System

Abstract

A system is provided.SOLUTION: A system comprising: means for generating a demand prediction model based on a purchase history, a browsing history, and a search history of a user; means for optimizing an inventory using the generated demand prediction model; means for generating a product recommendation based on the browsing history and the search history of the user; means for notifying a terminal of the user of generated recommendation information; and means for receiving a purchase request from the user, selecting an optimal distribution route, and giving a delivery instruction.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] In modern e-commerce, users expect to receive products quickly. However, logistics lead times and the decision-making process leading to a purchase are often lengthy, which can detract from the user experience. Furthermore, inefficient inventory management often leads to unnecessary shortages and excesses. This increases costs for companies and reduces user satisfaction. The objective of this invention is to significantly shorten these lead times, optimize inventory management, and improve the user experience. [Means for solving the problem]

[0005] The present invention solves the above problems by using the following means.

[0006] 1. A means of generating a demand forecasting model based on a user's purchasing history, browsing history, and search history.

[0007] 2. A means of optimizing inventory using the generated demand forecast model.

[0008] 3. A means of generating product recommendations based on a user's browsing and search history.

[0009] 4. A means of notifying the generated recommendation information to the user's device.

[0010] 5. A means of receiving purchase requests from users, selecting the optimal logistics route, and issuing delivery instructions.

[0011] As a result, the present invention significantly shortens logistics lead times and user decision-making lead times, enabling efficient inventory management and improving the user experience. Furthermore, by periodically updating the generated demand forecast model and performing dynamic inventory management, even more accurate supply and demand adjustments can be achieved. Furthermore, by collecting user action logs and utilizing them in generating recommendations from the next time onwards, the accuracy of recommendations can be improved.

[0012] A "demand forecasting model" is a statistical algorithm or machine learning model used to predict future demand based on a user's past purchasing history, browsing history, and search history.

[0013] "Inventory optimization" is a method of adjusting inventory levels based on a demand forecasting model so that products are stored in the right amount.

[0014] "Recommendation information" is information that suggests optimal products and services to a specific user based on the user's browsing history and search history.

[0015] A "terminal" is an electronic device used by a user, such as a computer or smartphone.

[0016] A "logistics route" is the delivery route that a product takes from the warehouse to the user.

[0017] "Delivery instructions" refers to the server selecting the optimal logistics route and issuing instructions on the procedures for actually delivering the product to the user.

[0018] An "action log" is a record of behavioral data when a user interacts, and is used to generate recommendations from the next time onwards. [Brief explanation of the drawings]

[0019] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0020] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0021] First, the terms used in the following description will be explained.

[0022] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0023] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0024] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0025] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0026] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0027] [First embodiment]

[0028] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0029] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0030] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0031] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0032] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0033] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0034] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0035] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0036] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0037] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0038] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0039] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0040] The present invention is a system that performs demand forecasting based on a user's purchase history, browsing history, and search history, and provides inventory optimization and product recommendations. Below, we will create a program for this system and explain the specific processing content.

[0041] Demand forecast data collection and analysis

[0042] The server collects data such as past purchase history, browsing history, and search history from the user database. This includes detailed information such as purchase date and time, purchased product category, purchased quantity, and user ID. The server preprocesses the acquired data and generates a demand forecasting model using machine learning algorithms (e.g., regression analysis and time series analysis). Future demand is estimated based on the generated forecasting model.

[0043] Examples:

[0044] The server analyzes "User A's" purchasing data for the past six months and predicts that "User A" is likely to purchase a particular electronic device again in the next month.

[0045] Inventory management

[0046] The server uses the generated demand forecast model to calculate the necessary inventory levels and allocate optimal stock to each warehouse. It periodically updates the inventory status and dynamically manages it to prevent shortages and excesses.

[0047] Examples:

[0048] The server predicts that demand for electronic devices will increase within the month and instructs the Tokyo warehouse to prepare additional inventory to accommodate this.

[0049] Recommendation function

[0050] The server collects users' browsing and search histories in real time, and uses a recommendation engine to select the most suitable products based on that information.The server then notifies the user of the selected recommendation information on their device.

[0051] Examples:

[0052] If "User B" searches for "smartphone cases" on a website, the server will recommend new products in the same case category.

[0053] User Notifications and Interactions

[0054] The device notifies the user of the recommendation information received from the server in real time. The user then moves to the page with detailed information and, if interested, proceeds to the purchase process.

[0055] Examples:

[0056] User C's device displays a notification for "Recommended Product: Waterproof Smartphone Case," and User C clicks on the notification to go to the details page.

[0057] Checkout and logistics

[0058] When the user presses the purchase button, the device sends the purchase information to the server. The server then checks the received purchase information, selects the optimal logistics route, and issues delivery instructions. The server also monitors the delivery status in real time and updates the delivery status on the user's device.

[0059] Examples:

[0060] When "User D" presses the purchase button, the server retrieves the product from the nearest warehouse, selects the fastest delivery method, and begins delivery to the user's address. During delivery, the server monitors the delivery status and notifies User D's device of the status as it progresses.

[0061] By executing such a program, the present invention aims to minimize logistics lead time and user decision-making lead time, further optimizing inventory and improving the user experience. Furthermore, by periodically updating the generated demand forecast model and performing dynamic inventory management, more accurate supply and demand adjustments become possible. Furthermore, by collecting user action logs and using them to generate recommendations for future purchases, more personalized services can be provided.

[0062] The processing flow will be explained below.

[0063] Step 1:

[0064] The server collects data such as past purchase history, browsing history, and search history from the user database. For example, it comprehensively obtains each user's purchase date and time, purchased product category, purchased quantity, user ID, etc.

[0065] Step 2:

[0066] The server preprocesses the acquired data, specifically by cleaning the data, filling in missing data, and normalizing the data, thereby converting the data into a state that can be analyzed.

[0067] Step 3:

[0068] The server then inputs the preprocessed data into a machine learning algorithm to generate a demand forecasting model, for example, using regression analysis or time series analysis to build a model that predicts future demand.

[0069] Step 4:

[0070] The server uses the generated demand forecasting model to predict future demand for specific products, calculates the necessary inventory levels based on the forecast, and allocates optimal inventory to each warehouse.

[0071] Step 5:

[0072] The server collects users' browsing and search history in real time, for example, capturing their actions when they search for a specific product on a website or view a detail page.

[0073] Step 6:

[0074] The server uses the collected real-time data to run a recommendation engine, for example, using collaborative filtering or content-based filtering to select the best products for the user.

[0075] Step 7:

[0076] The server sends the generated recommendation information to the user's device, which receives the information and notifies the user in real time.

[0077] Step 8:

[0078] The user checks the recommendation notification displayed on the device and clicks on the detailed information page if they are interested. For example, they receive a notification about a new product that is perfect for them: a waterproof smartphone case.

[0079] Step 9:

[0080] When the user presses the purchase button, the terminal sends the purchase information to the server, including the product ID, user ID, and delivery address information.

[0081] Step 10:

[0082] The server confirms the received purchasing information and checks the inventory status. If the inventory is available, it confirms the order and selects the optimal logistics route.

[0083] Step 11:

[0084] The server issues delivery instructions to the warehouse, for example, shipping the product from the nearest warehouse and selecting the fastest delivery method.

[0085] Step 12:

[0086] The server monitors the delivery status in real time and updates the status to the user's device, such as when the product has been shipped, when it is in transit, or when delivery has been completed.

[0087] Example 1

[0088] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0089] Conventional systems do not adequately forecast demand or optimize inventory based on users' purchasing, browsing, and search histories, making it difficult to detect inventory surpluses and shortages and to recommend appropriate products to users.Furthermore, there is a problem of reduced user convenience due to the lack of real-time monitoring of delivery status and the lack of proper notification of delivery status.

[0090] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0091] In this invention, the server includes means for generating a demand forecasting model based on a user's purchase history, browsing history, and search history, means for optimizing inventory using the generated demand forecasting model, means for generating product recommendations based on the user's browsing history and search history, means for notifying the user's terminal of the generated recommendation information, means for receiving purchase requests from the user, selecting an optimal logistics route, and issuing delivery instructions, and means for monitoring delivery status in real time and notifying the user's terminal of the delivery status. This prevents inventory surpluses and shortages, recommends optimal products to the user, and enables efficient delivery.

[0092] "User" refers to a person who uses the system and provides data such as purchase history, browsing history, and search history.

[0093] "Purchase history" includes information about products purchased by a user in the past, and refers to detailed information such as the purchase date and time, purchased product category, purchased quantity, and user ID.

[0094] "Browser history" refers to data that includes information about products viewed by a user on websites or within applications.

[0095] "Search history" refers to data that includes keywords that users searched for within the system and information about search results.

[0096] "Demand forecasting model" refers to a machine learning algorithm used to predict future demand for a product based on data such as purchase history, browsing history, and search history.

[0097] "Inventory optimization" refers to the use of demand forecasting models to allocate optimal amounts of inventory in each warehouse and prevent shortages and excess inventory.

[0098] "Product recommendation" refers to a method of selecting and suggesting optimal products to users based on their browsing history and search history.

[0099] "Notification" refers to the act of sending information such as recommendation information and delivery status from the server to the user's terminal.

[0100] The "logistics route" refers to the optimal logistics route for efficiently delivering the products purchased by the user.

[0101] "Delivery status" refers to the current status of the product purchased by the user in the delivery process, and includes states such as shipped, in delivery, and delivery completed.

[0102] The present invention provides a system for performing demand forecasting based on a user's purchase history, browsing history, and search history, optimizing inventory, and providing product recommendations. Specific embodiments of this system are described below.

[0103] First, the server collects data on past purchase history, browsing history, and search history from the user database. This data includes detailed information such as purchase date and time, purchased product category, purchased quantity, and user ID. The server then preprocesses the acquired data, for example, removing duplicates and filling in missing values. After preprocessing is complete, the server uses machine learning algorithms (e.g., regression analysis and time series analysis) to generate a demand forecasting model. This demand forecasting model is saved and used to forecast future demand.

[0104] As a specific example, by analyzing the purchasing data of "User A" for the past six months, it is possible to predict the likelihood that "User A" will purchase a particular electronic device again in the next month.

[0105] The server then uses the generated demand forecast model to calculate the required inventory and allocate optimal inventory to each warehouse. It periodically updates inventory status and dynamically manages it to prevent shortages and excesses. For example, the server may predict that demand for electronic devices will increase within the current month for the Tokyo warehouse and instruct it to allocate additional inventory.

[0106] Furthermore, the server collects the user's browsing history and search history in real time and uses a recommendation engine to select the most suitable product. This recommendation information is then sent to the user's device. For example, if "User B" is searching for a smartphone case, the server will recommend a new product.

[0107] When a user receives the recommendation information, they go to the detailed information page and, if they are interested, proceed to the purchase process. For example, "User C"'s device may display a notification for "Recommended Product: Waterproof Smartphone Case," and User C may go to the detailed information page.

[0108] Additionally, when a user presses the purchase button, the terminal sends the purchase information to the server. The server confirms the received purchase information, selects the optimal logistics route, and issues delivery instructions. It monitors the delivery situation in real time and updates the delivery status on the user's terminal. As a specific example, when "User D" presses the purchase button, the server retrieves the product from the nearest warehouse, selects the fastest delivery method, and notifies "User D's" terminal of the delivery status as the process progresses.

[0109] An example of a prompt for a generative AI model is as follows:

[0110] "Based on User A's purchasing data from the past six months, please predict the products that he is likely to purchase in the next month."

[0111] "If User B is searching for a smartphone case, please recommend related products."

[0112] "Please forecast which warehouses will see increased demand for electronic equipment this month and issue instructions for deploying the necessary inventory."

[0113] As described above, the system of the present invention aims to minimize logistics lead time and user decision-making lead time, optimize inventory, and improve the user experience. By periodically updating the generated demand forecast model and performing dynamic inventory management, more accurate supply and demand adjustments become possible. Furthermore, by collecting user action logs and using them to generate recommendations for future purchases, more personalized services can be provided.

[0114] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0115] Step 1:

[0116] The server collects purchase history, browsing history, and search history data from a user database.

[0117] Input: User database

[0118] Output: Purchase history, browsing history, search history

[0119] Specific operation: The server uses an SQL query to retrieve the purchase history, browsing history, and search history of "User A" and "User B" from the user database for the past six months.

[0120] Step 2:

[0121] The server pre-processes the collected data.

[0122] Input: purchase history, browsing history, search history

[0123] Output: Preprocessed data

[0124] Specific operation: The server removes duplicate data, imputes missing values, and converts the data into the required format. For example, if there are missing values, it imputes them with the average value.

[0125] Step 3:

[0126] The server uses the pre-processed data to train a demand forecasting model.

[0127] Input: Preprocessed data

[0128] Output: Demand forecast model

[0129] Specific operation: The server uses Python's scikit-learn to perform regression analysis and time series analysis to train a demand forecasting model, and saves this model for use in future demand forecasts.

[0130] Step 4:

[0131] The server uses the generated demand forecast model to optimize inventory.

[0132] Input: Demand forecast model

[0133] Output: Inventory allocation plan

[0134] Specific operation: The server calculates the amount of inventory required for each warehouse based on the demand forecast model. For example, it instructs the Tokyo warehouse to increase the inventory of electronic devices.

[0135] Step 5:

[0136] The server generates product recommendations based on the user's browsing history and search history.

[0137] Input: Real-time collected browsing history, search history

[0138] Output: Recommendation information

[0139] Specific operation: When a user searches for "smartphone case," the server selects new products in the same category and generates recommendation information.

[0140] Step 6:

[0141] The terminal notifies the user of the recommendation information received from the server.

[0142] Input: Recommendation information

[0143] Output: Notification message

[0144] Specific operation: User C's device displays a notification such as "Recommended product: Waterproof smartphone case," and User C moves to the details page.

[0145] Step 7:

[0146] The terminal transmits a purchase request from the user to the server.

[0147] Input: Purchase Request

[0148] Output: Purchase information

[0149] Specific operation: When the user presses the purchase button, the device sends that information to the server, which then selects the optimal logistics route and issues delivery instructions.

[0150] Step 8:

[0151] The server monitors the delivery status in real time and notifies the user's terminal of the delivery status.

[0152] Input: Delivery status from logistics company

[0153] Output: Delivery status update

[0154] Specific operation: The server uses the logistics company's API to monitor delivery status and notifies the user's device of statuses such as "shipped," "in transit," and "delivery completed."

[0155] (Application example 1)

[0156] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0157] The present invention relates to a system for improving the accuracy of demand forecasting and product recommendations based on users' purchase history, browsing history, and search history, optimizing inventory, and improving the user experience. With conventional technologies, demand forecasting and inventory management are time-consuming and often result in delayed notification of recommendation information. Therefore, a system combining efficient inventory management with rapid notification of recommendations is needed.

[0158] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0159] In this invention, the server includes: means for generating a demand forecasting model based on a user's purchase history, browsing history, and search history; means for optimizing inventory using the generated demand forecasting model; means for generating product recommendations based on the user's browsing history and search history; means for notifying the user's terminal of the generated recommendation information; means for receiving purchase requests from the user, selecting an optimal logistics route, and issuing delivery instructions; and means for providing appropriate product recommendations based on the user's shopping behavior and sending push notifications in real time using the smart device to increase purchasing motivation. This enables both inventory management and recommendation notifications to be performed with high accuracy and efficiency, improving the user experience.

[0160] "User purchase history" is a record of information about products purchased by the user in the past, such as purchase date and time, and purchase quantity.

[0161] A "demand forecasting model" is a mathematical model for predicting future demand based on a user's past purchasing history, browsing history, and search history.

[0162] "Inventory optimization" is the process of adjusting product inventory at each warehouse or store to meet demand using a demand forecasting model.

[0163] "Product recommendation" is a function that suggests products that may be of interest to users based on their browsing history and search history.

[0164] "User terminal" refers to a digital device used by a user, such as a smartphone or tablet.

[0165] A "purchase request" is information sent to the system when a user indicates their intention to purchase a product.

[0166] "Logistics route" is a concept that refers to the optimal route or means for delivering goods.

[0167] "Delivery Instructions" refers to the process of issuing specific instructions regarding the delivery of goods.

[0168] "Shopping behavior" refers to a series of actions taken by a user, such as browsing, searching, and purchasing products.

[0169] "Smart devices" is a general term for devices that can connect to the Internet, such as smartphones, tablets, and smartwatches.

[0170] "Push notification" is a function that sends information from a server to a user's smart device in real time.

[0171] The system for implementing this invention encompasses a series of processes including automated data collection, demand forecasting, inventory management, product recommendations, and user notifications. Specifically, it is realized in the following manner.

[0172] The server first collects users' purchase, browsing, and search histories from a database. This data includes detailed information such as user ID, purchased items, purchase date and time, and purchase quantity. Next, the collected data is preprocessed to generate a demand forecasting model. This model generation uses machine learning algorithms (e.g., linear regression and time series analysis).

[0173] The generated demand forecasting model predicts future demand, and inventory is optimized based on that. The server periodically updates the demand forecasting model to dynamically adjust inventory levels. This prevents overstocking and shortages and enables efficient resource management.

[0174] Furthermore, the server monitors the user's browsing and search history in real time and generates optimal product recommendations. This recommendation information is then pushed to the user via their smart device. For example, if "User A" searches for "smartphone cases" on a website, the server will recommend new products in the same category and notify the user's device.

[0175] When a user submits a purchase request, the server receives the information, selects the optimal logistics route, and issues delivery instructions. Once delivery begins, the server monitors the delivery status in real time and updates the delivery status on the user's device, allowing the user to track the progress of the product as it is delivered.

[0176] The hardware used includes servers, smart devices (smartphones, tablets, etc.), and database systems. The software includes machine learning libraries (e.g., scikit-learn), Python scripts for data collection and preprocessing, and notification services such as Firebase.

[0177] To generate a specific program, you can use the following prompt:

[0178] 1. "Generate Python code to forecast demand for the next month based on purchasing data from the past six months."

[0179] 2. "Create a program using Firebase to send recommendation notifications to users based on their purchase and browsing history."

[0180] This allows the server, terminal, and user to interact with each other, enabling both inventory management and recommendation notifications to be performed with high accuracy and efficiency, improving the user experience.

[0181] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0182] Step 1:

[0183] The server collects the user's purchase history, browsing history, and search history from a database. The input is the user ID, and the output is the user's various historical data (purchased items, purchase date and time, purchased quantity, viewed items, search keywords, etc.). This allows the past behavioral data of a specific user to be obtained.

[0184] Step 2:

[0185] The server pre-processes the collected data, cleaning the data (e.g., removing incomplete data and correcting outliers) and standardizing it. The input is the user history data collected in step 1, and the output is formatted data, which facilitates subsequent data analysis.

[0186] Step 3:

[0187] The server uses the preprocessed data to generate a demand forecasting model. It trains the model using machine learning algorithms (e.g., linear regression or time series analysis). The input is the preprocessed data, and the output is the demand forecasting model. This creates a foundation for forecasting future demand.

[0188] Step 4:

[0189] The server optimizes the planned inventory based on the generated demand forecast model. This involves a process of calculating how much of a specific product is needed at a specific time. The input is the demand forecast model, and the output is the calculation result of the optimal inventory amount. This makes inventory management more efficient.

[0190] Step 5:

[0191] The server monitors the user's real-time browsing and search history and generates product recommendations. The input is real-time user behavior data, and the output is a list of recommended products. This allows the server to recommend the most suitable products to the user.

[0192] Step 6:

[0193] The terminal notifies the user of the recommendation information sent from the server. This notification is performed using the push notification function of the smart device. The input is a list of recommended products, and the output is a notification message. This allows the user to receive new and recommended products in real time.

[0194] Step 7:

[0195] The user submits a purchase request, which is sent to the server with the purchase request data as input, thereby conveying the user's intention to purchase to the system.

