system

Abstract

We provide the system. [Solution] A means of collecting product data from multiple commercial facilities, A method for predicting demand using machine learning techniques, A means for optimally allocating goods among multiple commercial facilities based on predicted demand, A means of monitoring and updating information on changes in product inventory in real time, A means of analyzing user purchasing behavior and formulating sales strategies, A means of providing information on product inventory status within commercial facilities via a smartphone app, A system that includes means for instructing commercial facilities to replenish their stock based on demand forecasts that take weather information and event information into account.

Description

Technical Field

[0004] , , , ,

[0005] , , , , ,

[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 a character of the chatbot, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the modern retail industry, inventory management between e-commerce sites and physical stores tends to be fragmented, resulting in problems of overstock or understock, and missing sales opportunities. Furthermore, since it is unable to quickly respond to fluctuations in demand, there is a risk of reducing customer satisfaction and business profits. An integrated inventory management system for efficiently solving such problems is required.

Means for Solving the Problems

[0005] <0This invention provides a system that collects inventory data from multiple sales channels and predicts demand using a machine learning algorithm. Based on the predicted demand, it optimally allocates inventory across multiple sales channels, monitors inventory fluctuations in real time, and updates the data. Furthermore, by analyzing customer purchasing behavior and formulating marketing strategies, it provides a means to mitigate inventory oversupply and undersupply, and improve customer satisfaction.

[0006] "Multiple sales channels" refers to several different routes for selling products, such as e-commerce sites and physical stores.

[0007] "Means for collecting inventory data" refers to devices or programs that have the function of acquiring and aggregating inventory levels and sales status of products from each channel.

[0008] A "machine learning algorithm" is a computational method that automatically learns patterns and rules from past data to predict future demand.

[0009] "Methods for predicting demand" refers to functions that use machine learning algorithms to estimate future sales figures and sales trends.

[0010] "Means for optimally allocating inventory" refers to the function of calculating and executing the allocation of inventory efficiently across each sales channel based on demand forecasts.

[0011] "Means of monitoring inventory fluctuations in real time" refers to systems and equipment that allow for immediate confirmation of inventory status across all channels and immediate identification of any changes.

[0012] "Means for analyzing customer purchasing behavior" refer to software and methods that collect and analyze consumer purchase history and behavioral data to reveal purchasing trends and use them to formulate strategies.

[0013] "Object recognition technology" is a technology that uses cameras and sensors to automatically identify objects in images and videos, and to determine their attributes and quantities.

[0014] A "promotional campaign" is a market strategy or advertising activity conducted to promote the sale of a product or service. [Brief explanation of the drawing]

[0015] [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] This is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] This is a sequence diagram showing the processing flow of the data processing system in Example 2, which incorporates an emotion engine. [Figure 14]It is a sequence diagram showing the processing flow of a data processing system in Application Example 2 when a sentiment engine is combined.

Embodiments for Carrying out the Invention

[0016] 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.

[0017] First, the terms used in the following description will be explained.

[0018] In the following embodiments, a numbered 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 a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), etc.

[0019] In the following embodiments, a numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0020] In the following embodiments, a numbered storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, etc.

[0021] 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).

[0022] 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."

[0023] [First Embodiment]

[0024] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0025] 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.

[0026] 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).

[0027] 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.

[0028] 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.

[0029] 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.

[0030] 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.

[0031] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0032] 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.

[0033] 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.

[0034] 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.

[0035] 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".

[0036] This invention provides a system that collects inventory data from multiple sales channels and predicts demand using machine learning algorithms. To achieve this, the server, terminal, and user components cooperate to perform the following operations.

[0037] The server automatically collects inventory data from e-commerce sites and physical stores daily. To achieve this, it accesses databases for each channel to retrieve current inventory levels and recent sales figures. The collected data is integrated and stored in a database on the server. The server then uses machine learning algorithms to predict future customer demand, taking into account historical sales data and external factors (such as weather and event information). Based on this prediction, an optimization algorithm calculates how much inventory to allocate to each channel.

[0038] The terminal refers to an AI camera installed in a physical store that monitors shelf inventory in real time. The camera uses object recognition technology to instantly identify the type and quantity of products on the shelf and sends this information to a server, ensuring that inventory data is always up-to-date. For example, if an AI camera installed on a shelf scans the inventory of T-shirts and detects that the quantity has fallen below a threshold, that data is immediately sent to the server, triggering a command to replenish the inventory.

[0039] Users can view real-time inventory status and demand forecast data from the management dashboard. This dashboard is designed for intuitive operation and supports strategic decision-making regarding inventory placement and allocation. Users can also review inventory allocations and promotional plans suggested by the system and make manual adjustments as needed. For example, if a surge in demand is predicted at a particular store, users can prioritize replenishing inventory at that store.

[0040] This invention enables efficient inventory management and reduces problems of supply shortages and oversupply. As a result, lost sales opportunities are minimized, customer satisfaction improves, and profits can be optimized.

[0041] The following describes the processing flow.

[0042] Step 1:

[0043] The server collects inventory data from e-commerce sites and physical stores. After obtaining current inventory levels and recent sales history through APIs or database connections for each channel, it stores this information in a unified database.

[0044] Step 2:

[0045] The server uses collected inventory data and historical sales history to train a machine learning model. This model is designed to predict future demand, and the trained model generates sales forecasts.

[0046] Step 3:

[0047] The server optimizes inventory allocation based on predicted demand. Using an optimization algorithm (e.g., linear programming), it calculates the appropriate inventory allocation for each sales channel and sends the result as a command to each channel.

[0048] Step 4:

[0049] The terminal (an AI camera installed in the store) uses image recognition technology to monitor the inventory status of the shelves in real time. The camera identifies the type and number of products on the shelf and sends that numerical data to the server.

[0050] Step 5:

[0051] The server updates inventory information based on data sent from the terminals, adjusting it to prevent stockouts and excess inventory. It also issues automatic replenishment instructions as needed.

[0052] Step 6:

[0053] Users can view inventory data and forecasts in real time through the management dashboard. Based on the system's suggestions, users can consider inventory allocation and promotional strategies and make adjustments as needed.

[0054] (Example 1)

[0055] 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."

[0056] Traditional inventory management systems struggled to comprehensively track inventory trends across multiple sales channels in real time and accurately forecast demand. Furthermore, they lacked mechanisms to quickly reflect inventory surpluses and shortages, leading to lost sales opportunities and increased costs due to excess inventory. Additionally, it was difficult to implement strategic inventory allocation and promotions that effectively leveraged customer purchasing behavior.

[0057] 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.

[0058] In this invention, the server includes means for acquiring inventory information from multiple sales channels, means for estimating demand using machine learning techniques, and means for optimally allocating inventory among sales channels based on the estimated demand. This enables real-time optimization of inventory, formulation of sales strategies based on demand, and improvement of customer satisfaction.

[0059] "Multiple sales channels" refers to multiple distribution routes for providing goods or services, and these include different environments such as e-commerce sites and physical stores.

[0060] "Inventory information" refers to data regarding the types and quantities of products held by each sales channel.

[0061] "To obtain" means to retrieve necessary information from a database or other source of information.

[0062] "Machine learning techniques" are statistical and algorithmic methods that allow computers to learn patterns from data and use that knowledge to predict future actions and outcomes.

[0063] "Estimating demand" means predicting future demand for a product by taking into account past data and external factors.

[0064] "Optimal allocation" means efficiently distributing limited resources to achieve the objective to the greatest extent possible.

[0065] "Real-time" refers to a time frame in which processing or responses occur immediately without delay.

[0066] "Object recognition technology" is a technology that recognizes specific objects in images and videos and determines their type and attributes.

[0067] A "prompt" is a phrase or message that serves as a starting point or trigger for a system or user to input instructions or information.

[0068] This invention is a system that accurately forecasts demand and efficiently manages inventory based on inventory information obtained from multiple sales channels. The system consists of three main components: a server, terminals, and users.

[0069] The server first automatically collects inventory information from each sales channel. Specifically, it accesses the database of each channel and uses APIs to obtain current inventory levels and recent sales figures. This data is integrated into a central database on the server. Next, the server uses machine learning techniques to estimate future demand based on the collected data. This uses standard libraries in Python and R. Based on the estimation results, it optimizes inventory allocation using libraries such as Scikit-learn and TENSORFLOW®.

[0070] The terminal refers to an AI camera installed in a physical store that uses object recognition technology to determine the inventory status within the store in real time. For example, the AI ​​camera analyzes image data of products in real time and sends its type and quantity to a server. This information is immediately reflected in the server's database, contributing to efficient inventory management.

[0071] Users can access real-time inventory data and forecast information through the management screen. This management screen features data visualization capabilities and enables intuitive user operation. Based on the information displayed on the screen, users can make strategic inventory allocation decisions. Users can also use prompts when entering instructions into the system, such as text-based instructions like, "Optimize inventory based on the next demand forecast."

[0072] By implementing this invention, inventory management will become more efficient, reducing issues such as supply shortages and excess inventory, and enabling the maximization of sales opportunities and improvement of profits.

[0073] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0074] Step 1:

[0075] The server collects inventory information from multiple sales channels. The server accesses each channel's database via an API to retrieve current inventory levels and recent sales figures. This collected data becomes input, and data processing, specifically integration into the database, takes place. The integrated inventory data in the database becomes the output. Specifically, the server issues SQL queries to the database to extract information.

[0076] Step 2:

[0077] The server performs data preprocessing based on the integrated data. It removes outliers and standardizes data formats from the collected inventory data, and then combines it with external data (weather and event information). This preprocessed data becomes the input, and a new dataset called "clean data" is output. The Python Pandas library is used for data cleansing.

[0078] Step 3:

[0079] The server estimates demand using machine learning techniques. It takes clean data as input and uses models trained with Scikit-learn or TensorFlow to forecast demand. The forecast results are the output. Specifically, it feeds data into a machine learning model and calculates the demand value.

[0080] Step 4:

[0081] The server optimizes inventory allocation based on demand forecasts. It takes the forecast results and current inventory data as input and performs optimization calculations using linear programming and other methods. The output is an inventory allocation plan. The calculation process using the optimization algorithm is the specific operation.

[0082] Step 5:

[0083] The terminal's AI camera monitors the inventory status of shelves in physical stores and transmits the information to a server in real time. Image data of the shelves acquired by the camera serves as input, and inventory information analyzed using object recognition technology is output. This output is sent to the server, and the inventory information is updated in real time.

[0084] Step 6:

[0085] Users view demand forecasts and inventory allocation data through the management screen and enter prompts as needed. Real-time data provided by the server serves as input, and the user's strategic decisions are the output. Specifically, users analyze the data displayed on the screen and enter prompts such as, "Optimize inventory based on the next demand forecast."

[0086] (Application Example 1)

[0087] 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."

[0088] Inventory management in modern commercial facilities requires accommodating diverse sales channels and external factors, making it prone to inventory shortages and surpluses. Furthermore, efficient inventory replenishment and the development of appropriate sales strategies are essential, but effective systems to meet these needs are currently lacking. There is also a need for automated inventory management, including real-time inventory monitoring and demand forecasting using smart devices.

[0089] 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.

[0090] In this invention, the server includes means for collecting product data from multiple commercial facilities, means for predicting demand using machine learning techniques, and means for providing the status of product inventory within commercial facilities via a smartphone application. This enables proper management and efficient replenishment of product inventory, as well as automation of sales promotion activities.

[0091] "Multiple commercial facilities" refers to a collection of multiple retail stores or shops located in different places or locations.

[0092] "Product data" refers to a collection of information related to inventory levels, sales performance, and product characteristics.

[0093] "Machine learning techniques" refer to a group of algorithms that learn patterns from data and use them to make predictions and classifications about the future.

[0094] "Forecasting demand" is the process of predicting future purchasing activity based on past sales data and external factors.

[0095] A "smartphone app" is a computer program that runs on a portable computer device and provides specific functions or services.

[0096] "Providing product inventory status within a commercial facility" means presenting users with information that allows them to understand the inventory levels and placement of products within a store in real time.

[0097] "Object recognition technology" is a technology that uses cameras and sensors to identify specific objects in video footage and analyze their characteristics.

[0098] "Sales promotion activities" refer to marketing activities conducted to increase product awareness and purchasing intent.

[0099] This invention provides a system for efficiently managing inventory and forecasting demand within commercial facilities. Specific embodiments of this system are described below.

[0100] 1. Server Role

[0101] The server is responsible for collecting product data from multiple commercial facilities. The collected data includes inventory levels, sales data, weather information, and event information, which are then integrated and stored in a database. The server uses machine learning libraries such as TensorFlow to analyze the collected data and perform demand forecasting. Demand forecasting predicts sales trends for each commercial facility and serves as the basis for calculating the optimal allocation of products at each facility.

