System

Abstract

A system is provided.SOLUTION: A system, comprising: means for receiving textual information input by a user; means for analyzing the textual information using a natural language process engine to extract keywords and property information; means for obtaining merchandise information from a plurality of e-commerce sites based on the extracted keywords; means for analyzing images of the obtained merchandise information using a multi-modal AI model to identify merchandise that meets the requirements of the user; and means for suggesting the identified merchandise information to the user.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] Currently, many users struggle with product searches, and it takes a great deal of time and effort to find the product they are looking for. For busy users, the product search process itself can be a significant burden. Furthermore, existing search functions on e-commerce sites often fail to adequately reflect the specific criteria users desire, preventing users from finding products that meet their needs. Furthermore, there is a lack of ways to automate the thought process involved in product selection, and there is a need to improve the user purchasing experience. [Means for solving the problem]

[0005] In order to solve the above-mentioned problems, the present invention proposes the following means. The system includes a means for receiving text data input by a user, a means for analyzing the text data using a natural language processing engine, and a means for extracting important keywords and attribute information. The system further includes a means for acquiring product data from multiple e-commerce sites based on the extracted keywords. The system also includes a means for analyzing images of the acquired product data using a multimodal AI model, a means for identifying products that meet the user's requirements, and a means for suggesting identified product information to the user. The above means enable users to efficiently find products that meet their requirements and smoothly complete the purchasing process.

[0006] "User" refers to an individual or entity who utilizes the chat window to input product requirements.

[0007] "Text data" refers to data in the form of sentences that a user enters into a chat window.

[0008] A "natural language processing engine" refers to software or algorithms that analyze text data and extract important information such as keywords and attribute information.

[0009] "Keywords" refer to important words or phrases extracted from text data and used for subsequent searches and analysis.

[0010] "Attribute information" refers to information that indicates the characteristics and features of a product, extracted from text data.

[0011] "E-commerce site" refers to a website that sells goods and services over the Internet.

[0012] "Product Data" refers to information about a product (e.g., images, prices, links, etc.) obtained from an e-commerce site.

[0013] A "multimodal AI model" refers to an artificial intelligence model that analyzes multiple data formats (e.g., images, text) and understands the relationships between them.

[0014] "Product image" refers to an image file that is part of the product data and visually represents the product.

[0015] "Means of suggestion" refers to a method or mechanism for providing users with information about recommended products based on the analysis results.

[0016] "Purchase API" refers to an application programming interface provided by an e-commerce site to automate the product purchase process. [Brief explanation of the drawings]

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

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

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

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

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

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

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

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

[0025] [First embodiment]

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

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

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

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

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

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

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

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

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

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

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

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

[0038] The present invention is a system that allows users to easily and quickly find and purchase the products they desire. The system operates via a chat interface, allowing users to input requirements for the products they want, and then suggests products that match those requirements.

[0039] Program processing and specific examples

[0040] 1. Accepting input from the user

[0041] The terminal receives text data in which the user enters "I want a new sofa" in the chat window.

[0042] 2. Text interpretation and requirement extraction

[0043] The server passes the received text data to a natural language processing engine for analysis, which extracts the keyword "sofa" and the attribute information "new."

[0044] 3. Providing product images

[0045] The server retrieves product data (product images, price information, links, etc.) from APIs of multiple e-commerce sites based on the extracted keyword "sofa."

[0046] 4. Image Classification and Requirements Compliance Assessment

[0047] The server inputs the captured product images into a multimodal AI model to identify the product that best suits the user's requirements, for example, evaluating color and design features to select a modern sofa that meets the "new" requirements.

[0048] 5. Generate product suggestions and present them to the user

[0049] The server formats the identified product information and provides the user with a message via chat saying, "How about this sofa?" along with an image, price, and link to the selected product.

[0050] Specific examples

[0051] For example, if the user types "Show me more gray sofas," the device sends this new text to the server, which again uses its NLP engine to extract the new keyword "gray" and performs similar search and image analysis to generate new suggestions for gray sofas, allowing the user to find products that match their more specific requirements.

[0052] Purchase procedure

[0053] When the user selects a product and enters "I want to purchase this," the device sends this purchase intention to the server. The server then sends the user's account information to the e-commerce site's purchase API, automatically starting the purchase process. The e-commerce site then completes the purchase process and returns a confirmation message to the server. The server then notifies the user of the purchase completion message.

[0054] This system allows users to efficiently search for and purchase products, providing a superior purchasing experience.

[0055] The processing flow will be explained below.

[0056] Step 1:

[0057] The user types "I want a new sofa" into the chat window.

[0058] Step 2:

[0059] The terminal receives this text input and transmits it to the server in real time.

[0060] Step 3:

[0061] The server passes the received text data to a natural language processing engine for analysis.

[0062] Step 4:

[0063] The server uses a natural language processing engine to extract the keyword "sofa" and the attribute information "new" from the input text.

[0064] Step 5:

[0065] Based on the extracted keyword "sofa," the server sends requests to APIs of multiple e-commerce sites to retrieve related product data (product images, price information, links, etc.).

[0066] Step 6:

[0067] In response to the request, the server of the e-commerce site returns product data that matches the conditions to the server.

[0068] Step 7:

[0069] The server stores the acquired product data in a local database or memory.

[0070] Step 8:

[0071] The server inputs the saved product image data into a multimodal AI model for analysis.

[0072] Step 9:

[0073] The server uses a multimodal AI model to identify the product that best suits the user's requirements, evaluating color and design features to determine whether the product meets the user's requirements.

[0074] Step 10:

[0075] The server lists the identified product information and generates a suggestion message for the user, including product images, price information, and product links.

[0076] Step 11:

[0077] The server transmits the generated proposal message to the user through the terminal.

[0078] Step 12:

[0079] The user reviews the list of suggested products and provides additional input if more information is desired.

[0080] Step 13:

[0081] The terminal receives new text input and sends it back to the server.

[0082] Step 14:

[0083] The server passes the new text data to the natural language processing engine again to extract keywords and attribute information.

[0084] Step 15:

[0085] The server retrieves the product data again based on the new information extracted and analyzes it using a multimodal AI model.

[0086] Step 16:

[0087] The server generates an updated product list and again sends a suggestion message to the user.

[0088] Step 17:

[0089] The user selects a particular product and enters "I would like to purchase this."

[0090] Step 18:

[0091] The terminal transmits text data of the purchase intention to the server.

[0092] Step 19:

[0093] The server calls the purchase API of the e-commerce site and initiates the purchase process using the user's account information.

[0094] Step 20:

[0095] The e-commerce site's server processes the purchase and sends a confirmation message back to the server.

[0096] Step 21:

[0097] The server generates a purchase completion message and notifies the user through the terminal.

[0098] By going through each step, users can efficiently find products that meet their requirements and complete the purchasing process smoothly.

[0099] Example 1

[0100] 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."

[0101] Conventional online shopping systems have made it difficult for users to quickly and accurately find the products they are looking for. In particular, text-based searches have been problematic because they are unable to efficiently suggest products due to their inability to accommodate vague keywords and diverse user needs. Furthermore, there has been no system that can retrieve product information from multiple online markets and present products that best meet the user's requirements. As a result, users are forced to spend a long time searching for products, which often discourages them from making a purchase.

[0102] 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.

[0103] In this invention, the server includes means for receiving text data input by a user, means for analyzing the text data using a natural language processing engine to extract keywords and attribute information, means for acquiring product data from multiple online markets based on the extracted keywords, means for analyzing images of the acquired product data using a multi-function AI model to identify products that meet the user's requirements, and means for suggesting the identified product information to the user, thereby enabling the user to quickly and accurately find the product they are looking for.

[0104] "User" refers to an individual or corporation that uses the system to search for and purchase products.

[0105] "Text data" refers to character string information entered by the user, and refers to sentences used to explain searches and requirements.

[0106] A "natural language processing engine" refers to software or algorithms that analyze text data and extract keywords and attribute information.

[0107] "Keywords" refer to the main words or phrases that describe the product or its characteristics that a user is searching for.

[0108] "Attribute information" refers to information that indicates specific characteristics or requirements related to a keyword.

[0109] "Online marketplace" refers to platforms and websites that sell products over the internet.

[0110] "Product Data" means product information obtained from an online marketplace, including images, prices, links, etc.

[0111] A "multi-functional AI model" refers to an artificial intelligence model that integrates and analyzes data in multiple different formats to identify the products that best suit the user's requirements.

[0112] "Product information" is detailed information about a product suggested to a user, including an image, price, link, and the like.