[0196] Step 8:

[0197] The server selects the optimal logistics route based on the received purchase request and issues delivery instructions. The input is the purchase request data, and the output is logistics instructions and a delivery route. This ensures that purchased items are delivered to the user in the most efficient manner.

[0198] Step 9:

[0199] The server monitors the delivery status in real time and updates the delivery status to the user's device. The input is delivery progress data and the output is the updated delivery status, allowing the user to always know what stage their product is in.

[0200] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0201] This invention combines a system that performs demand forecasting based on users' purchase history, browsing history, and search history, optimizes inventory, and provides product recommendations with an emotion engine that recognizes users' emotions. Below, we will create a program for this system and explain the specific processing content.

[0202] Demand forecast data collection and analysis

[0203] The server collects past purchase history, browsing history, and search history from the user database. This includes detailed information such as purchase date and time, purchased product category, purchase quantity, and user ID. The server then preprocesses the collected data, cleaning the data, filling in missing values, and normalizing the data. The preprocessed data is then input into a machine learning algorithm (e.g., regression analysis or time series analysis) to generate a demand forecasting model. Future demand is predicted based on the generated forecasting model.

[0204] Examples:

[0205] The server analyzes "User A's" purchasing data for the past six months and predicts that "User A" is likely to purchase a particular electronic device again in the next month.

[0206] Collecting and analyzing emotional data

[0207] The server uses an emotion engine to recognize the user's emotions. It analyzes the user's facial expression data, voice tone, keyboard input, etc. sent from the device to identify the user's emotional state. This allows it to understand how the user is feeling at that time.

[0208] Examples:

[0209] While "User B" is browsing products on his smartphone, emotional data is collected using the camera and microphone. The emotion engine recognizes that User B is in an "excited" state.

[0210] Inventory management

[0211] The server combines the generated demand forecast model with user emotion data to calculate the necessary inventory levels. Based on this information, it allocates optimal inventory to each warehouse and regularly updates the inventory status to prevent shortages and overstocks.

[0212] Examples:

[0213] The server predicts that demand for electronic devices will increase within this month, and since user sentiment data indicates a positive reaction, it instructs the Tokyo warehouse to stock additional stock of those electronic devices.

[0214] Recommendation function

[0215] The server runs a recommendation engine based on the user's browsing history, search history, and emotional data. For example, it uses collaborative filtering or content-based filtering to select the most suitable products for the user. The generated recommendation information is sent to the user's device.

[0216] Examples:

[0217] When "User C" searches for "smartphone cases" on a website, the server analyzes the emotional data of "excitement." Taking that reaction into consideration, the server recommends "waterproof smartphone cases" as a product in the same category.

[0218] User Notifications and Interactions

[0219] The device will notify the user of the recommendation information received from the server in real time. The user can check the notification, and if they are interested, they can go to the detailed information page and proceed with the purchase process.

[0220] Examples:

[0221] User D's device displays a notification for "Recommended Product: Waterproof Smartphone Case," and User D clicks on the notification to go to the details page.

[0222] Checkout and logistics

[0223] When the user presses the purchase button, the terminal sends the purchase information to the server. The sent information includes the product ID, user ID, delivery address information, etc. The server confirms the received purchase information and checks the inventory status. If inventory is available, the order is confirmed and the optimal logistics route is selected. The server issues delivery instructions to the warehouse and updates the status on the user's terminal while monitoring the delivery status in real time.

[0224] Examples:

[0225] When "User E" presses the purchase button, the server retrieves the product from the nearest warehouse, selects the fastest delivery method, and begins delivery. During delivery, the server monitors the delivery status and notifies User E's device of the status as it progresses.

[0226] Through this series of processes, the present invention minimizes logistics lead time and user decision-making lead time, and further optimizes inventory. Furthermore, by periodically updating the generated demand forecast model and performing dynamic inventory management, more accurate supply and demand adjustments become possible. By utilizing user emotion data, more personalized services can be provided, improving the user experience.

[0227] The processing flow will be explained below.

[0228] Step 1:

[0229] The server collects data such as past purchase history, browsing history, and search history from the user database. The data collected includes purchase date and time, purchased product category, purchased quantity, and user ID.

[0230] Step 2:

[0231] The server preprocesses the acquired data, which includes data cleaning (e.g., removing incomplete data), imputing missing values ​​(e.g., imputing with the mean), and normalizing the data (e.g., scaling).

[0232] Step 3:

[0233] The server then inputs the preprocessed data into a machine learning algorithm to generate a demand forecasting model, for example, using regression analysis or time series analysis to build a model for predicting future demand.

[0234] Step 4:

[0235] The server periodically uses the generated demand forecasting model to predict future demand for specific products, calculates the necessary inventory levels based on this, and allocates optimal inventory to each warehouse.

[0236] Step 5:

[0237] The server uses an emotion engine to recognize the user's emotions and collects emotion data from the user's device, specifically analyzing facial expression data from the camera, voice tone from the microphone, and keyboard input speed and strength.

[0238] Step 6:

[0239] The server uses an emotion engine to analyze the collected emotional data and identify the emotional state the user is in at that time, categorizing emotions such as "excitement," "sadness," and "happiness."

[0240] Step 7:

[0241] The server incorporates the collected emotion data into the demand forecasting model to improve the accuracy of the forecast, thereby more accurately predicting product demand, which is strongly influenced by user emotions.

[0242] Step 8:

[0243] The server runs a recommendation engine based on the user's browsing history, search history, and emotional data, using collaborative filtering and content-based filtering, for example, to select the best products for the user.

[0244] Step 9:

[0245] The server transmits the generated recommendation information to the user's terminal, which notifies the user of the recommendation information in real time.

[0246] Step 10:

[0247] The user checks the recommendation notification displayed on the device and clicks on the detailed information page if they are interested. For example, a notification such as "Recommended product: Waterproof smartphone case" will be displayed.

[0248] Step 11:

[0249] When a user presses the purchase button, the device sends the purchase information to the server, including the product ID, user ID, and delivery address information.

[0250] Step 12:

[0251] The server confirms the received purchase information and checks the inventory status. If the item is available, it confirms the order and selects the optimal logistics route.

[0252] Step 13:

[0253] The server issues delivery instructions to the warehouse, for example, shipping the product from the nearest warehouse and selecting the fastest delivery method.

[0254] Step 14:

[0255] The server monitors the delivery status in real time and updates the status to the user's device, for example, notifying the user of the status when the product has been shipped, when it is being delivered, and when delivery has been completed.

[0256] Through these processing steps, the present invention minimizes logistics lead time and user decision-making lead time, optimizing inventory. By utilizing user emotion data, it is possible to provide more personalized services and improve the user experience.

[0257] Example 2

[0258] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0259] Many modern inventory management systems and product recommendation systems support demand forecasting and recommendations based on users' purchase, browsing, and search histories. However, they are unable to consider the user's emotional state, limiting their ability to optimize the user experience. Furthermore, inventory optimization is static, making it difficult to respond to dynamic demand fluctuations, leading to problems such as inventory shortages and overstocks. Furthermore, there is a need for more personalized recommendations by utilizing user emotional data.

[0260] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0261] In this invention, the server includes means for generating a demand forecasting model based on a user's purchase history, browsing history, and search history, means for optimizing inventory using the generated demand forecasting model, means for analyzing emotion data collected from the user's terminal, means for optimizing inventory and product recommendations using the analyzed emotion data, means for notifying the user's terminal of the generated recommendation information, and means for receiving purchase requests from the user, selecting an optimal logistics route, and issuing delivery instructions, thereby enabling personalized services and dynamic inventory management based on the user's emotions.

[0262] "User purchase history" is a record of products purchased by a user in the past, and specifically includes detailed information such as the purchase date and time, purchased product category, purchased quantity, and user ID.

[0263] "Browse history" is a record of web pages and products that a user has viewed in the past, and is data that indicates information such as the date and time of browsing, the number of times of browsing, and the products that have been viewed.

[0264] "Search history" is a record of a user's past search queries and search results, and is data that includes information such as search date and time, search keywords, and search result click history.

[0265] A "demand forecasting model" is a machine learning model for predicting future demand based on collected data, and is generated using algorithms such as regression analysis and time series analysis.

[0266] "Inventory optimization" is the process of using demand forecasting models to adjust inventory levels in anticipation of future demand, preventing shortages and overstocks.

[0267] "Emotion data" refers to data that reflects the user's emotional state, and specifically refers to information obtained from facial expression data, voice tone, keyboard input, and the like.

[0268] "Product recommendations" are recommendation information that suggests appropriate products to users, and are generated based on the user's purchasing history, emotional data, etc.

[0269] "Generated recommendation information" is information that suggests optimal products and services to a user, generated based on the user's purchase history, browsing history, search history, and emotional data.

[0270] A "purchase request" is information indicating a user's intention to purchase a product or service, and specifically includes a product ID, user ID, delivery address information, and the like.

[0271] The "optimal logistics route" is the most effective delivery route for efficiently delivering orders, and is selected taking into consideration factors such as cost, time, and inventory status.

[0272] This invention combines a system that performs demand forecasting based on a user's purchase history, browsing history, and search history, optimizes inventory, and provides product recommendations with an emotion engine that recognizes the user's emotions. Specific embodiments of this system are described below.

[0273] First, the server collects purchase history, browsing history, and search history from the user database. The collected data includes details such as purchase date and time, purchased product category, purchased quantity, and user ID. This data is then temporarily stored in cloud storage (e.g., Amazon S3). The software used at this stage includes database queries and cloud storage APIs.

[0274] Next, the server preprocesses the collected data. Specific processes include filling in missing values, removing outliers, and normalizing the data. This process uses Python's Pandas library and NumPy. The preprocessed data is then input into a machine learning algorithm (e.g., regression analysis, time series analysis) to generate a demand forecasting model. Machine learning libraries such as TensorFlow and scikit-learn are used at this stage.

[0275] The device uses a camera and microphone to collect user emotional data and transmits it to a server. Emotional data includes facial expressions, voice tone, and keyboard input. The collected data is analyzed by an emotion engine (e.g., NVIDIA DeepStream, Face++) to identify the user's emotional state. This information is used for demand forecasting and recommendation optimization.

[0276] The server optimizes inventory based on the generated demand forecasting model and emotion data. Specifically, it uses the demand forecasting model to forecast future demand and calculate the necessary inventory volume. Based on the calculation results, inventory is appropriately allocated through an inventory management system or ERP system (e.g., SAP, Oracle).

[0277] The server then recommends products based on the user's purchase history, browsing history, search history, and emotional data. This process uses recommendation engines (e.g., Amazon Personalize, Google RE) that use collaborative filtering or content-based filtering. The generated recommendation information is sent to the device in real time.

[0278] When a user purchases a product based on the recommended information, the device sends the purchase information to the server. This information includes the product ID, user ID, delivery address information, etc. The server then verifies the received purchase information and checks the inventory status. For example, if the product is available in stock, the server confirms the order and selects the optimal logistics route. This is done using a delivery management system (e.g., UPS WorldShip, FedEx Ship Manager).

[0279] Once delivery begins, the server monitors the delivery status in real time and notifies the user of the status as it progresses. This process improves the user experience and streamlines inventory management.

[0280] Prompt Sentence Examples

[0281] "Please explain a system that forecasts demand based on users' past purchasing history and optimizes inventory."

[0282] "Please explain a system that uses user emotional data to recommend products."

[0283] This invention enables personalized service and dynamic inventory management based on user emotions.

[0284] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0285] Step 1:

[0286] The server collects purchase history, browsing history, and search history data from the user database. The input data includes the purchase date and time, purchased product category, purchased quantity, user ID, etc. This data is stored in cloud storage (e.g., Amazon S3). Specifically, it executes a database query and uploads the retrieved data to cloud storage. The output of this step is the data stored in cloud storage.

[0287] Step 2:

[0288] The server preprocesses the collected data. Specifically, it imputes missing values, removes outliers, and normalizes the data. The software used includes Python's Pandas library and NumPy. The input data is stored in cloud storage, and the output data is the preprocessed, clean data.

[0289] Step 3:

[0290] The server inputs the preprocessed data into a machine learning algorithm to generate a demand forecasting model. This uses algorithms such as regression analysis and time series analysis. This process uses machine learning libraries such as TensorFlow and scikit-learn. The input data is preprocessed and clean, and the output is a demand forecasting model.

[0291] Step 4:

[0292] The device uses a camera and microphone to collect the user's emotional data and transmits it to a server. Emotional data includes facial expressions, voice tones, keyboard input, etc. The input data is emotional data acquired through the camera and microphone, and the output is emotional data transmitted to the server. Specific operations include collecting and transmitting sensor data.

[0293] Step 5:

[0294] The server analyzes the emotion data sent from the device. This analysis is performed using an emotion engine (e.g., NVIDIA DeepStream, Face++). The input data is the emotion data sent from the device, and the output is the analyzed emotional state.

[0295] Step 6:

[0296] The server combines the generated demand forecasting model with the analyzed emotion data to optimize inventory. Specifically, it uses the demand forecasting model to forecast future demand and calculates the required inventory volume. Based on the calculation results, inventory is appropriately allocated via an inventory management system or ERP system (e.g., SAP, Oracle). The input data are the demand forecasting model and emotion data, and the output is an optimized inventory list.

[0297] Step 7:

[0298] The server recommends products based on the user's purchase history, browsing history, search history, and emotional data. A recommendation engine (e.g., Amazon Personalize, Google RE) using collaborative filtering or content-based filtering is used. The generated recommendation information is sent to the device in real time. The input data are the aforementioned history data and emotional data, and the output is recommendation information.

[0299] Step 8:

[0300] When a user purchases a product based on the recommendation information, the terminal sends the purchase information to the server. The information sent includes the product ID, user ID, delivery address information, etc. A specific action is to press the purchase button. The input data is the execution data of the purchase page, and the output is the purchase information sent to the server.

[0301] Step 9:

[0302] The server confirms the received purchase information and checks the inventory status. If inventory is available, it confirms the order and selects the optimal logistics route. This is done using a delivery management system (e.g., UPS WorldShip, FedEx Ship Manager). The input data are purchase information and inventory information, and the output is the confirmed order and the selected logistics route.

[0303] Step 10:

[0304] The server monitors the delivery status in real time and notifies the user's device of the status as it progresses. Utilizing the delivery management system, it tracks each stage of delivery and updates the status as appropriate. The input data is delivery status information from the delivery management system, and the output is a status notification sent to the user's device.

[0305] (Application example 2)

[0306] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0307] Conventional demand forecasting systems are based solely on a user's purchasing history, browsing history, and search history, and do not take into account the user's real-time emotional state, making it difficult to provide personalized recommendations. Furthermore, they have the problem of not being able to adequately address the user's decision-making process or improve the user experience. The purpose of this invention is to solve these problems.

[0308] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for generating a demand forecasting model based on a user's purchase history, browsing history, and search history, means for optimizing inventory using the generated demand forecasting model, means for generating content recommendations based on the user's browsing history, search history, and emotional data, means for notifying the user's visual display device of the generated recommendation information, and means for receiving a selection request from the user and selecting the optimal content to be displayed next. This enables personalized recommendations and inventory management that take into account the user's real-time emotional state.

[0309] "User purchase history" is a collection of information about products and services that a user has purchased in the past.

[0310] "Browsing history" is a record of pages and content that a user has previously viewed on the Internet or in applications.

[0311] "Search history" is a history of keywords and phrases that a user has searched for on the Internet or in an application.

[0312] A "demand forecasting model" is an algorithm or mathematical model for predicting future demand based on past user behavior data.

[0313] "Inventory optimization" is the process of optimizing supply-side inventory levels based on demand forecasts.

[0314] "Emotion data" is information that identifies a user's emotional state in real time by analyzing their facial expressions and vocal tone.

[0315] "Content recommendation" is a function that suggests appropriate content based on a user's behavioral history and emotional data.

[0316] A "visual display device" is a device that visually displays information to a user, and includes smart glasses and digital displays.

[0317] A "selection request" is a request for a user to make a selection or designation from the presented recommendation information.

[0318] This invention is a system that forecasts demand and recommends content based on users' purchasing history, browsing history, search history, and emotional data. This system consists of three elements: a server, a terminal, and a user.

[0319] First, the server collects users' purchasing, browsing, and search histories and generates a demand forecasting model based on this information. This forecasting model is used to analyze users' consumption behavior and forecast future demand. The server also has functions for preprocessing the collected data, cleaning the data, imputing missing values, and normalizing the data. For this purpose, the server mainly uses machine learning libraries such as Scikit-learn.

[0320] The server also combines the user's browsing history, search history, and emotion engine to collect real-time emotional data. This emotional data is collected via the camera and microphone and analyzed using an emotion recognition model built with Keras. The server identifies the user's emotions and recommends content based on the results.

[0321] The device used by the user includes smart glasses that can capture the user's facial expressions and tone of voice in real time, and the captured emotional data is sent to a server where it is incorporated into a demand forecasting model and recommendation engine.

[0322] For example, if a user is watching a movie through smart glasses, the emotion engine can determine that the user is "excited" based on their facial expressions and voice. Based on the user's viewing history, the server then recommends movies and content that the user is likely to watch next, and displays them on the smart glasses' display. This process improves the user experience and provides more accurate recommendations.

[0323] In addition, the demand forecasting model is regularly updated, enabling dynamic inventory management. For example, by combining past data with the latest sentiment data, demand can be forecast more accurately and appropriate inventory levels can be maintained. This shortens logistics lead times and minimizes user purchasing lead times.

[0324] The main hardware and software required to implement this system are as follows:

[0325] Hardware:

[0326] Smart glasses (with camera and microphone)

[0327] Server (cloud environment recommended)

[0328] software:

[0329] OpenCV (face detection and facial expression data collection)

[0330] Scikit-learn (data preprocessing and demand forecasting model)

[0331] Keras (emotion recognition model)

[0332] Server-side API (obtaining user data)

[0333] For an example of how to use a generative AI model, the prompt would be:

[0334] "By aligning a user's viewing history with their current emotional state, we can recommend content they're likely to watch next. For example, if a user is excited and watching an action movie, what specific action movie should we recommend?"

[0335] This will create a system that integrates user behavioral data and emotional data to provide more personalized recommendations.

[0336] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0337] Step 1:

[0338] The server collects the user's purchase history, browsing history, and search history. For this, detailed information such as purchase date and time, purchased product category, and purchase quantity is extracted from the database based on the user ID. The input data is the user's behavior history information, and the output data is preprocessed user behavior history data.

[0339] Step 2:

[0340] The server preprocesses the collected data. Specifically, it performs data cleaning, missing value completion, data normalization, etc. This process formats the input data for the machine learning algorithm. The input data is the collected user behavior history data, and the output data is the preprocessed data.

[0341] Step 3:

[0342] The server uses the preprocessed data to generate a demand forecasting model. During this process, it uses a machine learning library such as Scikit-learn to perform regression analysis and time series analysis. The input data is the preprocessed data, and the output data is the demand forecasting model. The generated model is used to predict future demand.

[0343] Step 4:

[0344] The device captures the user's browsing history and search history, as well as facial and voice data in real time. This data is collected through a camera and microphone. The input data is real-time facial and voice data, and the output data is the captured raw data.

[0345] Step 5:

[0346] The server receives the facial expression and voice data sent from the device and analyzes them using an emotion recognition engine. Using an emotion recognition model built with Keras, the server identifies the user's emotional state. The input data is the captured facial expression and voice data, and the output data is the identified emotion data.

[0347] Step 6:

[0348] The server operates a recommendation engine based on the identified emotion data and the user's browsing and search history. Collaborative filtering and content-based filtering are used to generate optimal content. The input data are emotion data and behavioral history data, and the output data is recommendation information.

[0349] Step 7:

[0350] The terminal notifies the visual display device of the recommendation information received from the server. The user can check the notified recommendation information and move to a detailed information page if they are interested. The input data is the recommendation information, and the output data is the information displayed visually to the user.