[0102] 2. The role of the terminal

[0103] The terminal receives real-time video data from AI cameras installed within the commercial facility and monitors product inventory levels using object recognition technology. This recognition technology utilizes technologies such as OpenCV and YOLO. The recognition results are immediately sent to the server, and product data is updated. For example, if the terminal detects that the inventory of a particular product has fallen below a threshold, that data is sent to the server, and a restocking order is issued.

[0104] 3. User Roles

[0105] Users can view the inventory status of products in commercial facilities through a smartphone app. The app is developed using React Native and features an intuitive interface. Users can also receive inventory replenishment instructions based on demand forecasts generated by machine learning models, enabling efficient inventory management.

[0106] As a concrete example, a commercial facility is predicted to experience increased demand for ice cream over the weekend. Through this app, facility managers can order additional ice cream in advance, maximizing sales opportunities.

[0107] An example of a prompt to provide to the generating AI model is, "Please propose an inventory optimization strategy that takes into account demand fluctuations within a commercial facility."

[0108] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0109] Step 1:

[0110] The server collects product data from commercial facility databases. The input is the database of each facility, and the output is integrated product data. This process uses SQL queries to retrieve product inventory levels and sales history, and then integrates and stores this data in the server's database.

[0111] Step 2:

[0112] The terminal monitors product inventory in real time using AI cameras installed in physical stores. The input is images of shelves captured by the cameras, and the output is the type and quantity of recognized products. Object recognition is performed using OpenCV and YOLO, and the results are sent to the server. If the number of recognized products falls below a threshold, data on products that need to be restocked is also output.

[0113] Step 3:

[0114] The server combines collected product data with external weather and event information and uses TensorFlow to forecast demand. The input is integrated product data and external factors, and the output is forecast data of future demand for products at each commercial facility. This process calculates demand using a model trained on historical data.

[0115] Step 4:

[0116] Users can view inventory status and demand forecasts for each facility via a smartphone app and confirm replenishment instructions notified by the system. Input is inventory and demand forecast data retrieved from the server, and output is the user's decision based on that data. Users can intuitively operate the system through an interface built with React Native and issue necessary replenishment instructions.

[0117] Step 5:

[0118] The server calculates the optimal distribution of goods between commercial facilities and assists in optimizing inventory replenishment. Inputs are demand forecast data and inventory status data, and output is the optimal replenishment quantity for each product. An optimization algorithm is used to generate product distribution instructions for each facility.

[0119] 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.

[0120] This invention is a system that combines inventory data collection from multiple sales channels, demand forecasting, inventory allocation optimization, real-time inventory fluctuation monitoring, customer purchasing behavior analysis, and marketing strategy formulation with an emotion engine that recognizes user emotions. This enables the provision of a more personalized customer experience.

[0121] The server collects inventory data from e-commerce sites and physical stores and uses machine learning algorithms to forecast demand. Furthermore, it analyzes customer purchasing behavior to develop marketing strategies and uses an emotion engine to recognize users' emotional states. Based on this information, it can personalize promotions and advertisements and deliver suggestions to customers at the optimal time. For example, the server can analyze a user's emotions from their facial expressions and voice while they are online shopping, detecting signs of interest or dissatisfaction. It can then provide appropriate product suggestions and discount information in real time.

[0122] The terminal monitors inventory levels in physical stores in conjunction with AI cameras, while also observing the emotional state of customers visiting the store. An emotion engine analyzes customers' facial expressions and actions in real time and transmits the resulting emotional data to a server. Simultaneously, data on inventory fluctuations is updated, and inventory is replenished or its placement optimized as needed.

[0123] Users can view reports analyzed by the emotion engine on a management dashboard, gaining insights to fine-tune their sales and inventory management strategies. For example, if customer response to a particular product is positive, they can decide to prioritize replenishing that product's inventory. Furthermore, they can leverage the emotion engine's information to design customized campaigns and conduct sales promotion activities tailored to customer preferences.

[0124] Through this system, users can implement more effective inventory management and marketing strategies while personalizing the customer experience. This is expected to improve customer satisfaction and increase sales.

[0125] The following describes the processing flow.

[0126] Step 1:

[0127] The server collects inventory and sales data from e-commerce sites and physical stores. This includes using APIs to access databases for each channel to retrieve the latest inventory information and store it in a unified database.

[0128] Step 2:

[0129] The server uses the collected data to run machine learning algorithms. This allows for demand forecasting that takes into account past sales trends and external factors, and then calculates the optimal inventory allocation for each channel based on these forecasts.

[0130] Step 3:

[0131] The server utilizes an emotion engine to analyze customer interactions on e-commerce sites and in physical stores. Specifically, it uses emotion analysis tools to evaluate customer facial expressions and voices, generates emotion data, and combines this with purchasing behavior data to extract insights.

[0132] Step 4:

[0133] The terminal uses AI cameras installed in physical stores to monitor in-store inventory and customers' facial expressions. Using object recognition technology, it analyzes the number of items on shelves and the emotional state of customers in real time and transmits that information to a server.

[0134] Step 5:

[0135] The server adjusts inventory management and optimizes sales strategies based on inventory and sentiment data transmitted from terminals. If inventory replenishment or reassignment is necessary, it automatically notifies the relevant departments.

[0136] Step 6:

[0137] Users can use the management dashboard to review inventory allocations and customer sentiment-based promotions suggested by the system. Furthermore, they can manually fine-tune and develop specific marketing campaigns based on the insights displayed on the dashboard.

[0138] Step 7:

[0139] The server deploys the finalized promotional campaign to the e-commerce site and store management systems, and executes necessary customer notifications. These notifications include personalized messages to enhance the customer experience.

[0140] (Example 2)

[0141] 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".

[0142] In today's commercial environment, with the diversification of distribution channels, proper inventory management and efficient marketing strategies based on customer behavior are required. Providing personalized purchasing experiences that consider consumer emotions is also a crucial element. However, collecting and analyzing the necessary data in real time to meet these requirements is not easy, and systems to improve efficiency are needed.

[0143] 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.

[0144] In this invention, the server includes means for collecting inventory information from multiple distribution channels, means for predicting demand using computational processing, and means for analyzing consumer emotions using emotion recognition functionality. This enables inventory optimization and personalized recommendations based on consumer emotions.

[0145] "Distribution channels" refer to the routes through which a product or service is distributed from the manufacturer to the consumer.

[0146] "Inventory information" refers to data regarding the quantity and location of products available for sale at a specific point in time.

[0147] "Computational processing" refers to the procedure of using computers to analyze data and generate information that is useful for future predictions and problem solving.

[0148] "Predicting demand" refers to estimating the future purchase volume of a product or service based on past data.

[0149] "Emotion recognition function" refers to technology that analyzes audio and video data to infer a person's emotional state.

[0150] "Analyzing consumer emotions" refers to the process of identifying a customer's emotional state from their facial expressions and voice.

[0151] "Inventory optimization" refers to adjusting supply to match demand and managing inventory efficiently at the lowest possible cost.

[0152] "Personalized recommendations" refer to recommendations for products and services that are customized based on the specific needs and preferences of consumers.

[0153] The embodiments for carrying out this invention are described below.

[0154] The server collects inventory information from multiple distribution channels by aggregating data from each channel using programming languages ​​and system APIs. It retrieves information in real time from e-commerce sites and physical store sales management systems via HTTP protocols and database queries. The server also analyzes historical sales data and uses computational processing to predict demand. This process uses scikit-learn, a machine learning library, to build a predictive model.

[0155] The terminal uses AI cameras placed in physical stores to recognize consumers' emotions in real time. TensorFlow is used as the deep learning framework for emotion recognition, processing video data acquired from the cameras. The resulting emotion data is sent from the terminal to a server for further analysis.

[0156] Users view the aggregated and analyzed data from the server on a management dashboard. This dashboard utilizes a web interface built with a front-end framework such as React. On this interface, users can make decisions to optimize inventory and provide personalized recommendations based on consumer sentiment.

[0157] As a concrete example, by predicting demand when a new product is released on an e-commerce site and measuring customers' initial reactions through emotion recognition, appropriate sales promotion becomes possible. Furthermore, instructions can be given to the generative AI model in the form of a prompt such as, "Design a system that performs demand forecasting and personalized promotions based on data from e-commerce sites and physical stores."

[0158] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0159] Step 1:

[0160] The server collects inventory data from distribution channels. It receives inventory information obtained from APIs and databases of each channel as input. Data processing involves unifying data in different formats and storing it in the database. The output is consistent inventory information. In this step, the program sends requests to each channel's API to aggregate inventory quantity and location information.

[0161] Step 2:

[0162] The server performs demand forecasting based on the collected data. Historical sales data and the latest inventory data are used as input. For data calculation, a machine learning algorithm is used to build a model that predicts future demand. The output is a demand forecast value for each product. In this step, a regression model is created using the Python scikit-learn library to predict future sales.

[0163] Step 3:

[0164] The device recognizes customer emotions in a physical store. It uses video data acquired from an AI camera as input. Data processing involves analyzing customer facial expressions in the video and quantifying their emotional state. The output is the analyzed emotional data. This step utilizes TensorFlow and a deep learning model to process image data in real time.

[0165] Step 4:

[0166] The server combines demand forecasts and sentiment data to provide personalized recommendations. It uses demand forecasts and customer sentiment data as input. As a data calculation, it determines promotional content based on this data and generates recommendations best suited to a specific customer group. The output is personalized promotional information. This step considers the customer's purchase history and sentiment data to recommend the most suitable products.

[0167] Step 5:

[0168] Users view data provided by the server on a management dashboard. They receive consistent inventory information and analysis results as input. Output provides insights into inventory optimization strategies and sales promotion strategies. In this step, users visualize various data using graphs and other visualizations through an interface built with the React framework, enabling rapid decision-making.

[0169] (Application Example 2)

[0170] 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 device 14 will be referred to as the "terminal."

[0171] Traditionally, it has been difficult to accurately forecast demand and manage inventory across multiple sales channels. Furthermore, it is not easy to individually analyze customer purchasing behavior and provide personalized marketing strategies at the right time, thus failing to improve the customer experience. Moreover, in physical stores, the technology to accurately understand customer emotions and provide services based on them is still immature. Solving these challenges is essential.

[0172] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0173] In this invention, the server includes means for collecting product data from multiple sales channels, means for predicting demand using a learning algorithm, and means for evaluating user emotions using an emotion recognition engine. This makes it possible to analyze customer emotions and purchasing activities while monitoring fluctuations in products in real time, and to individually implement optimized sales strategies and service provision.

[0174] "Sales channels" refer to the routes through which goods and services reach consumers, encompassing a wide range of channels including online and offline retail stores and e-commerce sites.

[0175] "Item data" refers to data concerning inventory status, price, sales status, and related attribute information.

[0176] A "learning algorithm" is a computational method that computers use to find patterns in past data and make new predictions or decisions.

[0177] "To predict" means to estimate future demand and trends based on past data and trends.

[0178] An "emotion recognition engine" is a technology or software that analyzes a user's facial expressions and behavior to identify and evaluate their emotions.

[0179] "Evaluating user emotions" refers to understanding and quantitatively evaluating emotions such as joy, anger, and surprise based on the user's facial expressions, voice, and actions.

[0180] A "sales strategy" is a plan or policy designed to effectively sell products and services based on market analysis and customer behavior.

[0181] The system for implementing this invention mainly consists of a server, a terminal, and a user interface. The server collects goods data from multiple sales channels and uses a learning algorithm to predict demand. This allows the server to optimize the supply of goods in each sales channel.

[0182] Furthermore, the server utilizes an emotion recognition engine to analyze the user's facial expressions and voice to evaluate their emotions. This process employs machine learning software such as TensorFlow and Keras to achieve real-time, highly accurate emotion analysis. Based on this data, the server provides personalized promotional and discount information to each user.

[0183] The terminals are installed in physical environments such as retail stores and continuously monitor changes in goods using object recognition technology. The data acquired from the terminals is also sent to a server, and the supply status of goods is reflected immediately.

[0184] Users can adjust their sales strategies and inventory management based on feedback from the emotion recognition engine, delivered through the user interface. The user interface also functions as a dashboard, visually displaying analysis results and reports.

[0185] As a concrete example, emotion recognition technology is implemented in a physical store. If a customer shows a confused expression in front of a product, the system automatically sends a notification to the staff and provides assistance, thereby improving the customer experience.

[0186] An example of a prompt for a generative AI model could be, "How can we detect customer smiles and recommend new products based on those results?"

[0187] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0188] Step 1:

[0189] The server collects item data from multiple sales channels. It takes online and offline retail databases as input and analyzes this data to determine real-time inventory levels. The output is a list of inventory status for each sales channel. The server automatically collects data using an API and stores it in a unified database.

[0190] Step 2:

[0191] The server predicts demand based on data collected using a learning algorithm. The input is inventory data collected in step 1, and the output is the demand forecast result for each item. This process uses a prediction model based on TensorFlow to calculate future demand, taking seasonality and trends into account.