[0113] The present invention is a system that helps users quickly and accurately find and purchase the products they are looking for. The system operates via a chat interface and suggests products based on product requirements entered by the user.

[0114] Hardware and Software

[0115] This system consists of a user terminal, a server, and multiple APIs. The main hardware and software used are as follows:

[0116] User device: Smartphone, tablet, PC, etc. Used by users to input text data through the chat window.

[0117] Server: A virtual server provided by a cloud service provider. For example, commonly used cloud platforms include Amazon Web Services (AWS) and Google Cloud Platform (GCP).

[0118] Natural language processing engine: Software used to analyze text data. Google Cloud Natural Language API is one such example.

[0119] Multimodal AI model: An AI model that analyzes image and text data to recommend optimal products. Typically, OpenAI's CLIP (Contrastive Language-Image Pretraining) model is used.

[0120] Online marketplace API: An interface for obtaining product data from multiple online marketplaces. For example, Amazon Marketplace API or Rakuten Marketplace API are used.

[0121] Specific operation of the system

[0122] When a user opens a chat window and types "I want a new sofa," the device receives this text data, stores it in its internal memory, and then sends it to the server.

[0123] The server passes the received text data to a natural language processing engine for analysis, which extracts the keyword "sofa" and the attribute information "new."

[0124] The server then sends requests to multiple online marketplace APIs based on these extracted keywords. For example, it uses the APIs of Amazon and Rakuten to perform a product search using the keyword "sofa." The product data retrieved from these online marketplaces includes images, prices, links, etc.

[0125] The acquired product data is input by the server into a multi-function AI model, which identifies the most suitable product based on the product image and extracted attribute information. For example, using the CLIP model, it matches the image with the keyword "new sofa" and ranks the most relevant products.

[0126] The identified product information is converted into JSON format and then reformatted to fit the chat interface. The user is then provided with an image, price, and link to the suggested product along with the message, "How about this sofa?"

[0127] Specific examples

[0128] For example, if the user types "Show me more gray sofas," the device receives this new text data and sends it to the server. The server then uses its NLP engine to extract the new keyword "gray" and requests a new search from the online marketplace API. It then performs image analysis on the new product data and suggests the gray sofa that best suits the user's requirements.

[0129] Prompt Sentence Examples

[0130] A user types the following into a chat window:

[0131] "I want a new sofa"

[0132] "Show me more gray sofas."

[0133] This system allows users to efficiently search for and purchase products, providing a superior purchasing experience.

[0134] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0135] Step 1:

[0136] The terminal receives text data entered into the chat window by the user.

[0137] Input: A user types "I want a new sofa" into a chat window.

[0138] Operation: The terminal stores this text data in its internal memory and generates a data packet.

[0139] Output: Send the generated data packet to the server.

[0140] Step 2:

[0141] The server receives the text data sent from the terminal.

[0142] Input: Data packets from the terminal.

[0143] How it works: The server takes this text data and prepares it to be passed to a natural language processing (NLP) engine.

[0144] Output: Text data ready for analysis.

[0145] Step 3:

[0146] The server analyzes the text data using a natural language processing engine to extract keywords and attribute information.

[0147] Input: Text data ready for analysis.

[0148] How it works: The server parses the text data using the Google Cloud Natural Language API.

[0149] Output: Keyword "sofa" and attribute information "new".

[0150] Step 4:

[0151] The server retrieves product data from multiple online marketplaces based on the extracted keywords.

[0152] Input: keyword "sofa".

[0153] Operation: The server sends an HTTP request to the Amazon Marketplace API and the Rakuten Marketplace API to search for products using the keyword "sofa."

[0154] Output: Retrieved product data (images, prices, links, etc.).

[0155] Step 5:

[0156] The server analyzes the acquired product data images using a multi-function AI model and identifies products that meet the user's requirements.

[0157] Input: Retrieved product data.

[0158] How it works: The server uses OpenAI's CLIP model to match product images with the attribute "new" and rank the products that best match the user's requirements.

[0159] Output: A ranked list of products.

[0160] Step 6:

[0161] The server proposes the identified product information to the user.

[0162] Input: A ranked list of products.

[0163] What it does: The server converts this information into JSON format and reformats it into a format suitable for a chat interface with the message "How would you like this sofa?"

[0164] Output: Send the formatted product suggestion message to the terminal.

[0165] Step 7:

[0166] The terminal displays the product suggestion message sent from the server to the user.

[0167] Input: A formatted product suggestion message from the server.

[0168] What it does: The device displays this message in a chat window, providing the user with a picture of the product, its price, and a link.

[0169] Output: Product suggestions to the user.

[0170] Step 8:

[0171] The user enters additional requests for the suggested products, for example, "Show me more gray sofas."

[0172] Input: Additional input from the user.

[0173] Operation: The user enters a new request into the chat window of the terminal, which receives it and prepares to retransmit it to the server.

[0174] Output: New data packet.

[0175] Step 9:

[0176] The server receives the new text data again and again uses the NLP engine to extract keywords and attribute information.

[0177] Input: A new data packet.

[0178] How it works: The server uses the Google Cloud Natural Language API to parse newly received text data and extract the keyword "gray."

[0179] Output: New keyword and attribute information.

[0180] Step 10:

[0181] The server retrieves product data again based on the new keywords, and identifies and suggests more suitable products.

[0182] Input: new keyword "gray".

[0183] Operation: The server requests a re-search from the Amazon Marketplace API and Rakuten Market API to obtain product data for "gray sofas." It then analyzes the product information using a multi-function AI model.

[0184] Output: Send new product suggestion message to the terminal.

[0185] (Application example 1)

[0186] 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."

[0187] Conventional e-commerce systems have faced challenges in making it difficult for users to efficiently search for products and quickly find those that meet their requirements. Furthermore, product search and suggestion functions using voice input and augmented reality (AR) technology interfaces were inadequate, leaving a need for an improved user experience.

[0188] 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.

[0189] In this invention, the server includes means for receiving text data input by a user, means for analyzing the text data using a natural language processing engine to extract keywords and attribute information, means for acquiring product data from multiple e-commerce sites based on the extracted keywords, means for analyzing images of the acquired product data using a multimodal AI model to identify products that meet the user's requirements, means for suggesting the identified product information to the user, means for converting voice data into text data, and means for presenting product information in augmented reality on a visual display device, thereby enabling users to easily search for products through voice input, visually confirm product information using AR technology, and complete the purchase process efficiently and intuitively.

[0190] "User" refers to a consumer who searches for and purchases products using this system.

[0191] "Text data" refers to character information including requirements and conditions regarding products entered by a user.

[0192] "Natural language processing engine" refers to software and algorithms for understanding and analyzing human language.

[0193] "Keywords" refer to the main words necessary for searching for products, extracted from the user's input text.

[0194] "Attribute information" refers to specific characteristics or conditions related to a keyword.

[0195] "E-commerce site" refers to a website that sells goods and services over the Internet.

[0196] "Product Data" refers to information about a product obtained from an e-commerce site, primarily including the name, image, price, and link.

[0197] A "multimodal AI model" refers to an artificial intelligence model that integrates and analyzes multiple types of data (e.g., text data, image data).

[0198] A "visual display device" is a device that presents information to a user's field of vision, and primarily refers to smart glasses and augmented reality devices.

[0199] "Augmented reality (AR)" refers to the technology of overlaying digital information onto a real-world environment.

[0200] "Voice data" refers to sound information obtained from the user's speech.

[0201] "Product suggestion" refers to the act of presenting the most suitable product to the user based on analyzed product data.

[0202] The present invention is a system for enabling users to easily and quickly find and purchase desired products, and is particularly directed to a shopping assistant application that uses smart glasses. The system is configured as follows.

[0203] The server receives the text data entered by the user and analyzes it using a natural language processing engine. When the user enters product requirements by voice, the smart glasses' microphone picks up the voice data and converts it into text data. The converted text data is then analyzed by the natural language processing engine to extract keywords and attribute information.

[0204] Once the keywords are extracted, the server accesses the APIs of multiple e-commerce sites to retrieve the corresponding product data. An e-commerce site is a website that sells products and services over the Internet, and the product data includes the product name, image, price, and purchase link.

[0205] The acquired product data images are analyzed using a multimodal AI model, which is an artificial intelligence model that integrates and analyzes multiple types of data (e.g., text data, image data, etc.). This allows the system to identify the product that best suits the user's requirements.

[0206] The identified product information is then formatted and suggested to the user. The suggested product information is displayed as augmented reality (AR) on the smart glasses display. Augmented reality is a technology that overlays digital information on the real-world environment, allowing the user to intuitively check the product information superimposed on their field of vision.