[0351] Step 8:

[0352] When the user selects a content, a selection request is sent to the server. The input data is the user's selection request, and the output data is the request information sent to the server.

[0353] Step 9:

[0354] Based on the received selection request, the server selects the optimal content to display next and notifies the terminal again. This allows the user to seamlessly view the next content. The input data is the request information, and the output data is the content information to be displayed next.

[0355] In this way, through specific actions at each step, the system provides users with personalized content recommendations and performs real-time sentiment analysis and demand forecasting.

[0356] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0357] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0358] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0359] [Second embodiment]

[0360] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0361] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0362] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0363] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0364] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0365] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0366] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0367] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0368] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0369] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0370] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0371] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0372] The present invention is a system that performs demand forecasting based on a user's purchase history, browsing history, and search history, and provides inventory optimization and product recommendations. Below, we will create a program for this system and explain the specific processing content.

[0373] Demand forecast data collection and analysis

[0374] The server collects data such as past purchase history, browsing history, and search history from the user database. This includes detailed information such as purchase date and time, purchased product category, purchased quantity, and user ID. The server preprocesses the acquired data and generates a demand forecasting model using machine learning algorithms (e.g., regression analysis and time series analysis). Future demand is estimated based on the generated forecasting model.

[0375] Examples:

[0376] The server analyzes "User A's" purchasing data for the past six months and predicts that "User A" is likely to purchase a particular electronic device again in the next month.

[0377] Inventory management

[0378] The server uses the generated demand forecast model to calculate the necessary inventory levels and allocate optimal stock to each warehouse. It periodically updates the inventory status and dynamically manages it to prevent shortages and excesses.

[0379] Examples:

[0380] The server predicts that demand for electronic devices will increase within the month and instructs the Tokyo warehouse to prepare additional inventory to accommodate this.

[0381] Recommendation function

[0382] The server collects users' browsing and search histories in real time, and uses a recommendation engine to select the most suitable products based on that information.The server then notifies the user of the selected recommendation information on their device.

[0383] Examples:

[0384] If "User B" searches for "smartphone cases" on a website, the server will recommend new products in the same case category.

[0385] User Notifications and Interactions

[0386] The device notifies the user of the recommendation information received from the server in real time. The user then moves to the page with detailed information and, if interested, proceeds to the purchase process.

[0387] Examples:

[0388] User C's device displays a notification for "Recommended Product: Waterproof Smartphone Case," and User C clicks on the notification to go to the details page.

[0389] Checkout and logistics

[0390] When the user presses the purchase button, the device sends the purchase information to the server. The server then checks the received purchase information, selects the optimal logistics route, and issues delivery instructions. The server also monitors the delivery status in real time and updates the delivery status on the user's device.

[0391] Examples:

[0392] When "User D" presses the purchase button, the server retrieves the product from the nearest warehouse, selects the fastest delivery method, and begins delivery to the user's address. During delivery, the server monitors the delivery status and notifies User D's device of the status as it progresses.

[0393] By executing such a program, the present invention aims to minimize logistics lead time and user decision-making lead time, further optimizing inventory and improving the user experience. Furthermore, by periodically updating the generated demand forecast model and performing dynamic inventory management, more accurate supply and demand adjustments become possible. Furthermore, by collecting user action logs and using them to generate recommendations for future purchases, more personalized services can be provided.

[0394] The processing flow will be explained below.

[0395] Step 1:

[0396] The server collects data such as past purchase history, browsing history, and search history from the user database. For example, it comprehensively obtains each user's purchase date and time, purchased product category, purchased quantity, user ID, etc.

[0397] Step 2:

[0398] The server preprocesses the acquired data, specifically by cleaning the data, filling in missing data, and normalizing the data, thereby converting the data into a state that can be analyzed.

[0399] Step 3:

[0400] The server then inputs the preprocessed data into a machine learning algorithm to generate a demand forecasting model, for example, using regression analysis or time series analysis to build a model that predicts future demand.

[0401] Step 4:

[0402] The server uses the generated demand forecasting model to predict future demand for specific products, calculates the necessary inventory levels based on the forecast, and allocates optimal inventory to each warehouse.

[0403] Step 5:

[0404] The server collects users' browsing and search history in real time, for example, capturing their actions when they search for a specific product on a website or view a detail page.

[0405] Step 6:

[0406] The server uses the collected real-time data to run a recommendation engine, for example, using collaborative filtering or content-based filtering to select the best products for the user.

[0407] Step 7:

[0408] The server sends the generated recommendation information to the user's device, which receives the information and notifies the user in real time.

[0409] Step 8:

[0410] The user checks the recommendation notification displayed on the device and clicks on the detailed information page if they are interested. For example, they receive a notification about a new product that is perfect for them: a waterproof smartphone case.

[0411] Step 9:

[0412] When the user presses the purchase button, the terminal sends the purchase information to the server, including the product ID, user ID, and delivery address information.

[0413] Step 10:

[0414] The server confirms the received purchasing information and checks the inventory status. If the inventory is available, it confirms the order and selects the optimal logistics route.

[0415] Step 11:

[0416] The server issues delivery instructions to the warehouse, for example, shipping the product from the nearest warehouse and selecting the fastest delivery method.

[0417] Step 12:

[0418] The server monitors the delivery status in real time and updates the status to the user's device, such as when the product has been shipped, when it is in transit, or when delivery has been completed.

[0419] Example 1

[0420] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0421] Conventional systems do not adequately forecast demand or optimize inventory based on users' purchasing, browsing, and search histories, making it difficult to detect inventory surpluses and shortages and to recommend appropriate products to users.Furthermore, there is a problem of reduced user convenience due to the lack of real-time monitoring of delivery status and the lack of proper notification of delivery status.

[0422] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0423] In this invention, the server includes means for generating a demand forecasting model based on a user's purchase history, browsing history, and search history, means for optimizing inventory using the generated demand forecasting model, means for generating product recommendations based on the user's browsing history and search history, means for notifying the user's terminal of the generated recommendation information, means for receiving purchase requests from the user, selecting an optimal logistics route, and issuing delivery instructions, and means for monitoring delivery status in real time and notifying the user's terminal of the delivery status. This prevents inventory surpluses and shortages, recommends optimal products to the user, and enables efficient delivery.

[0424] "User" refers to a person who uses the system and provides data such as purchase history, browsing history, and search history.

[0425] "Purchase history" includes information about products purchased by a user in the past, and refers to detailed information such as the purchase date and time, purchased product category, purchased quantity, and user ID.

[0426] "Browser history" refers to data that includes information about products viewed by a user on websites or within applications.

[0427] "Search history" refers to data that includes keywords that users searched for within the system and information about search results.

[0428] "Demand forecasting model" refers to a machine learning algorithm used to predict future demand for a product based on data such as purchase history, browsing history, and search history.

[0429] "Inventory optimization" refers to the use of demand forecasting models to allocate optimal amounts of inventory in each warehouse and prevent shortages and excess inventory.

[0430] "Product recommendation" refers to a method of selecting and suggesting optimal products to users based on their browsing history and search history.

[0431] "Notification" refers to the act of sending information such as recommendation information and delivery status from the server to the user's terminal.

[0432] The "logistics route" refers to the optimal logistics route for efficiently delivering the products purchased by the user.

[0433] "Delivery status" refers to the current status of the product purchased by the user in the delivery process, and includes states such as shipped, in delivery, and delivery completed.

[0434] The present invention provides a system for performing demand forecasting based on a user's purchase history, browsing history, and search history, optimizing inventory, and providing product recommendations. Specific embodiments of this system are described below.

[0435] First, the server collects data on past purchase history, browsing history, and search history from the user database. This data includes detailed information such as purchase date and time, purchased product category, purchased quantity, and user ID. The server then preprocesses the acquired data, for example, removing duplicates and filling in missing values. After preprocessing is complete, the server uses machine learning algorithms (e.g., regression analysis and time series analysis) to generate a demand forecasting model. This demand forecasting model is saved and used to forecast future demand.

[0436] As a specific example, by analyzing the purchasing data of "User A" for the past six months, it is possible to predict the likelihood that "User A" will purchase a particular electronic device again in the next month.

[0437] The server then uses the generated demand forecast model to calculate the required inventory and allocate optimal inventory to each warehouse. It periodically updates inventory status and dynamically manages it to prevent shortages and excesses. For example, the server may predict that demand for electronic devices will increase within the current month for the Tokyo warehouse and instruct it to allocate additional inventory.

[0438] Furthermore, the server collects the user's browsing history and search history in real time and uses a recommendation engine to select the most suitable product. This recommendation information is then sent to the user's device. For example, if "User B" is searching for a smartphone case, the server will recommend a new product.

[0439] When a user receives the recommendation information, they go to the detailed information page and, if they are interested, proceed to the purchase process. For example, "User C"'s device may display a notification for "Recommended Product: Waterproof Smartphone Case," and User C may go to the detailed information page.

[0440] Additionally, when a user presses the purchase button, the terminal sends the purchase information to the server. The server confirms the received purchase information, selects the optimal logistics route, and issues delivery instructions. It monitors the delivery situation in real time and updates the delivery status on the user's terminal. As a specific example, when "User D" presses the purchase button, the server retrieves the product from the nearest warehouse, selects the fastest delivery method, and notifies "User D's" terminal of the delivery status as the process progresses.

[0441] An example of a prompt for a generative AI model is as follows:

[0442] "Based on User A's purchasing data from the past six months, please predict the products that he is likely to purchase in the next month."

[0443] "If User B is searching for a smartphone case, please recommend related products."

[0444] "Please forecast which warehouses will see increased demand for electronic equipment this month and issue instructions for deploying the necessary inventory."

[0445] As described above, the system of the present invention aims to minimize logistics lead time and user decision-making lead time, optimize inventory, and improve the user experience. By periodically updating the generated demand forecast model and performing dynamic inventory management, more accurate supply and demand adjustments become possible. Furthermore, by collecting user action logs and using them to generate recommendations for future purchases, more personalized services can be provided.

[0446] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0447] Step 1:

[0448] The server collects purchase history, browsing history, and search history data from a user database.

[0449] Input: User database

[0450] Output: Purchase history, browsing history, search history

[0451] Specific operation: The server uses an SQL query to retrieve the purchase history, browsing history, and search history of "User A" and "User B" from the user database for the past six months.

[0452] Step 2:

[0453] The server pre-processes the collected data.

[0454] Input: purchase history, browsing history, search history

[0455] Output: Preprocessed data

[0456] Specific operation: The server removes duplicate data, imputes missing values, and converts the data into the required format. For example, if there are missing values, it imputes them with the average value.

[0457] Step 3:

[0458] The server uses the pre-processed data to train a demand forecasting model.

[0459] Input: Preprocessed data

[0460] Output: Demand forecast model

[0461] Specific operation: The server uses Python's scikit-learn to perform regression analysis and time series analysis to train a demand forecasting model, and saves this model for use in future demand forecasts.

[0462] Step 4:

[0463] The server uses the generated demand forecast model to optimize inventory.

[0464] Input: Demand forecast model

[0465] Output: Inventory allocation plan

[0466] Specific operation: The server calculates the amount of inventory required for each warehouse based on the demand forecast model. For example, it instructs the Tokyo warehouse to increase the inventory of electronic devices.

[0467] Step 5:

[0468] The server generates product recommendations based on the user's browsing history and search history.

[0469] Input: Real-time collected browsing history, search history

[0470] Output: Recommendation information

[0471] Specific operation: When a user searches for "smartphone case," the server selects new products in the same category and generates recommendation information.

[0472] Step 6:

[0473] The terminal notifies the user of the recommendation information received from the server.

[0474] Input: Recommendation information

[0475] Output: Notification message

[0476] Specific operation: User C's device displays a notification such as "Recommended product: Waterproof smartphone case," and User C moves to the details page.

[0477] Step 7:

[0478] The terminal transmits a purchase request from the user to the server.

[0479] Input: Purchase Request

[0480] Output: Purchase information

[0481] Specific operation: When the user presses the purchase button, the device sends that information to the server, which then selects the optimal logistics route and issues delivery instructions.

[0482] Step 8:

[0483] The server monitors the delivery status in real time and notifies the user's terminal of the delivery status.

[0484] Input: Delivery status from logistics company

[0485] Output: Delivery status update

[0486] Specific operation: The server uses the logistics company's API to monitor delivery status and notifies the user's device of statuses such as "shipped," "in transit," and "delivery completed."

[0487] (Application example 1)

[0488] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0489] The present invention relates to a system for improving the accuracy of demand forecasting and product recommendations based on users' purchase history, browsing history, and search history, optimizing inventory, and improving the user experience. With conventional technologies, demand forecasting and inventory management are time-consuming and often result in delayed notification of recommendation information. Therefore, a system combining efficient inventory management with rapid notification of recommendations is needed.

[0490] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0491] In this invention, the server includes: means for generating a demand forecasting model based on a user's purchase history, browsing history, and search history; means for optimizing inventory using the generated demand forecasting model; means for generating product recommendations based on the user's browsing history and search history; means for notifying the user's terminal of the generated recommendation information; means for receiving purchase requests from the user, selecting an optimal logistics route, and issuing delivery instructions; and means for providing appropriate product recommendations based on the user's shopping behavior and sending push notifications in real time using the smart device to increase purchasing motivation. This enables both inventory management and recommendation notifications to be performed with high accuracy and efficiency, improving the user experience.

[0492] "User purchase history" is a record of information about products purchased by the user in the past, such as purchase date and time, and purchase quantity.

[0493] A "demand forecasting model" is a mathematical model for predicting future demand based on a user's past purchasing history, browsing history, and search history.

[0494] "Inventory optimization" is the process of adjusting product inventory at each warehouse or store to meet demand using a demand forecasting model.

[0495] "Product recommendation" is a function that suggests products that may be of interest to users based on their browsing history and search history.

[0496] "User terminal" refers to a digital device used by a user, such as a smartphone or tablet.

[0497] A "purchase request" is information sent to the system when a user indicates their intention to purchase a product.

[0498] "Logistics route" is a concept that refers to the optimal route or means for delivering goods.

[0499] "Delivery Instructions" refers to the process of issuing specific instructions regarding the delivery of goods.

[0500] "Shopping behavior" refers to a series of actions taken by a user, such as browsing, searching, and purchasing products.

[0501] "Smart devices" is a general term for devices that can connect to the Internet, such as smartphones, tablets, and smartwatches.

[0502] "Push notification" is a function that sends information from a server to a user's smart device in real time.

[0503] The system for implementing this invention encompasses a series of processes including automated data collection, demand forecasting, inventory management, product recommendations, and user notifications. Specifically, it is realized in the following manner.

[0504] The server first collects users' purchase, browsing, and search histories from a database. This data includes detailed information such as user ID, purchased items, purchase date and time, and purchase quantity. Next, the collected data is preprocessed to generate a demand forecasting model. This model generation uses machine learning algorithms (e.g., linear regression and time series analysis).

[0505] The generated demand forecasting model predicts future demand, and inventory is optimized based on that. The server periodically updates the demand forecasting model to dynamically adjust inventory levels. This prevents overstocking and shortages and enables efficient resource management.

[0506] Furthermore, the server monitors the user's browsing and search history in real time and generates optimal product recommendations. This recommendation information is then pushed to the user via their smart device. For example, if "User A" searches for "smartphone cases" on a website, the server will recommend new products in the same category and notify the user's device.

[0507] When a user submits a purchase request, the server receives the information, selects the optimal logistics route, and issues delivery instructions. Once delivery begins, the server monitors the delivery status in real time and updates the delivery status on the user's device, allowing the user to track the progress of the product as it is delivered.

[0508] The hardware used includes servers, smart devices (smartphones, tablets, etc.), and database systems. The software includes machine learning libraries (e.g., scikit-learn), Python scripts for data collection and preprocessing, and notification services such as Firebase.

[0509] To generate a specific program, you can use the following prompt:

[0510] 1. "Generate Python code to forecast demand for the next month based on purchasing data from the past six months."

[0511] 2. "Create a program using Firebase to send recommendation notifications to users based on their purchase and browsing history."

[0512] This allows the server, terminal, and user to interact with each other, enabling both inventory management and recommendation notifications to be performed with high accuracy and efficiency, improving the user experience.

[0513] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0514] Step 1:

[0515] The server collects the user's purchase history, browsing history, and search history from a database. The input is the user ID, and the output is the user's various historical data (purchased items, purchase date and time, purchased quantity, viewed items, search keywords, etc.). This allows the past behavioral data of a specific user to be obtained.

[0516] Step 2:

[0517] The server pre-processes the collected data, cleaning the data (e.g., removing incomplete data and correcting outliers) and standardizing it. The input is the user history data collected in step 1, and the output is formatted data, which facilitates subsequent data analysis.

[0518] Step 3:

[0519] The server uses the preprocessed data to generate a demand forecasting model. It trains the model using machine learning algorithms (e.g., linear regression or time series analysis). The input is the preprocessed data, and the output is the demand forecasting model. This creates a foundation for forecasting future demand.

[0520] Step 4:

[0521] The server optimizes the planned inventory based on the generated demand forecast model. This involves a process of calculating how much of a specific product is needed at a specific time. The input is the demand forecast model, and the output is the calculation result of the optimal inventory amount. This makes inventory management more efficient.

[0522] Step 5:

[0523] The server monitors the user's real-time browsing and search history and generates product recommendations. The input is real-time user behavior data, and the output is a list of recommended products. This allows the server to recommend the most suitable products to the user.

[0524] Step 6:

[0525] The terminal notifies the user of the recommendation information sent from the server. This notification is performed using the push notification function of the smart device. The input is a list of recommended products, and the output is a notification message. This allows the user to receive new and recommended products in real time.

[0526] Step 7:

[0527] The user submits a purchase request, which is sent to the server with the purchase request data as input, thereby conveying the user's intention to purchase to the system.

[0528] Step 8:

[0529] The server selects the optimal logistics route based on the received purchase request and issues delivery instructions. The input is the purchase request data, and the output is logistics instructions and a delivery route. This ensures that purchased items are delivered to the user in the most efficient manner.

[0530] Step 9:

[0531] The server monitors the delivery status in real time and updates the delivery status to the user's device. The input is delivery progress data and the output is the updated delivery status, allowing the user to always know what stage their product is in.

[0532] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0533] This invention combines a system that performs demand forecasting based on users' purchase history, browsing history, and search history, optimizes inventory, and provides product recommendations with an emotion engine that recognizes users' emotions. Below, we will create a program for this system and explain the specific processing content.

[0534] Demand forecast data collection and analysis

[0535] The server collects past purchase history, browsing history, and search history from the user database. This includes detailed information such as purchase date and time, purchased product category, purchase quantity, and user ID. The server then preprocesses the collected data, cleaning the data, filling in missing values, and normalizing the data. The preprocessed data is then input into a machine learning algorithm (e.g., regression analysis or time series analysis) to generate a demand forecasting model. Future demand is predicted based on the generated forecasting model.

[0536] Examples:

[0537] The server analyzes "User A's" purchasing data for the past six months and predicts that "User A" is likely to purchase a particular electronic device again in the next month.

[0538] Collecting and analyzing emotional data

[0539] The server uses an emotion engine to recognize the user's emotions. It analyzes the user's facial expression data, voice tone, keyboard input, etc. sent from the device to identify the user's emotional state. This allows it to understand how the user is feeling at that time.

[0540] Examples:

[0541] While "User B" is browsing products on his smartphone, emotional data is collected using the camera and microphone. The emotion engine recognizes that User B is in an "excited" state.

[0542] Inventory management

[0543] The server combines the generated demand forecast model with user emotion data to calculate the necessary inventory levels. Based on this information, it allocates optimal inventory to each warehouse and regularly updates the inventory status to prevent shortages and overstocks.