[0192] Step 3:

[0193] The server evaluates the user's emotions using an emotion recognition engine. The input is video data from cameras installed in the store. The output is the analyzed emotion data. The video data is broken down frame by frame, and for each frame, the emotion recognition algorithm extracts facial features and determines the type of emotion.

[0194] Step 4:

[0195] The terminal uses object recognition technology to detect the status of items in a physical store. Its input is camera footage from inside the store, and its output is data on the presence and placement of items. Object recognition software identifies items from the video and sends this data to a server for inventory management.

[0196] Step 5:

[0197] The server generates and provides users with appropriate promotional and discount information based on sentiment and purchase data. Inputs are demand forecasts from step 2 and sentiment data from step 3. Output is customized promotional content. A generative AI model automatically generates promotional strategies based on forecasts and sentiment data, and notifies users via a dashboard.

[0198] Step 6:

[0199] Users receive feedback from the server and adjust sales strategies and inventory management through a dashboard. Inputs are promotional information and inventory data obtained in step 5. Outputs are improved sales strategy proposals. Users interact with the interface to view results in real time and update strategies as needed.

[0200] 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.

[0201] 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.

[0202] 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.

[0203] [Second Embodiment]

[0204] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0205] 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.

[0206] 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).

[0207] 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.

[0208] 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.

[0209] 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).

[0210] 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.

[0211] 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.

[0212] 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.

[0213] 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.

[0214] 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.

[0215] 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".

[0216] This invention provides a system that collects inventory data from multiple sales channels and predicts demand using machine learning algorithms. To achieve this, the server, terminal, and user components cooperate to perform the following operations.

[0217] The server automatically collects inventory data from e-commerce sites and physical stores daily. To achieve this, it accesses databases for each channel to retrieve current inventory levels and recent sales figures. The collected data is integrated and stored in a database on the server. The server then uses machine learning algorithms to predict future customer demand, taking into account historical sales data and external factors (such as weather and event information). Based on this prediction, an optimization algorithm calculates how much inventory to allocate to each channel.

[0218] The terminal refers to an AI camera installed in a physical store that monitors shelf inventory in real time. The camera uses object recognition technology to instantly identify the type and quantity of products on the shelf and sends this information to a server, ensuring that inventory data is always up-to-date. For example, if an AI camera installed on a shelf scans the inventory of T-shirts and detects that the quantity has fallen below a threshold, that data is immediately sent to the server, triggering a command to replenish the inventory.

[0219] Users can view real-time inventory status and demand forecast data from the management dashboard. This dashboard is designed for intuitive operation and supports strategic decision-making regarding inventory placement and allocation. Users can also review inventory allocations and promotional plans suggested by the system and make manual adjustments as needed. For example, if a surge in demand is predicted at a particular store, users can prioritize replenishing inventory at that store.

[0220] This invention enables efficient inventory management and reduces problems of supply shortages and oversupply. As a result, lost sales opportunities are minimized, customer satisfaction improves, and profits can be optimized.

[0221] The following describes the processing flow.

[0222] Step 1:

[0223] The server collects inventory data from e-commerce sites and physical stores. After obtaining current inventory levels and recent sales history through APIs or database connections for each channel, it stores this information in a unified database.

[0224] Step 2:

[0225] The server uses collected inventory data and historical sales history to train a machine learning model. This model is designed to predict future demand, and the trained model generates sales forecasts.

[0226] Step 3:

[0227] The server optimizes inventory allocation based on predicted demand. Using an optimization algorithm (e.g., linear programming), it calculates the appropriate inventory allocation for each sales channel and sends the result as a command to each channel.

[0228] Step 4:

[0229] The terminal (an AI camera installed in the store) uses image recognition technology to monitor the inventory status of the shelves in real time. The camera identifies the type and number of products on the shelf and sends that numerical data to the server.

[0230] Step 5:

[0231] The server updates inventory information based on data sent from the terminals, adjusting it to prevent stockouts and excess inventory. It also issues automatic replenishment instructions as needed.

[0232] Step 6:

[0233] Users can view inventory data and forecasts in real time through the management dashboard. Based on the system's suggestions, users can consider inventory allocation and promotional strategies and make adjustments as needed.

[0234] (Example 1)

[0235] 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".

[0236] Traditional inventory management systems struggled to comprehensively track inventory trends across multiple sales channels in real time and accurately forecast demand. Furthermore, they lacked mechanisms to quickly reflect inventory surpluses and shortages, leading to lost sales opportunities and increased costs due to excess inventory. Additionally, it was difficult to implement strategic inventory allocation and promotions that effectively leveraged customer purchasing behavior.

[0237] 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.

[0238] In this invention, the server includes means for acquiring inventory information from multiple sales channels, means for estimating demand using machine learning techniques, and means for optimally allocating inventory among sales channels based on the estimated demand. This enables real-time optimization of inventory, formulation of sales strategies based on demand, and improvement of customer satisfaction.

[0239] "Multiple sales channels" refers to multiple distribution routes for providing goods or services, and these include different environments such as e-commerce sites and physical stores.

[0240] "Inventory information" refers to data regarding the types and quantities of products held by each sales channel.

[0241] "To obtain" means to retrieve necessary information from a database or other source of information.

[0242] "Machine learning techniques" are statistical and algorithmic methods that allow computers to learn patterns from data and use that knowledge to predict future actions and outcomes.

[0243] "Estimating demand" means predicting future demand for a product by taking into account past data and external factors.

[0244] "Optimal allocation" means efficiently distributing limited resources to achieve the objective to the greatest extent possible.

[0245] "Real-time" refers to a time frame in which processing or responses occur immediately without delay.

[0246] "Object recognition technology" is a technology that recognizes specific objects in images and videos and determines their type and attributes.

[0247] A "prompt" is a phrase or message that serves as a starting point or trigger for a system or user to input instructions or information.

[0248] This invention is a system that accurately forecasts demand and efficiently manages inventory based on inventory information obtained from multiple sales channels. The system consists of three main components: a server, terminals, and users.

[0249] The server first automatically collects inventory information from each sales channel. Specifically, it accesses the database of each channel and uses APIs to retrieve current inventory levels and recent sales figures. This data is integrated into a central database on the server. Next, the server uses machine learning techniques to estimate future demand based on the collected data. This is done using standard libraries in Python and R. Based on the estimation results, it optimizes inventory allocation using libraries such as Scikit-learn and TensorFlow.

[0250] The terminal refers to an AI camera installed in a physical store that uses object recognition technology to determine the inventory status within the store in real time. For example, the AI ​​camera analyzes image data of products in real time and sends its type and quantity to a server. This information is immediately reflected in the server's database, contributing to efficient inventory management.

[0251] Users can access real-time inventory data and forecast information through the management screen. This management screen features data visualization capabilities and enables intuitive user operation. Based on the information displayed on the screen, users can make strategic inventory allocation decisions. Users can also use prompts when entering instructions into the system, such as text-based instructions like, "Optimize inventory based on the next demand forecast."

[0252] By implementing this invention, inventory management will become more efficient, reducing issues such as supply shortages and excess inventory, and enabling the maximization of sales opportunities and improvement of profits.

[0253] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0254] Step 1:

[0255] The server collects inventory information from multiple sales channels. The server accesses each channel's database via an API to retrieve current inventory levels and recent sales figures. This collected data becomes input, and data processing, specifically integration into the database, takes place. The integrated inventory data in the database becomes the output. Specifically, the server issues SQL queries to the database to extract information.

[0256] Step 2:

[0257] The server performs data preprocessing based on the integrated data. It removes outliers and standardizes data formats from the collected inventory data, and then combines it with external data (weather and event information). This preprocessed data becomes the input, and a new dataset called "clean data" is output. The Python Pandas library is used for data cleansing.

[0258] Step 3:

[0259] The server estimates demand using machine learning techniques. It takes clean data as input and uses models trained with Scikit-learn or TensorFlow to forecast demand. The forecast results are the output. Specifically, it feeds data into a machine learning model and calculates the demand value.

[0260] Step 4:

[0261] The server optimizes inventory allocation based on demand forecasts. It takes the forecast results and current inventory data as input and performs optimization calculations using linear programming and other methods. The output is an inventory allocation plan. The calculation process using the optimization algorithm is the specific operation.

[0262] Step 5:

[0263] The terminal's AI camera monitors the inventory status of shelves in physical stores and transmits the information to a server in real time. Image data of the shelves acquired by the camera serves as input, and inventory information analyzed using object recognition technology is output. This output is sent to the server, and the inventory information is updated in real time.

[0264] Step 6:

[0265] Users view demand forecasts and inventory allocation data through the management screen and enter prompts as needed. Real-time data provided by the server serves as input, and the user's strategic decisions are the output. Specifically, users analyze the data displayed on the screen and enter prompts such as, "Optimize inventory based on the next demand forecast."

[0266] (Application Example 1)

[0267] 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."

[0268] Inventory management in modern commercial facilities requires accommodating diverse sales channels and external factors, making it prone to inventory shortages and surpluses. Furthermore, efficient inventory replenishment and the development of appropriate sales strategies are essential, but effective systems to meet these needs are currently lacking. There is also a need for automated inventory management, including real-time inventory monitoring and demand forecasting using smart devices.

[0269] 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.

[0270] In this invention, the server includes means for collecting product data from multiple commercial facilities, means for predicting demand using machine learning techniques, and means for providing the status of product inventory within commercial facilities via a smartphone application. This enables proper management and efficient replenishment of product inventory, as well as automation of sales promotion activities.

[0271] "Multiple commercial facilities" refers to a collection of multiple retail stores or shops located in different places or locations.

[0272] "Product data" refers to a collection of information related to inventory levels, sales performance, and product characteristics.

[0273] "Machine learning techniques" refer to a group of algorithms that learn patterns from data and use them to make predictions and classifications about the future.

[0274] "Forecasting demand" is the process of predicting future purchasing activity based on past sales data and external factors.

[0275] A "smartphone app" is a computer program that runs on a portable computer device and provides specific functions or services.

[0276] "Providing product inventory status within a commercial facility" means presenting users with information that allows them to understand the inventory levels and placement of products within a store in real time.

[0277] "Object recognition technology" is a technology that uses cameras and sensors to identify specific objects in video footage and analyze their characteristics.

[0278] "Sales promotion activities" refer to marketing activities conducted to increase product awareness and purchasing intent.

[0279] This invention provides a system for efficiently managing inventory and forecasting demand within commercial facilities. Specific embodiments of this system are described below.

[0280] 1. Server Role

[0281] The server is responsible for collecting product data from multiple commercial facilities. The collected data includes inventory levels, sales data, weather information, and event information, which are then integrated and stored in a database. The server uses machine learning libraries such as TensorFlow to analyze the collected data and perform demand forecasting. Demand forecasting predicts sales trends for each commercial facility and serves as the basis for calculating the optimal allocation of products at each facility.

[0282] 2. The role of the terminal

[0283] The terminal receives real-time video data from an AI camera installed in a commercial facility and monitors the inventory status of products using object recognition technology. Technologies such as OpenCV and YOLO are used for this recognition technology. The recognition results are immediately sent to the server, and the product data is updated. For example, when the terminal detects that the inventory of a specific product has fallen below the threshold, the data is sent to the server, and a replenishment of the product is instructed.

[0284] 3. Role of the User

[0285] Users can view the inventory status of products in a commercial facility through a smartphone app. The app is developed using React Native and has an intuitive operation interface. Users can also receive inventory replenishment instructions based on demand predictions by a machine learning model and conduct inventory management without waste.

[0286] As a specific example, in a certain commercial facility, a rise in temperature on the weekend is predicted, suggesting an increase in the demand for ice cream in that facility. Through this app, the facility manager can place an order for additional ice cream in advance and maximize sales opportunities.

[0287] As an example of a prompt sentence provided to the generative AI model, "Please propose an inventory optimization strategy considering demand fluctuations in a commercial facility" can be considered.

[0288] The flow of specific processing in Application Example 1 will be described using FIG. 12.

[0289] Step 1:

[0290] The server collects product data from the databases of commercial facilities. The input is the database of each facility, and the output is the integrated product data. In this process, SQL queries are used to obtain the inventory quantity and sales history of products and integrate and store them in the server's database.

[0291] Step 2:

[0292] The terminal monitors product inventory in real time using AI cameras installed in physical stores. The input is images of shelves captured by the cameras, and the output is the type and quantity of recognized products. Object recognition is performed using OpenCV and YOLO, and the results are sent to the server. If the number of recognized products falls below a threshold, data on products that need to be restocked is also output.

[0293] Step 3:

[0294] The server combines collected product data with external weather and event information and uses TensorFlow to forecast demand. The input is integrated product data and external factors, and the output is forecast data of future demand for products at each commercial facility. This process calculates demand using a model trained on historical data.

[0295] Step 4:

[0296] Users can view inventory status and demand forecasts for each facility via a smartphone app and confirm replenishment instructions notified by the system. Input is inventory and demand forecast data retrieved from the server, and output is the user's decision based on that data. Users can intuitively operate the system through an interface built with React Native and issue necessary replenishment instructions.

