to communicating prediction results or information to the user through specific means.

[1310] A "smart device" is an electronic device that has communication capabilities and computing power, and allows users to display and operate information.

[1311] A "dashboard display" is an interface that visually organizes and presents information.

[1312] "Email notification" refers to a method of sending information in the form of an email.

[1313] A "remote database" is a data storage system located remotely and accessible via a network.

[1314] "External data sources" refer to data obtained from data providers outside the company, such as public databases.

[1315] System Overview

[1316] This invention is a system that streamlines inventory management by precisely predicting the production quantity of promotional items for physical stores. The system collects historical sales and order data, preprocesses and normalizes this data, then uses a generative artificial intelligence model for training and integrates real-time data to predict future demand. The prediction results are then communicated to the user's smart device.

[1317] Program Processing Description

[1318] Data collection

[1319] The server accesses the "sales management system" and "order management system" to collect historical sales and order data. This data is obtained from remote databases and external data sources.

[1320] Data Integration and Preprocessing

[1321] The server removes duplicate data from the collected data and performs imputation on incomplete data. It also normalizes the data and converts it into a format that is easy for generative artificial intelligence models to learn. Python and general-purpose data analysis libraries (such as Pandas and NumPy) are used in this process.

[1322] Learning of generative artificial intelligence models

[1323] The server uses preprocessed data to train generative artificial intelligence models (for example, LSTM or Transformer models using Keras). These models extract patterns from past data and predict future demand.

[1324] Real-time data integration

[1325] The server collects daily sales performance and new order status data in real time. This data is obtained from POS systems and sensor devices. The real-time data is integrated with existing datasets and then cleaned and formatted again.

[1326] Prediction and calculation

[1327] The server uses a generative artificial intelligence model to input the latest data and predict the production quantity of future promotional items. This prediction is stored in a database and can be updated and recalculated instantly as needed.

[1328] Providing results

[1329] The server notifies the user of the prediction results on their smart device. Notification methods include a dashboard display on the smart device and email notifications. Users can check the prediction results using their smartphones or tablets.

[1330] Specific example

[1331] For example, when a new product is launched, the user enters product information into the "sales management system." The server collects necessary data based on past data of similar products and market trends, and uses a generative artificial intelligence model to predict future demand. As a result, the predicted production quantity is notified to the user's smartphone. Based on that quantity, the user can proceed with the production of promotional materials for the new product. This entire process significantly reduces the risk of inventory shortages or surpluses, enabling the efficient production of promotional materials.

[1332] Examples of prompt sentences:

[1333] Collect new sales data and feed it into an AI model to predict future demand. We want to build an application that notifies us of the prediction results in real time. Historical sales data is located here: [sales_data.csv]. Order data is located here: [order_data.csv]. Real-time data is located here: [new_data.csv].

[1334] As described above, this system significantly improves the accuracy of predicting the production quantity of promotional items and is a powerful tool for creating efficient production plans for promotional items.

[1335] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1336] Step 1:

[1337] The server collects historical sales and order data from the "sales management system" and "order management system." This data includes information such as product name, sales quantity, sales date, order quantity, and order date. The collected data is stored in a database.

[1338] Input: Sales data, order data

[1339] Output: Raw data stored in the database

[1340] The server also accesses remote databases and external data sources to obtain market trends and competitive information.

[1341] Step 2:

[1342] The server preprocesses the collected data. It removes duplicate data and imputes incomplete data, for example, by filling in missing values ​​with the median. Furthermore, it normalizes the data and converts it into a format that is easy for AI models to learn from.

[1343] Input: Collected raw data

[1344] Output: Normalized preprocessed data

[1345] Specifically, we will use Python's Pandas and NumPy libraries to perform data cleaning and imputation.

[1346] Step 3:

[1347] The server uses preprocessed data to train a generative artificial intelligence model. This model consists of components such as LSTM (Long Short-Term Memory) and Transformer. The training process is executed using Keras and TensorFlow.

[1348] Input: Normalized preprocessed data

[1349] Output: Trained generative artificial intelligence model

[1350] The server splits the dataset into a training set and a test set, and then trains and evaluates the model.

[1351] Step 4:

[1352] The server collects daily sales performance and new order status data in real time. This data is obtained from POS systems and sensor devices. The real-time data is integrated with existing datasets and then cleaned and formatted again.