[0207] For example, if a user says "I'm looking for red sneakers," the voice data is picked up through the smart glasses' microphone and converted into text data. The server then uses a natural language processing engine to analyze the converted text data and extract the keyword "red sneakers." Product data is then retrieved using the API of the e-commerce site, and analyzed using a multimodal AI model to identify the most suitable product.

[0208] As a result, product information for "red sneakers" is displayed in AR within the user's field of vision. When the user finds a specific product and says "I want to buy this," the server automatically initiates the purchase process through the e-commerce site's purchase API and sends a confirmation message to the user.

[0209] This system allows users to search for and purchase products efficiently and intuitively, providing an excellent purchasing experience.

[0210] Example prompt sentence:

[0211] The user speaks, "I'm looking for red sneakers."

[0212] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0213] Step 1:

[0214] The user speaks into the microphone of the smart glasses to input information about a product. For example, they might say, "I'm looking for red sneakers." This voice data is picked up by the microphone of the smart glasses. The input is voice data, and the output is voice data.

[0215] Step 2:

[0216] The device converts the acquired voice data into text data. This is done using a voice recognition library (e.g., Google's voice recognition API). This converts the voice data into text data. The input is voice data, and the output is text data.

[0217] Step 3:

[0218] Text data is sent to the server. The server analyzes the text data using a natural language processing engine (e.g., the BERT model) and extracts keywords and attribute information. This yields a keyword, for example, "red sneakers." The input is text data, and the output is keywords and attribute information.

[0219] Step 4:

[0220] The server accesses the APIs of multiple e-commerce sites based on the extracted keywords to retrieve related product data, including product names, images, prices, and purchase links. The input is the keywords, and the output is the product data.

[0221] Step 5:

[0222] The server inputs the acquired product data images into a multimodal AI model (e.g., CLIP model) and analyzes them. This identifies products that meet the user's requirements. The input is the product image, and the output is the identified product information.

[0223] Step 6:

[0224] The server formats the identified product information and presents it as augmented reality (AR) on the smart glasses display, allowing the user to view the product information directly in their field of vision. The input is the identified product information, and the output is the AR-displayed product information.

[0225] Step 7:

[0226] The user checks the display on the smart glasses, selects a specific product, and then voice-inputs, "I would like to purchase this." The input is voice data, and the output is text data indicating the purchase intention.

[0227] Step 8:

[0228] The terminal receives the voice data again, converts it into text data, and sends it to the server. The input is voice data, and the output is text data.

[0229] Step 9:

[0230] The server analyzes the received text data, confirms the user's purchase intention, and executes the purchase procedure via the e-commerce site's purchase API. The input is the text data of the purchase intention, and the output is purchase confirmation information.

[0231] Step 10:

[0232] The server confirms that the purchase has been completed and notifies the user of the result. The user receives a confirmation message on their smart glasses. The input is the purchase confirmation information, and the output is a user notification.

[0233] 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.

[0234] The present invention combines an emotion engine with a system that allows users to easily and quickly find and purchase the products they desire. The system operates via a chat interface, allowing users to input their desired product requirements and then suggests products that match those requirements. The system can also recognize the user's emotional state and adjust product suggestions based on the results.

[0235] Program processing and specific examples

[0236] 1. Accepting input from the user

[0237] The terminal receives text data in which the user enters "I want a new sofa" in the chat window.

[0238] 2. Text Interpretation and Emotion Recognition

[0239] The server passes the received text data to a natural language processing engine, analyzes the text, and uses an emotion engine to recognize the user's emotional state. For example, the emotion engine classifies the user's text as an emotion such as "joy" or "dissatisfaction."

[0240] 3. Extracting keywords and attribute information

[0241] The server uses a natural language processing engine to extract the keyword "sofa" and the attribute information "new" from the input text.

[0242] 4. Providing product images

[0243] The server retrieves product data (product images, price information, links, etc.) from APIs of multiple e-commerce sites based on the extracted keyword "sofa."

[0244] 5. Image Classification and Requirements Compliance Assessment

[0245] The server inputs the acquired product image data into a multimodal AI model for analysis. This analysis also takes into account the user's emotional state. For example, if a user's emotional state is "dissatisfied," the server will prioritize suggesting highly rated products and discount information.

[0246] 6. Generate product suggestions and present them to the user

[0247] The server then lists the identified product information and generates a suggestion message by adjusting the wording and tone of the message according to the user's emotional state. For example, it might provide a message such as "This sofa is recommended!" along with an image, price, and link to the selected product via chat.

[0248] Specific examples

[0249] For example, if the user again types "Show me more gray sofas," the device sends this new text to the server, which again uses the NLP engine to extract the new keyword "gray" and performs similar search and image analysis to generate new suggestions for gray sofas, again using the emotion engine to adjust the suggestions based on the user's latest emotional state.

[0250] Purchase procedure

[0251] When the user selects a product and enters "I want to purchase this," the device sends this purchase intention to the server. The server then sends the user's account information to the e-commerce site's purchase API, automatically starting the purchase process. The e-commerce site then completes the purchase process and returns a confirmation message to the server. The server then generates a purchase completion message and notifies the user via the device.

[0252] This system allows users to efficiently search for and purchase products, and provides an excellent purchasing experience with appropriate suggestions based on their emotional state.

[0253] The processing flow will be explained below.

[0254] Step 1:

[0255] The user types "I want a new sofa" into the chat window.

[0256] Step 2:

[0257] The terminal receives this text input and transmits it to the server in real time.

[0258] Step 3:

[0259] The server passes the received text data to a natural language processing engine and analyzes the text.

[0260] Step 4:

[0261] The server uses a natural language processing engine to extract the keyword "sofa" and the attribute information "new" from the input text.

[0262] Step 5:

[0263] The server inputs text data into an emotion engine to recognize the user's emotional state (e.g., "joy," "dissatisfaction," "excitement," etc.).

[0264] Step 6:

[0265] Based on the extracted keyword "sofa," the server sends requests to APIs of multiple e-commerce sites to retrieve related product data (product images, price information, links, etc.).

[0266] Step 7:

[0267] In response to the request, the server of the e-commerce site returns product data that matches the conditions to the server.

[0268] Step 8:

[0269] The server stores the acquired product data in a local database or memory.

[0270] Step 9:

[0271] The server inputs the saved product image data into a multimodal AI model for analysis.

[0272] Step 10:

[0273] The server uses a multimodal AI model to identify the products that best fit the user's requirements, taking into account not only color and design features but also the user's emotional state as recognized by an emotion engine.

[0274] Step 11:

[0275] The server then lists the identified product information and generates a suggested message by adjusting the wording and tone of the message according to the user's emotional state. For example, if the user is in a "dissatisfied" state, it adds an encouraging message such as "This sofa has been highly rated by many users!"

[0276] Step 12:

[0277] The server transmits the generated proposal message to the user through the terminal.

[0278] Step 13:

[0279] The user reviews the list of suggested products and provides additional input if they require more information (e.g., "Show me more gray sofas").

[0280] Step 14:

[0281] The terminal receives new text input and sends it back to the server.

[0282] Step 15:

[0283] The server again uses the NLP engine to extract the new keyword "gray" and similarly performs emotion recognition using the emotion engine.

[0284] Step 16:

[0285] The server retrieves product data again based on the new keywords and analyzes it using a multimodal AI model.

[0286] Step 17:

[0287] The server generates an updated product list and again creates and sends suggestion messages based on the user's latest emotional state.

[0288] Step 18:

[0289] The user selects a particular product and enters "I would like to purchase this."

[0290] Step 19:

[0291] The terminal transmits text data of the purchase intention to the server.

[0292] Step 20:

[0293] The server calls the purchase API of the e-commerce site and initiates the purchase process using the user's account information.

[0294] Step 21:

[0295] The e-commerce site's server processes the purchase and sends a confirmation message back to the server.

[0296] Step 22:

[0297] The server generates a purchase completion message and notifies the user through the terminal.

[0298] Through these specific processing steps, users can efficiently find products that meet their requirements and smoothly complete the purchasing process.The use of the emotion engine allows users to receive appropriate product suggestions based on their emotional state, significantly improving the purchasing experience.

[0299] Example 2

[0300] 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."

[0301] Conventional e-commerce systems have had difficulty in quickly and accurately searching for and suggesting products that users want. Furthermore, they were unable to suggest products that corresponded to the user's emotional state, resulting in low satisfaction. Furthermore, the purchasing process was cumbersome, making it difficult to improve the user experience.

[0302] 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.