[0297] Step 5:

[0298] The server calculates the optimal distribution of goods between commercial facilities and assists in optimizing inventory replenishment. Inputs are demand forecast data and inventory status data, and output is the optimal replenishment quantity for each product. An optimization algorithm is used to generate product distribution instructions for each facility.

[0299] 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.

[0300] This invention is a system that combines inventory data collection from multiple sales channels, demand forecasting, inventory allocation optimization, real-time inventory fluctuation monitoring, customer purchasing behavior analysis, and marketing strategy formulation with an emotion engine that recognizes user emotions. This enables the provision of a more personalized customer experience.

[0301] The server collects inventory data from e-commerce sites and physical stores and uses machine learning algorithms to forecast demand. Furthermore, it analyzes customer purchasing behavior to develop marketing strategies and uses an emotion engine to recognize users' emotional states. Based on this information, it can personalize promotions and advertisements and deliver suggestions to customers at the optimal time. For example, the server can analyze a user's emotions from their facial expressions and voice while they are online shopping, detecting signs of interest or dissatisfaction. It can then provide appropriate product suggestions and discount information in real time.

[0302] The terminal monitors inventory levels in physical stores in conjunction with AI cameras, while also observing the emotional state of customers visiting the store. An emotion engine analyzes customers' facial expressions and actions in real time and transmits the resulting emotional data to a server. Simultaneously, data on inventory fluctuations is updated, and inventory is replenished or its placement optimized as needed.

[0303] Users can view reports analyzed by the emotion engine on a management dashboard, gaining insights to fine-tune their sales and inventory management strategies. For example, if customer response to a particular product is positive, they can decide to prioritize replenishing that product's inventory. Furthermore, they can leverage the emotion engine's information to design customized campaigns and conduct sales promotion activities tailored to customer preferences.

[0304] Through this system, users can implement more effective inventory management and marketing strategies while personalizing the customer experience. As a result, customer satisfaction can be improved and an increase in sales can be expected.

[0305] The following describes the processing flow.

[0306] Step 1:

[0307] The server collects inventory data and sales data from the e-commerce site and physical stores. This includes accessing the APIs of the databases of each channel to obtain the latest inventory information and storing it in the integrated database.

[0308] Step 2:

[0309] The server executes a machine learning algorithm using the collected data. As a result, demand forecasting considering past sales trends and external factors is performed, and calculations for optimizing inventory allocation to each channel are performed based on this.

[0310] Step 3:

[0311] The server utilizes an emotion engine to analyze customer interactions on the e-commerce site and physical stores. Specifically, it evaluates the expressions and voices of customers using an emotion analysis tool to generate emotion data, and extracts insights by combining this with purchase behavior data.

[0312] Step 4:

[0313] The terminal monitors the in-store inventory and customers' expressions using an AI camera installed in the physical store. Using object recognition technology, it analyzes the number of products on the shelves and the emotional state of customers in real time and transmits that information to the server.

[0314] Step 5:

[0315] The server adjusts inventory management and optimizes sales strategies based on inventory and sentiment data transmitted from terminals. If inventory replenishment or reassignment is necessary, it automatically notifies the relevant departments.

[0316] Step 6:

[0317] Users can use the management dashboard to review inventory allocations and customer sentiment-based promotions suggested by the system. Furthermore, they can manually fine-tune and develop specific marketing campaigns based on the insights displayed on the dashboard.

[0318] Step 7:

[0319] The server deploys the finalized promotional campaign to the e-commerce site and store management systems, and executes necessary customer notifications. These notifications include personalized messages to enhance the customer experience.

[0320] (Example 2)

[0321] 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".

[0322] In today's commercial environment, with the diversification of distribution channels, proper inventory management and efficient marketing strategies based on customer behavior are required. Providing personalized purchasing experiences that consider consumer emotions is also a crucial element. However, collecting and analyzing the necessary data in real time to meet these requirements is not easy, and systems to improve efficiency are needed.

[0323] 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.

[0324] In this invention, the server includes means for collecting inventory information from multiple distribution channels, means for predicting demand using computational processing, and means for analyzing consumer emotions using emotion recognition functionality. This enables inventory optimization and personalized recommendations based on consumer emotions.

[0325] "Distribution channels" refer to the routes through which a product or service is distributed from the manufacturer to the consumer.

[0326] "Inventory information" refers to data regarding the quantity and location of products available for sale at a specific point in time.

[0327] "Computational processing" refers to the procedure of using computers to analyze data and generate information that is useful for future predictions and problem solving.

[0328] "Predicting demand" refers to estimating the future purchase volume of a product or service based on past data.

[0329] "Emotion recognition function" refers to technology that analyzes audio and video data to infer a person's emotional state.

[0330] "Analyzing consumer emotions" refers to the process of identifying a customer's emotional state from their facial expressions and voice.

[0331] "Inventory optimization" refers to adjusting supply to match demand and managing inventory efficiently at the lowest possible cost.

[0332] "Personalized recommendations" refer to recommendations for products and services that are customized based on the specific needs and preferences of consumers.

[0333] The embodiments for carrying out this invention are described below.

[0334] The server collects inventory information from multiple distribution channels by aggregating data from each channel using programming languages ​​and system APIs. It retrieves information in real time from e-commerce sites and physical store sales management systems via HTTP protocols and database queries. The server also analyzes historical sales data and uses computational processing to predict demand. This process uses scikit-learn, a machine learning library, to build a predictive model.

[0335] The terminal uses AI cameras placed in physical stores to recognize consumers' emotions in real time. TensorFlow is used as the deep learning framework for emotion recognition, processing video data acquired from the cameras. The resulting emotion data is sent from the terminal to a server for further analysis.

[0336] Users view the aggregated and analyzed data from the server on a management dashboard. This dashboard utilizes a web interface built with a front-end framework such as React. On this interface, users can make decisions to optimize inventory and provide personalized recommendations based on consumer sentiment.

[0337] As a concrete example, by predicting demand when a new product is released on an e-commerce site and measuring customers' initial reactions through emotion recognition, appropriate sales promotion becomes possible. Furthermore, instructions can be given to the generative AI model in the form of a prompt such as, "Design a system that performs demand forecasting and personalized promotions based on data from e-commerce sites and physical stores."

[0338] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0339] Step 1:

[0340] The server collects inventory data from distribution channels. It receives inventory information obtained from APIs and databases of each channel as input. Data processing involves unifying data in different formats and storing it in the database. The output is consistent inventory information. In this step, the program sends requests to each channel's API to aggregate inventory quantity and location information.

[0341] Step 2:

[0342] The server performs demand forecasting based on the collected data. Historical sales data and the latest inventory data are used as input. For data calculation, a machine learning algorithm is used to build a model that predicts future demand. The output is a demand forecast value for each product. In this step, a regression model is created using the Python scikit-learn library to predict future sales.

[0343] Step 3:

[0344] The device recognizes customer emotions in a physical store. It uses video data acquired from an AI camera as input. Data processing involves analyzing customer facial expressions in the video and quantifying their emotional state. The output is the analyzed emotional data. This step utilizes TensorFlow and a deep learning model to process image data in real time.

[0345] Step 4:

[0346] The server combines demand forecasts and sentiment data to provide personalized recommendations. It uses demand forecasts and customer sentiment data as input. As a data calculation, it determines promotional content based on this data and generates recommendations best suited to a specific customer group. The output is personalized promotional information. This step considers the customer's purchase history and sentiment data to recommend the most suitable products.

[0347] Step 5:

[0348] Users view data provided by the server on a management dashboard. They receive consistent inventory information and analysis results as input. Output provides insights into inventory optimization strategies and sales promotion strategies. In this step, users visualize various data using graphs and other visualizations through an interface built with the React framework, enabling rapid decision-making.

[0349] (Application Example 2)

[0350] 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."

[0351] Traditionally, it has been difficult to accurately forecast demand and manage inventory across multiple sales channels. Furthermore, it is not easy to individually analyze customer purchasing behavior and provide personalized marketing strategies at the right time, thus failing to improve the customer experience. Moreover, in physical stores, the technology to accurately understand customer emotions and provide services based on them is still immature. Solving these challenges is essential.

[0352] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0353] In this invention, the server includes means for collecting product data from multiple sales channels, means for predicting demand using a learning algorithm, and means for evaluating user emotions using an emotion recognition engine. This makes it possible to analyze customer emotions and purchasing activities while monitoring fluctuations in products in real time, and to individually implement optimized sales strategies and service provision.

[0354] "Sales channels" refer to the routes through which goods and services reach consumers, encompassing a wide range of channels including online and offline retail stores and e-commerce sites.

[0355] "Item data" refers to data concerning inventory status, price, sales status, and related attribute information.

[0356] A "learning algorithm" is a computational method that computers use to find patterns in past data and make new predictions or decisions.

[0357] "To predict" means to estimate future demand and trends based on past data and trends.

[0358] An "emotion recognition engine" is a technology or software that analyzes a user's facial expressions and behavior to identify and evaluate their emotions.

[0359] "Evaluating user emotions" refers to understanding and quantitatively evaluating emotions such as joy, anger, and surprise based on the user's facial expressions, voice, and actions.

[0360] A "sales strategy" is a plan or policy designed to effectively sell products and services based on market analysis and customer behavior.

[0361] The system for implementing this invention mainly consists of a server, a terminal, and a user interface. The server collects goods data from multiple sales channels and uses a learning algorithm to predict demand. This allows the server to optimize the supply of goods in each sales channel.

[0362] Furthermore, the server utilizes an emotion recognition engine to analyze the user's facial expressions and voice to evaluate their emotions. This process employs machine learning software such as TensorFlow and Keras to achieve real-time, highly accurate emotion analysis. Based on this data, the server provides personalized promotional and discount information to each user.

[0363] The terminals are installed in physical environments such as retail stores and continuously monitor changes in goods using object recognition technology. The data acquired from the terminals is also sent to a server, and the supply status of goods is reflected immediately.

[0364] Users can adjust their sales strategies and inventory management based on feedback from the emotion recognition engine, delivered through the user interface. The user interface also functions as a dashboard, visually displaying analysis results and reports.

[0365] As a concrete example, emotion recognition technology is implemented in a physical store. If a customer shows a confused expression in front of a product, the system automatically sends a notification to the staff and provides assistance, thereby improving the customer experience.

[0366] An example of a prompt for a generative AI model could be, "How can we detect customer smiles and recommend new products based on those results?"

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

[0368] Step 1:

[0369] The server collects item data from multiple sales channels. It takes online and offline retail databases as input and analyzes this data to determine real-time inventory levels. The output is a list of inventory status for each sales channel. The server automatically collects data using an API and stores it in a unified database.

[0370] Step 2:

[0371] The server predicts demand based on data collected using a learning algorithm. The input is inventory data collected in step 1, and the output is the demand forecast result for each item. This process uses a prediction model based on TensorFlow to calculate future demand, taking seasonality and trends into account.

[0372] Step 3:

[0373] The server evaluates the user's emotions using an emotion recognition engine. The input is video data from cameras installed in the store. The output is the analyzed emotion data. The video data is broken down frame by frame, and for each frame, the emotion recognition algorithm extracts facial features and determines the type of emotion.

[0374] Step 4:

[0375] The terminal uses object recognition technology to detect the status of items in a physical store. Its input is camera footage from inside the store, and its output is data on the presence and placement of items. Object recognition software identifies items from the video and sends this data to a server for inventory management.

[0376] Step 5:

[0377] The server generates and provides users with appropriate promotional and discount information based on sentiment and purchase data. Inputs are demand forecasts from step 2 and sentiment data from step 3. Output is customized promotional content. A generative AI model automatically generates promotional strategies based on forecasts and sentiment data, and notifies users via a dashboard.

[0378] Step 6:

[0379] Users receive feedback from the server and adjust sales strategies and inventory management through a dashboard. Inputs are promotional information and inventory data obtained in step 5. Outputs are improved sales strategy proposals. Users interact with the interface to view results in real time and update strategies as needed.

[0380] 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.

[0381] 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.

[0382] 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.

[0383] [Third Embodiment]

[0384] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0385] 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.

[0386] 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).

[0387] 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.

[0388] 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.

[0389] 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).

[0390] 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.

[0391] 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.

[0392] 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.

[0393] 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.

[0394] 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.

[0395] 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".

[0396] This invention provides a system that collects inventory data from multiple sales channels and predicts demand using machine learning algorithms. To achieve this, the server, terminal, and user components cooperate to perform the following operations.

[0397] The server automatically collects inventory data from e-commerce sites and physical stores daily. To achieve this, it accesses databases for each channel to retrieve current inventory levels and recent sales figures. The collected data is integrated and stored in a database on the server. The server then uses machine learning algorithms to predict future customer demand, taking into account historical sales data and external factors (such as weather and event information). Based on this prediction, an optimization algorithm calculates how much inventory to allocate to each channel.