[1353] Input: Real-time data

[1354] Output: Updated dataset

[1355] The server processes real-time data quickly and incorporates it into model updates.

[1356] Step 5:

[1357] The server uses a generative artificial intelligence model to predict the production quantities of future promotional items. The predicted data is stored in a database and can be updated and recalculated instantly.

[1358] Input: Updated dataset

[1359] Output: Predicted production quantity

[1360] The server runs a prediction algorithm to calculate production quantities based on future demand.

[1361] Step 6:

[1362] The server notifies the user of the prediction results on their smart device. Notification methods include a dashboard display on the smart device and email notifications. Users can check the prediction results using their smartphone or tablet.

[1363] Input: Predicted production quantity

[1364] Output: Notification to the user

[1365] The server either sends emails using the SMTP protocol or displays results on a web-based dashboard.

[1366] 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.

[1367] System Overview

[1368] This invention provides a system that combines an emotion engine with a system for precisely calculating the production quantity of in-store promotional items, thereby recognizing and reflecting user emotions to achieve even more accurate predictions. This system has means for collecting past sales data and order data, and preprocesses and normalizes the collected data. Furthermore, it uses the preprocessed data to train a generative artificial intelligence model, integrates newly collected data in real time, and predicts future demand. Finally, it incorporates an emotion engine that recognizes user emotions, and reflects user emotion data in the predictions.

[1369] Program Processing Description

[1370] Data collection

[1371] The server connects to the "Sales Management System" and "Order Management System" to collect historical sales and order data. Furthermore, it acquires market trends and competitor information from external data sources to ensure comprehensive data richness and accuracy.

[1372] Data Integration and Preprocessing

[1373] The server stores the collected data in a database and performs duplicate data removal and missing data completion. For example, it applies the historical mean or median to missing items. It also normalizes the data and converts it into a format that is easy for generative artificial intelligence models to learn from.

[1374] Learning of generative artificial intelligence models

[1375] The server uses pre-processed data to train a generative artificial intelligence model (e.g., LSTM or Transformer). The model recognizes past patterns and adjusts the parameters necessary for forecasting future demand.

[1376] Real-time data integration

[1377] The server periodically acquires real-time data such as daily acquisition results and new order status, and integrates it with the existing database. This real-time data is also properly preprocessed and normalized to maintain a consistent dataset.

[1378] Collection of user sentiment data

[1379] The user connects to the server via their device, and the emotion engine recognizes the user's emotions. The emotion engine uses facial recognition and voice analysis to understand the user's emotional state. The user's emotional data is sent to the server and stored in a database.

[1380] Analysis and integration of emotional data

[1381] The server analyzes the collected user sentiment data and incorporates the results into a generative artificial intelligence model. This enables highly accurate predictions that take into account the impact of sentiment data on demand forecasting.

[1382] Prediction and calculation

[1383] The server uses a generative artificial intelligence model to input the latest integrated data (sales data, order data, real-time data, and sentiment data) and predict the production quantity of future promotional items. The prediction results include an approximate production quantity and range of variation, taking sentiment data into account.

[1384] Providing results

[1385] The server saves the calculated prediction results to a database and notifies the user's device. Notification methods include dashboard display and email notifications. Users can check the prediction results through their device and make adjustments as needed.

[1386] Specific example

[1387] For example, when a new product is launched, the user enters product information into the "sales management system." The server collects necessary data based on past data of similar products, market trends, and user sentiment data, and makes predictions using a generative artificial intelligence model. For instance, if a user shows very positive sentiment, this is analyzed as an indication of increased demand, and the predicted production quantity will also increase. The results are notified to the user, who can then plan and execute a production plan for promotional items based on this information. This significantly reduces the risk of inventory shortages or surpluses, enabling the efficient production of promotional items.

[1388] As described above, this system significantly improves the accuracy of predicting the production quantity of promotional items, and by incorporating user sentiment data into the prediction process, it achieves more precise and adaptable predictions.

[1389] The following describes the processing flow.

[1390] Step 1:

[1391] The server connects to the sales management system and order management system, collecting historical sales and order data. Furthermore, it also obtains market trends and competitor information from external data sources.

[1392] Step 2:

[1393] The server stores the collected data in a database and performs duplicate removal and missing data insertion. For missing data, it applies the historical mean or median.