[0544] Examples:

[0545] The server predicts that demand for electronic devices will increase within this month, and since user sentiment data indicates a positive reaction, it instructs the Tokyo warehouse to stock additional stock of those electronic devices.

[0546] Recommendation function

[0547] The server runs a recommendation engine based on the user's browsing history, search history, and emotional data. For example, it uses collaborative filtering or content-based filtering to select the most suitable products for the user. The generated recommendation information is sent to the user's device.

[0548] Examples:

[0549] When "User C" searches for "smartphone cases" on a website, the server analyzes the emotional data of "excitement." Taking that reaction into consideration, the server recommends "waterproof smartphone cases" as a product in the same category.

[0550] User Notifications and Interactions

[0551] The device will notify the user of the recommendation information received from the server in real time. The user can check the notification, and if they are interested, they can go to the detailed information page and proceed with the purchase process.

[0552] Examples:

[0553] User D's device displays a notification for "Recommended Product: Waterproof Smartphone Case," and User D clicks on the notification to go to the details page.

[0554] Checkout and logistics

[0555] When the user presses the purchase button, the terminal sends the purchase information to the server. The sent information includes the product ID, user ID, delivery address information, etc. The server confirms the received purchase information and checks the inventory status. If inventory is available, the order is confirmed and the optimal logistics route is selected. The server issues delivery instructions to the warehouse and updates the status on the user's terminal while monitoring the delivery status in real time.

[0556] Examples:

[0557] When "User E" presses the purchase button, the server retrieves the product from the nearest warehouse, selects the fastest delivery method, and begins delivery. During delivery, the server monitors the delivery status and notifies User E's device of the status as it progresses.

[0558] Through this series of processes, the present invention minimizes logistics lead time and user decision-making lead time, and further optimizes inventory. Furthermore, by periodically updating the generated demand forecast model and performing dynamic inventory management, more accurate supply and demand adjustments become possible. By utilizing user emotion data, more personalized services can be provided, improving the user experience.

[0559] The processing flow will be explained below.

[0560] Step 1:

[0561] The server collects data such as past purchase history, browsing history, and search history from the user database. The data collected includes purchase date and time, purchased product category, purchased quantity, and user ID.

[0562] Step 2:

[0563] The server preprocesses the acquired data, which includes data cleaning (e.g., removing incomplete data), imputing missing values ​​(e.g., imputing with the mean), and normalizing the data (e.g., scaling).

[0564] Step 3:

[0565] The server then inputs the preprocessed data into a machine learning algorithm to generate a demand forecasting model, for example, using regression analysis or time series analysis to build a model for predicting future demand.

[0566] Step 4:

[0567] The server periodically uses the generated demand forecasting model to predict future demand for specific products, calculates the necessary inventory levels based on this, and allocates optimal inventory to each warehouse.

[0568] Step 5:

[0569] The server uses an emotion engine to recognize the user's emotions and collects emotion data from the user's device, specifically analyzing facial expression data from the camera, voice tone from the microphone, and keyboard input speed and strength.

[0570] Step 6:

[0571] The server uses an emotion engine to analyze the collected emotional data and identify the emotional state the user is in at that time, categorizing emotions such as "excitement," "sadness," and "happiness."

[0572] Step 7:

[0573] The server incorporates the collected emotion data into the demand forecasting model to improve the accuracy of the forecast, thereby more accurately predicting product demand, which is strongly influenced by user emotions.

[0574] Step 8:

[0575] The server runs a recommendation engine based on the user's browsing history, search history, and emotional data, using collaborative filtering and content-based filtering, for example, to select the best products for the user.

[0576] Step 9:

[0577] The server transmits the generated recommendation information to the user's terminal, which notifies the user of the recommendation information in real time.

[0578] Step 10:

[0579] The user checks the recommendation notification displayed on the device and clicks on the detailed information page if they are interested. For example, a notification such as "Recommended product: Waterproof smartphone case" will be displayed.

[0580] Step 11:

[0581] When a user presses the purchase button, the device sends the purchase information to the server, including the product ID, user ID, and delivery address information.

[0582] Step 12:

[0583] The server confirms the received purchase information and checks the inventory status. If the item is available, it confirms the order and selects the optimal logistics route.

[0584] Step 13:

[0585] The server issues delivery instructions to the warehouse, for example, shipping the product from the nearest warehouse and selecting the fastest delivery method.

[0586] Step 14:

[0587] The server monitors the delivery status in real time and updates the status to the user's device, for example, notifying the user of the status when the product has been shipped, when it is being delivered, and when delivery has been completed.

[0588] Through these processing steps, the present invention minimizes logistics lead time and user decision-making lead time, optimizing inventory. By utilizing user emotion data, it is possible to provide more personalized services and improve the user experience.

[0589] Example 2

[0590] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0591] Many modern inventory management systems and product recommendation systems support demand forecasting and recommendations based on users' purchase, browsing, and search histories. However, they are unable to consider the user's emotional state, limiting their ability to optimize the user experience. Furthermore, inventory optimization is static, making it difficult to respond to dynamic demand fluctuations, leading to problems such as inventory shortages and overstocks. Furthermore, there is a need for more personalized recommendations by utilizing user emotional data.

[0592] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0593] In this invention, the server includes means for generating a demand forecasting model based on a user's purchase history, browsing history, and search history, means for optimizing inventory using the generated demand forecasting model, means for analyzing emotion data collected from the user's terminal, means for optimizing inventory and product recommendations using the analyzed emotion data, means for notifying the user's terminal of the generated recommendation information, and means for receiving purchase requests from the user, selecting an optimal logistics route, and issuing delivery instructions, thereby enabling personalized services and dynamic inventory management based on the user's emotions.

[0594] "User purchase history" is a record of products purchased by a user in the past, and specifically includes detailed information such as the purchase date and time, purchased product category, purchased quantity, and user ID.

[0595] "Browse history" is a record of web pages and products that a user has viewed in the past, and is data that indicates information such as the date and time of browsing, the number of times of browsing, and the products that have been viewed.

[0596] "Search history" is a record of a user's past search queries and search results, and is data that includes information such as search date and time, search keywords, and search result click history.

[0597] A "demand forecasting model" is a machine learning model for predicting future demand based on collected data, and is generated using algorithms such as regression analysis and time series analysis.

[0598] "Inventory optimization" is the process of using demand forecasting models to adjust inventory levels in anticipation of future demand, preventing shortages and overstocks.

[0599] "Emotion data" refers to data that reflects the user's emotional state, and specifically refers to information obtained from facial expression data, voice tone, keyboard input, and the like.

[0600] "Product recommendations" are recommendation information that suggests appropriate products to users, and are generated based on the user's purchasing history, emotional data, etc.

[0601] "Generated recommendation information" is information that suggests optimal products and services to a user, generated based on the user's purchase history, browsing history, search history, and emotional data.

[0602] A "purchase request" is information indicating a user's intention to purchase a product or service, and specifically includes a product ID, user ID, delivery address information, and the like.

[0603] The "optimal logistics route" is the most effective delivery route for efficiently delivering orders, and is selected taking into consideration factors such as cost, time, and inventory status.

[0604] This invention combines a system that performs demand forecasting based on a user's purchase history, browsing history, and search history, optimizes inventory, and provides product recommendations with an emotion engine that recognizes the user's emotions. Specific embodiments of this system are described below.

[0605] First, the server collects purchase history, browsing history, and search history from the user database. The collected data includes details such as purchase date and time, purchased product category, purchased quantity, and user ID. This data is then temporarily stored in cloud storage (e.g., Amazon S3). The software used at this stage includes database queries and cloud storage APIs.

[0606] Next, the server preprocesses the collected data. Specific processes include filling in missing values, removing outliers, and normalizing the data. This process uses Python's Pandas library and NumPy. The preprocessed data is then input into a machine learning algorithm (e.g., regression analysis, time series analysis) to generate a demand forecasting model. Machine learning libraries such as TensorFlow and scikit-learn are used at this stage.

[0607] The device uses a camera and microphone to collect user emotional data and transmits it to a server. Emotional data includes facial expressions, voice tone, and keyboard input. The collected data is analyzed by an emotion engine (e.g., NVIDIA DeepStream, Face++) to identify the user's emotional state. This information is used for demand forecasting and recommendation optimization.

[0608] The server optimizes inventory based on the generated demand forecasting model and emotion data. Specifically, it uses the demand forecasting model to forecast future demand and calculate the necessary inventory volume. Based on the calculation results, inventory is appropriately allocated through an inventory management system or ERP system (e.g., SAP, Oracle).

[0609] The server then recommends products based on the user's purchase history, browsing history, search history, and emotional data. This process uses recommendation engines (e.g., Amazon Personalize, Google RE) that use collaborative filtering or content-based filtering. The generated recommendation information is sent to the device in real time.

[0610] When a user purchases a product based on the recommended information, the device sends the purchase information to the server. This information includes the product ID, user ID, delivery address information, etc. The server then verifies the received purchase information and checks the inventory status. For example, if the product is available in stock, the server confirms the order and selects the optimal logistics route. This is done using a delivery management system (e.g., UPS WorldShip, FedEx Ship Manager).

[0611] Once delivery begins, the server monitors the delivery status in real time and notifies the user of the status as it progresses. This process improves the user experience and streamlines inventory management.

[0612] Prompt Sentence Examples

[0613] "Please explain a system that forecasts demand based on users' past purchasing history and optimizes inventory."

[0614] "Please explain a system that uses user emotional data to recommend products."

[0615] This invention enables personalized service and dynamic inventory management based on user emotions.

[0616] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0617] Step 1:

[0618] The server collects purchase history, browsing history, and search history data from the user database. The input data includes the purchase date and time, purchased product category, purchased quantity, user ID, etc. This data is stored in cloud storage (e.g., Amazon S3). Specifically, it executes a database query and uploads the retrieved data to cloud storage. The output of this step is the data stored in cloud storage.

[0619] Step 2:

[0620] The server preprocesses the collected data. Specifically, it imputes missing values, removes outliers, and normalizes the data. The software used includes Python's Pandas library and NumPy. The input data is stored in cloud storage, and the output data is the preprocessed, clean data.

[0621] Step 3:

[0622] The server inputs the preprocessed data into a machine learning algorithm to generate a demand forecasting model. This uses algorithms such as regression analysis and time series analysis. This process uses machine learning libraries such as TensorFlow and scikit-learn. The input data is preprocessed and clean, and the output is a demand forecasting model.

[0623] Step 4:

[0624] The device uses a camera and microphone to collect the user's emotional data and transmits it to a server. Emotional data includes facial expressions, voice tones, keyboard input, etc. The input data is emotional data acquired through the camera and microphone, and the output is emotional data transmitted to the server. Specific operations include collecting and transmitting sensor data.

[0625] Step 5:

[0626] The server analyzes the emotion data sent from the device. This analysis is performed using an emotion engine (e.g., NVIDIA DeepStream, Face++). The input data is the emotion data sent from the device, and the output is the analyzed emotional state.

[0627] Step 6:

[0628] The server combines the generated demand forecasting model with the analyzed emotion data to optimize inventory. Specifically, it uses the demand forecasting model to forecast future demand and calculates the required inventory volume. Based on the calculation results, inventory is appropriately allocated via an inventory management system or ERP system (e.g., SAP, Oracle). The input data are the demand forecasting model and emotion data, and the output is an optimized inventory list.

[0629] Step 7:

[0630] The server recommends products based on the user's purchase history, browsing history, search history, and emotional data. A recommendation engine (e.g., Amazon Personalize, Google RE) using collaborative filtering or content-based filtering is used. The generated recommendation information is sent to the device in real time. The input data are the aforementioned history data and emotional data, and the output is recommendation information.

[0631] Step 8:

[0632] When a user purchases a product based on the recommendation information, the terminal sends the purchase information to the server. The information sent includes the product ID, user ID, delivery address information, etc. A specific action is to press the purchase button. The input data is the execution data of the purchase page, and the output is the purchase information sent to the server.

[0633] Step 9:

[0634] The server confirms the received purchase information and checks the inventory status. If inventory is available, it confirms the order and selects the optimal logistics route. This is done using a delivery management system (e.g., UPS WorldShip, FedEx Ship Manager). The input data are purchase information and inventory information, and the output is the confirmed order and the selected logistics route.

[0635] Step 10:

[0636] The server monitors the delivery status in real time and notifies the user's device of the status as it progresses. Utilizing the delivery management system, it tracks each stage of delivery and updates the status as appropriate. The input data is delivery status information from the delivery management system, and the output is a status notification sent to the user's device.

[0637] (Application example 2)

[0638] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0639] Conventional demand forecasting systems are based solely on a user's purchasing history, browsing history, and search history, and do not take into account the user's real-time emotional state, making it difficult to provide personalized recommendations. Furthermore, they have the problem of not being able to adequately address the user's decision-making process or improve the user experience. The purpose of this invention is to solve these problems.

[0640] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for generating a demand forecasting model based on a user's purchase history, browsing history, and search history, means for optimizing inventory using the generated demand forecasting model, means for generating content recommendations based on the user's browsing history, search history, and emotional data, means for notifying the user's visual display device of the generated recommendation information, and means for receiving a selection request from the user and selecting the optimal content to be displayed next. This enables personalized recommendations and inventory management that take into account the user's real-time emotional state.

[0641] "User purchase history" is a collection of information about products and services that a user has purchased in the past.

[0642] "Browsing history" is a record of pages and content that a user has previously viewed on the Internet or in applications.

[0643] "Search history" is a history of keywords and phrases that a user has searched for on the Internet or in an application.

[0644] A "demand forecasting model" is an algorithm or mathematical model for predicting future demand based on past user behavior data.

[0645] "Inventory optimization" is the process of optimizing supply-side inventory levels based on demand forecasts.

[0646] "Emotion data" is information that identifies a user's emotional state in real time by analyzing their facial expressions and vocal tone.

[0647] "Content recommendation" is a function that suggests appropriate content based on a user's behavioral history and emotional data.

[0648] A "visual display device" is a device that visually displays information to a user, and includes smart glasses and digital displays.

[0649] A "selection request" is a request for a user to make a selection or designation from the presented recommendation information.

[0650] This invention is a system that forecasts demand and recommends content based on users' purchasing history, browsing history, search history, and emotional data. This system consists of three elements: a server, a terminal, and a user.

[0651] First, the server collects users' purchasing, browsing, and search histories and generates a demand forecasting model based on this information. This forecasting model is used to analyze users' consumption behavior and forecast future demand. The server also has functions for preprocessing the collected data, cleaning the data, imputing missing values, and normalizing the data. For this purpose, the server mainly uses machine learning libraries such as Scikit-learn.

[0652] The server also combines the user's browsing history, search history, and emotion engine to collect real-time emotional data. This emotional data is collected via the camera and microphone and analyzed using an emotion recognition model built with Keras. The server identifies the user's emotions and recommends content based on the results.

[0653] The device used by the user includes smart glasses that can capture the user's facial expressions and tone of voice in real time, and the captured emotional data is sent to a server where it is incorporated into a demand forecasting model and recommendation engine.

[0654] For example, if a user is watching a movie through smart glasses, the emotion engine can determine that the user is "excited" based on their facial expressions and voice. Based on the user's viewing history, the server then recommends movies and content that the user is likely to watch next, and displays them on the smart glasses' display. This process improves the user experience and provides more accurate recommendations.

[0655] In addition, the demand forecasting model is regularly updated, enabling dynamic inventory management. For example, by combining past data with the latest sentiment data, demand can be forecast more accurately and appropriate inventory levels can be maintained. This shortens logistics lead times and minimizes user purchasing lead times.

[0656] The main hardware and software required to implement this system are as follows:

[0657] Hardware:

[0658] Smart glasses (with camera and microphone)

[0659] Server (cloud environment recommended)

[0660] software:

[0661] OpenCV (face detection and facial expression data collection)

[0662] Scikit-learn (data preprocessing and demand forecasting model)

[0663] Keras (emotion recognition model)

[0664] Server-side API (obtaining user data)

[0665] For an example of how to use a generative AI model, the prompt would be:

[0666] "By aligning a user's viewing history with their current emotional state, we can recommend content they're likely to watch next. For example, if a user is excited and watching an action movie, what specific action movie should we recommend?"

[0667] This will create a system that integrates user behavioral data and emotional data to provide more personalized recommendations.

[0668] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0669] Step 1:

[0670] The server collects the user's purchase history, browsing history, and search history. For this, detailed information such as purchase date and time, purchased product category, and purchase quantity is extracted from the database based on the user ID. The input data is the user's behavior history information, and the output data is preprocessed user behavior history data.

[0671] Step 2:

[0672] The server preprocesses the collected data. Specifically, it performs data cleaning, missing value completion, data normalization, etc. This process formats the input data for the machine learning algorithm. The input data is the collected user behavior history data, and the output data is the preprocessed data.

[0673] Step 3:

[0674] The server uses the preprocessed data to generate a demand forecasting model. During this process, it uses a machine learning library such as Scikit-learn to perform regression analysis and time series analysis. The input data is the preprocessed data, and the output data is the demand forecasting model. The generated model is used to predict future demand.

[0675] Step 4:

[0676] The device captures the user's browsing history and search history, as well as facial and voice data in real time. This data is collected through a camera and microphone. The input data is real-time facial and voice data, and the output data is the captured raw data.

[0677] Step 5:

[0678] The server receives the facial expression and voice data sent from the device and analyzes them using an emotion recognition engine. Using an emotion recognition model built with Keras, the server identifies the user's emotional state. The input data is the captured facial expression and voice data, and the output data is the identified emotion data.

[0679] Step 6:

[0680] The server operates a recommendation engine based on the identified emotion data and the user's browsing and search history. Collaborative filtering and content-based filtering are used to generate optimal content. The input data are emotion data and behavioral history data, and the output data is recommendation information.

[0681] Step 7:

[0682] The terminal notifies the visual display device of the recommendation information received from the server. The user can check the notified recommendation information and move to a detailed information page if they are interested. The input data is the recommendation information, and the output data is the information displayed visually to the user.

[0683] Step 8:

[0684] When the user selects a content, a selection request is sent to the server. The input data is the user's selection request, and the output data is the request information sent to the server.

[0685] Step 9:

[0686] Based on the received selection request, the server selects the optimal content to display next and notifies the terminal again. This allows the user to seamlessly view the next content. The input data is the request information, and the output data is the content information to be displayed next.

[0687] In this way, through specific actions at each step, the system provides users with personalized content recommendations and performs real-time sentiment analysis and demand forecasting.

[0688] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0689] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0690] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0691] [Third embodiment]

[0692] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0693] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0694] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0695] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0696] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0697] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0698] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0699] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0700] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0701] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0702] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0703] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0704] The present invention is a system that performs demand forecasting based on a user's purchase history, browsing history, and search history, and provides inventory optimization and product recommendations. Below, we will create a program for this system and explain the specific processing content.

[0705] Demand forecast data collection and analysis

[0706] The server collects data such as past purchase history, browsing history, and search history from the user database. This includes detailed information such as purchase date and time, purchased product category, purchased quantity, and user ID. The server preprocesses the acquired data and generates a demand forecasting model using machine learning algorithms (e.g., regression analysis and time series analysis). Future demand is estimated based on the generated forecasting model.

[0707] Examples:

[0708] The server analyzes "User A's" purchasing data for the past six months and predicts that "User A" is likely to purchase a particular electronic device again in the next month.

[0709] Inventory management

[0710] The server uses the generated demand forecast model to calculate the necessary inventory levels and allocate optimal stock to each warehouse. It periodically updates the inventory status and dynamically manages it to prevent shortages and excesses.

[0711] Examples:

[0712] The server predicts that demand for electronic devices will increase within the month and instructs the Tokyo warehouse to prepare additional inventory to accommodate this.

[0713] Recommendation function

[0714] The server collects users' browsing and search histories in real time, and uses a recommendation engine to select the most suitable products based on that information.The server then notifies the user of the selected recommendation information on their device.