[0398] The terminal refers to an AI camera installed in a physical store that monitors shelf inventory in real time. The camera uses object recognition technology to instantly identify the type and quantity of products on the shelf and sends this information to a server, ensuring that inventory data is always up-to-date. For example, if an AI camera installed on a shelf scans the inventory of T-shirts and detects that the quantity has fallen below a threshold, that data is immediately sent to the server, triggering a command to replenish the inventory.

[0399] Users can view real-time inventory status and demand forecast data from the management dashboard. This dashboard is designed for intuitive operation and supports strategic decision-making regarding inventory placement and allocation. Users can also review inventory allocations and promotional plans suggested by the system and make manual adjustments as needed. For example, if a surge in demand is predicted at a particular store, users can prioritize replenishing inventory at that store.

[0400] This invention enables efficient inventory management and reduces problems of supply shortages and oversupply. As a result, lost sales opportunities are minimized, customer satisfaction improves, and profits can be optimized.

[0401] The following describes the processing flow.

[0402] Step 1:

[0403] The server collects inventory data from e-commerce sites and physical stores. After obtaining current inventory levels and recent sales history through APIs or database connections for each channel, it stores this information in a unified database.

[0404] Step 2:

[0405] The server uses collected inventory data and historical sales history to train a machine learning model. This model is designed to predict future demand, and the trained model generates sales forecasts.

[0406] Step 3:

[0407] The server optimizes inventory allocation based on predicted demand. Using an optimization algorithm (e.g., linear programming), it calculates the appropriate inventory allocation for each sales channel and sends the result as a command to each channel.

[0408] Step 4:

[0409] The terminal (an AI camera installed in the store) uses image recognition technology to monitor the inventory status of the shelves in real time. The camera identifies the type and number of products on the shelf and sends that numerical data to the server.

[0410] Step 5:

[0411] The server updates inventory information based on data sent from the terminals, adjusting it to prevent stockouts and excess inventory. It also issues automatic replenishment instructions as needed.

[0412] Step 6:

[0413] Users can view inventory data and forecasts in real time through the management dashboard. Based on the system's suggestions, users can consider inventory allocation and promotional strategies and make adjustments as needed.

[0414] (Example 1)

[0415] 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."

[0416] Traditional inventory management systems struggled to comprehensively track inventory trends across multiple sales channels in real time and accurately forecast demand. Furthermore, they lacked mechanisms to quickly reflect inventory surpluses and shortages, leading to lost sales opportunities and increased costs due to excess inventory. Additionally, it was difficult to implement strategic inventory allocation and promotions that effectively leveraged customer purchasing behavior.

[0417] 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.

[0418] In this invention, the server includes means for acquiring inventory information from multiple sales channels, means for estimating demand using machine learning techniques, and means for optimally allocating inventory among sales channels based on the estimated demand. This enables real-time optimization of inventory, formulation of sales strategies based on demand, and improvement of customer satisfaction.

[0419] "Multiple sales channels" refers to multiple distribution routes for providing goods or services, and these include different environments such as e-commerce sites and physical stores.

[0420] "Inventory information" refers to data regarding the types and quantities of products held by each sales channel.

[0421] "To obtain" means to retrieve necessary information from a database or other source of information.

[0422] "Machine learning techniques" are statistical and algorithmic methods that allow computers to learn patterns from data and use that knowledge to predict future actions and outcomes.

[0423] "Estimating demand" means predicting future demand for a product by taking into account past data and external factors.

[0424] "Optimal allocation" means efficiently distributing limited resources to achieve the objective to the greatest extent possible.

[0425] "Real-time" refers to a time frame in which processing or responses occur immediately without delay.

[0426] "Object recognition technology" is a technology that recognizes specific objects in images and videos and determines their type and attributes.

[0427] A "prompt" is a phrase or message that serves as a starting point or trigger for a system or user to input instructions or information.

[0428] This invention is a system that accurately forecasts demand and efficiently manages inventory based on inventory information obtained from multiple sales channels. The system consists of three main components: a server, terminals, and users.

[0429] The server first automatically collects inventory information from each sales channel. Specifically, it accesses the database of each channel and uses APIs to retrieve current inventory levels and recent sales figures. This data is integrated into a central database on the server. Next, the server uses machine learning techniques to estimate future demand based on the collected data. This is done using standard libraries in Python and R. Based on the estimation results, it optimizes inventory allocation using libraries such as Scikit-learn and TensorFlow.

[0430] The terminal refers to an AI camera installed in a physical store that uses object recognition technology to determine the inventory status within the store in real time. For example, the AI ​​camera analyzes image data of products in real time and sends its type and quantity to a server. This information is immediately reflected in the server's database, contributing to efficient inventory management.

[0431] Users can access real-time inventory data and forecast information through the management screen. This management screen features data visualization capabilities and enables intuitive user operation. Based on the information displayed on the screen, users can make strategic inventory allocation decisions. Users can also use prompts when entering instructions into the system, such as text-based instructions like, "Optimize inventory based on the next demand forecast."

[0432] By implementing this invention, inventory management will become more efficient, reducing issues such as supply shortages and excess inventory, and enabling the maximization of sales opportunities and improvement of profits.

[0433] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0434] Step 1:

[0435] The server collects inventory information from multiple sales channels. The server accesses each channel's database via an API to retrieve current inventory levels and recent sales figures. This collected data becomes input, and data processing, specifically integration into the database, takes place. The integrated inventory data in the database becomes the output. Specifically, the server issues SQL queries to the database to extract information.

[0436] Step 2:

[0437] The server performs data preprocessing based on the integrated data. It removes outliers and standardizes data formats from the collected inventory data, and then combines it with external data (weather and event information). This preprocessed data becomes the input, and a new dataset called "clean data" is output. The Python Pandas library is used for data cleansing.

[0438] Step 3:

[0439] The server estimates demand using machine learning techniques. It takes clean data as input and uses models trained with Scikit-learn or TensorFlow to forecast demand. The forecast results are the output. Specifically, it feeds data into a machine learning model and calculates the demand value.

[0440] Step 4:

[0441] The server optimizes inventory allocation based on demand forecasts. It takes the forecast results and current inventory data as input and performs optimization calculations using linear programming and other methods. The output is an inventory allocation plan. The calculation process using the optimization algorithm is the specific operation.

[0442] Step 5:

[0443] The terminal's AI camera monitors the inventory status of shelves in physical stores and transmits the information to a server in real time. Image data of the shelves acquired by the camera serves as input, and inventory information analyzed using object recognition technology is output. This output is sent to the server, and the inventory information is updated in real time.

[0444] Step 6:

[0445] Users view demand forecasts and inventory allocation data through the management screen and enter prompts as needed. Real-time data provided by the server serves as input, and the user's strategic decisions are the output. Specifically, users analyze the data displayed on the screen and enter prompts such as, "Optimize inventory based on the next demand forecast."

[0446] (Application Example 1)

[0447] 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."

[0448] Inventory management in modern commercial facilities requires accommodating diverse sales channels and external factors, making it prone to inventory shortages and surpluses. Furthermore, efficient inventory replenishment and the development of appropriate sales strategies are essential, but effective systems to meet these needs are currently lacking. There is also a need for automated inventory management, including real-time inventory monitoring and demand forecasting using smart devices.

[0449] 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.

[0450] In this invention, the server includes means for collecting product data from multiple commercial facilities, means for predicting demand using machine learning techniques, and means for providing the status of product inventory within commercial facilities via a smartphone application. This enables proper management and efficient replenishment of product inventory, as well as automation of sales promotion activities.

[0451] "Multiple commercial facilities" refers to a collection of multiple retail stores or shops located in different places or locations.

[0452] "Product data" refers to a collection of information related to inventory levels, sales performance, and product characteristics.

[0453] "Machine learning techniques" refer to a group of algorithms that learn patterns from data and use them to make predictions and classifications about the future.

[0454] "Forecasting demand" is the process of predicting future purchasing activity based on past sales data and external factors.

[0455] A "smartphone app" is a computer program that runs on a portable computer device and provides specific functions or services.

[0456] "Providing product inventory status within a commercial facility" means presenting users with information that allows them to understand the inventory levels and placement of products within a store in real time.

[0457] "Object recognition technology" is a technology that uses cameras and sensors to identify specific objects in video footage and analyze their characteristics.

[0458] "Sales promotion activities" refer to marketing activities conducted to increase product awareness and purchasing intent.

[0459] This invention provides a system for efficiently managing inventory and forecasting demand within commercial facilities. Specific embodiments of this system are described below.

[0460] 1. Server Role

[0461] The server is responsible for collecting product data from multiple commercial facilities. The collected data includes inventory levels, sales data, weather information, and event information, which are then integrated and stored in a database. The server uses machine learning libraries such as TensorFlow to analyze the collected data and perform demand forecasting. Demand forecasting predicts sales trends for each commercial facility and serves as the basis for calculating the optimal allocation of products at each facility.

[0462] 2. The role of the terminal

[0463] The terminal receives real-time video data from AI cameras installed within the commercial facility and monitors product inventory levels using object recognition technology. This recognition technology utilizes technologies such as OpenCV and YOLO. The recognition results are immediately sent to the server, and product data is updated. For example, if the terminal detects that the inventory of a particular product has fallen below a threshold, that data is sent to the server, and a restocking order is issued.

[0464] 3. User Roles

[0465] Users can view the inventory status of products in commercial facilities through a smartphone app. The app is developed using React Native and features an intuitive interface. Users can also receive inventory replenishment instructions based on demand forecasts generated by machine learning models, enabling efficient inventory management.

[0466] As a concrete example, a commercial facility is predicted to experience increased demand for ice cream over the weekend. Through this app, facility managers can order additional ice cream in advance, maximizing sales opportunities.

[0467] An example of a prompt to provide to the generating AI model is, "Please propose an inventory optimization strategy that takes into account demand fluctuations within a commercial facility."

[0468] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0469] Step 1:

[0470] The server collects product data from commercial facility databases. The input is the database of each facility, and the output is integrated product data. This process uses SQL queries to retrieve product inventory levels and sales history, and then integrates and stores this data in the server's database.

[0471] Step 2:

[0472] The terminal monitors product inventory in real time using AI cameras installed in physical stores. The input is images of shelves captured by the cameras, and the output is the type and quantity of recognized products. Object recognition is performed using OpenCV and YOLO, and the results are sent to the server. If the number of recognized products falls below a threshold, data on products that need to be restocked is also output.

[0473] Step 3:

[0474] The server combines collected product data with external weather and event information and uses TensorFlow to forecast demand. The input is integrated product data and external factors, and the output is forecast data of future demand for products at each commercial facility. This process calculates demand using a model trained on historical data.

[0475] Step 4:

[0476] Users can view inventory status and demand forecasts for each facility via a smartphone app and confirm replenishment instructions notified by the system. Input is inventory and demand forecast data retrieved from the server, and output is the user's decision based on that data. Users can intuitively operate the system through an interface built with React Native and issue necessary replenishment instructions.

[0477] Step 5:

[0478] The server calculates the optimal distribution of goods between commercial facilities and assists in optimizing inventory replenishment. Inputs are demand forecast data and inventory status data, and output is the optimal replenishment quantity for each product. An optimization algorithm is used to generate product distribution instructions for each facility.

[0479] 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.

[0480] This invention is a system that combines inventory data collection from multiple sales channels, demand forecasting, inventory allocation optimization, real-time inventory fluctuation monitoring, customer purchasing behavior analysis, and marketing strategy formulation with an emotion engine that recognizes user emotions. This enables the provision of a more personalized customer experience.

[0481] The server collects inventory data from e-commerce sites and physical stores and uses machine learning algorithms to forecast demand. Furthermore, it analyzes customer purchasing behavior to develop marketing strategies and uses an emotion engine to recognize users' emotional states. Based on this information, it can personalize promotions and advertisements and deliver suggestions to customers at the optimal time. For example, the server can analyze a user's emotions from their facial expressions and voice while they are online shopping, detecting signs of interest or dissatisfaction. It can then provide appropriate product suggestions and discount information in real time.

[0482] The terminal monitors inventory levels in physical stores in conjunction with AI cameras, while also observing the emotional state of customers visiting the store. An emotion engine analyzes customers' facial expressions and actions in real time and transmits the resulting emotional data to a server. Simultaneously, data on inventory fluctuations is updated, and inventory is replenished or its placement optimized as needed.

[0483] Users can view reports analyzed by the emotion engine on a management dashboard, gaining insights to fine-tune their sales and inventory management strategies. For example, if customer response to a particular product is positive, they can decide to prioritize replenishing that product's inventory. Furthermore, they can leverage the emotion engine's information to design customized campaigns and conduct sales promotion activities tailored to customer preferences.

[0484] Through this system, users can implement more effective inventory management and marketing strategies while personalizing the customer experience. This is expected to improve customer satisfaction and increase sales.