[1394] Step 3:

[1395] The server normalizes the pre-processed data and transforms it into a format that is easy for generative artificial intelligence models to learn. Numerical data is scaled, and categorical data is encoded.

[1396] Step 4:

[1397] The server uses pre-processed data to train generative artificial intelligence models (e.g., LSTM or Transformer). In the process, it extracts patterns from past data.

[1398] Step 5:

[1399] The server evaluates the trained generative artificial intelligence model on a validation dataset to verify the model's accuracy. It then adjusts hyperparameters as needed to build an optimal predictive model.

[1400] Step 6:

[1401] The server periodically acquires real-time data such as daily acquisition results and new order status, and this data is similarly pre-processed and normalized. The real-time data is then integrated into the database.

[1402] Step 7:

[1403] The user connects to the server via their device and activates the emotion engine. The emotion engine collects the user's emotional data through facial recognition and voice analysis, and sends this data to the server.

[1404] Step 8:

[1405] The server analyzes the collected user sentiment data and integrates the results into a generative artificial intelligence model. It analyzes the impact of the sentiment data on predictions and reflects this in the model.

[1406] Step 9:

[1407] The server uses a generative artificial intelligence model to input integrated, up-to-date data (sales data, order data, real-time data, and sentiment data) and predict the production quantity of future promotional items.

[1408] Step 10:

[1409] The server saves the prediction results to a database and notifies the user's device. Notifications are made through methods such as dashboard displays and email notifications.

[1410] Step 11:

[1411] Users access the server from their terminal and view the predicted production quantity on the dashboard. If necessary, they adjust and determine the actual production quantity based on the prediction results.

[1412] Step 12:

[1413] The user sends the determined production quantity from their terminal to the server. The server then saves this information back to the database and sends the necessary production instructions to the order management system.

[1414] Specific example

[1415] For example, when a new product launch is decided, the user enters the new product information into the sales management system. The server collects historical data on similar products and market information, as well as user sentiment data. If the user shows positive sentiment, it is analyzed that this positive sentiment will lead to increased demand. The server inputs this data into a generative artificial intelligence model to predict the production quantity of promotional items with high accuracy. The prediction results are notified to the user via a dashboard or email, and the user then decides on the appropriate production quantity based on that information.

[1416] (Example 2)

[1417] 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".

[1418] Currently, many companies predict the production quantity of promotional items based on past sales and order data, but the accuracy of this method is limited. Furthermore, even when utilizing real-time data, the prediction accuracy does not improve sufficiently because qualitative data such as user sentiment is not considered. Additionally, there is a lack of timely and effective methods for notifying users of the prediction results. To address these challenges, a more accurate and flexible prediction system is needed.

[1419] 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.

[1420] In this invention, the server includes means for collecting past sales data and order data; means for preprocessing and normalizing the collected data; means for training a generative artificial intelligence model using the preprocessed data; means for integrating newly collected data in real time into the generative artificial intelligence model; means for collecting user sentiment data; means for analyzing the collected user sentiment data and integrating it into the generative artificial intelligence model; means for predicting future demand using the generative artificial intelligence model and automatically calculating production quantities; and means for notifying the user of the calculated prediction results. This enables highly accurate demand forecasting that includes user sentiment data.

[1421] "Past sales data" refers to data that records the past sales performance of a particular product or service.

[1422] "Order data" refers to data related to orders placed by companies to receive the supply of products or services.

[1423] "Means of collection" refers to methods or devices for acquiring and storing specific information.

[1424] "Preprocessing" refers to the process of shaping, cleaning, and normalizing data, which is necessary for analysis and model training.

[1425] "Normalization" is the process of unifying the variability of data and bringing it within a specific range.

[1426] A "generative artificial intelligence model" is a type of AI algorithm that learns from collected data and uses it to predict future situations.

[1427] "Real-time data" refers to the latest information on ongoing events and situations.

[1428] "Emotional data" refers to data that records information about a user's emotional state.

[1429] "Analysis" is a scientific method for finding patterns and rules based on collected data.

[1430] "Prediction" is the process of estimating future situations or events based on collected data.

[1431] "Notification" refers to an action or mechanism for informing a user of specific information.

[1432] A "dashboard display" is an interface that provides information to users visually.

[1433] "Email notification" refers to a method of sending information to users via email.

[1434] A "cloud database" is an online database that can be accessed via the internet.