[0715] Examples:

[0716] If "User B" searches for "smartphone cases" on a website, the server will recommend new products in the same case category.

[0717] User Notifications and Interactions

[0718] The device notifies the user of the recommendation information received from the server in real time. The user then moves to the page with detailed information and, if interested, proceeds to the purchase process.

[0719] Examples:

[0720] User C's device displays a notification for "Recommended Product: Waterproof Smartphone Case," and User C clicks on the notification to go to the details page.

[0721] Checkout and logistics

[0722] When the user presses the purchase button, the device sends the purchase information to the server. The server then checks the received purchase information, selects the optimal logistics route, and issues delivery instructions. The server also monitors the delivery status in real time and updates the delivery status on the user's device.

[0723] Examples:

[0724] When "User D" presses the purchase button, the server retrieves the product from the nearest warehouse, selects the fastest delivery method, and begins delivery to the user's address. During delivery, the server monitors the delivery status and notifies User D's device of the status as it progresses.

[0725] By executing such a program, the present invention aims to minimize logistics lead time and user decision-making lead time, further optimizing inventory and improving the user experience. Furthermore, by periodically updating the generated demand forecast model and performing dynamic inventory management, more accurate supply and demand adjustments become possible. Furthermore, by collecting user action logs and using them to generate recommendations for future purchases, more personalized services can be provided.

[0726] The processing flow will be explained below.

[0727] Step 1:

[0728] The server collects data such as past purchase history, browsing history, and search history from the user database. For example, it comprehensively obtains each user's purchase date and time, purchased product category, purchased quantity, user ID, etc.

[0729] Step 2:

[0730] The server preprocesses the acquired data, specifically by cleaning the data, filling in missing data, and normalizing the data, thereby converting the data into a state that can be analyzed.

[0731] Step 3:

[0732] The server then inputs the preprocessed data into a machine learning algorithm to generate a demand forecasting model, for example, using regression analysis or time series analysis to build a model that predicts future demand.

[0733] Step 4:

[0734] The server uses the generated demand forecasting model to predict future demand for specific products, calculates the necessary inventory levels based on the forecast, and allocates optimal inventory to each warehouse.

[0735] Step 5:

[0736] The server collects users' browsing and search history in real time, for example, capturing their actions when they search for a specific product on a website or view a detail page.

[0737] Step 6:

[0738] The server uses the collected real-time data to run a recommendation engine, for example, using collaborative filtering or content-based filtering to select the best products for the user.

[0739] Step 7:

[0740] The server sends the generated recommendation information to the user's device, which receives the information and notifies the user in real time.

[0741] Step 8:

[0742] The user checks the recommendation notification displayed on the device and clicks on the detailed information page if they are interested. For example, they receive a notification about a new product that is perfect for them: a waterproof smartphone case.

[0743] Step 9:

[0744] When the user presses the purchase button, the terminal sends the purchase information to the server, including the product ID, user ID, and delivery address information.

[0745] Step 10:

[0746] The server confirms the received purchasing information and checks the inventory status. If the inventory is available, it confirms the order and selects the optimal logistics route.

[0747] Step 11:

[0748] The server issues delivery instructions to the warehouse, for example, shipping the product from the nearest warehouse and selecting the fastest delivery method.

[0749] Step 12:

[0750] The server monitors the delivery status in real time and updates the status to the user's device, such as when the product has been shipped, when it is in transit, or when delivery has been completed.

[0751] Example 1

[0752] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0753] Conventional systems do not adequately forecast demand or optimize inventory based on users' purchasing, browsing, and search histories, making it difficult to detect inventory surpluses and shortages and to recommend appropriate products to users.Furthermore, there is a problem of reduced user convenience due to the lack of real-time monitoring of delivery status and the lack of proper notification of delivery status.

[0754] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0755] In this invention, the server includes means for generating a demand forecasting model based on a user's purchase history, browsing history, and search history, means for optimizing inventory using the generated demand forecasting model, means for generating product recommendations based on the user's browsing history and search history, means for notifying the user's terminal of the generated recommendation information, means for receiving purchase requests from the user, selecting an optimal logistics route, and issuing delivery instructions, and means for monitoring delivery status in real time and notifying the user's terminal of the delivery status. This prevents inventory surpluses and shortages, recommends optimal products to the user, and enables efficient delivery.

[0756] "User" refers to a person who uses the system and provides data such as purchase history, browsing history, and search history.

[0757] "Purchase history" includes information about products purchased by a user in the past, and refers to detailed information such as the purchase date and time, purchased product category, purchased quantity, and user ID.

[0758] "Browser history" refers to data that includes information about products viewed by a user on websites or within applications.

[0759] "Search history" refers to data that includes keywords that users searched for within the system and information about search results.

[0760] "Demand forecasting model" refers to a machine learning algorithm used to predict future demand for a product based on data such as purchase history, browsing history, and search history.

[0761] "Inventory optimization" refers to the use of demand forecasting models to allocate optimal amounts of inventory in each warehouse and prevent shortages and excess inventory.

[0762] "Product recommendation" refers to a method of selecting and suggesting optimal products to users based on their browsing history and search history.

[0763] "Notification" refers to the act of sending information such as recommendation information and delivery status from the server to the user's terminal.

[0764] The "logistics route" refers to the optimal logistics route for efficiently delivering the products purchased by the user.

[0765] "Delivery status" refers to the current status of the product purchased by the user in the delivery process, and includes states such as shipped, in delivery, and delivery completed.

[0766] The present invention provides a system for performing demand forecasting based on a user's purchase history, browsing history, and search history, optimizing inventory, and providing product recommendations. Specific embodiments of this system are described below.

[0767] First, the server collects data on past purchase history, browsing history, and search history from the user database. This data includes detailed information such as purchase date and time, purchased product category, purchased quantity, and user ID. The server then preprocesses the acquired data, for example, removing duplicates and filling in missing values. After preprocessing is complete, the server uses machine learning algorithms (e.g., regression analysis and time series analysis) to generate a demand forecasting model. This demand forecasting model is saved and used to forecast future demand.

[0768] As a specific example, by analyzing the purchasing data of "User A" for the past six months, it is possible to predict the likelihood that "User A" will purchase a particular electronic device again in the next month.

[0769] The server then uses the generated demand forecast model to calculate the required inventory and allocate optimal inventory to each warehouse. It periodically updates inventory status and dynamically manages it to prevent shortages and excesses. For example, the server may predict that demand for electronic devices will increase within the current month for the Tokyo warehouse and instruct it to allocate additional inventory.

[0770] Furthermore, the server collects the user's browsing history and search history in real time and uses a recommendation engine to select the most suitable product. This recommendation information is then sent to the user's device. For example, if "User B" is searching for a smartphone case, the server will recommend a new product.

[0771] When a user receives the recommendation information, they go to the detailed information page and, if they are interested, proceed to the purchase process. For example, "User C"'s device may display a notification for "Recommended Product: Waterproof Smartphone Case," and User C may go to the detailed information page.

[0772] Additionally, when a user presses the purchase button, the terminal sends the purchase information to the server. The server confirms the received purchase information, selects the optimal logistics route, and issues delivery instructions. It monitors the delivery situation in real time and updates the delivery status on the user's terminal. As a specific example, when "User D" presses the purchase button, the server retrieves the product from the nearest warehouse, selects the fastest delivery method, and notifies "User D's" terminal of the delivery status as the process progresses.

[0773] An example of a prompt for a generative AI model is as follows:

[0774] "Based on User A's purchasing data from the past six months, please predict the products that he is likely to purchase in the next month."

[0775] "If User B is searching for a smartphone case, please recommend related products."

[0776] "Please forecast which warehouses will see increased demand for electronic equipment this month and issue instructions for deploying the necessary inventory."

[0777] As described above, the system of the present invention aims to minimize logistics lead time and user decision-making lead time, optimize inventory, and improve the user experience. By periodically updating the generated demand forecast model and performing dynamic inventory management, more accurate supply and demand adjustments become possible. Furthermore, by collecting user action logs and using them to generate recommendations for future purchases, more personalized services can be provided.

[0778] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0779] Step 1:

[0780] The server collects purchase history, browsing history, and search history data from a user database.

[0781] Input: User database

[0782] Output: Purchase history, browsing history, search history

[0783] Specific operation: The server uses an SQL query to retrieve the purchase history, browsing history, and search history of "User A" and "User B" from the user database for the past six months.

[0784] Step 2:

[0785] The server pre-processes the collected data.

[0786] Input: purchase history, browsing history, search history

[0787] Output: Preprocessed data

[0788] Specific operation: The server removes duplicate data, imputes missing values, and converts the data into the required format. For example, if there are missing values, it imputes them with the average value.

[0789] Step 3:

[0790] The server uses the pre-processed data to train a demand forecasting model.

[0791] Input: Preprocessed data

[0792] Output: Demand forecast model

[0793] Specific operation: The server uses Python's scikit-learn to perform regression analysis and time series analysis to train a demand forecasting model, and saves this model for use in future demand forecasts.

[0794] Step 4:

[0795] The server uses the generated demand forecast model to optimize inventory.

[0796] Input: Demand forecast model

[0797] Output: Inventory allocation plan

[0798] Specific operation: The server calculates the amount of inventory required for each warehouse based on the demand forecast model. For example, it instructs the Tokyo warehouse to increase the inventory of electronic devices.

[0799] Step 5:

[0800] The server generates product recommendations based on the user's browsing history and search history.

[0801] Input: Real-time collected browsing history, search history

[0802] Output: Recommendation information

[0803] Specific operation: When a user searches for "smartphone case," the server selects new products in the same category and generates recommendation information.

[0804] Step 6:

[0805] The terminal notifies the user of the recommendation information received from the server.

[0806] Input: Recommendation information

[0807] Output: Notification message

[0808] Specific operation: User C's device displays a notification such as "Recommended product: Waterproof smartphone case," and User C moves to the details page.

[0809] Step 7:

[0810] The terminal transmits a purchase request from the user to the server.

[0811] Input: Purchase Request

[0812] Output: Purchase information

[0813] Specific operation: When the user presses the purchase button, the device sends that information to the server, which then selects the optimal logistics route and issues delivery instructions.

[0814] Step 8:

[0815] The server monitors the delivery status in real time and notifies the user's terminal of the delivery status.

[0816] Input: Delivery status from logistics company

[0817] Output: Delivery status update

[0818] Specific operation: The server uses the logistics company's API to monitor delivery status and notifies the user's device of statuses such as "shipped," "in transit," and "delivery completed."

[0819] (Application example 1)

[0820] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0821] The present invention relates to a system for improving the accuracy of demand forecasting and product recommendations based on users' purchase history, browsing history, and search history, optimizing inventory, and improving the user experience. With conventional technologies, demand forecasting and inventory management are time-consuming and often result in delayed notification of recommendation information. Therefore, a system combining efficient inventory management with rapid notification of recommendations is needed.

[0822] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0823] In this invention, the server includes: means for generating a demand forecasting model based on a user's purchase history, browsing history, and search history; means for optimizing inventory using the generated demand forecasting model; means for generating product recommendations based on the user's browsing history and search history; means for notifying the user's terminal of the generated recommendation information; means for receiving purchase requests from the user, selecting an optimal logistics route, and issuing delivery instructions; and means for providing appropriate product recommendations based on the user's shopping behavior and sending push notifications in real time using the smart device to increase purchasing motivation. This enables both inventory management and recommendation notifications to be performed with high accuracy and efficiency, improving the user experience.

[0824] "User purchase history" is a record of information about products purchased by the user in the past, such as purchase date and time, and purchase quantity.

[0825] A "demand forecasting model" is a mathematical model for predicting future demand based on a user's past purchasing history, browsing history, and search history.

[0826] "Inventory optimization" is the process of adjusting product inventory at each warehouse or store to meet demand using a demand forecasting model.

[0827] "Product recommendation" is a function that suggests products that may be of interest to users based on their browsing history and search history.

[0828] "User terminal" refers to a digital device used by a user, such as a smartphone or tablet.

[0829] A "purchase request" is information sent to the system when a user indicates their intention to purchase a product.

[0830] "Logistics route" is a concept that refers to the optimal route or means for delivering goods.

[0831] "Delivery Instructions" refers to the process of issuing specific instructions regarding the delivery of goods.

[0832] "Shopping behavior" refers to a series of actions taken by a user, such as browsing, searching, and purchasing products.

[0833] "Smart devices" is a general term for devices that can connect to the Internet, such as smartphones, tablets, and smartwatches.

[0834] "Push notification" is a function that sends information from a server to a user's smart device in real time.

[0835] The system for implementing this invention encompasses a series of processes including automated data collection, demand forecasting, inventory management, product recommendations, and user notifications. Specifically, it is realized in the following manner.

[0836] The server first collects users' purchase, browsing, and search histories from a database. This data includes detailed information such as user ID, purchased items, purchase date and time, and purchase quantity. Next, the collected data is preprocessed to generate a demand forecasting model. This model generation uses machine learning algorithms (e.g., linear regression and time series analysis).

[0837] The generated demand forecasting model predicts future demand, and inventory is optimized based on that. The server periodically updates the demand forecasting model to dynamically adjust inventory levels. This prevents overstocking and shortages and enables efficient resource management.

[0838] Furthermore, the server monitors the user's browsing and search history in real time and generates optimal product recommendations. This recommendation information is then pushed to the user via their smart device. For example, if "User A" searches for "smartphone cases" on a website, the server will recommend new products in the same category and notify the user's device.

[0839] When a user submits a purchase request, the server receives the information, selects the optimal logistics route, and issues delivery instructions. Once delivery begins, the server monitors the delivery status in real time and updates the delivery status on the user's device, allowing the user to track the progress of the product as it is delivered.

[0840] The hardware used includes servers, smart devices (smartphones, tablets, etc.), and database systems. The software includes machine learning libraries (e.g., scikit-learn), Python scripts for data collection and preprocessing, and notification services such as Firebase.

[0841] To generate a specific program, you can use the following prompt:

[0842] 1. "Generate Python code to forecast demand for the next month based on purchasing data from the past six months."

[0843] 2. "Create a program using Firebase to send recommendation notifications to users based on their purchase and browsing history."

[0844] This allows the server, terminal, and user to interact with each other, enabling both inventory management and recommendation notifications to be performed with high accuracy and efficiency, improving the user experience.

[0845] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0846] Step 1:

[0847] The server collects the user's purchase history, browsing history, and search history from a database. The input is the user ID, and the output is the user's various historical data (purchased items, purchase date and time, purchased quantity, viewed items, search keywords, etc.). This allows the past behavioral data of a specific user to be obtained.

[0848] Step 2:

[0849] The server pre-processes the collected data, cleaning the data (e.g., removing incomplete data and correcting outliers) and standardizing it. The input is the user history data collected in step 1, and the output is formatted data, which facilitates subsequent data analysis.

[0850] Step 3:

[0851] The server uses the preprocessed data to generate a demand forecasting model. It trains the model using machine learning algorithms (e.g., linear regression or time series analysis). The input is the preprocessed data, and the output is the demand forecasting model. This creates a foundation for forecasting future demand.

[0852] Step 4:

[0853] The server optimizes the planned inventory based on the generated demand forecast model. This involves a process of calculating how much of a specific product is needed at a specific time. The input is the demand forecast model, and the output is the calculation result of the optimal inventory amount. This makes inventory management more efficient.

[0854] Step 5:

[0855] The server monitors the user's real-time browsing and search history and generates product recommendations. The input is real-time user behavior data, and the output is a list of recommended products. This allows the server to recommend the most suitable products to the user.

[0856] Step 6:

[0857] The terminal notifies the user of the recommendation information sent from the server. This notification is performed using the push notification function of the smart device. The input is a list of recommended products, and the output is a notification message. This allows the user to receive new and recommended products in real time.

[0858] Step 7:

[0859] The user submits a purchase request, which is sent to the server with the purchase request data as input, thereby conveying the user's intention to purchase to the system.

[0860] Step 8:

[0861] The server selects the optimal logistics route based on the received purchase request and issues delivery instructions. The input is the purchase request data, and the output is logistics instructions and a delivery route. This ensures that purchased items are delivered to the user in the most efficient manner.

[0862] Step 9:

[0863] The server monitors the delivery status in real time and updates the delivery status to the user's device. The input is delivery progress data and the output is the updated delivery status, allowing the user to always know what stage their product is in.

[0864] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0865] This invention combines a system that performs demand forecasting based on users' purchase history, browsing history, and search history, optimizes inventory, and provides product recommendations with an emotion engine that recognizes users' emotions. Below, we will create a program for this system and explain the specific processing content.

[0866] Demand forecast data collection and analysis

[0867] The server collects past purchase history, browsing history, and search history from the user database. This includes detailed information such as purchase date and time, purchased product category, purchase quantity, and user ID. The server then preprocesses the collected data, cleaning the data, filling in missing values, and normalizing the data. The preprocessed data is then input into a machine learning algorithm (e.g., regression analysis or time series analysis) to generate a demand forecasting model. Future demand is predicted based on the generated forecasting model.

[0868] Examples:

[0869] The server analyzes "User A's" purchasing data for the past six months and predicts that "User A" is likely to purchase a particular electronic device again in the next month.

[0870] Collecting and analyzing emotional data

[0871] The server uses an emotion engine to recognize the user's emotions. It analyzes the user's facial expression data, voice tone, keyboard input, etc. sent from the device to identify the user's emotional state. This allows it to understand how the user is feeling at that time.

[0872] Examples:

[0873] While "User B" is browsing products on his smartphone, emotional data is collected using the camera and microphone. The emotion engine recognizes that User B is in an "excited" state.

[0874] Inventory management

[0875] The server combines the generated demand forecast model with user emotion data to calculate the necessary inventory levels. Based on this information, it allocates optimal inventory to each warehouse and regularly updates the inventory status to prevent shortages and overstocks.

[0876] Examples:

[0877] The server predicts that demand for electronic devices will increase within this month, and since user sentiment data indicates a positive reaction, it instructs the Tokyo warehouse to stock additional stock of those electronic devices.

[0878] Recommendation function

[0879] The server runs a recommendation engine based on the user's browsing history, search history, and emotional data. For example, it uses collaborative filtering or content-based filtering to select the most suitable products for the user. The generated recommendation information is sent to the user's device.

[0880] Examples:

[0881] When "User C" searches for "smartphone cases" on a website, the server analyzes the emotional data of "excitement." Taking that reaction into consideration, the server recommends "waterproof smartphone cases" as a product in the same category.

[0882] User Notifications and Interactions

[0883] The device will notify the user of the recommendation information received from the server in real time. The user can check the notification, and if they are interested, they can go to the detailed information page and proceed with the purchase process.

[0884] Examples:

[0885] User D's device displays a notification for "Recommended Product: Waterproof Smartphone Case," and User D clicks on the notification to go to the details page.

[0886] Checkout and logistics

[0887] When the user presses the purchase button, the terminal sends the purchase information to the server. The sent information includes the product ID, user ID, delivery address information, etc. The server confirms the received purchase information and checks the inventory status. If inventory is available, the order is confirmed and the optimal logistics route is selected. The server issues delivery instructions to the warehouse and updates the status on the user's terminal while monitoring the delivery status in real time.

[0888] Examples:

[0889] When "User E" presses the purchase button, the server retrieves the product from the nearest warehouse, selects the fastest delivery method, and begins delivery. During delivery, the server monitors the delivery status and notifies User E's device of the status as it progresses.

[0890] Through this series of processes, the present invention minimizes logistics lead time and user decision-making lead time, and further optimizes inventory. Furthermore, by periodically updating the generated demand forecast model and performing dynamic inventory management, more accurate supply and demand adjustments become possible. By utilizing user emotion data, more personalized services can be provided, improving the user experience.

[0891] The processing flow will be explained below.