[0485] The following describes the processing flow.

[0486] Step 1:

[0487] The server collects inventory and sales data from e-commerce sites and physical stores. This includes using APIs to access databases for each channel to retrieve the latest inventory information and store it in a unified database.

[0488] Step 2:

[0489] The server uses the collected data to run machine learning algorithms. This allows for demand forecasting that takes into account past sales trends and external factors, and then calculates the optimal inventory allocation for each channel based on these forecasts.

[0490] Step 3:

[0491] The server utilizes an emotion engine to analyze customer interactions on e-commerce sites and in physical stores. Specifically, it uses emotion analysis tools to evaluate customer facial expressions and voices, generates emotion data, and combines this with purchasing behavior data to extract insights.

[0492] Step 4:

[0493] The terminal uses AI cameras installed in physical stores to monitor in-store inventory and customers' facial expressions. Using object recognition technology, it analyzes the number of items on shelves and the emotional state of customers in real time and transmits that information to a server.

[0494] Step 5:

[0495] The server adjusts inventory management and optimizes sales strategies based on inventory and sentiment data transmitted from terminals. If inventory replenishment or reassignment is necessary, it automatically notifies the relevant departments.

[0496] Step 6:

[0497] Users can use the management dashboard to review inventory allocations and customer sentiment-based promotions suggested by the system. Furthermore, they can manually fine-tune and develop specific marketing campaigns based on the insights displayed on the dashboard.

[0498] Step 7:

[0499] The server deploys the finalized promotional campaign to the e-commerce site and store management systems, and executes necessary customer notifications. These notifications include personalized messages to enhance the customer experience.

[0500] (Example 2)

[0501] 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."

[0502] In today's commercial environment, with the diversification of distribution channels, proper inventory management and efficient marketing strategies based on customer behavior are required. Providing personalized purchasing experiences that consider consumer emotions is also a crucial element. However, collecting and analyzing the necessary data in real time to meet these requirements is not easy, and systems to improve efficiency are needed.

[0503] 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.

[0504] In this invention, the server includes means for collecting inventory information from multiple distribution channels, means for predicting demand using computational processing, and means for analyzing consumer emotions using emotion recognition functionality. This enables inventory optimization and personalized recommendations based on consumer emotions.

[0505] "Distribution channels" refer to the routes through which a product or service is distributed from the manufacturer to the consumer.

[0506] "Inventory information" refers to data regarding the quantity and location of products available for sale at a specific point in time.

[0507] "Computational processing" refers to the procedure of using computers to analyze data and generate information that is useful for future predictions and problem solving.

[0508] "Predicting demand" refers to estimating the future purchase volume of a product or service based on past data.

[0509] "Emotion recognition function" refers to technology that analyzes audio and video data to infer a person's emotional state.

[0510] "Analyzing consumer emotions" refers to the process of identifying a customer's emotional state from their facial expressions and voice.

[0511] "Inventory optimization" refers to adjusting supply to match demand and managing inventory efficiently at the lowest possible cost.

[0512] "Personalized recommendations" refer to recommendations for products and services that are customized based on the specific needs and preferences of consumers.

[0513] The embodiments for carrying out this invention are described below.

[0514] The server collects inventory information from multiple distribution channels by aggregating data from each channel using programming languages ​​and system APIs. It retrieves information in real time from e-commerce sites and physical store sales management systems via HTTP protocols and database queries. The server also analyzes historical sales data and uses computational processing to predict demand. This process uses scikit-learn, a machine learning library, to build a predictive model.

[0515] The terminal uses AI cameras placed in physical stores to recognize consumers' emotions in real time. TensorFlow is used as the deep learning framework for emotion recognition, processing video data acquired from the cameras. The resulting emotion data is sent from the terminal to a server for further analysis.

[0516] Users view the aggregated and analyzed data from the server on a management dashboard. This dashboard utilizes a web interface built with a front-end framework such as React. On this interface, users can make decisions to optimize inventory and provide personalized recommendations based on consumer sentiment.

[0517] As a concrete example, by predicting demand when a new product is released on an e-commerce site and measuring customers' initial reactions through emotion recognition, appropriate sales promotion becomes possible. Furthermore, instructions can be given to the generative AI model in the form of a prompt such as, "Design a system that performs demand forecasting and personalized promotions based on data from e-commerce sites and physical stores."

[0518] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0519] Step 1:

[0520] The server collects inventory data from distribution channels. It receives inventory information obtained from APIs and databases of each channel as input. Data processing involves unifying data in different formats and storing it in the database. The output is consistent inventory information. In this step, the program sends requests to each channel's API to aggregate inventory quantity and location information.

[0521] Step 2:

[0522] The server performs demand forecasting based on the collected data. Historical sales data and the latest inventory data are used as input. For data calculation, a machine learning algorithm is used to build a model that predicts future demand. The output is a demand forecast value for each product. In this step, a regression model is created using the Python scikit-learn library to predict future sales.

[0523] Step 3:

[0524] The device recognizes customer emotions in a physical store. It uses video data acquired from an AI camera as input. Data processing involves analyzing customer facial expressions in the video and quantifying their emotional state. The output is the analyzed emotional data. This step utilizes TensorFlow and a deep learning model to process image data in real time.

[0525] Step 4:

[0526] The server combines demand forecasts and sentiment data to provide personalized recommendations. It uses demand forecasts and customer sentiment data as input. As a data calculation, it determines promotional content based on this data and generates recommendations best suited to a specific customer group. The output is personalized promotional information. This step considers the customer's purchase history and sentiment data to recommend the most suitable products.

[0527] Step 5:

[0528] Users view data provided by the server on a management dashboard. They receive consistent inventory information and analysis results as input. Output provides insights into inventory optimization strategies and sales promotion strategies. In this step, users visualize various data using graphs and other visualizations through an interface built with the React framework, enabling rapid decision-making.

[0529] (Application Example 2)

[0530] 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."

[0531] Traditionally, it has been difficult to accurately forecast demand and manage inventory across multiple sales channels. Furthermore, it is not easy to individually analyze customer purchasing behavior and provide personalized marketing strategies at the right time, thus failing to improve the customer experience. Moreover, in physical stores, the technology to accurately understand customer emotions and provide services based on them is still immature. Solving these challenges is essential.

[0532] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0533] In this invention, the server includes means for collecting product data from multiple sales channels, means for predicting demand using a learning algorithm, and means for evaluating user emotions using an emotion recognition engine. This makes it possible to analyze customer emotions and purchasing activities while monitoring fluctuations in products in real time, and to individually implement optimized sales strategies and service provision.

[0534] "Sales channels" refer to the routes through which goods and services reach consumers, encompassing a wide range of channels including online and offline retail stores and e-commerce sites.

[0535] "Item data" refers to data concerning inventory status, price, sales status, and related attribute information.

[0536] A "learning algorithm" is a computational method that computers use to find patterns in past data and make new predictions or decisions.

[0537] "To predict" means to estimate future demand and trends based on past data and trends.

[0538] An "emotion recognition engine" is a technology or software that analyzes a user's facial expressions and behavior to identify and evaluate their emotions.

[0539] "Evaluating user emotions" refers to understanding and quantitatively evaluating emotions such as joy, anger, and surprise based on the user's facial expressions, voice, and actions.

[0540] A "sales strategy" is a plan or policy designed to effectively sell products and services based on market analysis and customer behavior.

[0541] The system for implementing this invention mainly consists of a server, a terminal, and a user interface. The server collects goods data from multiple sales channels and uses a learning algorithm to predict demand. This allows the server to optimize the supply of goods in each sales channel.

[0542] Furthermore, the server utilizes an emotion recognition engine to analyze the user's facial expressions and voice to evaluate their emotions. This process employs machine learning software such as TensorFlow and Keras to achieve real-time, highly accurate emotion analysis. Based on this data, the server provides personalized promotional and discount information to each user.

[0543] The terminals are installed in physical environments such as retail stores and continuously monitor changes in goods using object recognition technology. The data acquired from the terminals is also sent to a server, and the supply status of goods is reflected immediately.

[0544] Users can adjust their sales strategies and inventory management based on feedback from the emotion recognition engine, delivered through the user interface. The user interface also functions as a dashboard, visually displaying analysis results and reports.

[0545] As a concrete example, emotion recognition technology is implemented in a physical store. If a customer shows a confused expression in front of a product, the system automatically sends a notification to the staff and provides assistance, thereby improving the customer experience.

[0546] An example of a prompt for a generative AI model could be, "How can we detect customer smiles and recommend new products based on those results?"

[0547] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0548] Step 1:

[0549] The server collects item data from multiple sales channels. It takes online and offline retail databases as input and analyzes this data to determine real-time inventory levels. The output is a list of inventory status for each sales channel. The server automatically collects data using an API and stores it in a unified database.

[0550] Step 2:

[0551] The server predicts demand based on data collected using a learning algorithm. The input is inventory data collected in step 1, and the output is the demand forecast result for each item. This process uses a prediction model based on TensorFlow to calculate future demand, taking seasonality and trends into account.

[0552] Step 3:

[0553] The server evaluates the user's emotions using an emotion recognition engine. The input is video data from cameras installed in the store. The output is the analyzed emotion data. The video data is broken down frame by frame, and for each frame, the emotion recognition algorithm extracts facial features and determines the type of emotion.

[0554] Step 4:

[0555] The terminal uses object recognition technology to detect the status of items in a physical store. Its input is camera footage from inside the store, and its output is data on the presence and placement of items. Object recognition software identifies items from the video and sends this data to a server for inventory management.

[0556] Step 5:

[0557] The server generates and provides users with appropriate promotional and discount information based on sentiment and purchase data. Inputs are demand forecasts from step 2 and sentiment data from step 3. Output is customized promotional content. A generative AI model automatically generates promotional strategies based on forecasts and sentiment data, and notifies users via a dashboard.

[0558] Step 6:

[0559] Users receive feedback from the server and adjust sales strategies and inventory management through a dashboard. Inputs are promotional information and inventory data obtained in step 5. Outputs are improved sales strategy proposals. Users interact with the interface to view results in real time and update strategies as needed.

[0560] 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.

[0561] 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.

[0562] 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.

[0563] [Fourth Embodiment]

[0564] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0565] 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.

[0566] 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).

[0567] 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.

[0568] 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.

[0569] 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).

[0570] 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.

[0571] 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.

[0572] 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.

[0573] 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.

[0574] 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.

[0575] 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.

[0576] 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".

[0577] This invention provides a system that collects inventory data from multiple sales channels and predicts demand using machine learning algorithms. To achieve this, the server, terminal, and user components cooperate to perform the following operations.

[0578] The server automatically collects inventory data from e-commerce sites and physical stores daily. To achieve this, it accesses databases for each channel to retrieve current inventory levels and recent sales figures. The collected data is integrated and stored in a database on the server. The server then uses machine learning algorithms to predict future customer demand, taking into account historical sales data and external factors (such as weather and event information). Based on this prediction, an optimization algorithm calculates how much inventory to allocate to each channel.

[0579] The terminal refers to an AI camera installed in a physical store that monitors shelf inventory in real time. The camera uses object recognition technology to instantly identify the type and quantity of products on the shelf and sends this information to a server, ensuring that inventory data is always up-to-date. For example, if an AI camera installed on a shelf scans the inventory of T-shirts and detects that the quantity has fallen below a threshold, that data is immediately sent to the server, triggering a command to replenish the inventory.

[0580] Users can view real-time inventory status and demand forecast data from the management dashboard. This dashboard is designed for intuitive operation and supports strategic decision-making regarding inventory placement and allocation. Users can also review inventory allocations and promotional plans suggested by the system and make manual adjustments as needed. For example, if a surge in demand is predicted at a particular store, users can prioritize replenishing inventory at that store.

[0581] This invention enables efficient inventory management and reduces problems of supply shortages and oversupply. As a result, lost sales opportunities are minimized, customer satisfaction improves, and profits can be optimized.

[0582] The following describes the processing flow.

[0583] Step 1:

[0584] The server collects inventory data from e-commerce sites and physical stores. After obtaining current inventory levels and recent sales history through APIs or database connections for each channel, it stores this information in a unified database.

[0585] Step 2:

[0586] The server uses collected inventory data and historical sales history to train a machine learning model. This model is designed to predict future demand, and the trained model generates sales forecasts.

[0587] Step 3:

[0588] The server optimizes inventory allocation based on predicted demand. Using an optimization algorithm (e.g., linear programming), it calculates the appropriate inventory allocation for each sales channel and sends the result as a command to each channel.

[0589] Step 4:

[0590] The terminal (an AI camera installed in the store) uses image recognition technology to monitor the inventory status of the shelves in real time. The camera identifies the type and number of products on the shelf and sends that numerical data to the server.

[0591] Step 5:

[0592] The server updates inventory information based on data sent from the terminals, adjusting it to prevent stockouts and excess inventory. It also issues automatic replenishment instructions as needed.