[1435] "External data sources" refer to data provided by external services or data providers, rather than data from within a company.

[1436] This invention provides a system for accurately calculating the production quantity of in-store promotional items. This system collects historical sales and order data, preprocesses and normalizes this data, and then trains a generative artificial intelligence model. Furthermore, it integrates newly collected data and user sentiment data in real time to achieve advanced demand forecasting and production quantity calculation.

[1437] Data collection

[1438] Sales and order data collection via server

[1439] The server periodically connects to the "sales management system" and "order management system" via specific APIs to automatically collect the necessary data. API calls, for example, save sales and order data for the past year to the database.

[1440] Server-based market trend and competitor information gathering

[1441] The server automatically retrieves market trends and competitor information from external data sources (e.g., Google Trends and social media analysis tools). This ensures the richness and accuracy of the data.

[1442] Data Integration and Preprocessing

[1443] Server-based data storage and cleaning

[1444] The server stores the collected data in a database and performs duplicate data removal and missing data imputation. Missing data is filled in using historical mean and median values.

[1445] Data normalization by the server

[1446] The server converts the data into a unified format and shapes it so that it can be easily trained by generative artificial intelligence models. Specifically, it scales sales quantities to a range of 0 to 1 and performs one-hot encoding on categorical data.

[1447] Learning of generative artificial intelligence models

[1448] Server-based AI model training

[1449] The server uses the pre-processed data to train generative artificial intelligence models (e.g., LSTM or Transformer). This training process improves the accuracy of predictions.

[1450] Real-time data integration

[1451] Real-time data collection and integration by the server.

[1452] The server acquires real-time data such as daily sales performance and order status, and integrates it with the existing database. The new data is appropriately preprocessed and merged into the existing dataset.

[1453] Collection of user sentiment data

[1454] User provision of sentiment data

[1455] The user, through their device, allows the emotion engine to acquire their emotional state using facial recognition and voice analysis. The user provides emotional information to the emotion engine, for example, using the camera or microphone.

[1456] Storage of emotional data by a server

[1457] The server stores the acquired user sentiment data in a database and integrates it with other data.

[1458] Analysis and integration of emotional data

[1459] Server-based analysis of emotional data

[1460] The server analyzes emotional data and incorporates it into a generative artificial intelligence model to improve prediction accuracy. For example, positive emotions may indicate increased demand, thus influencing the predicted values.

[1461] Prediction and calculation

[1462] Server-based future demand forecasting

[1463] The server uses a generative artificial intelligence model to calculate the future production quantity of promotional items based on integrated, up-to-date data. The prediction results include sales data, order data, real-time data, and sentiment data.

[1464] Providing results

[1465] Server-based notification of results

[1466] The server saves the calculated prediction results to a database and notifies the user's device. Notifications are made via a dashboard display or email.

[1467] User review and adjustment of results

[1468] Users can view the prediction results through their devices and make adjustments as needed. This allows for the creation of efficient promotional material production plans. For example, if a user expresses positive emotions, the predicted production quantity will also increase.

[1469] Specific example

[1470] For example, when a new product is launched, the user enters product information into the "sales management system." The server uses a generative artificial intelligence model to make predictions based on past data of similar products, market trends, and user sentiment data. For instance, if a user expresses very positive sentiment, an increase in demand is predicted, and the production quantity will also increase. The results are notified to the user, who can then plan and execute a promotional item production plan based on that information.

[1471] Example of a prompt

[1472] User: I have positive feelings about the new product A. Please provide a sales forecast.

[1473] Server: Based on past sales data, order data, market trend information, and user sentiment data, a generative AI model predicts the demand for new product A.

[1474] This system enables highly accurate demand forecasting, including user sentiment data, leading to more efficient promotional activities.

[1475] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1476] Step 1: Data Collection

[1477] The server periodically connects to the "Sales Management System" and "Order Management System" to collect historical sales and order data. Specifically, it sends data collection requests via API and stores the past year's worth of data in the database. The input is the API request, and the output is the retrieved sales and order data.

[1478] Step 2: Gather market trends and competitive information

[1479] The server connects to external data sources (e.g., Google Trends and social media analysis tools) to collect market trends and competitor information. Specifically, it sends requests via APIs and stores the collected data in a database. The input is information requests from external data sources, and the output is the retrieved market trends and competitor information.