[0303] In this invention, the server includes: means for receiving text data entered by a user; means for analyzing the text data using a natural language processing engine to extract keywords and attribute information; means for recognizing the user's emotional state from the text data; means for acquiring product data from multiple e-commerce sites based on the extracted keywords; means for analyzing images of the acquired product data using a multimodal generative AI model to identify products that meet the user's requirements and emotional state; and means for suggesting the identified product information to the user in a message corresponding to the user's emotional state. This allows for quick and accurate suggestions of products desired by the user, and further enables optimal product suggestions based on the user's emotional state. In addition, a smoother purchase process is expected to improve the user experience.

[0304] "User" means an individual or corporation that uses the system.

[0305] "Text data" is text information entered by the user.

[0306] A "natural language processing engine" is a software tool or algorithm for analyzing text data and extracting keywords and attribute information.

[0307] A "keyword" is an important word or phrase that identifies the product or service a user is looking for.

[0308] "Attribute information" is data that indicates characteristics and conditions that embody a user's request.

[0309] The "emotional state" is a recognition of the psychological state of the user when he or she inputs text data.

[0310] An "e-commerce site" is a website that sells and buys goods and services online.

[0311] "Product data" is information about a product, including images, prices, links, etc.

[0312] A "multimodal generative AI model" is an artificial intelligence model that integrates and analyzes multiple data formats (text, images, etc.).

[0313] "Analysis" is the processing of collected data to extract meaning and information.

[0314] "Proposal" is the act of providing appropriate product or service information to a user.

[0315] A "message" is text containing notifications or information sent from the system to a user.

[0316] A "purchase interface" is a software component that executes a purchase transaction on an e-commerce site.

[0317] The "purchase procedure" is a series of steps for purchasing the product or service selected by the user.

[0318] The present invention combines emotion recognition with a system that helps users quickly find and purchase the products they desire. The system operates via a chat interface and makes product suggestions based on text data entered by the user.

[0319] The system mainly consists of a server and a terminal. The terminal receives text data entered by the user in the chat window, and the server analyzes the data. The specific operation of the system is explained below.

[0320] First, the user enters the product requirements into the chat window on the device. For example, they might type, "I want a new sofa." This text data is received by the device and sent to the server.

[0321] The server passes the received text data to a natural language processing engine (e.g., Google Cloud Natural Language API) to extract keywords and attribute information. At the same time, it uses an emotion recognition engine (e.g., open-source emotion recognition software) to analyze the user's emotional state (e.g., joy, dissatisfaction). The analysis results allow the user's psychological state to be understood.

[0322] Next, the server retrieves product data (e.g., product images, price information, links, etc.) from multiple e-commerce sites (e.g., large online marketplaces) based on the extracted keywords, using the APIs of the e-commerce sites (e.g., widely used e-commerce APIs).

[0323] The server inputs the acquired product data into a multimodal generative AI model (e.g., CLIP model) and performs an integrated analysis of image and text elements. The system also takes the user's emotional state into consideration. For example, if the emotional state is "dissatisfied," it will prioritize suggestions of highly rated products and discount information.

[0324] The server generates product suggestions appropriate for the user based on the analysis results. The message is generated by adjusting the wording and tone of the message according to the user's emotional state. The message includes product images, prices, and links, and displays a message such as "We recommend this sofa!"

[0325] This allows users to view product details in the chat window and receive highly satisfying recommendations. For example, if the user types "Show me more gray sofas," the device sends this new text to the server, which again uses its natural language processing engine to extract the keyword "gray" and performs the same search and analysis. The emotion engine is also utilized again to adjust the recommendations based on the user's latest emotional state.

[0326] Finally, if the user selects the product and enters "I want to purchase this," the terminal sends this purchase intention to the server. The server starts the purchase procedure in cooperation with the purchase interface of the e-commerce site, and receives a confirmation message from the e-commerce site. The server then generates a purchase completion message and notifies the user through the terminal.

[0327] This system allows users to efficiently search for and purchase products, and provides an excellent purchasing experience by receiving appropriate suggestions based on their emotional state.

[0328] Prompt Sentence Examples

[0329] For example, if a user types "I'm looking for a red dress," the device sends the text data to the server, which then retrieves and suggests product data in a similar manner, using a search query based on the keyword "red dress" and further adjusting it according to the user's emotional state.

[0330] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0331] Step 1:

[0332] The user types "I want a new sofa" into the chat window. The device receives this text data and stores it in its internal memory. The input is text data based on the user's request, and the output is the stored text data. A specific example of this behavior is when the user types "I want a new sofa" and presses the SEND button.

[0333] Step 2:

[0334] The device sends the received text data to the server, which then analyzes the text data using a natural language processing engine (for example, a natural language processing API). The input is the received text data, and the output is the extracted keyword "sofa" and attribute information "new." The analysis results are saved in internal memory. A specific example of operation is when the server sends text to a natural language processing API and receives the analysis results.

[0335] Step 3:

[0336] The server passes the text data to an emotion recognition engine (e.g., emotion recognition software) to recognize the user's emotional state (e.g., "joy"). The input is text data, and the output is the emotion recognition result (type of emotion). A specific example of operation is sending text data to an emotion recognition engine to obtain the emotional state.

[0337] Step 4:

[0338] The server retrieves product data using the APIs of multiple e-commerce sites (e.g., large online marketplaces) based on the extracted keyword "sofa." The input is the keyword "sofa," and the output is the retrieved product data (product images, price information, and links). A specific example of the operation is to send a keyword to the API of an e-commerce site and receive product data.

[0339] Step 5:

[0340] The server inputs the acquired product data images into a multimodal generative AI model (e.g., an AI model) and analyzes the suitability of the product. The input is the product data image, and the output is the set product feature vector. The user's emotional state is also taken into consideration here. A specific example of operation is inputting a product image into the AI ​​model and obtaining a feature vector.

[0341] Step 6:

[0342] Based on the analysis results, the server selects product information appropriate for the user and generates a message according to their emotional state. The input is the user's emotional state and the product's feature vector, and the output is the suggested product information and the generated message. As a specific example of operation, if the emotional state is "dissatisfied," a message containing highly rated products and discount information is generated.

[0343] Step 7:

[0344] The server sends a proposal message to the terminal, and the terminal displays the message to the user. The input is the proposal message sent from the server, and the output is the proposal message displayed in the user's chat window. A specific example of the operation is to display the proposal message in the chat window.

[0345] Step 8:

[0346] The user inputs (or clicks) "I want to buy this" for the selected product, and that information is sent to the server by the terminal. The input is the user's intention to purchase, and the output is a purchase request sent to the server. A specific example of this behavior is when the user clicks the product purchase button.

[0347] Step 9:

[0348] The server sends the user's purchase information to the e-commerce site's purchase interface and begins the purchase process. The input is the user's purchase information, and the output is a notification that the purchase process has been completed. A specific example of this operation is sending the user's account information and product information to the purchase API.

[0349] Step 10:

[0350] The e-commerce site completes the purchase process and sends a confirmation message to the server. The server generates a purchase completion message and sends it to the terminal, which then notifies the user. The input is the purchase completion notification and the output is the purchase completion message. A specific example of this behavior is when the terminal displays "Purchase completed!" in a chat window.

[0351] In this way, the system can suggest optimal products based on the user's desires and emotional state, and efficiently process the purchase.

[0352] (Application example 2)

[0353] 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."

[0354] Conventional e-commerce systems provide insufficient support for users to find products quickly and appropriately, and in particular do not provide product suggestions that take into account the user's emotional state. Furthermore, they are unable to respond flexibly to the user's emotions, which can lead to a decrease in satisfaction. Furthermore, when a user makes additional input to a product suggestion, a fast and accurate response to that request is required, but current systems do not adequately achieve this.

[0355] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0356] In this invention, the server includes: means for receiving text data entered by a user; means for analyzing the text data using a natural language processing engine to extract keywords and attribute information; means for acquiring product data from multiple e-commerce sites based on the extracted keywords; means for analyzing images of the acquired product data using a multimodal AI model to identify products that meet the user's requirements; means for recognizing the user's emotional state and adjusting product suggestions based on the recognized emotional state; and means for suggesting the identified product information to the user using a message with wording and tone that matches the user's emotional state. This allows the server to appropriately determine the user's emotional state and make optimal product suggestions based on that state, thereby improving the user's purchasing experience. Furthermore, the server can quickly respond to additional user input and re-suggest suitable products, thereby flexibly meeting user requests.

[0357] A "user" is a person who uses the system to search for products and consider purchasing them.

[0358] "Text data" refers to character information entered by the user.

[0359] A "natural language processing engine" is a computer program that analyzes and interprets text data and extracts keywords and attribute information.