[0593] Step 6:

[0594] Users can view inventory data and forecasts in real time through the management dashboard. Based on the system's suggestions, users can consider inventory allocation and promotional strategies and make adjustments as needed.

[0595] (Example 1)

[0596] 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".

[0597] Traditional inventory management systems struggled to comprehensively track inventory trends across multiple sales channels in real time and accurately forecast demand. Furthermore, they lacked mechanisms to quickly reflect inventory surpluses and shortages, leading to lost sales opportunities and increased costs due to excess inventory. Additionally, it was difficult to implement strategic inventory allocation and promotions that effectively leveraged customer purchasing behavior.

[0598] 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.

[0599] In this invention, the server includes means for acquiring inventory information from multiple sales channels, means for estimating demand using machine learning techniques, and means for optimally allocating inventory among sales channels based on the estimated demand. This enables real-time optimization of inventory, formulation of sales strategies based on demand, and improvement of customer satisfaction.

[0600] "Multiple sales channels" refers to multiple distribution routes for providing goods or services, and these include different environments such as e-commerce sites and physical stores.

[0601] "Inventory information" refers to data regarding the types and quantities of products held by each sales channel.

[0602] "To obtain" means to retrieve necessary information from a database or other source of information.

[0603] "Machine learning techniques" are statistical and algorithmic methods that allow computers to learn patterns from data and use that knowledge to predict future actions and outcomes.

[0604] "Estimating demand" means predicting future demand for a product by taking into account past data and external factors.

[0605] "Optimal allocation" means efficiently distributing limited resources to achieve the objective to the greatest extent possible.

[0606] "Real-time" refers to a time frame in which processing or responses occur immediately without delay.

[0607] "Object recognition technology" is a technology that recognizes specific objects in images and videos and determines their type and attributes.

[0608] A "prompt" is a phrase or message that serves as a starting point or trigger for a system or user to input instructions or information.

[0609] This invention is a system that accurately forecasts demand and efficiently manages inventory based on inventory information obtained from multiple sales channels. The system consists of three main components: a server, terminals, and users.

[0610] The server first automatically collects inventory information from each sales channel. Specifically, it accesses the database of each channel and uses APIs to retrieve current inventory levels and recent sales figures. This data is integrated into a central database on the server. Next, the server uses machine learning techniques to estimate future demand based on the collected data. This is done using standard libraries in Python and R. Based on the estimation results, it optimizes inventory allocation using libraries such as Scikit-learn and TensorFlow.

[0611] The terminal refers to an AI camera installed in a physical store that uses object recognition technology to determine the inventory status within the store in real time. For example, the AI ​​camera analyzes image data of products in real time and sends its type and quantity to a server. This information is immediately reflected in the server's database, contributing to efficient inventory management.

[0612] Users can access real-time inventory data and forecast information through the management screen. This management screen features data visualization capabilities and enables intuitive user operation. Based on the information displayed on the screen, users can make strategic inventory allocation decisions. Users can also use prompts when entering instructions into the system, such as text-based instructions like, "Optimize inventory based on the next demand forecast."

[0613] By implementing this invention, inventory management will become more efficient, reducing issues such as supply shortages and excess inventory, and enabling the maximization of sales opportunities and improvement of profits.

[0614] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0615] Step 1:

[0616] The server collects inventory information from multiple sales channels. The server accesses each channel's database via an API to retrieve current inventory levels and recent sales figures. This collected data becomes input, and data processing, specifically integration into the database, takes place. The integrated inventory data in the database becomes the output. Specifically, the server issues SQL queries to the database to extract information.

[0617] Step 2:

[0618] The server performs data preprocessing based on the integrated data. It removes outliers and standardizes data formats from the collected inventory data, and then combines it with external data (weather and event information). This preprocessed data becomes the input, and a new dataset called "clean data" is output. The Python Pandas library is used for data cleansing.

[0619] Step 3:

[0620] The server estimates demand using machine learning techniques. It takes clean data as input and uses models trained with Scikit-learn or TensorFlow to forecast demand. The forecast results are the output. Specifically, it feeds data into a machine learning model and calculates the demand value.

[0621] Step 4:

[0622] The server optimizes inventory allocation based on demand forecasts. It takes the forecast results and current inventory data as input and performs optimization calculations using linear programming and other methods. The output is an inventory allocation plan. The calculation process using the optimization algorithm is the specific operation.

[0623] Step 5:

[0624] The terminal's AI camera monitors the inventory status of shelves in physical stores and transmits the information to a server in real time. Image data of the shelves acquired by the camera serves as input, and inventory information analyzed using object recognition technology is output. This output is sent to the server, and the inventory information is updated in real time.

[0625] Step 6:

[0626] Users view demand forecasts and inventory allocation data through the management screen and enter prompts as needed. Real-time data provided by the server serves as input, and the user's strategic decisions are the output. Specifically, users analyze the data displayed on the screen and enter prompts such as, "Optimize inventory based on the next demand forecast."

[0627] (Application Example 1)

[0628] 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".

[0629] Inventory management in modern commercial facilities requires accommodating diverse sales channels and external factors, making it prone to inventory shortages and surpluses. Furthermore, efficient inventory replenishment and the development of appropriate sales strategies are essential, but effective systems to meet these needs are currently lacking. There is also a need for automated inventory management, including real-time inventory monitoring and demand forecasting using smart devices.

[0630] 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.

[0631] In this invention, the server includes means for collecting product data from multiple commercial facilities, means for predicting demand using machine learning techniques, and means for providing the status of product inventory within commercial facilities via a smartphone application. This enables proper management and efficient replenishment of product inventory, as well as automation of sales promotion activities.

[0632] "Multiple commercial facilities" refers to a collection of multiple retail stores or shops located in different places or locations.

[0633] "Product data" refers to a collection of information related to inventory levels, sales performance, and product characteristics.

[0634] "Machine learning techniques" refer to a group of algorithms that learn patterns from data and use them to make predictions and classifications about the future.

[0635] "Forecasting demand" is the process of predicting future purchasing activity based on past sales data and external factors.

[0636] A "smartphone app" is a computer program that runs on a portable computer device and provides specific functions or services.

[0637] "Providing product inventory status within a commercial facility" means presenting users with information that allows them to understand the inventory levels and placement of products within a store in real time.

[0638] "Object recognition technology" is a technology that uses cameras and sensors to identify specific objects in video footage and analyze their characteristics.

[0639] "Sales promotion activities" refer to marketing activities conducted to increase product awareness and purchasing intent.

[0640] This invention provides a system for efficiently managing inventory and forecasting demand within commercial facilities. Specific embodiments of this system are described below.

[0641] 1. Server Role

[0642] The server is responsible for collecting product data from multiple commercial facilities. The collected data includes inventory levels, sales data, weather information, and event information, which are then integrated and stored in a database. The server uses machine learning libraries such as TensorFlow to analyze the collected data and perform demand forecasting. Demand forecasting predicts sales trends for each commercial facility and serves as the basis for calculating the optimal allocation of products at each facility.

[0643] 2. The role of the terminal

[0644] The terminal receives real-time video data from AI cameras installed within the commercial facility and monitors product inventory levels using object recognition technology. This recognition technology utilizes technologies such as OpenCV and YOLO. The recognition results are immediately sent to the server, and product data is updated. For example, if the terminal detects that the inventory of a particular product has fallen below a threshold, that data is sent to the server, and a restocking order is issued.

[0645] 3. User Roles

[0646] Users can view the inventory status of products in commercial facilities through a smartphone app. The app is developed using React Native and features an intuitive interface. Users can also receive inventory replenishment instructions based on demand forecasts generated by machine learning models, enabling efficient inventory management.

[0647] As a concrete example, a commercial facility is predicted to experience increased demand for ice cream over the weekend. Through this app, facility managers can order additional ice cream in advance, maximizing sales opportunities.

[0648] An example of a prompt to provide to the generating AI model is, "Please propose an inventory optimization strategy that takes into account demand fluctuations within a commercial facility."

[0649] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0650] Step 1:

[0651] The server collects product data from commercial facility databases. The input is the database of each facility, and the output is integrated product data. This process uses SQL queries to retrieve product inventory levels and sales history, and then integrates and stores this data in the server's database.

[0652] Step 2:

[0653] The terminal monitors product inventory in real time using AI cameras installed in physical stores. The input is images of shelves captured by the cameras, and the output is the type and quantity of recognized products. Object recognition is performed using OpenCV and YOLO, and the results are sent to the server. If the number of recognized products falls below a threshold, data on products that need to be restocked is also output.

[0654] Step 3:

[0655] The server combines collected product data with external weather and event information and uses TensorFlow to forecast demand. The input is integrated product data and external factors, and the output is forecast data of future demand for products at each commercial facility. This process calculates demand using a model trained on historical data.

[0656] Step 4:

[0657] Users can view inventory status and demand forecasts for each facility via a smartphone app and confirm replenishment instructions notified by the system. Input is inventory and demand forecast data retrieved from the server, and output is the user's decision based on that data. Users can intuitively operate the system through an interface built with React Native and issue necessary replenishment instructions.

[0658] Step 5:

[0659] The server calculates the optimal distribution of goods between commercial facilities and assists in optimizing inventory replenishment. Inputs are demand forecast data and inventory status data, and output is the optimal replenishment quantity for each product. An optimization algorithm is used to generate product distribution instructions for each facility.

[0660] 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.

[0661] This invention is a system that combines inventory data collection from multiple sales channels, demand forecasting, inventory allocation optimization, real-time inventory fluctuation monitoring, customer purchasing behavior analysis, and marketing strategy formulation with an emotion engine that recognizes user emotions. This enables the provision of a more personalized customer experience.

[0662] The server collects inventory data from e-commerce sites and physical stores and uses machine learning algorithms to forecast demand. Furthermore, it analyzes customer purchasing behavior to develop marketing strategies and uses an emotion engine to recognize users' emotional states. Based on this information, it can personalize promotions and advertisements and deliver suggestions to customers at the optimal time. For example, the server can analyze a user's emotions from their facial expressions and voice while they are online shopping, detecting signs of interest or dissatisfaction. It can then provide appropriate product suggestions and discount information in real time.

[0663] The terminal monitors inventory levels in physical stores in conjunction with AI cameras, while also observing the emotional state of customers visiting the store. An emotion engine analyzes customers' facial expressions and actions in real time and transmits the resulting emotional data to a server. Simultaneously, data on inventory fluctuations is updated, and inventory is replenished or its placement optimized as needed.

[0664] Users can view reports analyzed by the emotion engine on a management dashboard, gaining insights to fine-tune their sales and inventory management strategies. For example, if customer response to a particular product is positive, they can decide to prioritize replenishing that product's inventory. Furthermore, they can leverage the emotion engine's information to design customized campaigns and conduct sales promotion activities tailored to customer preferences.

[0665] Through this system, users can implement more effective inventory management and marketing strategies while personalizing the customer experience. This is expected to improve customer satisfaction and increase sales.

[0666] The following describes the processing flow.

[0667] Step 1:

[0668] The server collects inventory and sales data from e-commerce sites and physical stores. This includes using APIs to access databases for each channel to retrieve the latest inventory information and store it in a unified database.

[0669] Step 2:

[0670] The server uses the collected data to run machine learning algorithms. This allows for demand forecasting that takes into account past sales trends and external factors, and then calculates the optimal inventory allocation for each channel based on these forecasts.

[0671] Step 3:

[0672] The server utilizes an emotion engine to analyze customer interactions on e-commerce sites and in physical stores. Specifically, it uses emotion analysis tools to evaluate customer facial expressions and voices, generates emotion data, and combines this with purchasing behavior data to extract insights.

[0673] Step 4:

[0674] The terminal uses AI cameras installed in physical stores to monitor in-store inventory and customers' facial expressions. Using object recognition technology, it analyzes the number of items on shelves and the emotional state of customers in real time and transmits that information to a server.

[0675] Step 5:

[0676] The server adjusts inventory management and optimizes sales strategies based on inventory and sentiment data transmitted from terminals. If inventory replenishment or reassignment is necessary, it automatically notifies the relevant departments.

[0677] Step 6:

[0678] Users can use the management dashboard to review inventory allocations and customer sentiment-based promotions suggested by the system. Furthermore, they can manually fine-tune and develop specific marketing campaigns based on the insights displayed on the dashboard.

[0679] Step 7:

[0680] The server deploys the finalized promotional campaign to the e-commerce site and store management systems, and executes necessary customer notifications. These notifications include personalized messages to enhance the customer experience.

[0681] (Example 2)

[0682] 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".

[0683] In today's commercial environment, with the diversification of distribution channels, proper inventory management and efficient marketing strategies based on customer behavior are required. Providing personalized purchasing experiences that consider consumer emotions is also a crucial element. However, collecting and analyzing the necessary data in real time to meet these requirements is not easy, and systems to improve efficiency are needed.

[0684] 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.