[1480] Step 3: Data preprocessing and normalization

[1481] The server stores the collected data in a database and performs duplicate data removal and missing data imputation. Specifically, it removes duplicate data and imputes missing data with historical mean or median values. Furthermore, it converts the data into a unified format (e.g., numerical scaling, one-hot encoding of categorical data) to make it easier for AI models to learn. The input is the various data acquired, and the output is the pre-processed dataset.

[1482] Step 4: Training a generative artificial intelligence model

[1483] The server uses pre-processed data to train generative artificial intelligence models (e.g., LSTM and Transformer). Specifically, it inputs data into the AI ​​model and adjusts the model's parameters while training. The input is the pre-processed dataset, and the output is the optimized AI model.

[1484] Step 5: Collect and integrate real-time data

[1485] The server periodically retrieves daily real-time data (new sales figures and order status) and integrates it with the existing database. Specifically, it collects real-time data via an API and merges the new data into the existing dataset. The input is real-time data, and the output is the updated integrated dataset.

[1486] Step 6: Collecting user sentiment data

[1487] The user accesses the emotion engine through their device and acquires their emotional state using facial recognition and voice analysis. Specifically, the user uses the camera and microphone to provide emotional information to the emotion engine. The input is the user's facial expressions and voice data, and the output is the analyzed emotional data.

[1488] Step 7: Analysis and integration of emotional data

[1489] The server analyzes user sentiment data and incorporates it into a generative artificial intelligence model. Specifically, it quantifies sentiment data, cross-analyzes it with sales data, and inputs it into the AI ​​model to improve prediction accuracy. The input is sentiment data and an integrated dataset, and the output is the final dataset with the sentiment data integrated.

[1490] Step 8: Forecast future demand and calculate production quantities.

[1491] The server uses a generative artificial intelligence model based on the latest integrated data to predict the production quantity of future promotional items. Specifically, it inputs the integrated data into the AI ​​model and performs demand forecasting. As a result, a demand forecast value is output. The input is the integrated dataset, and the output is the predicted production quantity of promotional items.

[1492] Step 9: Notification of Results

[1493] The server saves the calculated prediction results to a database and notifies the user's terminal. Notifications are made via a dashboard display or email. Specifically, the server displays the prediction results on a dashboard or sends them to the user via email. The input is the prediction result data, and the output is the notification to the user.

[1494] Step 10: Review and adjust results

[1495] Users can view prediction results via their devices and make adjustments as needed. Specifically, users view prediction results on a dashboard and then create a production plan for promotional items based on those results. The input is the prediction result, and the output is the adjusted production plan.

[1496] This completes the entire processing flow, enabling highly accurate demand forecasting and production quantity calculation for promotional items.

[1497] (Application Example 2)

[1498] 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".

[1499] Traditional demand forecasting systems make predictions based on sales and order data, but they often lack accuracy because they don't take into account customer emotions and reactions at the actual point of sale. As a result, they can either overestimate or underestimate the quantity of promotional items to be produced. This can lead to inefficient inventory management and potentially lost business opportunities.

[1500] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for collecting past sales data and order data, means for preprocessing and normalizing the collected data, means for training a generative artificial intelligence model using the preprocessed data, means for integrating newly collected data in real time into the generative artificial intelligence model, means for predicting future demand using the generative artificial intelligence model and automatically calculating the production quantity, means for notifying the user of the calculated prediction results, and means for collecting user sentiment data and integrating it into the generative artificial intelligence model. As a result, customer sentiment data can be incorporated into demand forecasting, improving forecasting accuracy and enabling more accurate estimation of the production quantity of promotional items.

[1501] "Past sales data" refers to data that shows the history of sales transactions that have taken place in the past, and includes details such as quantity, date, and product name.

[1502] "Order data" refers to data related to orders for goods and services, including information such as the order date, order quantity, and ordering source information.

[1503] "Collection means" refers to methods or devices for collecting specific data, and includes means of acquiring information using sensors, APIs, databases, etc.

[1504] "Preprocessing means" refers to methods and techniques for converting collected data into an appropriate format, making it easy to use for analysis and model training.

[1505] "Methods of normalization" refer to methods and techniques for unifying data variability and bringing it within a certain range, and include techniques such as standardization and scaling.

[1506] "Generative artificial intelligence models" refer to machine learning algorithms and deep learning models that learn patterns from large amounts of data and predict future outcomes.