[0892] Step 1:

[0893] The server collects data such as past purchase history, browsing history, and search history from the user database. The data collected includes purchase date and time, purchased product category, purchased quantity, and user ID.

[0894] Step 2:

[0895] The server preprocesses the acquired data, which includes data cleaning (e.g., removing incomplete data), imputing missing values ​​(e.g., imputing with the mean), and normalizing the data (e.g., scaling).

[0896] Step 3:

[0897] The server then inputs the preprocessed data into a machine learning algorithm to generate a demand forecasting model, for example, using regression analysis or time series analysis to build a model for predicting future demand.

[0898] Step 4:

[0899] The server periodically uses the generated demand forecasting model to predict future demand for specific products, calculates the necessary inventory levels based on this, and allocates optimal inventory to each warehouse.

[0900] Step 5:

[0901] The server uses an emotion engine to recognize the user's emotions and collects emotion data from the user's device, specifically analyzing facial expression data from the camera, voice tone from the microphone, and keyboard input speed and strength.

[0902] Step 6:

[0903] The server uses an emotion engine to analyze the collected emotional data and identify the emotional state the user is in at that time, categorizing emotions such as "excitement," "sadness," and "happiness."

[0904] Step 7:

[0905] The server incorporates the collected emotion data into the demand forecasting model to improve the accuracy of the forecast, thereby more accurately predicting product demand, which is strongly influenced by user emotions.

[0906] Step 8:

[0907] The server runs a recommendation engine based on the user's browsing history, search history, and emotional data, using collaborative filtering and content-based filtering, for example, to select the best products for the user.

[0908] Step 9:

[0909] The server transmits the generated recommendation information to the user's terminal, which notifies the user of the recommendation information in real time.

[0910] Step 10:

[0911] The user checks the recommendation notification displayed on the device and clicks on the detailed information page if they are interested. For example, a notification such as "Recommended product: Waterproof smartphone case" will be displayed.

[0912] Step 11:

[0913] When a user presses the purchase button, the device sends the purchase information to the server, including the product ID, user ID, and delivery address information.

[0914] Step 12:

[0915] The server confirms the received purchase information and checks the inventory status. If the item is available, it confirms the order and selects the optimal logistics route.

[0916] Step 13:

[0917] The server issues delivery instructions to the warehouse, for example, shipping the product from the nearest warehouse and selecting the fastest delivery method.

[0918] Step 14:

[0919] The server monitors the delivery status in real time and updates the status to the user's device, for example, notifying the user of the status when the product has been shipped, when it is being delivered, and when delivery has been completed.

[0920] Through these processing steps, the present invention minimizes logistics lead time and user decision-making lead time, optimizing inventory. By utilizing user emotion data, it is possible to provide more personalized services and improve the user experience.

[0921] Example 2

[0922] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0923] Many modern inventory management systems and product recommendation systems support demand forecasting and recommendations based on users' purchase, browsing, and search histories. However, they are unable to consider the user's emotional state, limiting their ability to optimize the user experience. Furthermore, inventory optimization is static, making it difficult to respond to dynamic demand fluctuations, leading to problems such as inventory shortages and overstocks. Furthermore, there is a need for more personalized recommendations by utilizing user emotional data.

[0924] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0925] In this invention, the server includes means for generating a demand forecasting model based on a user's purchase history, browsing history, and search history, means for optimizing inventory using the generated demand forecasting model, means for analyzing emotion data collected from the user's terminal, means for optimizing inventory and product recommendations using the analyzed emotion data, means for notifying the user's terminal of the generated recommendation information, and means for receiving purchase requests from the user, selecting an optimal logistics route, and issuing delivery instructions, thereby enabling personalized services and dynamic inventory management based on the user's emotions.

[0926] "User purchase history" is a record of products purchased by a user in the past, and specifically includes detailed information such as the purchase date and time, purchased product category, purchased quantity, and user ID.

[0927] "Browse history" is a record of web pages and products that a user has viewed in the past, and is data that indicates information such as the date and time of browsing, the number of times of browsing, and the products that have been viewed.

[0928] "Search history" is a record of a user's past search queries and search results, and is data that includes information such as search date and time, search keywords, and search result click history.

[0929] A "demand forecasting model" is a machine learning model for predicting future demand based on collected data, and is generated using algorithms such as regression analysis and time series analysis.

[0930] "Inventory optimization" is the process of using demand forecasting models to adjust inventory levels in anticipation of future demand, preventing shortages and overstocks.

[0931] "Emotion data" refers to data that reflects the user's emotional state, and specifically refers to information obtained from facial expression data, voice tone, keyboard input, and the like.

[0932] "Product recommendations" are recommendation information that suggests appropriate products to users, and are generated based on the user's purchasing history, emotional data, etc.

[0933] "Generated recommendation information" is information that suggests optimal products and services to a user, generated based on the user's purchase history, browsing history, search history, and emotional data.

[0934] A "purchase request" is information indicating a user's intention to purchase a product or service, and specifically includes a product ID, user ID, delivery address information, and the like.

[0935] The "optimal logistics route" is the most effective delivery route for efficiently delivering orders, and is selected taking into consideration factors such as cost, time, and inventory status.

[0936] This invention combines a system that performs demand forecasting based on a user's purchase history, browsing history, and search history, optimizes inventory, and provides product recommendations with an emotion engine that recognizes the user's emotions. Specific embodiments of this system are described below.

[0937] First, the server collects purchase history, browsing history, and search history from the user database. The collected data includes details such as purchase date and time, purchased product category, purchased quantity, and user ID. This data is then temporarily stored in cloud storage (e.g., Amazon S3). The software used at this stage includes database queries and cloud storage APIs.

[0938] Next, the server preprocesses the collected data. Specific processes include filling in missing values, removing outliers, and normalizing the data. This process uses Python's Pandas library and NumPy. The preprocessed data is then input into a machine learning algorithm (e.g., regression analysis, time series analysis) to generate a demand forecasting model. Machine learning libraries such as TensorFlow and scikit-learn are used at this stage.

[0939] The device uses a camera and microphone to collect user emotional data and transmits it to a server. Emotional data includes facial expressions, voice tone, and keyboard input. The collected data is analyzed by an emotion engine (e.g., NVIDIA DeepStream, Face++) to identify the user's emotional state. This information is used for demand forecasting and recommendation optimization.

[0940] The server optimizes inventory based on the generated demand forecasting model and emotion data. Specifically, it uses the demand forecasting model to forecast future demand and calculate the necessary inventory volume. Based on the calculation results, inventory is appropriately allocated through an inventory management system or ERP system (e.g., SAP, Oracle).

[0941] The server then recommends products based on the user's purchase history, browsing history, search history, and emotional data. This process uses recommendation engines (e.g., Amazon Personalize, Google RE) that use collaborative filtering or content-based filtering. The generated recommendation information is sent to the device in real time.

[0942] When a user purchases a product based on the recommended information, the device sends the purchase information to the server. This information includes the product ID, user ID, delivery address information, etc. The server then verifies the received purchase information and checks the inventory status. For example, if the product is available in stock, the server confirms the order and selects the optimal logistics route. This is done using a delivery management system (e.g., UPS WorldShip, FedEx Ship Manager).

[0943] Once delivery begins, the server monitors the delivery status in real time and notifies the user of the status as it progresses. This process improves the user experience and streamlines inventory management.

[0944] Prompt Sentence Examples

[0945] "Please explain a system that forecasts demand based on users' past purchasing history and optimizes inventory."

[0946] "Please explain a system that uses user emotional data to recommend products."

[0947] This invention enables personalized service and dynamic inventory management based on user emotions.

[0948] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0949] Step 1:

[0950] The server collects purchase history, browsing history, and search history data from the user database. The input data includes the purchase date and time, purchased product category, purchased quantity, user ID, etc. This data is stored in cloud storage (e.g., Amazon S3). Specifically, it executes a database query and uploads the retrieved data to cloud storage. The output of this step is the data stored in cloud storage.

[0951] Step 2:

[0952] The server preprocesses the collected data. Specifically, it imputes missing values, removes outliers, and normalizes the data. The software used includes Python's Pandas library and NumPy. The input data is stored in cloud storage, and the output data is the preprocessed, clean data.

[0953] Step 3:

[0954] The server inputs the preprocessed data into a machine learning algorithm to generate a demand forecasting model. This uses algorithms such as regression analysis and time series analysis. This process uses machine learning libraries such as TensorFlow and scikit-learn. The input data is preprocessed and clean, and the output is a demand forecasting model.

[0955] Step 4:

[0956] The device uses a camera and microphone to collect the user's emotional data and transmits it to a server. Emotional data includes facial expressions, voice tones, keyboard input, etc. The input data is emotional data acquired through the camera and microphone, and the output is emotional data transmitted to the server. Specific operations include collecting and transmitting sensor data.

[0957] Step 5:

[0958] The server analyzes the emotion data sent from the device. This analysis is performed using an emotion engine (e.g., NVIDIA DeepStream, Face++). The input data is the emotion data sent from the device, and the output is the analyzed emotional state.

[0959] Step 6:

[0960] The server combines the generated demand forecasting model with the analyzed emotion data to optimize inventory. Specifically, it uses the demand forecasting model to forecast future demand and calculates the required inventory volume. Based on the calculation results, inventory is appropriately allocated via an inventory management system or ERP system (e.g., SAP, Oracle). The input data are the demand forecasting model and emotion data, and the output is an optimized inventory list.

[0961] Step 7:

[0962] The server recommends products based on the user's purchase history, browsing history, search history, and emotional data. A recommendation engine (e.g., Amazon Personalize, Google RE) using collaborative filtering or content-based filtering is used. The generated recommendation information is sent to the device in real time. The input data are the aforementioned history data and emotional data, and the output is recommendation information.

[0963] Step 8:

[0964] When a user purchases a product based on the recommendation information, the terminal sends the purchase information to the server. The information sent includes the product ID, user ID, delivery address information, etc. A specific action is to press the purchase button. The input data is the execution data of the purchase page, and the output is the purchase information sent to the server.

[0965] Step 9:

[0966] The server confirms the received purchase information and checks the inventory status. If inventory is available, it confirms the order and selects the optimal logistics route. This is done using a delivery management system (e.g., UPS WorldShip, FedEx Ship Manager). The input data are purchase information and inventory information, and the output is the confirmed order and the selected logistics route.

[0967] Step 10:

[0968] The server monitors the delivery status in real time and notifies the user's device of the status as it progresses. Utilizing the delivery management system, it tracks each stage of delivery and updates the status as appropriate. The input data is delivery status information from the delivery management system, and the output is a status notification sent to the user's device.

[0969] (Application example 2)

[0970] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0971] Conventional demand forecasting systems are based solely on a user's purchasing history, browsing history, and search history, and do not take into account the user's real-time emotional state, making it difficult to provide personalized recommendations. Furthermore, they have the problem of not being able to adequately address the user's decision-making process or improve the user experience. The purpose of this invention is to solve these problems.

[0972] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for generating a demand forecasting model based on a user's purchase history, browsing history, and search history, means for optimizing inventory using the generated demand forecasting model, means for generating content recommendations based on the user's browsing history, search history, and emotional data, means for notifying the user's visual display device of the generated recommendation information, and means for receiving a selection request from the user and selecting the optimal content to be displayed next. This enables personalized recommendations and inventory management that take into account the user's real-time emotional state.

[0973] "User purchase history" is a collection of information about products and services that a user has purchased in the past.

[0974] "Browsing history" is a record of pages and content that a user has previously viewed on the Internet or in applications.

[0975] "Search history" is a history of keywords and phrases that a user has searched for on the Internet or in an application.

[0976] A "demand forecasting model" is an algorithm or mathematical model for predicting future demand based on past user behavior data.

[0977] "Inventory optimization" is the process of optimizing supply-side inventory levels based on demand forecasts.

[0978] "Emotion data" is information that identifies a user's emotional state in real time by analyzing their facial expressions and vocal tone.

[0979] "Content recommendation" is a function that suggests appropriate content based on a user's behavioral history and emotional data.

[0980] A "visual display device" is a device that visually displays information to a user, and includes smart glasses and digital displays.

[0981] A "selection request" is a request for a user to make a selection or designation from the presented recommendation information.

[0982] This invention is a system that forecasts demand and recommends content based on users' purchasing history, browsing history, search history, and emotional data. This system consists of three elements: a server, a terminal, and a user.

[0983] First, the server collects users' purchasing, browsing, and search histories and generates a demand forecasting model based on this information. This forecasting model is used to analyze users' consumption behavior and forecast future demand. The server also has functions for preprocessing the collected data, cleaning the data, imputing missing values, and normalizing the data. For this purpose, the server mainly uses machine learning libraries such as Scikit-learn.

[0984] The server also combines the user's browsing history, search history, and emotion engine to collect real-time emotional data. This emotional data is collected via the camera and microphone and analyzed using an emotion recognition model built with Keras. The server identifies the user's emotions and recommends content based on the results.

[0985] The device used by the user includes smart glasses that can capture the user's facial expressions and tone of voice in real time, and the captured emotional data is sent to a server where it is incorporated into a demand forecasting model and recommendation engine.

[0986] For example, if a user is watching a movie through smart glasses, the emotion engine can determine that the user is "excited" based on their facial expressions and voice. Based on the user's viewing history, the server then recommends movies and content that the user is likely to watch next, and displays them on the smart glasses' display. This process improves the user experience and provides more accurate recommendations.

[0987] In addition, the demand forecasting model is regularly updated, enabling dynamic inventory management. For example, by combining past data with the latest sentiment data, demand can be forecast more accurately and appropriate inventory levels can be maintained. This shortens logistics lead times and minimizes user purchasing lead times.

[0988] The main hardware and software required to implement this system are as follows:

[0989] Hardware:

[0990] Smart glasses (with camera and microphone)

[0991] Server (cloud environment recommended)

[0992] software:

[0993] OpenCV (face detection and facial expression data collection)

[0994] Scikit-learn (data preprocessing and demand forecasting model)

[0995] Keras (emotion recognition model)

[0996] Server-side API (obtaining user data)

[0997] For an example of how to use a generative AI model, the prompt would be:

[0998] "By aligning a user's viewing history with their current emotional state, we can recommend content they're likely to watch next. For example, if a user is excited and watching an action movie, what specific action movie should we recommend?"

[0999] This will create a system that integrates user behavioral data and emotional data to provide more personalized recommendations.

[1000] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1001] Step 1:

[1002] The server collects the user's purchase history, browsing history, and search history. For this, detailed information such as purchase date and time, purchased product category, and purchase quantity is extracted from the database based on the user ID. The input data is the user's behavior history information, and the output data is preprocessed user behavior history data.

[1003] Step 2:

[1004] The server preprocesses the collected data. Specifically, it performs data cleaning, missing value completion, data normalization, etc. This process formats the input data for the machine learning algorithm. The input data is the collected user behavior history data, and the output data is the preprocessed data.

[1005] Step 3:

[1006] The server uses the preprocessed data to generate a demand forecasting model. During this process, it uses a machine learning library such as Scikit-learn to perform regression analysis and time series analysis. The input data is the preprocessed data, and the output data is the demand forecasting model. The generated model is used to predict future demand.

[1007] Step 4:

[1008] The device captures the user's browsing history and search history, as well as facial and voice data in real time. This data is collected through a camera and microphone. The input data is real-time facial and voice data, and the output data is the captured raw data.

[1009] Step 5:

[1010] The server receives the facial expression and voice data sent from the device and analyzes them using an emotion recognition engine. Using an emotion recognition model built with Keras, the server identifies the user's emotional state. The input data is the captured facial expression and voice data, and the output data is the identified emotion data.

[1011] Step 6:

[1012] The server operates a recommendation engine based on the identified emotion data and the user's browsing and search history. Collaborative filtering and content-based filtering are used to generate optimal content. The input data are emotion data and behavioral history data, and the output data is recommendation information.

[1013] Step 7:

[1014] The terminal notifies the visual display device of the recommendation information received from the server. The user can check the notified recommendation information and move to a detailed information page if they are interested. The input data is the recommendation information, and the output data is the information displayed visually to the user.

[1015] Step 8:

[1016] When the user selects a content, a selection request is sent to the server. The input data is the user's selection request, and the output data is the request information sent to the server.

[1017] Step 9:

[1018] Based on the received selection request, the server selects the optimal content to display next and notifies the terminal again. This allows the user to seamlessly view the next content. The input data is the request information, and the output data is the content information to be displayed next.

[1019] In this way, through specific actions at each step, the system provides users with personalized content recommendations and performs real-time sentiment analysis and demand forecasting.

[1020] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1021] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1022] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1023] [Fourth embodiment]

[1024] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1025] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1026] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1027] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1028] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1029] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1030] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1031] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1032] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1033] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1034] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1035] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1036] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1037] The present invention is a system that performs demand forecasting based on a user's purchase history, browsing history, and search history, and provides inventory optimization and product recommendations. Below, we will create a program for this system and explain the specific processing content.

[1038] Demand forecast data collection and analysis

[1039] The server collects data such as past purchase history, browsing history, and search history from the user database. This includes detailed information such as purchase date and time, purchased product category, purchased quantity, and user ID. The server preprocesses the acquired data and generates a demand forecasting model using machine learning algorithms (e.g., regression analysis and time series analysis). Future demand is estimated based on the generated forecasting model.

[1040] Examples:

[1041] The server analyzes "User A's" purchasing data for the past six months and predicts that "User A" is likely to purchase a particular electronic device again in the next month.

[1042] Inventory management

[1043] The server uses the generated demand forecast model to calculate the necessary inventory levels and allocate optimal stock to each warehouse. It periodically updates the inventory status and dynamically manages it to prevent shortages and excesses.

[1044] Examples:

[1045] The server predicts that demand for electronic devices will increase within the month and instructs the Tokyo warehouse to prepare additional inventory to accommodate this.

[1046] Recommendation function

[1047] The server collects users' browsing and search histories in real time, and uses a recommendation engine to select the most suitable products based on that information.The server then notifies the user of the selected recommendation information on their device.

[1048] Examples:

[1049] If "User B" searches for "smartphone cases" on a website, the server will recommend new products in the same case category.

[1050] User Notifications and Interactions

[1051] The device notifies the user of the recommendation information received from the server in real time. The user then moves to the page with detailed information and, if interested, proceeds to the purchase process.

[1052] Examples:

[1053] User C's device displays a notification for "Recommended Product: Waterproof Smartphone Case," and User C clicks on the notification to go to the details page.

[1054] Checkout and logistics

[1055] When the user presses the purchase button, the device sends the purchase information to the server. The server then checks the received purchase information, selects the optimal logistics route, and issues delivery instructions. The server also monitors the delivery status in real time and updates the delivery status on the user's device.

[1056] Examples:

[1057] When "User D" presses the purchase button, the server retrieves the product from the nearest warehouse, selects the fastest delivery method, and begins delivery to the user's address. During delivery, the server monitors the delivery status and notifies User D's device of the status as it progresses.

[1058] By executing such a program, the present invention aims to minimize logistics lead time and user decision-making lead time, further optimizing inventory and improving the user experience. Furthermore, by periodically updating the generated demand forecast model and performing dynamic inventory management, more accurate supply and demand adjustments become possible. Furthermore, by collecting user action logs and using them to generate recommendations for future purchases, more personalized services can be provided.

[1059] The processing flow will be explained below.

[1060] Step 1:

[1061] The server collects data such as past purchase history, browsing history, and search history from the user database. For example, it comprehensively obtains each user's purchase date and time, purchased product category, purchased quantity, user ID, etc.

[1062] Step 2:

[1063] The server preprocesses the acquired data, specifically by cleaning the data, filling in missing data, and normalizing the data, thereby converting the data into a state that can be analyzed.

[1064] Step 3:

[1065] The server then inputs the preprocessed data into a machine learning algorithm to generate a demand forecasting model, for example, using regression analysis or time series analysis to build a model that predicts future demand.