[0685] In this invention, the server includes means for collecting inventory information from multiple distribution channels, means for predicting demand using computational processing, and means for analyzing consumer emotions using emotion recognition functionality. This enables inventory optimization and personalized recommendations based on consumer emotions.

[0686] "Distribution channels" refer to the routes through which a product or service is distributed from the manufacturer to the consumer.

[0687] "Inventory information" refers to data regarding the quantity and location of products available for sale at a specific point in time.

[0688] "Computational processing" refers to the procedure of using computers to analyze data and generate information that is useful for future predictions and problem solving.

[0689] "Predicting demand" refers to estimating the future purchase volume of a product or service based on past data.

[0690] "Emotion recognition function" refers to technology that analyzes audio and video data to infer a person's emotional state.

[0691] "Analyzing consumer emotions" refers to the process of identifying a customer's emotional state from their facial expressions and voice.

[0692] "Inventory optimization" refers to adjusting supply to match demand and managing inventory efficiently at the lowest possible cost.

[0693] "Personalized recommendations" refer to recommendations for products and services that are customized based on the specific needs and preferences of consumers.

[0694] The embodiments for carrying out this invention are described below.

[0695] The server collects inventory information from multiple distribution channels by aggregating data from each channel using programming languages ​​and system APIs. It retrieves information in real time from e-commerce sites and physical store sales management systems via HTTP protocols and database queries. The server also analyzes historical sales data and uses computational processing to predict demand. This process uses scikit-learn, a machine learning library, to build a predictive model.

[0696] The terminal uses AI cameras placed in physical stores to recognize consumers' emotions in real time. TensorFlow is used as the deep learning framework for emotion recognition, processing video data acquired from the cameras. The resulting emotion data is sent from the terminal to a server for further analysis.

[0697] Users view the aggregated and analyzed data from the server on a management dashboard. This dashboard utilizes a web interface built with a front-end framework such as React. On this interface, users can make decisions to optimize inventory and provide personalized recommendations based on consumer sentiment.

[0698] As a concrete example, by predicting demand when a new product is released on an e-commerce site and measuring customers' initial reactions through emotion recognition, appropriate sales promotion becomes possible. Furthermore, instructions can be given to the generative AI model in the form of a prompt such as, "Design a system that performs demand forecasting and personalized promotions based on data from e-commerce sites and physical stores."

[0699] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0700] Step 1:

[0701] The server collects inventory data from distribution channels. It receives inventory information obtained from APIs and databases of each channel as input. Data processing involves unifying data in different formats and storing it in the database. The output is consistent inventory information. In this step, the program sends requests to each channel's API to aggregate inventory quantity and location information.

[0702] Step 2:

[0703] The server performs demand forecasting based on the collected data. Historical sales data and the latest inventory data are used as input. For data calculation, a machine learning algorithm is used to build a model that predicts future demand. The output is a demand forecast value for each product. In this step, a regression model is created using the Python scikit-learn library to predict future sales.

[0704] Step 3:

[0705] The device recognizes customer emotions in a physical store. It uses video data acquired from an AI camera as input. Data processing involves analyzing customer facial expressions in the video and quantifying their emotional state. The output is the analyzed emotional data. This step utilizes TensorFlow and a deep learning model to process image data in real time.

[0706] Step 4:

[0707] The server combines demand forecasts and sentiment data to provide personalized recommendations. It uses demand forecasts and customer sentiment data as input. As a data calculation, it determines promotional content based on this data and generates recommendations best suited to a specific customer group. The output is personalized promotional information. This step considers the customer's purchase history and sentiment data to recommend the most suitable products.

[0708] Step 5:

[0709] Users view data provided by the server on a management dashboard. They receive consistent inventory information and analysis results as input. Output provides insights into inventory optimization strategies and sales promotion strategies. In this step, users visualize various data using graphs and other visualizations through an interface built with the React framework, enabling rapid decision-making.

[0710] (Application Example 2)

[0711] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0712] Traditionally, it has been difficult to accurately forecast demand and manage inventory across multiple sales channels. Furthermore, it is not easy to individually analyze customer purchasing behavior and provide personalized marketing strategies at the right time, thus failing to improve the customer experience. Moreover, in physical stores, the technology to accurately understand customer emotions and provide services based on them is still immature. Solving these challenges is essential.

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

[0714] In this invention, the server includes means for collecting product data from multiple sales channels, means for predicting demand using a learning algorithm, and means for evaluating user emotions using an emotion recognition engine. This makes it possible to analyze customer emotions and purchasing activities while monitoring fluctuations in products in real time, and to individually implement optimized sales strategies and service provision.

[0715] "Sales channels" refer to the routes through which goods and services reach consumers, encompassing a wide range of channels including online and offline retail stores and e-commerce sites.

[0716] "Item data" refers to data concerning inventory status, price, sales status, and related attribute information.

[0717] A "learning algorithm" is a computational method that computers use to find patterns in past data and make new predictions or decisions.

[0718] "To predict" means to estimate future demand and trends based on past data and trends.

[0719] An "emotion recognition engine" is a technology or software that analyzes a user's facial expressions and behavior to identify and evaluate their emotions.

[0720] "Evaluating user emotions" refers to understanding and quantitatively evaluating emotions such as joy, anger, and surprise based on the user's facial expressions, voice, and actions.

[0721] A "sales strategy" is a plan or policy designed to effectively sell products and services based on market analysis and customer behavior.

[0722] The system for implementing this invention mainly consists of a server, a terminal, and a user interface. The server collects goods data from multiple sales channels and uses a learning algorithm to predict demand. This allows the server to optimize the supply of goods in each sales channel.

[0723] Furthermore, the server utilizes an emotion recognition engine to analyze the user's facial expressions and voice to evaluate their emotions. This process employs machine learning software such as TensorFlow and Keras to achieve real-time, highly accurate emotion analysis. Based on this data, the server provides personalized promotional and discount information to each user.

[0724] The terminals are installed in physical environments such as retail stores and continuously monitor changes in goods using object recognition technology. The data acquired from the terminals is also sent to a server, and the supply status of goods is reflected immediately.

[0725] Users can adjust their sales strategies and inventory management based on feedback from the emotion recognition engine, delivered through the user interface. The user interface also functions as a dashboard, visually displaying analysis results and reports.

[0726] As a concrete example, emotion recognition technology is implemented in a physical store. If a customer shows a confused expression in front of a product, the system automatically sends a notification to the staff and provides assistance, thereby improving the customer experience.

[0727] An example of a prompt for a generative AI model could be, "How can we detect customer smiles and recommend new products based on those results?"

[0728] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0729] Step 1:

[0730] The server collects item data from multiple sales channels. It takes online and offline retail databases as input and analyzes this data to determine real-time inventory levels. The output is a list of inventory status for each sales channel. The server automatically collects data using an API and stores it in a unified database.

[0731] Step 2:

[0732] The server predicts demand based on data collected using a learning algorithm. The input is inventory data collected in step 1, and the output is the demand forecast result for each item. This process uses a prediction model based on TensorFlow to calculate future demand, taking seasonality and trends into account.

[0733] Step 3:

[0734] The server evaluates the user's emotions using an emotion recognition engine. The input is video data from cameras installed in the store. The output is the analyzed emotion data. The video data is broken down frame by frame, and for each frame, the emotion recognition algorithm extracts facial features and determines the type of emotion.

[0735] Step 4:

[0736] The terminal uses object recognition technology to detect the status of items in a physical store. Its input is camera footage from inside the store, and its output is data on the presence and placement of items. Object recognition software identifies items from the video and sends this data to a server for inventory management.

[0737] Step 5:

[0738] The server generates and provides users with appropriate promotional and discount information based on sentiment and purchase data. Inputs are demand forecasts from step 2 and sentiment data from step 3. Output is customized promotional content. A generative AI model automatically generates promotional strategies based on forecasts and sentiment data, and notifies users via a dashboard.

[0739] Step 6:

[0740] Users receive feedback from the server and adjust sales strategies and inventory management through a dashboard. Inputs are promotional information and inventory data obtained in step 5. Outputs are improved sales strategy proposals. Users interact with the interface to view results in real time and update strategies as needed.

[0741] 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.

[0742] 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.

[0743] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0744] 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.

[0745] 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.

[0746] 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.

[0747] 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.

[0748] 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.

[0749] 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."

[0750] 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.

[0751] 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.

[0752] 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.

[0753] 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.

[0754] 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.

[0755] 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.

[0756] 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.

[0757] 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.

[0758] 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.

[0759] 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.

[0760] 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.

[0761] 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.

[0762] The following is further disclosed regarding the embodiments described above.

[0763] (Claim 1)

[0764] A means of collecting inventory data from multiple sales channels,

[0765] A method for predicting demand using machine learning algorithms,

[0766] A means for optimally allocating inventory across multiple sales channels based on predicted demand,

[0767] A means of monitoring inventory fluctuations in real time and updating data,

[0768] A system that includes means for analyzing customer purchasing behavior and formulating marketing strategies.

[0769] (Claim 2)

[0770] The system according to claim 1, comprising means for detecting the inventory status of a physical store using object recognition technology.

[0771] (Claim 3)

[0772] The system according to claim 1, comprising means for automatically creating a promotional campaign based on analysis and notifying customers.

[0773] "Example 1"

[0774] (Claim 1)

[0775] A means of obtaining inventory information from multiple sales channels,

[0776] A method for estimating demand using machine learning techniques,

[0777] A means for optimally allocating inventory across sales channels based on estimated demand,

[0778] A means of monitoring inventory changes in real time and updating information,

[0779] A system that includes means to enable data visualization and manipulation through an administration screen.

[0780] (Claim 2)

[0781] The system according to claim 1, which uses object recognition technology to identify the inventory status of a physical store.

[0782] (Claim 3)

[0783] The system according to claim 1, which uses prompts based on the generated demand forecast to adjust inventory allocation.

[0784] "Application Example 1"

[0785] (Claim 1)

[0786] A means of collecting product data from multiple commercial facilities,

[0787] A method for predicting demand using machine learning techniques,

[0788] A means for optimally allocating goods among multiple commercial facilities based on predicted demand,

[0789] A means of monitoring and updating information on changes in product inventory in real time,

[0790] A means of analyzing user purchasing behavior and formulating sales strategies,

[0791] A means of providing information on product inventory status within commercial facilities via a smartphone app,

[0792] A system that includes means for instructing commercial facilities to replenish their stock based on demand forecasts that take weather information and event information into account.

[0793] (Claim 2)

[0794] The system according to claim 1, which uses object recognition technology to detect the status of goods in a commercial facility.

[0795] (Claim 3)

[0796] The system according to claim 1, which automatically creates sales promotion activities based on analysis and notifies the user.

[0797] "Example 2 of combining an emotion engine"

[0798] (Claim 1)

[0799] A means of collecting inventory information from multiple distribution channels,

[0800] A means of predicting demand using computational processing,

[0801] A means for optimally allocating inventory across multiple distribution channels based on predicted demand,

[0802] A means of monitoring inventory fluctuations in real time and updating information,

[0803] A means of analyzing consumer purchasing behavior and formulating sales promotion strategies,

[0804] A system that includes means for analyzing consumer emotions using emotion recognition functions.

[0805] (Claim 2)

[0806] The system according to claim 1, which uses image analysis technology to detect the inventory status of a physical store.

[0807] (Claim 3)

[0808] The system according to claim 1, which automatically creates and presents sales promotion activities to consumers based on analysis.

[0809] "Application example 2 when combining with an emotional engine"

[0810] (Claim 1)

[0811] A means of collecting product data from multiple sales channels,

[0812] A means of predicting demand using a learning algorithm,

[0813] A means for optimally allocating goods across multiple sales channels based on predicted demand,

[0814] A means of instantly monitoring changes in goods and updating information,

[0815] A means of analyzing customer purchasing behavior and formulating sales strategies,

[0816] A means of evaluating a user's emotions using an emotion recognition engine,

[0817] A means of providing appropriate product suggestions and discount information based on evaluated emotions,

[0818] A system that includes this.

[0819] (Claim 2)

[0820] The system according to claim 1, which uses object recognition technology to detect the status of items in a physical store.

[0821] (Claim 3)

[0822] The system according to claim 1, which automatically creates and notifies customers of sales promotion activities based on evaluated emotions. [Explanation of Symbols]

[0823] 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 product data from multiple commercial facilities, A method for predicting demand using machine learning techniques, A means for optimally allocating goods among multiple commercial facilities based on predicted demand, A means of monitoring and updating information on changes in product inventory in real time, A means of analyzing user purchasing behavior and formulating sales strategies, A means of providing information on product inventory status within commercial facilities via a smartphone app, A system that includes means for instructing commercial facilities to replenish their stock based on demand forecasts that take weather information and event information into account.

2. The system according to claim 1, which uses object recognition technology to detect the status of goods in a commercial facility.

3. The system according to claim 1, which automatically creates sales promotion activities based on analysis and notifies the user.

Images