[1507] "Real-time data" refers to data about events currently in progress or data generated at that moment, including data that is collected and processed immediately.

[1508] "Integration methods" refer to methods and techniques for combining information obtained from multiple data sources to create a unified dataset.

[1509] "Predictive means" refers to methods and techniques that use generative artificial intelligence models to estimate future demand and determine the required production quantity.

[1510] "Notification means" refers to methods and technologies for informing users of the calculated prediction results, and includes means such as dashboards, smart device displays, and email notifications.

[1511] "Emotional data" refers to information that indicates a user's emotional state, and includes data obtained from facial expressions, voice, behavioral patterns, etc.

[1512] System Overview

[1513] The present invention is a system for predicting the production quantity of promotional items using user sentiment data, and mainly includes the following elements: means for collecting past sales data and order data, preprocessing means, normalization means, real-time data integration means, generative artificial intelligence model learning means, demand forecasting means, notification means, and sentiment data collection means.

[1514] Program Embodiment

[1515] 1. Data Collection

[1516] The server connects to the "sales management system" and "order management system" to collect historical sales and order data. Furthermore, it acquires market trends and competitor information from external data sources to ensure comprehensive data richness and accuracy. Using smartphones and smart glasses, it collects user emotion data in real time from customers' facial expressions and voices in stores.

[1517] 2. Data Integration and Preprocessing

[1518] The server stores the collected data in a database and uses Pandas and NumPy to remove duplicate data and impute missing data. For example, it applies the historical mean or median to missing items. It also normalizes the data and converts it into a format that is easy for generative artificial intelligence models to learn.

[1519] 3. Training of generative artificial intelligence models

[1520] The server uses preprocessed data to train generative artificial intelligence models (e.g., LSTM or Transformer) using TensorFlow or Keras. The models recognize past patterns and adjust the parameters necessary for future demand forecasting.

[1521] 4. Integration of real-time data

[1522] The server periodically acquires real-time data such as daily acquisition results and new order status, and integrates it with existing databases using Apache Kafka or RabbitMQ. This real-time data is also preprocessed and normalized to maintain a consistent dataset.

[1523] 5. Collection of emotional data

[1524] The user's smartphone or smart glasses use camera APIs and microphone APIs to recognize the user's emotional state from their facial expressions and voice via an emotion engine, and then send that data to the server.

[1525] 6. Analysis and Integration of Emotional Data

[1526] The server analyzes the collected user sentiment data using Scikit-learn and TensorFlow, and incorporates the results into a generative artificial intelligence model. This enables highly accurate predictions that take into account the impact of sentiment data on demand forecasting.

[1527] 7. Prediction and Calculation

[1528] The server inputs the latest integrated data (sales data, order data, real-time data, sentiment data) into a generative artificial intelligence model to predict the production quantity of future promotional items. For example, if a user expresses very positive sentiment, this is analyzed as an indication of increased demand, and the predicted production quantity will also increase.

[1529] 8. Providing the results

[1530] The server notifies users of the calculated prediction results through a smartphone app using React Native or a smart glasses app using ARKit / ARCore. Based on this, users can plan and execute promotional product production plans.

[1531] Specific example

[1532] For example, imagine a store selling a new product, with staff wearing smart glasses. When a customer shows interest in the product, emotional data is collected. This data is sent to a server in real time and integrated with past sales data and market trends. A generative artificial intelligence model analyzes this data to predict demand and calculate the quantity of promotional items to produce for the new product. Staff can instantly check the information through the smart glasses' HUD and take appropriate action.

[1533] Example of a prompt

[1534] Prompt: "To accurately predict the quantity of promotional items to be produced at stores, collect historical sales data, order data, and user sentiment data in real time, and use a generative artificial intelligence model to make predictions."

[1535] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1536] Step 1:

[1537] Data collection

[1538] The server connects to the "sales management system" and "order management system" to collect historical sales and order data. Specifically, it retrieves data from the database using a REST API and also collects market trends and competitor information via external APIs. In addition, smart devices (smartphones and smart glasses) collect customer facial expressions and voices in real time. This data is sent to the server as raw data and stored in the database.