[0360] "Keywords" are important words that represent products or requirements identified from the text entered by the user.

[0361] "Attribute information" refers to additional information that accompanies a keyword, such as adjectives like "new" or "cheap."

[0362] An "e-commerce site" is a website that sells and buys goods over the Internet.

[0363] "Product data" is information about a product, specifically, product name, price, image, description, etc.

[0364] A "multimodal AI model" is an artificial intelligence model that can simultaneously analyze multiple forms of data, such as text and images.

[0365] "User requirements" refer to the product characteristics and conditions desired by the user.

[0366] "Emotional state" refers to the psychological state (e.g., joy, dissatisfaction, etc.) that a user experiences when searching for a product.

[0367] "Product suggestion" refers to the system recommending products selected by the system to the user.

[0368] "Message wording and tone" refers to the content of the product proposal message, which is adjusted according to the emotional state, and how it is delivered.

[0369] "Additional input" refers to new text information added by the user after the initial input.

[0370] The "purchase procedure" refers to a series of steps required for the user to actually purchase the product selected by the user.

[0371] A "purchase API" is an application programming interface that works with e-commerce sites to automate the product purchasing process.

[0372] The present invention is a system that allows users to easily and quickly find and purchase the products they desire, and is combined with an emotion engine. This system operates via a chat interface, and when users input their product requests, it suggests products that match those requests.

[0373] System Configuration

[0374] The system mainly includes the following elements:

[0375] Terminal: A device that allows users to input text data. This includes smartphones and PCs.

[0376] Server: A central system that analyzes text data, acquires product data, recognizes emotions, and generates product suggestions.

[0377] Natural language processing engine: Software for analyzing text data and extracting keywords and attribute information.

[0378] Multimodal AI model: An artificial intelligence model for analyzing data in multiple formats, such as text data and image data.

[0379] Emotion Engine: An engine for recognizing and analyzing the user's emotional state.

[0380] Process Flow

[0381] 1. Accept user input:

[0382] The user uses the terminal to input text data into the chat window, for example, "I want a new sofa."

[0383] 2. Text data analysis:

[0384] The server passes the received text data to a natural language processing engine, which analyzes the text and extracts the important keyword "sofa" and attribute information "new" from the text.

[0385] 3. Emotion recognition:

[0386] The server uses an emotion engine to analyze the user's emotional state, for example classifying the user's text as an emotion such as "happy" or "frustrated."

[0387] 4. Product data acquisition:

[0388] The server retrieves product data from APIs of multiple e-commerce sites based on the extracted keywords, including product images, price information, and detailed descriptions.

[0389] 5. Multimodal AI analysis:

[0390] The server inputs the acquired product data into a multimodal AI model for analysis, and the analysis results take into account the user's emotional state.

[0391] 6. Product proposal generation:

[0392] The server then lists the identified product information and generates product suggestions with message wording and tone that correspond to the user's emotional state. For example, it could provide a message such as "This sofa is recommended!" along with product images, prices, and links via chat.

[0393] Hardware and software used

[0394] Hardware: smartphones, PCs, servers

[0395] software:

[0396] OpenAI API: An API for emotion recognition.

[0397] TextBlob: A tool that performs natural language processing and extracts keywords from text.

[0398] requests library: A library for obtaining product data using the API of an e-commerce site.

[0399] Specific examples

[0400] For example, if the user again types "Show me more gray sofas," the device sends this new text to the server, which again uses the NLP engine to extract the new keyword "gray" and performs similar search and image analysis to generate new suggestions for gray sofas, again using the emotion engine to adjust the suggestions based on the user's latest emotional state.

[0401] Prompt Sentence Examples

[0402] "I want a new sofa"

[0403] "Show me a cheaper sofa."

[0404] "I want to see a gray sofa."

[0405] As described above, this system realizes flexible product suggestions that take into account the user's emotional state, thereby significantly improving the user's purchasing experience.

[0406] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0407] Step 1:

[0408] Accepting user input

[0409] Operation: A user inputs a request for a product into a chat window on the terminal.

[0410] Input: Text data entered by the user: "I want a new sofa."

[0411] Output: The user's input text is sent from the terminal to the server.

[0412] Step 2:

[0413] Text data analysis

[0414] How it works: The server passes the text data to a natural language processing engine for analysis, extracting important keywords and attribute information.

[0415] Input: User input text "I want a new sofa."

[0416] Data processing: A natural language processing engine is used to analyze the text and extract the attribute information "new" and the keyword "sofa."

[0417] Output: The extracted keyword information of "new" and "sofa" is obtained.

[0418] Step 3:

[0419] emotion recognition

[0420] Operation: The server uses the emotion engine to analyze the user's emotional state.

[0421] Input: User input text "I want a new sofa."

[0422] Data processing: Using an emotion engine to classify emotions such as "happy" or "frustrated" from user text.

[0423] Output: The user's emotional state (e.g., "joy" is recognized).

[0424] Step 4:

[0425] Get product data

[0426] Operation: The server retrieves product data from the APIs of multiple e-commerce sites based on the extracted keywords.

[0427] Input: The extracted keyword "sofa" and attribute information "new".

[0428] Data processing: Call the API of the e-commerce site and obtain the relevant product information (product images, price information, detailed descriptions, etc.).

[0429] Output: A list of retrieved product data.

[0430] Step 5:

[0431] Multimodal AI analysis

[0432] Operation: The server inputs the acquired product data images into a multimodal AI model to identify products that meet the user's requirements.

[0433] Input: A list of retrieved product data and the user's emotional state.

[0434] Data processing: Using a multimodal AI model to analyze images and evaluate products based on emotional states.

[0435] Output: A list of products that match the user requirements.

[0436] Step 6:

[0437] Generate product suggestions

[0438] Operation: The server generates a suggestion message based on the identified product information with a message wording and tone that corresponds to the emotional state.

[0439] Input: A list of products that match the user requirements and the user's emotional state.

[0440] Data processing: Generate appropriate messages taking into account the emotional state and list the identified product information.

[0441] Output: Product suggestion message to be presented to the user.

[0442] Step 7:

[0443] Presenting product suggestions to users

[0444] Operation: The server sends the generated proposal message to the terminal and presents it to the user.

[0445] Input: The generated product suggestion message.

[0446] Output: Product suggestion message displayed in the device's chat window.

[0447] Example: "Show me more gray sofas."

[0448] When the user makes this input, "gray" is extracted as a new keyword, and data is acquired and analyzed again, generating new product proposals that match the user's requirements.

[0449] 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.

[0450] 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.

[0451] 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.

[0452] [Second embodiment]

[0453] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0454] 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.

[0455] 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).

[0456] 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.

[0457] 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.

[0458] 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).

[0459] 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.

[0460] 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.

[0461] 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.

[0462] 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.

[0463] 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.

[0464] 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."

[0465] The present invention is a system that allows users to easily and quickly find and purchase the products they desire. The system operates via a chat interface, allowing users to input requirements for the products they want, and then suggests products that match those requirements.

[0466] Program processing and specific examples

[0467] 1. Accepting input from the user

[0468] The terminal receives text data in which the user enters "I want a new sofa" in the chat window.

[0469] 2. Text interpretation and requirement extraction

[0470] The server passes the received text data to a natural language processing engine for analysis, which extracts the keyword "sofa" and the attribute information "new."

[0471] 3. Providing product images

[0472] The server retrieves product data (product images, price information, links, etc.) from APIs of multiple e-commerce sites based on the extracted keyword "sofa."

[0473] 4. Image Classification and Requirements Compliance Assessment

[0474] The server inputs the captured product images into a multimodal AI model to identify the product that best suits the user's requirements, for example, evaluating color and design features to select a modern sofa that meets the "new" requirements.

[0475] 5. Generate product suggestions and present them to the user

[0476] The server formats the identified product information and provides the user with a message via chat saying, "How about this sofa?" along with an image, price, and link to the selected product.

[0477] Specific examples

[0478] For example, if the user types "Show me more gray sofas," the device sends this new text to the server, which again uses its NLP engine to extract the new keyword "gray" and performs similar search and image analysis to generate new suggestions for gray sofas, allowing the user to find products that match their more specific requirements.

[0479] Purchase procedure

[0480] When the user selects a product and enters "I want to purchase this," the device sends this purchase intention to the server. The server then sends the user's account information to the e-commerce site's purchase API, automatically starting the purchase process. The e-commerce site then completes the purchase process and returns a confirmation message to the server. The server then notifies the user of the purchase completion message.

[0481] This system allows users to efficiently search for and purchase products, providing a superior purchasing experience.

[0482] The processing flow will be explained below.