[1066] Step 4:

[1067] The server uses the generated demand forecasting model to predict future demand for specific products, calculates the necessary inventory levels based on the forecast, and allocates optimal inventory to each warehouse.

[1068] Step 5:

[1069] The server collects users' browsing and search history in real time, for example, capturing their actions when they search for a specific product on a website or view a detail page.

[1070] Step 6:

[1071] The server uses the collected real-time data to run a recommendation engine, for example, using collaborative filtering or content-based filtering to select the best products for the user.

[1072] Step 7:

[1073] The server sends the generated recommendation information to the user's device, which receives the information and notifies the user in real time.

[1074] Step 8:

[1075] The user checks the recommendation notification displayed on the device and clicks on the detailed information page if they are interested. For example, they receive a notification about a new product that is perfect for them: a waterproof smartphone case.

[1076] Step 9:

[1077] When the user presses the purchase button, the terminal sends the purchase information to the server, including the product ID, user ID, and delivery address information.

[1078] Step 10:

[1079] The server confirms the received purchasing information and checks the inventory status. If the inventory is available, it confirms the order and selects the optimal logistics route.

[1080] Step 11:

[1081] The server issues delivery instructions to the warehouse, for example, shipping the product from the nearest warehouse and selecting the fastest delivery method.

[1082] Step 12:

[1083] The server monitors the delivery status in real time and updates the status to the user's device, such as when the product has been shipped, when it is in transit, or when delivery has been completed.

[1084] Example 1

[1085] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1086] Conventional systems do not adequately forecast demand or optimize inventory based on users' purchasing, browsing, and search histories, making it difficult to detect inventory surpluses and shortages and to recommend appropriate products to users.Furthermore, there is a problem of reduced user convenience due to the lack of real-time monitoring of delivery status and the lack of proper notification of delivery status.

[1087] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1088] In this invention, the server includes means for generating a demand forecasting model based on a user's purchase history, browsing history, and search history, means for optimizing inventory using the generated demand forecasting model, means for generating product recommendations based on the user's browsing history and search history, means for notifying the user's terminal of the generated recommendation information, means for receiving purchase requests from the user, selecting an optimal logistics route, and issuing delivery instructions, and means for monitoring delivery status in real time and notifying the user's terminal of the delivery status. This prevents inventory surpluses and shortages, recommends optimal products to the user, and enables efficient delivery.

[1089] "User" refers to a person who uses the system and provides data such as purchase history, browsing history, and search history.

[1090] "Purchase history" includes information about products purchased by a user in the past, and refers to detailed information such as the purchase date and time, purchased product category, purchased quantity, and user ID.

[1091] "Browser history" refers to data that includes information about products viewed by a user on websites or within applications.

[1092] "Search history" refers to data that includes keywords that users searched for within the system and information about search results.

[1093] "Demand forecasting model" refers to a machine learning algorithm used to predict future demand for a product based on data such as purchase history, browsing history, and search history.

[1094] "Inventory optimization" refers to the use of demand forecasting models to allocate optimal amounts of inventory in each warehouse and prevent shortages and excess inventory.

[1095] "Product recommendation" refers to a method of selecting and suggesting optimal products to users based on their browsing history and search history.

[1096] "Notification" refers to the act of sending information such as recommendation information and delivery status from the server to the user's terminal.

[1097] The "logistics route" refers to the optimal logistics route for efficiently delivering the products purchased by the user.

[1098] "Delivery status" refers to the current status of the product purchased by the user in the delivery process, and includes states such as shipped, in delivery, and delivery completed.

[1099] The present invention provides a system for performing demand forecasting based on a user's purchase history, browsing history, and search history, optimizing inventory, and providing product recommendations. Specific embodiments of this system are described below.

[1100] First, the server collects data on past purchase history, browsing history, and search history from the user database. This data includes detailed information such as purchase date and time, purchased product category, purchased quantity, and user ID. The server then preprocesses the acquired data, for example, removing duplicates and filling in missing values. After preprocessing is complete, the server uses machine learning algorithms (e.g., regression analysis and time series analysis) to generate a demand forecasting model. This demand forecasting model is saved and used to forecast future demand.

[1101] As a specific example, by analyzing the purchasing data of "User A" for the past six months, it is possible to predict the likelihood that "User A" will purchase a particular electronic device again in the next month.

[1102] The server then uses the generated demand forecast model to calculate the required inventory and allocate optimal inventory to each warehouse. It periodically updates inventory status and dynamically manages it to prevent shortages and excesses. For example, the server may predict that demand for electronic devices will increase within the current month for the Tokyo warehouse and instruct it to allocate additional inventory.

[1103] Furthermore, the server collects the user's browsing history and search history in real time and uses a recommendation engine to select the most suitable product. This recommendation information is then sent to the user's device. For example, if "User B" is searching for a smartphone case, the server will recommend a new product.

[1104] When a user receives the recommendation information, they go to the detailed information page and, if they are interested, proceed to the purchase process. For example, "User C"'s device may display a notification for "Recommended Product: Waterproof Smartphone Case," and User C may go to the detailed information page.

[1105] Additionally, when a user presses the purchase button, the terminal sends the purchase information to the server. The server confirms the received purchase information, selects the optimal logistics route, and issues delivery instructions. It monitors the delivery situation in real time and updates the delivery status on the user's terminal. As a specific example, when "User D" presses the purchase button, the server retrieves the product from the nearest warehouse, selects the fastest delivery method, and notifies "User D's" terminal of the delivery status as the process progresses.

[1106] An example of a prompt for a generative AI model is as follows:

[1107] "Based on User A's purchasing data from the past six months, please predict the products that he is likely to purchase in the next month."

[1108] "If User B is searching for a smartphone case, please recommend related products."

[1109] "Please forecast which warehouses will see increased demand for electronic equipment this month and issue instructions for deploying the necessary inventory."

[1110] As described above, the system of the present invention aims to minimize logistics lead time and user decision-making lead time, optimize inventory, and improve the user experience. By periodically updating the generated demand forecast model and performing dynamic inventory management, more accurate supply and demand adjustments become possible. Furthermore, by collecting user action logs and using them to generate recommendations for future purchases, more personalized services can be provided.

[1111] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1112] Step 1:

[1113] The server collects purchase history, browsing history, and search history data from a user database.

[1114] Input: User database

[1115] Output: Purchase history, browsing history, search history

[1116] Specific operation: The server uses an SQL query to retrieve the purchase history, browsing history, and search history of "User A" and "User B" from the user database for the past six months.

[1117] Step 2:

[1118] The server pre-processes the collected data.

[1119] Input: purchase history, browsing history, search history

[1120] Output: Preprocessed data

[1121] Specific operation: The server removes duplicate data, imputes missing values, and converts the data into the required format. For example, if there are missing values, it imputes them with the average value.

[1122] Step 3:

[1123] The server uses the pre-processed data to train a demand forecasting model.

[1124] Input: Preprocessed data

[1125] Output: Demand forecast model

[1126] Specific operation: The server uses Python's scikit-learn to perform regression analysis and time series analysis to train a demand forecasting model, and saves this model for use in future demand forecasts.

[1127] Step 4:

[1128] The server uses the generated demand forecast model to optimize inventory.

[1129] Input: Demand forecast model

[1130] Output: Inventory allocation plan

[1131] Specific operation: The server calculates the amount of inventory required for each warehouse based on the demand forecast model. For example, it instructs the Tokyo warehouse to increase the inventory of electronic devices.

[1132] Step 5:

[1133] The server generates product recommendations based on the user's browsing history and search history.

[1134] Input: Real-time collected browsing history, search history

[1135] Output: Recommendation information

[1136] Specific operation: When a user searches for "smartphone case," the server selects new products in the same category and generates recommendation information.

[1137] Step 6:

[1138] The terminal notifies the user of the recommendation information received from the server.

[1139] Input: Recommendation information

[1140] Output: Notification message

[1141] Specific operation: User C's device displays a notification such as "Recommended product: Waterproof smartphone case," and User C moves to the details page.

[1142] Step 7:

[1143] The terminal transmits a purchase request from the user to the server.

[1144] Input: Purchase Request

[1145] Output: Purchase information

[1146] Specific operation: When the user presses the purchase button, the device sends that information to the server, which then selects the optimal logistics route and issues delivery instructions.

[1147] Step 8:

[1148] The server monitors the delivery status in real time and notifies the user's terminal of the delivery status.

[1149] Input: Delivery status from logistics company

[1150] Output: Delivery status update

[1151] Specific operation: The server uses the logistics company's API to monitor delivery status and notifies the user's device of statuses such as "shipped," "in transit," and "delivery completed."

[1152] (Application example 1)

[1153] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1154] The present invention relates to a system for improving the accuracy of demand forecasting and product recommendations based on users' purchase history, browsing history, and search history, optimizing inventory, and improving the user experience. With conventional technologies, demand forecasting and inventory management are time-consuming and often result in delayed notification of recommendation information. Therefore, a system combining efficient inventory management with rapid notification of recommendations is needed.

[1155] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1156] In this invention, the server includes: means for generating a demand forecasting model based on a user's purchase history, browsing history, and search history; means for optimizing inventory using the generated demand forecasting model; means for generating product recommendations based on the user's browsing history and search history; means for notifying the user's terminal of the generated recommendation information; means for receiving purchase requests from the user, selecting an optimal logistics route, and issuing delivery instructions; and means for providing appropriate product recommendations based on the user's shopping behavior and sending push notifications in real time using the smart device to increase purchasing motivation. This enables both inventory management and recommendation notifications to be performed with high accuracy and efficiency, improving the user experience.

[1157] "User purchase history" is a record of information about products purchased by the user in the past, such as purchase date and time, and purchase quantity.

[1158] A "demand forecasting model" is a mathematical model for predicting future demand based on a user's past purchasing history, browsing history, and search history.

[1159] "Inventory optimization" is the process of adjusting product inventory at each warehouse or store to meet demand using a demand forecasting model.

[1160] "Product recommendation" is a function that suggests products that may be of interest to users based on their browsing history and search history.

[1161] "User terminal" refers to a digital device used by a user, such as a smartphone or tablet.

[1162] A "purchase request" is information sent to the system when a user indicates their intention to purchase a product.

[1163] "Logistics route" is a concept that refers to the optimal route or means for delivering goods.

[1164] "Delivery Instructions" refers to the process of issuing specific instructions regarding the delivery of goods.

[1165] "Shopping behavior" refers to a series of actions taken by a user, such as browsing, searching, and purchasing products.

[1166] "Smart devices" is a general term for devices that can connect to the Internet, such as smartphones, tablets, and smartwatches.

[1167] "Push notification" is a function that sends information from a server to a user's smart device in real time.

[1168] The system for implementing this invention encompasses a series of processes including automated data collection, demand forecasting, inventory management, product recommendations, and user notifications. Specifically, it is realized in the following manner.

[1169] The server first collects users' purchase, browsing, and search histories from a database. This data includes detailed information such as user ID, purchased items, purchase date and time, and purchase quantity. Next, the collected data is preprocessed to generate a demand forecasting model. This model generation uses machine learning algorithms (e.g., linear regression and time series analysis).

[1170] The generated demand forecasting model predicts future demand, and inventory is optimized based on that. The server periodically updates the demand forecasting model to dynamically adjust inventory levels. This prevents overstocking and shortages and enables efficient resource management.

[1171] Furthermore, the server monitors the user's browsing and search history in real time and generates optimal product recommendations. This recommendation information is then pushed to the user via their smart device. For example, if "User A" searches for "smartphone cases" on a website, the server will recommend new products in the same category and notify the user's device.

[1172] When a user submits a purchase request, the server receives the information, selects the optimal logistics route, and issues delivery instructions. Once delivery begins, the server monitors the delivery status in real time and updates the delivery status on the user's device, allowing the user to track the progress of the product as it is delivered.

[1173] The hardware used includes servers, smart devices (smartphones, tablets, etc.), and database systems. The software includes machine learning libraries (e.g., scikit-learn), Python scripts for data collection and preprocessing, and notification services such as Firebase.

[1174] To generate a specific program, you can use the following prompt:

[1175] 1. "Generate Python code to forecast demand for the next month based on purchasing data from the past six months."

[1176] 2. "Create a program using Firebase to send recommendation notifications to users based on their purchase and browsing history."

[1177] This allows the server, terminal, and user to interact with each other, enabling both inventory management and recommendation notifications to be performed with high accuracy and efficiency, improving the user experience.

[1178] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1179] Step 1:

[1180] The server collects the user's purchase history, browsing history, and search history from a database. The input is the user ID, and the output is the user's various historical data (purchased items, purchase date and time, purchased quantity, viewed items, search keywords, etc.). This allows the past behavioral data of a specific user to be obtained.

[1181] Step 2:

[1182] The server pre-processes the collected data, cleaning the data (e.g., removing incomplete data and correcting outliers) and standardizing it. The input is the user history data collected in step 1, and the output is formatted data, which facilitates subsequent data analysis.

[1183] Step 3:

[1184] The server uses the preprocessed data to generate a demand forecasting model. It trains the model using machine learning algorithms (e.g., linear regression or time series analysis). The input is the preprocessed data, and the output is the demand forecasting model. This creates a foundation for forecasting future demand.

[1185] Step 4:

[1186] The server optimizes the planned inventory based on the generated demand forecast model. This involves a process of calculating how much of a specific product is needed at a specific time. The input is the demand forecast model, and the output is the calculation result of the optimal inventory amount. This makes inventory management more efficient.

[1187] Step 5:

[1188] The server monitors the user's real-time browsing and search history and generates product recommendations. The input is real-time user behavior data, and the output is a list of recommended products. This allows the server to recommend the most suitable products to the user.

[1189] Step 6:

[1190] The terminal notifies the user of the recommendation information sent from the server. This notification is performed using the push notification function of the smart device. The input is a list of recommended products, and the output is a notification message. This allows the user to receive new and recommended products in real time.

[1191] Step 7:

[1192] The user submits a purchase request, which is sent to the server with the purchase request data as input, thereby conveying the user's intention to purchase to the system.

[1193] Step 8:

[1194] The server selects the optimal logistics route based on the received purchase request and issues delivery instructions. The input is the purchase request data, and the output is logistics instructions and a delivery route. This ensures that purchased items are delivered to the user in the most efficient manner.

[1195] Step 9:

[1196] The server monitors the delivery status in real time and updates the delivery status to the user's device. The input is delivery progress data and the output is the updated delivery status, allowing the user to always know what stage their product is in.

[1197] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1198] This invention combines a system that performs demand forecasting based on users' purchase history, browsing history, and search history, optimizes inventory, and provides product recommendations with an emotion engine that recognizes users' emotions. Below, we will create a program for this system and explain the specific processing content.

[1199] Demand forecast data collection and analysis

[1200] The server collects past purchase history, browsing history, and search history from the user database. This includes detailed information such as purchase date and time, purchased product category, purchase quantity, and user ID. The server then preprocesses the collected data, cleaning the data, filling in missing values, and normalizing the data. The preprocessed data is then input into a machine learning algorithm (e.g., regression analysis or time series analysis) to generate a demand forecasting model. Future demand is predicted based on the generated forecasting model.

[1201] Examples:

[1202] The server analyzes "User A's" purchasing data for the past six months and predicts that "User A" is likely to purchase a particular electronic device again in the next month.

[1203] Collecting and analyzing emotional data

[1204] The server uses an emotion engine to recognize the user's emotions. It analyzes the user's facial expression data, voice tone, keyboard input, etc. sent from the device to identify the user's emotional state. This allows it to understand how the user is feeling at that time.

[1205] Examples:

[1206] While "User B" is browsing products on his smartphone, emotional data is collected using the camera and microphone. The emotion engine recognizes that User B is in an "excited" state.

[1207] Inventory management

[1208] The server combines the generated demand forecast model with user emotion data to calculate the necessary inventory levels. Based on this information, it allocates optimal inventory to each warehouse and regularly updates the inventory status to prevent shortages and overstocks.

[1209] Examples:

[1210] The server predicts that demand for electronic devices will increase within this month, and since user sentiment data indicates a positive reaction, it instructs the Tokyo warehouse to stock additional stock of those electronic devices.

[1211] Recommendation function

[1212] The server runs a recommendation engine based on the user's browsing history, search history, and emotional data. For example, it uses collaborative filtering or content-based filtering to select the most suitable products for the user. The generated recommendation information is sent to the user's device.

[1213] Examples:

[1214] When "User C" searches for "smartphone cases" on a website, the server analyzes the emotional data of "excitement." Taking that reaction into consideration, the server recommends "waterproof smartphone cases" as a product in the same category.

[1215] User Notifications and Interactions

[1216] The device will notify the user of the recommendation information received from the server in real time. The user can check the notification, and if they are interested, they can go to the detailed information page and proceed with the purchase process.

[1217] Examples:

[1218] User D's device displays a notification for "Recommended Product: Waterproof Smartphone Case," and User D clicks on the notification to go to the details page.

[1219] Checkout and logistics

[1220] When the user presses the purchase button, the terminal sends the purchase information to the server. The sent information includes the product ID, user ID, delivery address information, etc. The server confirms the received purchase information and checks the inventory status. If inventory is available, the order is confirmed and the optimal logistics route is selected. The server issues delivery instructions to the warehouse and updates the status on the user's terminal while monitoring the delivery status in real time.

[1221] Examples:

[1222] When "User E" presses the purchase button, the server retrieves the product from the nearest warehouse, selects the fastest delivery method, and begins delivery. During delivery, the server monitors the delivery status and notifies User E's device of the status as it progresses.

[1223] Through this series of processes, the present invention minimizes logistics lead time and user decision-making lead time, and further optimizes inventory. Furthermore, by periodically updating the generated demand forecast model and performing dynamic inventory management, more accurate supply and demand adjustments become possible. By utilizing user emotion data, more personalized services can be provided, improving the user experience.

[1224] The processing flow will be explained below.

[1225] Step 1:

[1226] The server collects data such as past purchase history, browsing history, and search history from the user database. The data collected includes purchase date and time, purchased product category, purchased quantity, and user ID.

[1227] Step 2:

[1228] The server preprocesses the acquired data, which includes data cleaning (e.g., removing incomplete data), imputing missing values ​​(e.g., imputing with the mean), and normalizing the data (e.g., scaling).

[1229] Step 3:

[1230] The server then inputs the preprocessed data into a machine learning algorithm to generate a demand forecasting model, for example, using regression analysis or time series analysis to build a model for predicting future demand.

[1231] Step 4:

[1232] The server periodically uses the generated demand forecasting model to predict future demand for specific products, calculates the necessary inventory levels based on this, and allocates optimal inventory to each warehouse.

[1233] Step 5:

[1234] The server uses an emotion engine to recognize the user's emotions and collects emotion data from the user's device, specifically analyzing facial expression data from the camera, voice tone from the microphone, and keyboard input speed and strength.

[1235] Step 6:

[1236] The server uses an emotion engine to analyze the collected emotional data and identify the emotional state the user is in at that time, categorizing emotions such as "excitement," "sadness," and "happiness."

[1237] Step 7:

[1238] The server incorporates the collected emotion data into the demand forecasting model to improve the accuracy of the forecast, thereby more accurately predicting product demand, which is strongly influenced by user emotions.

[1239] Step 8:

[1240] The server runs a recommendation engine based on the user's browsing history, search history, and emotional data, using collaborative filtering and content-based filtering, for example, to select the best products for the user.

[1241] Step 9:

[1242] The server transmits the generated recommendation information to the user's terminal, which notifies the user of the recommendation information in real time.

[1243] Step 10:

[1244] The user checks the recommendation notification displayed on the device and clicks on the detailed information page if they are interested. For example, a notification such as "Recommended product: Waterproof smartphone case" will be displayed.

[1245] Step 11:

[1246] When a user presses the purchase button, the device sends the purchase information to the server, including the product ID, user ID, and delivery address information.