[1539] Inputs: Sales data, order data, market trend data, competitor information, sentiment data

[1540] Output: Collected raw dataset

[1541] Step 2:

[1542] Data Integration and Preprocessing

[1543] The server processes the collected raw data using Pandas and NumPy. Specifically, it removes duplicate data and imputes missing data with the mean or median. Next, it normalizes the data and converts it into a format suitable for training the model. For example, it scales numerical data and one-hot encodes categorical data.

[1544] Input: Collected raw dataset

[1545] Output: Preprocessed normalized dataset

[1546] Step 3:

[1547] Learning of generative artificial intelligence models

[1548] The server uses TensorFlow and Keras to train generative artificial intelligence models (such as LSTM and Transformer) on preprocessed datasets. Specifically, it splits the data into a training set and a test set, trains the model using the training set, and evaluates it using the test set. It adjusts parameters and sets the number of epochs to ensure the model learns properly.

[1549] Input: Preprocessed normalized dataset

[1550] Output: Trained generative artificial intelligence model

[1551] Step 4:

[1552] Real-time data integration

[1553] The server uses Apache Kafka and RabbitMQ to collect new sales data, order status, and other information in real time and integrate it with existing databases. This data is also preprocessed and normalized. Real-time data is sent to the server as soon as it is collected, ensuring that it is always up-to-date.

[1554] Input: Real-time dataset

[1555] Output: Integrated modern database

[1556] Step 5:

[1557] Collection of emotional data

[1558] The user's smart device uses camera and microphone APIs to collect customer facial expressions and voice data, and analyzes the user's emotional state through an emotion engine. This emotional data is sent to a server via WebSocket or REST API and stored in a database.

[1559] Input: Facial expression data, audio data

[1560] Output: Analyzed sentiment data

[1561] Step 6:

[1562] Analysis and integration of emotional data

[1563] The server uses Scikit-learn and TensorFlow to analyze the collected sentiment data. Specifically, it analyzes the sentiment data and extracts factors that influence demand forecasting. Then, this sentiment data is integrated into a generative artificial intelligence model to improve the model's accuracy.

[1564] Input: Analyzed sentiment data

[1565] Output: Integrated sentiment dataset

[1566] Step 7:

[1567] Prediction and calculation

[1568] The server inputs the integrated dataset into a generative artificial intelligence model to predict future demand. Specifically, it uses LSTM and Transformer models to calculate predicted sales and production quantities. For example, if there are many positive user sentiments, it analyzes this as an indication of increased demand and increases the predicted production quantity accordingly.

[1569] Input: Integrated dataset

[1570] Output: Estimated production quantity of promotional items

[1571] Step 8:

[1572] Providing results

[1573] The server notifies users of prediction results through a smartphone app using React Native or a smart glasses app using ARKit / ARCore. Specifically, the prediction results are displayed on a dashboard, allowing users to immediately check and take action.

[1574] Input: Estimated production quantity of promotional items

[1575] Output: Notifications to the user (dashboard display, smart device display)

[1576] Example of a prompt

[1577] Prompt: "To accurately predict the quantity of promotional items to be produced at stores, collect historical sales data, order data, and user sentiment data in real time, and use a generative artificial intelligence model to make predictions."

[1578] 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.

[1579] 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 the following. 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 indicated 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.

[1580] 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 robot 414.

[1581] 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.

[1582] 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.

[1583] 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.

[1584] 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.

[1585] 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, for example, based 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.

[1586] 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."

[1587] 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.

[1588] 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.

[1589] 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.

[1590] 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.

[1591] 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.

[1592] 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.

[1593] 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.

[1594] 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.

[1595] 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.

[1596] 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.

[1597] 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.

[1598] 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.

[1599] The following is further disclosed regarding the embodiments described above.

[1600] (Claim 1)

[1601] A means of collecting past sales data and order data,

[1602] Methods for preprocessing and normalizing the collected data,

[1603] A means for training a generative artificial intelligence model using preprocessed data,

[1604] A means of integrating newly collected data in real time into a generative artificial intelligence model,

[1605] A method for predicting future demand using a generative artificial intelligence model and automatically calculating production quantities,

[1606] A means of notifying the user of the calculated prediction results,

[1607] A system that includes this.

[1608] (Claim 2)

[1609] The system according to claim 1, comprising means of notifying the user of the calculated prediction results, such as displaying them on a dashboard or sending an email notification.

[1610] (Claim 3)

[1611] The system according to claim 1, comprising means for accessing a cloud database or an external data source as a means for collecting past sales data and order data.