[0483] Step 1:

[0484] The user types "I want a new sofa" into the chat window.

[0485] Step 2:

[0486] The terminal receives this text input and transmits it to the server in real time.

[0487] Step 3:

[0488] The server passes the received text data to a natural language processing engine for analysis.

[0489] Step 4:

[0490] The server uses a natural language processing engine to extract the keyword "sofa" and the attribute information "new" from the input text.

[0491] Step 5:

[0492] Based on the extracted keyword "sofa," the server sends requests to APIs of multiple e-commerce sites to retrieve related product data (product images, price information, links, etc.).

[0493] Step 6:

[0494] In response to the request, the server of the e-commerce site returns product data that matches the conditions to the server.

[0495] Step 7:

[0496] The server stores the acquired product data in a local database or memory.

[0497] Step 8:

[0498] The server inputs the saved product image data into a multimodal AI model for analysis.

[0499] Step 9:

[0500] The server uses a multimodal AI model to identify the product that best suits the user's requirements, evaluating color and design features to determine whether the product meets the user's requirements.

[0501] Step 10:

[0502] The server lists the identified product information and generates a suggestion message for the user, including product images, price information, and product links.

[0503] Step 11:

[0504] The server transmits the generated proposal message to the user through the terminal.

[0505] Step 12:

[0506] The user reviews the list of suggested products and provides additional input if more information is desired.

[0507] Step 13:

[0508] The terminal receives new text input and sends it back to the server.

[0509] Step 14:

[0510] The server passes the new text data to the natural language processing engine again to extract keywords and attribute information.

[0511] Step 15:

[0512] The server retrieves the product data again based on the new information extracted and analyzes it using a multimodal AI model.

[0513] Step 16:

[0514] The server generates an updated product list and again sends a suggestion message to the user.

[0515] Step 17:

[0516] The user selects a particular product and enters "I would like to purchase this."

[0517] Step 18:

[0518] The terminal transmits text data of the purchase intention to the server.

[0519] Step 19:

[0520] The server calls the purchase API of the e-commerce site and initiates the purchase process using the user's account information.

[0521] Step 20:

[0522] The e-commerce site's server processes the purchase and sends a confirmation message back to the server.

[0523] Step 21:

[0524] The server generates a purchase completion message and notifies the user through the terminal.

[0525] By going through each step, users can efficiently find products that meet their requirements and complete the purchasing process smoothly.

[0526] Example 1

[0527] 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."

[0528] Conventional online shopping systems have made it difficult for users to quickly and accurately find the products they are looking for. In particular, text-based searches have been problematic because they are unable to efficiently suggest products due to their inability to accommodate vague keywords and diverse user needs. Furthermore, there has been no system that can retrieve product information from multiple online markets and present products that best meet the user's requirements. As a result, users are forced to spend a long time searching for products, which often discourages them from making a purchase.

[0529] 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.

[0530] In this invention, the server includes means for receiving text data input by a user, means for analyzing the text data using a natural language processing engine to extract keywords and attribute information, means for acquiring product data from multiple online markets based on the extracted keywords, means for analyzing images of the acquired product data using a multi-function AI model to identify products that meet the user's requirements, and means for suggesting the identified product information to the user, thereby enabling the user to quickly and accurately find the product they are looking for.

[0531] "User" refers to an individual or corporation that uses the system to search for and purchase products.

[0532] "Text data" refers to character string information entered by the user, and refers to sentences used to explain searches and requirements.

[0533] A "natural language processing engine" refers to software or algorithms that analyze text data and extract keywords and attribute information.

[0534] "Keywords" refer to the main words or phrases that describe the product or its characteristics that a user is searching for.

[0535] "Attribute information" refers to information that indicates specific characteristics or requirements related to a keyword.

[0536] "Online marketplace" refers to platforms and websites that sell products over the internet.

[0537] "Product Data" means product information obtained from an online marketplace, including images, prices, links, etc.

[0538] A "multi-functional AI model" refers to an artificial intelligence model that integrates and analyzes data in multiple different formats to identify the products that best suit the user's requirements.

[0539] "Product information" is detailed information about a product suggested to a user, including an image, price, link, and the like.

[0540] The present invention is a system that helps users quickly and accurately find and purchase the products they are looking for. The system operates via a chat interface and suggests products based on product requirements entered by the user.

[0541] Hardware and Software

[0542] This system consists of a user terminal, a server, and multiple APIs. The main hardware and software used are as follows:

[0543] User device: Smartphone, tablet, PC, etc. Used by users to input text data through the chat window.

[0544] Server: A virtual server provided by a cloud service provider. For example, commonly used cloud platforms include Amazon Web Services (AWS) and Google Cloud Platform (GCP).

[0545] Natural language processing engine: Software used to analyze text data. Google Cloud Natural Language API is one such example.

[0546] Multimodal AI model: An AI model that analyzes image and text data to recommend optimal products. Typically, OpenAI's CLIP (Contrastive Language-Image Pretraining) model is used.

[0547] Online marketplace API: An interface for obtaining product data from multiple online marketplaces. For example, Amazon Marketplace API or Rakuten Marketplace API are used.

[0548] Specific operation of the system

[0549] When a user opens a chat window and types "I want a new sofa," the device receives this text data, stores it in its internal memory, and then sends it to the server.

[0550] The server passes the received text data to a natural language processing engine for analysis, which extracts the keyword "sofa" and the attribute information "new."

[0551] The server then sends requests to multiple online marketplace APIs based on these extracted keywords. For example, it uses the APIs of Amazon and Rakuten to perform a product search using the keyword "sofa." The product data retrieved from these online marketplaces includes images, prices, links, etc.

[0552] The acquired product data is input by the server into a multi-function AI model, which identifies the most suitable product based on the product image and extracted attribute information. For example, using the CLIP model, it matches the image with the keyword "new sofa" and ranks the most relevant products.

[0553] The identified product information is converted into JSON format and then reformatted to fit the chat interface. The user is then provided with an image, price, and link to the suggested product along with the message, "How about this sofa?"

[0554] Specific examples

[0555] For example, if the user types "Show me more gray sofas," the device receives this new text data and sends it to the server. The server then uses its NLP engine to extract the new keyword "gray" and requests a new search from the online marketplace API. It then performs image analysis on the new product data and suggests the gray sofa that best suits the user's requirements.

[0556] Prompt Sentence Examples

[0557] A user types the following into a chat window:

[0558] "I want a new sofa"

[0559] "Show me more gray sofas."

[0560] This system allows users to efficiently search for and purchase products, providing a superior purchasing experience.

[0561] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0562] Step 1:

[0563] The terminal receives text data entered into the chat window by the user.

[0564] Input: A user types "I want a new sofa" into a chat window.

[0565] Operation: The terminal stores this text data in its internal memory and generates a data packet.

[0566] Output: Send the generated data packet to the server.

[0567] Step 2:

[0568] The server receives the text data sent from the terminal.

[0569] Input: Data packets from the terminal.

[0570] How it works: The server takes this text data and prepares it to be passed to a natural language processing (NLP) engine.

[0571] Output: Text data ready for analysis.

[0572] Step 3:

[0573] The server analyzes the text data using a natural language processing engine to extract keywords and attribute information.

[0574] Input: Text data ready for analysis.

[0575] How it works: The server parses the text data using the Google Cloud Natural Language API.

[0576] Output: Keyword "sofa" and attribute information "new".

[0577] Step 4:

[0578] The server retrieves product data from multiple online marketplaces based on the extracted keywords.

[0579] Input: keyword "sofa".

[0580] Operation: The server sends an HTTP request to the Amazon Marketplace API and the Rakuten Marketplace API to search for products using the keyword "sofa."

[0581] Output: Retrieved product data (images, prices, links, etc.).

[0582] Step 5:

[0583] The server analyzes the acquired product data images using a multi-function AI model and identifies products that meet the user's requirements.

[0584] Input: Retrieved product data.

[0585] How it works: The server uses OpenAI's CLIP model to match product images with the attribute "new" and rank the products that best match the user's requirements.

[0586] Output: A ranked list of products.

[0587] Step 6:

[0588] The server proposes the identified product information to the user.

[0589] Input: A ranked list of products.

[0590] What it does: The server converts this information into JSON format and reformats it into a format suitable for a chat interface with the message "How would you like this sofa?"

[0591] Output: Send the formatted product suggestion message to the terminal.

[0592] Step 7:

[0593] The terminal displays the product suggestion message sent from the server to the user.

[0594] Input: A formatted product suggestion message from the server.

[0595] What it does: The device displays this message in a chat window, providing the user with a picture of the product, its price, and a link.