[1247] Step 12:

[1248] The server confirms the received purchase information and checks the inventory status. If the item is available, it confirms the order and selects the optimal logistics route.

[1249] Step 13:

[1250] The server issues delivery instructions to the warehouse, for example, shipping the product from the nearest warehouse and selecting the fastest delivery method.

[1251] Step 14:

[1252] The server monitors the delivery status in real time and updates the status to the user's device, for example, notifying the user of the status when the product has been shipped, when it is being delivered, and when delivery has been completed.

[1253] Through these processing steps, the present invention minimizes logistics lead time and user decision-making lead time, optimizing inventory. By utilizing user emotion data, it is possible to provide more personalized services and improve the user experience.

[1254] Example 2

[1255] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1256] Many modern inventory management systems and product recommendation systems support demand forecasting and recommendations based on users' purchase, browsing, and search histories. However, they are unable to consider the user's emotional state, limiting their ability to optimize the user experience. Furthermore, inventory optimization is static, making it difficult to respond to dynamic demand fluctuations, leading to problems such as inventory shortages and overstocks. Furthermore, there is a need for more personalized recommendations by utilizing user emotional data.

[1257] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1258] In this invention, the server includes means for generating a demand forecasting model based on a user's purchase history, browsing history, and search history, means for optimizing inventory using the generated demand forecasting model, means for analyzing emotion data collected from the user's terminal, means for optimizing inventory and product recommendations using the analyzed emotion data, means for notifying the user's terminal of the generated recommendation information, and means for receiving purchase requests from the user, selecting an optimal logistics route, and issuing delivery instructions, thereby enabling personalized services and dynamic inventory management based on the user's emotions.

[1259] "User purchase history" is a record of products purchased by a user in the past, and specifically includes detailed information such as the purchase date and time, purchased product category, purchased quantity, and user ID.

[1260] "Browse history" is a record of web pages and products that a user has viewed in the past, and is data that indicates information such as the date and time of browsing, the number of times of browsing, and the products that have been viewed.

[1261] "Search history" is a record of a user's past search queries and search results, and is data that includes information such as search date and time, search keywords, and search result click history.

[1262] A "demand forecasting model" is a machine learning model for predicting future demand based on collected data, and is generated using algorithms such as regression analysis and time series analysis.

[1263] "Inventory optimization" is the process of using demand forecasting models to adjust inventory levels in anticipation of future demand, preventing shortages and overstocks.

[1264] "Emotion data" refers to data that reflects the user's emotional state, and specifically refers to information obtained from facial expression data, voice tone, keyboard input, and the like.

[1265] "Product recommendations" are recommendation information that suggests appropriate products to users, and are generated based on the user's purchasing history, emotional data, etc.

[1266] "Generated recommendation information" is information that suggests optimal products and services to a user, generated based on the user's purchase history, browsing history, search history, and emotional data.

[1267] A "purchase request" is information indicating a user's intention to purchase a product or service, and specifically includes a product ID, user ID, delivery address information, and the like.

[1268] The "optimal logistics route" is the most effective delivery route for efficiently delivering orders, and is selected taking into consideration factors such as cost, time, and inventory status.

[1269] This invention combines a system that performs demand forecasting based on a user's purchase history, browsing history, and search history, optimizes inventory, and provides product recommendations with an emotion engine that recognizes the user's emotions. Specific embodiments of this system are described below.

[1270] First, the server collects purchase history, browsing history, and search history from the user database. The collected data includes details such as purchase date and time, purchased product category, purchased quantity, and user ID. This data is then temporarily stored in cloud storage (e.g., Amazon S3). The software used at this stage includes database queries and cloud storage APIs.

[1271] Next, the server preprocesses the collected data. Specific processes include filling in missing values, removing outliers, and normalizing the data. This process uses Python's Pandas library and NumPy. The preprocessed data is then input into a machine learning algorithm (e.g., regression analysis, time series analysis) to generate a demand forecasting model. Machine learning libraries such as TensorFlow and scikit-learn are used at this stage.

[1272] The device uses a camera and microphone to collect user emotional data and transmits it to a server. Emotional data includes facial expressions, voice tone, and keyboard input. The collected data is analyzed by an emotion engine (e.g., NVIDIA DeepStream, Face++) to identify the user's emotional state. This information is used for demand forecasting and recommendation optimization.

[1273] The server optimizes inventory based on the generated demand forecasting model and emotion data. Specifically, it uses the demand forecasting model to forecast future demand and calculate the necessary inventory volume. Based on the calculation results, inventory is appropriately allocated through an inventory management system or ERP system (e.g., SAP, Oracle).

[1274] The server then recommends products based on the user's purchase history, browsing history, search history, and emotional data. This process uses recommendation engines (e.g., Amazon Personalize, Google RE) that use collaborative filtering or content-based filtering. The generated recommendation information is sent to the device in real time.

[1275] When a user purchases a product based on the recommended information, the device sends the purchase information to the server. This information includes the product ID, user ID, delivery address information, etc. The server then verifies the received purchase information and checks the inventory status. For example, if the product is available in stock, the server confirms the order and selects the optimal logistics route. This is done using a delivery management system (e.g., UPS WorldShip, FedEx Ship Manager).

[1276] Once delivery begins, the server monitors the delivery status in real time and notifies the user of the status as it progresses. This process improves the user experience and streamlines inventory management.

[1277] Prompt Sentence Examples

[1278] "Please explain a system that forecasts demand based on users' past purchasing history and optimizes inventory."

[1279] "Please explain a system that uses user emotional data to recommend products."

[1280] This invention enables personalized service and dynamic inventory management based on user emotions.

[1281] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1282] Step 1:

[1283] The server collects purchase history, browsing history, and search history data from the user database. The input data includes the purchase date and time, purchased product category, purchased quantity, user ID, etc. This data is stored in cloud storage (e.g., Amazon S3). Specifically, it executes a database query and uploads the retrieved data to cloud storage. The output of this step is the data stored in cloud storage.

[1284] Step 2:

[1285] The server preprocesses the collected data. Specifically, it imputes missing values, removes outliers, and normalizes the data. The software used includes Python's Pandas library and NumPy. The input data is stored in cloud storage, and the output data is the preprocessed, clean data.

[1286] Step 3:

[1287] The server inputs the preprocessed data into a machine learning algorithm to generate a demand forecasting model. This uses algorithms such as regression analysis and time series analysis. This process uses machine learning libraries such as TensorFlow and scikit-learn. The input data is preprocessed and clean, and the output is a demand forecasting model.

[1288] Step 4:

[1289] The device uses a camera and microphone to collect the user's emotional data and transmits it to a server. Emotional data includes facial expressions, voice tones, keyboard input, etc. The input data is emotional data acquired through the camera and microphone, and the output is emotional data transmitted to the server. Specific operations include collecting and transmitting sensor data.

[1290] Step 5:

[1291] The server analyzes the emotion data sent from the device. This analysis is performed using an emotion engine (e.g., NVIDIA DeepStream, Face++). The input data is the emotion data sent from the device, and the output is the analyzed emotional state.

[1292] Step 6:

[1293] The server combines the generated demand forecasting model with the analyzed emotion data to optimize inventory. Specifically, it uses the demand forecasting model to forecast future demand and calculates the required inventory volume. Based on the calculation results, inventory is appropriately allocated via an inventory management system or ERP system (e.g., SAP, Oracle). The input data are the demand forecasting model and emotion data, and the output is an optimized inventory list.

[1294] Step 7:

[1295] The server recommends products based on the user's purchase history, browsing history, search history, and emotional data. A recommendation engine (e.g., Amazon Personalize, Google RE) using collaborative filtering or content-based filtering is used. The generated recommendation information is sent to the device in real time. The input data are the aforementioned history data and emotional data, and the output is recommendation information.

[1296] Step 8:

[1297] When a user purchases a product based on the recommendation information, the terminal sends the purchase information to the server. The information sent includes the product ID, user ID, delivery address information, etc. A specific action is to press the purchase button. The input data is the execution data of the purchase page, and the output is the purchase information sent to the server.

[1298] Step 9:

[1299] The server confirms the received purchase information and checks the inventory status. If inventory is available, it confirms the order and selects the optimal logistics route. This is done using a delivery management system (e.g., UPS WorldShip, FedEx Ship Manager). The input data are purchase information and inventory information, and the output is the confirmed order and the selected logistics route.

[1300] Step 10:

[1301] The server monitors the delivery status in real time and notifies the user's device of the status as it progresses. Utilizing the delivery management system, it tracks each stage of delivery and updates the status as appropriate. The input data is delivery status information from the delivery management system, and the output is a status notification sent to the user's device.

[1302] (Application example 2)

[1303] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1304] Conventional demand forecasting systems are based solely on a user's purchasing history, browsing history, and search history, and do not take into account the user's real-time emotional state, making it difficult to provide personalized recommendations. Furthermore, they have the problem of not being able to adequately address the user's decision-making process or improve the user experience. The purpose of this invention is to solve these problems.

[1305] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for generating a demand forecasting model based on a user's purchase history, browsing history, and search history, means for optimizing inventory using the generated demand forecasting model, means for generating content recommendations based on the user's browsing history, search history, and emotional data, means for notifying the user's visual display device of the generated recommendation information, and means for receiving a selection request from the user and selecting the optimal content to be displayed next. This enables personalized recommendations and inventory management that take into account the user's real-time emotional state.

[1306] "User purchase history" is a collection of information about products and services that a user has purchased in the past.

[1307] "Browsing history" is a record of pages and content that a user has previously viewed on the Internet or in applications.

[1308] "Search history" is a history of keywords and phrases that a user has searched for on the Internet or in an application.

[1309] A "demand forecasting model" is an algorithm or mathematical model for predicting future demand based on past user behavior data.

[1310] "Inventory optimization" is the process of optimizing supply-side inventory levels based on demand forecasts.

[1311] "Emotion data" is information that identifies a user's emotional state in real time by analyzing their facial expressions and vocal tone.

[1312] "Content recommendation" is a function that suggests appropriate content based on a user's behavioral history and emotional data.

[1313] A "visual display device" is a device that visually displays information to a user, and includes smart glasses and digital displays.

[1314] A "selection request" is a request for a user to make a selection or designation from the presented recommendation information.

[1315] This invention is a system that forecasts demand and recommends content based on users' purchasing history, browsing history, search history, and emotional data. This system consists of three elements: a server, a terminal, and a user.

[1316] First, the server collects users' purchasing, browsing, and search histories and generates a demand forecasting model based on this information. This forecasting model is used to analyze users' consumption behavior and forecast future demand. The server also has functions for preprocessing the collected data, cleaning the data, imputing missing values, and normalizing the data. For this purpose, the server mainly uses machine learning libraries such as Scikit-learn.

[1317] The server also combines the user's browsing history, search history, and emotion engine to collect real-time emotional data. This emotional data is collected via the camera and microphone and analyzed using an emotion recognition model built with Keras. The server identifies the user's emotions and recommends content based on the results.

[1318] The device used by the user includes smart glasses that can capture the user's facial expressions and tone of voice in real time, and the captured emotional data is sent to a server where it is incorporated into a demand forecasting model and recommendation engine.

[1319] For example, if a user is watching a movie through smart glasses, the emotion engine can determine that the user is "excited" based on their facial expressions and voice. Based on the user's viewing history, the server then recommends movies and content that the user is likely to watch next, and displays them on the smart glasses' display. This process improves the user experience and provides more accurate recommendations.

[1320] In addition, the demand forecasting model is regularly updated, enabling dynamic inventory management. For example, by combining past data with the latest sentiment data, demand can be forecast more accurately and appropriate inventory levels can be maintained. This shortens logistics lead times and minimizes user purchasing lead times.

[1321] The main hardware and software required to implement this system are as follows:

[1322] Hardware:

[1323] Smart glasses (with camera and microphone)

[1324] Server (cloud environment recommended)

[1325] software:

[1326] OpenCV (face detection and facial expression data collection)

[1327] Scikit-learn (data preprocessing and demand forecasting model)

[1328] Keras (emotion recognition model)

[1329] Server-side API (obtaining user data)

[1330] For an example of how to use a generative AI model, the prompt would be:

[1331] "By aligning a user's viewing history with their current emotional state, we can recommend content they're likely to watch next. For example, if a user is excited and watching an action movie, what specific action movie should we recommend?"

[1332] This will create a system that integrates user behavioral data and emotional data to provide more personalized recommendations.

[1333] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1334] Step 1:

[1335] The server collects the user's purchase history, browsing history, and search history. For this, detailed information such as purchase date and time, purchased product category, and purchase quantity is extracted from the database based on the user ID. The input data is the user's behavior history information, and the output data is preprocessed user behavior history data.

[1336] Step 2:

[1337] The server preprocesses the collected data. Specifically, it performs data cleaning, missing value completion, data normalization, etc. This process formats the input data for the machine learning algorithm. The input data is the collected user behavior history data, and the output data is the preprocessed data.

[1338] Step 3:

[1339] The server uses the preprocessed data to generate a demand forecasting model. During this process, it uses a machine learning library such as Scikit-learn to perform regression analysis and time series analysis. The input data is the preprocessed data, and the output data is the demand forecasting model. The generated model is used to predict future demand.

[1340] Step 4:

[1341] The device captures the user's browsing history and search history, as well as facial and voice data in real time. This data is collected through a camera and microphone. The input data is real-time facial and voice data, and the output data is the captured raw data.

[1342] Step 5:

[1343] The server receives the facial expression and voice data sent from the device and analyzes them using an emotion recognition engine. Using an emotion recognition model built with Keras, the server identifies the user's emotional state. The input data is the captured facial expression and voice data, and the output data is the identified emotion data.

[1344] Step 6:

[1345] The server operates a recommendation engine based on the identified emotion data and the user's browsing and search history. Collaborative filtering and content-based filtering are used to generate optimal content. The input data are emotion data and behavioral history data, and the output data is recommendation information.

[1346] Step 7:

[1347] The terminal notifies the visual display device of the recommendation information received from the server. The user can check the notified recommendation information and move to a detailed information page if they are interested. The input data is the recommendation information, and the output data is the information displayed visually to the user.

[1348] Step 8:

[1349] When the user selects a content, a selection request is sent to the server. The input data is the user's selection request, and the output data is the request information sent to the server.

[1350] Step 9:

[1351] Based on the received selection request, the server selects the optimal content to display next and notifies the terminal again. This allows the user to seamlessly view the next content. The input data is the request information, and the output data is the content information to be displayed next.

[1352] In this way, through specific actions at each step, the system provides users with personalized content recommendations and performs real-time sentiment analysis and demand forecasting.

[1353] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1354] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1355] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1356] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1357] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1358] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1359] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1360] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1361] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1362] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1363] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1364] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1365] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1366] 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.

[1367] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1368] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1369] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1370] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1371] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1372] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1373] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1374] The following is further disclosed regarding the above embodiment.

[1375] (Claim 1)

[1376] A means for generating a demand forecasting model based on a user's purchase history, browsing history, and search history;

[1377] A means for optimizing inventory using the generated demand forecast model;

[1378] means for generating product recommendations based on a user's browsing history and search history;

[1379] a means for notifying a user terminal of the generated recommendation information;

[1380] A means for receiving a purchase request from a user, selecting an optimal logistics route, and issuing delivery instructions;

[1381] A system including:

[1382] (Claim 2)

[1383] 2. The system of claim 1, further comprising means for periodically updating the generated demand forecast model to dynamically optimize inventory.

[1384] (Claim 3)

[1385] 2. The system according to claim 1, further comprising means for collecting an action log based on recommendation information sent from the user's terminal and reflecting the collected action log in subsequent recommendation generation.

[1386] "Example 1"

[1387] (Claim 1)

[1388] A means for generating a demand forecasting model based on a user's purchase history, browsing history, and search history;

[1389] A means for optimizing inventory using the generated demand forecast model;

[1390] means for generating product recommendations based on a user's browsing history and search history;

[1391] a means for notifying a user terminal of the generated recommendation information;

[1392] A means for receiving a purchase request from a user, selecting an optimal logistics route, and issuing delivery instructions;

[1393] A means for monitoring the delivery status in real time and notifying the user of the delivery status on their terminal;

[1394] A system including:

[1395] (Claim 2)

[1396] 2. The system of claim 1, further comprising means for periodically updating the generated demand forecast model to dynamically optimize inventory.

[1397] (Claim 3)

[1398] 2. The system according to claim 1, further comprising means for collecting an action log based on recommendation information sent from the user's terminal and reflecting the collected action log in subsequent recommendation generation.

[1399] "Application Example 1"

[1400] (Claim 1)

[1401] A means for generating a demand forecasting model based on a user's purchase history, browsing history, and search history;

[1402] A means for optimizing inventory using the generated demand forecast model;

[1403] means for generating product recommendations based on a user's browsing history and search history;

[1404] a means for notifying a user terminal of the generated recommendation information;

[1405] A means for receiving a purchase request from a user, selecting an optimal logistics route, and issuing delivery instructions;

[1406] A means for providing appropriate product recommendations based on a user's shopping behavior and sending push notifications in real time using a smartphone to increase purchasing motivation;

[1407] A system including:

[1408] (Claim 2)

[1409] 2. The system of claim 1, further comprising means for periodically updating the generated demand forecast model to dynamically optimize inventory.

[1410] (Claim 3)

[1411] 2. The system according to claim 1, further comprising means for collecting an action log based on recommendation information sent from the user's terminal and reflecting the collected action log in subsequent recommendation generation.

[1412] "Example 2: Combining Emotion Engines"

[1413] (Claim 1)

[1414] A means for generating a demand forecasting model based on a user's purchase history, browsing history, and search history;

[1415] A means for optimizing inventory using the generated demand forecast model;

[1416] A means for analyzing emotion data collected from a user's device;

[1417] a means for optimizing inventory and product recommendations using the analyzed sentiment data;

[1418] a means for notifying a user terminal of the generated recommendation information;

[1419] A means for receiving a purchase request from a user, selecting an optimal logistics route, and issuing delivery instructions;

[1420] A system including:

[1421] (Claim 2)

[1422] 2. The system of claim 1, further comprising means for periodically updating the generated demand forecast model to dynamically optimize inventory.

[1423] (Claim 3)

[1424] 2. The system according to claim 1, further comprising means for collecting an action log based on recommendation information sent from the user's terminal and reflecting the collected action log in subsequent recommendation generation.

[1425] "Application example 2 when combining emotion engines"

[1426] (Claim 1)

[1427] A means for generating a demand forecasting model based on a user's purchase history, browsing history, and search history;

[1428] A means for optimizing inventory using the generated demand forecast model;

[1429] means for generating content recommendations based on a user's browsing history, search history, and emotion data;

[1430] means for notifying the generated recommendation information to a visual display device of a user;

[1431] means for receiving a selection request from a user and selecting the most appropriate content to display next;

[1432] A system including:

[1433] (Claim 2)

[1434] 2. The system of claim 1, further comprising means for periodically updating the generated demand forecast model to dynamically optimize inventory.

[1435] (Claim 3)

[1436] 2. The system according to claim 1, further comprising means for collecting an action log based on the recommendation information transmitted from the user's visual display device and reflecting the collected action log in subsequent recommendation generation. [Explanation of symbols]

[1437] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means for generating a demand forecasting model based on a user's purchase history, browsing history, and search history; A means for optimizing inventory using the generated demand forecast model; means for generating product recommendations based on a user's browsing history and search history; a means for notifying a user terminal of the generated recommendation information; A means for receiving a purchase request from a user, selecting an optimal logistics route, and issuing delivery instructions; A system including:

2. The system according to claim 1 , further comprising means for periodically updating the generated demand forecast model to dynamically optimize inventory.

3. The system according to claim 1, further comprising means for collecting an action log based on the recommendation information transmitted from the user's terminal and reflecting the collected action log in subsequent recommendation generation.

Images