[1612] "Example 1"

[1613] (Claim 1)

[1614] A means of collecting past sales data and order data,

[1615] A means to eliminate duplicates in the collected data and impute missing values,

[1616] A method for normalizing data and converting it into a format that is easy for generative AI models to learn,

[1617] A means of training a generative AI model using preprocessed data,

[1618] A method for predicting future demand using a pre-trained generative AI model,

[1619] A means of integrating newly collected data in real time into a generating AI model,

[1620] A means to automatically calculate the predicted production quantity,

[1621] A means of notifying the user of the calculated prediction results,

[1622] A system that includes this.

[1623] (Claim 2)

[1624] The system according to claim 1, which has a dashboard display function or an email notification function as a method for notifying the user of the calculated prediction results.

[1625] (Claim 3)

[1626] The system according to claim 1, which includes a function to access a cloud database or an external data source as a means of collecting past sales data and order data.

[1627] "Application Example 1"

[1628] (Claim 1)

[1629] A means of collecting past sales data and order data,

[1630] Methods for preprocessing and normalizing the collected data,

[1631] A means for training a generative artificial intelligence model using preprocessed data,

[1632] A means of integrating newly collected data in real time into a generative artificial intelligence model,

[1633] A method for predicting future demand using a generative artificial intelligence model and automatically calculating production quantities,

[1634] A means of notifying the user of the calculated prediction results,

[1635] A means of displaying the notified prediction results on a smart device,

[1636] A system that includes this.

[1637] (Claim 2)

[1638] The system according to claim 1, which includes means for displaying the notified prediction results on a smart device dashboard or for sending email notifications.

[1639] (Claim 3)

[1640] The system according to claim 1, comprising means for accessing a remote database or an external data source as a means for collecting past sales data and order data.

[1641] "Example 2 of combining an emotion engine"

[1642] (Claim 1)

[1643] A means of collecting past sales data and order data,

[1644] Methods for preprocessing and normalizing the collected data,

[1645] A means for training a generative artificial intelligence model using preprocessed data,

[1646] A means of integrating newly collected data in real time into a generative artificial intelligence model,

[1647] Means for collecting user sentiment data,

[1648] A means for analyzing collected user sentiment data and integrating it into a generative artificial intelligence model,

[1649] A method for predicting future demand using a generative artificial intelligence model and automatically calculating production quantities,

[1650] A means of notifying the user of the calculated prediction results,

[1651] A system that includes this.

[1652] (Claim 2)

[1653] The system according to claim 1, comprising means of notifying the user of the calculated prediction results, such as displaying them on a dashboard or sending an email notification.

[1654] (Claim 3)

[1655] The system according to claim 1, comprising means for accessing a cloud database or an external data source as a means for collecting past sales data and order data.

[1656] "Application example 2 when combining with an emotional engine"

[1657] (Claim 1)

[1658] A means of collecting past sales data and order data,

[1659] Methods for preprocessing and normalizing the collected data,

[1660] A means for training a generative artificial intelligence model using preprocessed data,

[1661] A means of integrating newly collected data in real time into a generative artificial intelligence model,

[1662] A method for predicting future demand using a generative artificial intelligence model and automatically calculating production quantities,

[1663] A means of notifying the user of the calculated prediction results,

[1664] A means of collecting user emotion data and integrating it into a generative artificial intelligence model,

[1665] A system that includes this.

[1666] (Claim 2)

[1667] The system according to claim 1, comprising means of displaying the calculated prediction results on a dashboard or on a smart terminal as a method for notifying the user of the results.

[1668] (Claim 3)

[1669] The system according to claim 1, comprising means for accessing a cloud database or an external data source as a means for collecting past sales data and order data. [Explanation of Symbols]

[1670] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of collecting past sales data and order data, Methods for preprocessing and normalizing the collected data, A means for training a generative artificial intelligence model using preprocessed data, A means of integrating newly collected data in real time into a generative artificial intelligence model, A method for predicting future demand using a generative artificial intelligence model and automatically calculating production quantities, A means of notifying the user of the calculated prediction results, A system that includes this.

2. The system according to claim 1, comprising means of notifying the user of the calculated prediction results, such as displaying them on a dashboard or sending an email notification.

3. The system according to claim 1, comprising means for accessing a cloud database or an external data source as a means for collecting past sales data and order data.

Images