[0596] Output: Product suggestions to the user.

[0597] Step 8:

[0598] The user enters additional requests for the suggested products, for example, "Show me more gray sofas."

[0599] Input: Additional input from the user.

[0600] Operation: The user enters a new request into the chat window of the terminal, which receives it and prepares to retransmit it to the server.

[0601] Output: New data packet.

[0602] Step 9:

[0603] The server receives the new text data again and again uses the NLP engine to extract keywords and attribute information.

[0604] Input: A new data packet.

[0605] How it works: The server uses the Google Cloud Natural Language API to parse newly received text data and extract the keyword "gray."

[0606] Output: New keyword and attribute information.

[0607] Step 10:

[0608] The server retrieves product data again based on the new keywords, and identifies and suggests more suitable products.

[0609] Input: new keyword "gray".

[0610] Operation: The server requests a re-search from the Amazon Marketplace API and Rakuten Market API to obtain product data for "gray sofas." It then analyzes the product information using a multi-function AI model.

[0611] Output: Send new product suggestion message to the terminal.

[0612] (Application example 1)

[0613] 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."

[0614] Conventional e-commerce systems have faced challenges in making it difficult for users to efficiently search for products and quickly find those that meet their requirements. Furthermore, product search and suggestion functions using voice input and augmented reality (AR) technology interfaces were inadequate, leaving a need for an improved user experience.

[0615] 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.

[0616] In this invention, the server includes means for receiving text data input by a user, means for analyzing the text data using a natural language processing engine to extract keywords and attribute information, means for acquiring product data from multiple e-commerce sites based on the extracted keywords, means for analyzing images of the acquired product data using a multimodal AI model to identify products that meet the user's requirements, means for suggesting the identified product information to the user, means for converting voice data into text data, and means for presenting product information in augmented reality on a visual display device, thereby enabling users to easily search for products through voice input, visually confirm product information using AR technology, and complete the purchase process efficiently and intuitively.

[0617] "User" refers to a consumer who searches for and purchases products using this system.

[0618] "Text data" refers to character information including requirements and conditions regarding products entered by a user.

[0619] "Natural language processing engine" refers to software and algorithms for understanding and analyzing human language.

[0620] "Keywords" refer to the main words necessary for searching for products, extracted from the user's input text.

[0621] "Attribute information" refers to specific characteristics or conditions related to a keyword.

[0622] "E-commerce site" refers to a website that sells goods and services over the Internet.

[0623] "Product Data" refers to information about a product obtained from an e-commerce site, primarily including the name, image, price, and link.

[0624] A "multimodal AI model" refers to an artificial intelligence model that integrates and analyzes multiple types of data (e.g., text data, image data).

[0625] A "visual display device" is a device that presents information to a user's field of vision, and primarily refers to smart glasses and augmented reality devices.

[0626] "Augmented reality (AR)" refers to the technology of overlaying digital information onto a real-world environment.

[0627] "Voice data" refers to sound information obtained from the user's speech.

[0628] "Product suggestion" refers to the act of presenting the most suitable product to the user based on analyzed product data.

[0629] The present invention is a system for enabling users to easily and quickly find and purchase desired products, and is particularly directed to a shopping assistant application that uses smart glasses. The system is configured as follows.

[0630] The server receives the text data entered by the user and analyzes it using a natural language processing engine. When the user enters product requirements by voice, the smart glasses' microphone picks up the voice data and converts it into text data. The converted text data is then analyzed by the natural language processing engine to extract keywords and attribute information.

[0631] Once the keywords are extracted, the server accesses the APIs of multiple e-commerce sites to retrieve the corresponding product data. An e-commerce site is a website that sells products and services over the Internet, and the product data includes the product name, image, price, and purchase link.

[0632] The acquired product data images are analyzed using a multimodal AI model, which is an artificial intelligence model that integrates and analyzes multiple types of data (e.g., text data, image data, etc.). This allows the system to identify the product that best suits the user's requirements.

[0633] The identified product information is then formatted and suggested to the user. The suggested product information is displayed as augmented reality (AR) on the smart glasses display. Augmented reality is a technology that overlays digital information on the real-world environment, allowing the user to intuitively check the product information superimposed on their field of vision.

[0634] For example, if a user says "I'm looking for red sneakers," the voice data is picked up through the smart glasses' microphone and converted into text data. The server then uses a natural language processing engine to analyze the converted text data and extract the keyword "red sneakers." Product data is then retrieved using the API of the e-commerce site, and analyzed using a multimodal AI model to identify the most suitable product.

[0635] As a result, product information for "red sneakers" is displayed in AR within the user's field of vision. When the user finds a specific product and says "I want to buy this," the server automatically initiates the purchase process through the e-commerce site's purchase API and sends a confirmation message to the user.

[0636] This system allows users to search for and purchase products efficiently and intuitively, providing an excellent purchasing experience.

[0637] Example prompt sentence:

[0638] The user speaks, "I'm looking for red sneakers."

[0639] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0640] Step 1:

[0641] The user speaks into the microphone of the smart glasses to input information about a product. For example, they might say, "I'm looking for red sneakers." This voice data is picked up by the microphone of the smart glasses. The input is voice data, and the output is voice data.

[0642] Step 2:

[0643] The device converts the acquired voice data into text data. This is done using a voice recognition library (e.g., Google's voice recognition API). This converts the voice data into text data. The input is voice data, and the output is text data.

[0644] Step 3:

[0645] Text data is sent to the server. The server analyzes the text data using a natural language processing engine (e.g., the BERT model) and extracts keywords and attribute information. This yields a keyword, for example, "red sneakers." The input is text data, and the output is keywords and attribute information.

[0646] Step 4:

[0647] The server accesses the APIs of multiple e-commerce sites based on the extracted keywords to retrieve related product data, including product names, images, prices, and purchase links. The input is the keywords, and the output is the product data.

[0648] Step 5:

[0649] The server inputs the acquired product data images into a multimodal AI model (e.g., CLIP model) and analyzes them. This identifies products that meet the user's requirements. The input is the product image, and the output is the identified product information.

[0650] Step 6:

[0651] The server formats the identified product information and presents it as augmented reality (AR) on the smart glasses display, allowing the user to view the product information directly in their field of vision. The input is the identified product information, and the output is the AR-displayed product information.

[0652] Step 7:

[0653] The user checks the display on the smart glasses, selects a specific product, and then voice-inputs, "I would like to purchase this." The input is voice data, and the output is text data indicating the purchase intention.

[0654] Step 8:

[0655] The terminal receives the voice data again, converts it into text data, and sends it to the server. The input is voice data, and the output is text data.

[0656] Step 9:

[0657] The server analyzes the received text data, confirms the user's purchase intention, and executes the purchase procedure via the e-commerce site's purchase API. The input is the text data of the purchase intention, and the output is purchase confirmation information.

[0658] Step 10:

[0659] The server confirms that the purchase has been completed and notifies the user of the result. The user receives a confirmation message on their smart glasses. The input is the purchase confirmation information, and the output is a user notification.

[0660] 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.

[0661] The present invention combines an emotion engine with a system that allows users to easily and quickly find and purchase the products they desire. The system operates via a chat interface, allowing users to input their desired product requirements and then suggests products that match those requirements. The system can also recognize the user's emotional state and adjust product suggestions based on the results.

[0662] Program processing and specific examples

[0663] 1. Accepting input from the user

[0664] The terminal receives text data in which the user enters "I want a new sofa" in the chat window.

[0665] 2. Text Interpretation and Emotion Recognition

[0666] The server passes the received text data to a natural language processing engine, analyzes the text, and uses an emotion engine to recognize the user's emotional state. For example, the emotion engine classifies the user's text as an emotion such as "joy" or "dissatisfaction."

[0667] 3. Extracting keywords and attribute information

[0668] The server uses a natural language processing engine to extract the keyword "sofa" and the attribute information "new" from the input text.

[0669] 4. Providing product images

[0670] The server retrieves product data (product images, price information, links, etc.) from APIs of multiple e-commerce sites based on the extracted keyword "sofa."

[0671] 5. Image Classification and Requirements Compliance Assessment

[0672] The server inputs the acquired product image data into a multimodal AI model for analysis. This analysis also takes into account the user's emotional state. For example, if a user's emotional state is "dissatisfied," the server will prioritize suggesting highly rated products and discount information.

[0673] 6. Generate product suggestions and present them to the user

[0674] The server then lists the identified product information and generates a suggestion message by adjusting the wording and tone of the message according to the user's emotional state. For example, it might provide a message such as "This sofa is recommended!" along with an image, price, and link to the selected product via chat.

[0675] Specific examples

[0676] For example, if the user again types "Show me more gray sofas," the device sends this new text to the server, which again uses the NLP engine to extract the new keyword "gray" and performs similar search and image analysis to generate new suggestions for gray sofas, again using the emotion engine to adjust the suggestions based on the user's latest emotional state.

[0677] Purchase procedure

[0678] When the user selects a product and enters "I want to purchase this," the device sends this purchase intention to the server. The server then sends the user's account information to the e-commerce site's purchase API, automatically starting the purchase process. The e-commerce site then completes the purchase process and returns a confirmation message to the server. The server then generates a purchase completion message and notifies the user via the device.

[0679] This system allows users to efficiently search for and purchase products, and provides an excellent purchasing experience with appropriate suggestions based on their emotional state.

[0680] The processing flow will be explained below.

[0681] Step 1:

[0682] The user types "I want a new sofa" into the chat window.

[0683] Step 2:

[0684] The terminal receives this text input and transmits it to the server in real time.

[0685] Step 3:

[0686] The server passes the received text data to a natural language processing engine and analyzes the text.

[0687] Step 4:

[0688] The server uses a natural language processing engine to extract the keyword "sofa" and the attribute information "new" from the input text.

[0689] Step 5:

[0690] The server inputs text data into an emotion engine to recognize the user's emotional state (e.g., "joy," "dissatisfaction," "excitement," etc.).

[0691] Step 6:

[0692] Based on the extracted keyword "sofa," the server sends requests to APIs of multiple e-commerce sites to retrieve related product data (product images, price information, links, etc.).

[0693] Step 7:

[0694] In response to the request, the server of the e-commerce site returns product data that matches the conditions to the server.

[0695] Step 8:

[0696] The server stores the acquired product data in a local database or memory.

[0697] Step 9:

[0698] The server inputs the saved product image data into a multimodal AI model for analysis.

[0699] Step 10:

[0700] The server uses a multimodal AI model to identify the products that best fit the user's requirements, taking into account not only color and design features but also the user's emotional state as recognized by an emotion engine.

[0701] Step 11:

[0702] The server then lists the identified product information and generates a suggested message by adjusting the wording and tone of the message according to the user's emotional state. For example, if the user is in a "dissatisfied" state, it adds an encouraging message such as "This sofa has been highly rated by many users!"

[0703] Step 12:

[0704] The server transmits the generated proposal message to the user through the terminal.

[0705] Step 13:

[0706] The user reviews the list of suggested products and provides additional input if they require more information (e.g., "Show me more gray sofas").

[0707] Step 14:

[0708] The terminal receives new text input and sends it back to the server.

[0709] Step 15:

[0710] The server again uses the NLP engine to extract the new keyword "gray" and similarly performs emotion recognition using the emotion engine.

[0711] Step 16:

[0712] The server retrieves product data again based on the new keywords and analyzes it using a multimodal AI model.

[0713] Step 17:

[0714] The server generates an updated product list and again creates and sends suggestion messages based on the user's latest emotional state.

[0715] Step 18:

[0716] The user selects a particular product and enters "I would like to purchase this."

[0717] Step 19:

[0718] The terminal transmits text data of the purchase intention to the server.

[0719] Step 20:

[0720] The server calls the purchase API of the e-commerce site and initiates the purchase process using the user's account information.

[0721] Step 21:

[0722] The e-commerce site's server processes the purchase and sends a confirmation message back to the server.

[0723] Step 22:

[0724] The server generates a purchase completion message and notifies the user through the terminal.

[0725] Through these specific processing steps, users can efficiently find products that meet their requirements and smoothly complete the purchasing process.The use of the emotion engine allows users to receive appropriate product suggestions based on their emotional state, significantly improving the purchasing experience.

[0726] Example 2

[0727] 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."

[0728] Conventional e-commerce systems have had difficulty in quickly and accurately searching for and suggesting products that users want. Furthermore, they were unable to suggest products that corresponded to the user's emotional state, resulting in low satisfaction. Furthermore, the purchasing process was cumbersome, making it difficult to improve the user experience.

[0729] 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.

[0730] In this invention, the server includes: means for receiving text data entered by a user; means for analyzing the text data using a natural language processing engine to extract keywords and attribute information; means for recognizing the user's emotional state from the text data; means for acquiring product data from multiple e-commerce sites based on the extracted keywords; means for analyzing images of the acquired product data using a multimodal generative AI model to identify products that meet the user's requirements and emotional state; and means for suggesting the identified product information to the user in a message corresponding to the user's emotional state. This allows for quick and accurate suggestions of products desired by the user, and further enables optimal product suggestions based on the user's emotional state. In addition, a smoother purchase process is expected to improve the user experience.

[0731] "User" means an individual or corporation that uses the system.

[0732] "Text data" is text information entered by the user.

[0733] A "natural language processing engine" is a software tool or algorithm for analyzing text data and extracting keywords and attribute information.

[0734] A "keyword" is an important word or phrase that identifies the product or service a user is looking for.

[0735] "Attribute information" is data that indicates characteristics and conditions that embody a user's request.

[0736] The "emotional state" is a recognition of the psychological state of the user when he or she inputs text data.

[0737] An "e-commerce site" is a website that sells and buys goods and services online.

[0738] "Product data" is information about a product, including images, prices, links, etc.

[0739] A "multimodal generative AI model" is an artificial intelligence model that integrates and analyzes multiple data formats (text, images, etc.).

[0740] "Analysis" is the processing of collected data to extract meaning and information.

[0741] "Proposal" is the act of providing appropriate product or service information to a user.

[0742] A "message" is text containing notifications or information sent from the system to a user.

[0743] A "purchase interface" is a software component that executes a purchase transaction on an e-commerce site.

[0744] The "purchase procedure" is a series of steps for purchasing the product or service selected by the user.

[0745] The present invention combines emotion recognition with a system that helps users quickly find and purchase the products they desire. The system operates via a chat interface and makes product suggestions based on text data entered by the user.

[0746] The system mainly consists of a server and a terminal. The terminal receives text data entered by the user in the chat window, and the server analyzes the data. The specific operation of the system is explained below.

[0747] First, the user enters the product requirements into the chat window on the device. For example, they might type, "I want a new sofa." This text data is received by the device and sent to the server.

[0748] The server passes the received text data to a natural language processing engine (e.g., Google Cloud Natural Language API) to extract keywords and attribute information. At the same time, it uses an emotion recognition engine (e.g., open-source emotion recognition software) to analyze the user's emotional state (e.g., joy, dissatisfaction). The analysis results allow the user's psychological state to be understood.

[0749] Next, the server retrieves product data (e.g., product images, price information, links, etc.) from multiple e-commerce sites (e.g., large online marketplaces) based on the extracted keywords, using the APIs of the e-commerce sites (e.g., widely used e-commerce APIs).

[0750] The server inputs the acquired product data into a multimodal generative AI model (e.g., CLIP model) and performs an integrated analysis of image and text elements. The system also takes the user's emotional state into consideration. For example, if the emotional state is "dissatisfied," it will prioritize suggestions of highly rated products and discount information.

[0751] The server generates product suggestions appropriate for the user based on the analysis results. The message is generated by adjusting the wording and tone of the message according to the user's emotional state. The message includes product images, prices, and links, and displays a message such as "We recommend this sofa!"

[0752] This allows users to view product details in the chat window and receive highly satisfying recommendations. For example, if the user types "Show me more gray sofas," the device sends this new text to the server, which again uses its natural language processing engine to extract the keyword "gray" and performs the same search and analysis. The emotion engine is also utilized again to adjust the recommendations based on the user's latest emotional state.

[0753] Finally, if the user selects the product and enters "I want to purchase this," the terminal sends this purchase intention to the server. The server starts the purchase procedure in cooperation with the purchase interface of the e-commerce site, and receives a confirmation message from the e-commerce site. The server then generates a purchase completion message and notifies the user through the terminal.

[0754] This system allows users to efficiently search for and purchase products, and provides an excellent purchasing experience by receiving appropriate suggestions based on their emotional state.

[0755] Prompt Sentence Examples

[0756] For example, if a user types "I'm looking for a red dress," the device sends the text data to the server, which then retrieves and suggests product data in a similar manner, using a search query based on the keyword "red dress" and further adjusting it according to the user's emotional state.

[0757] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0758] Step 1:

[0759] The user types "I want a new sofa" into the chat window. The device receives this text data and stores it in its internal memory. The input is text data based on the user's request, and the output is the stored text data. A specific example of this behavior is when the user types "I want a new sofa" and presses the SEND button.

[0760] Step 2:

[0761] The device sends the received text data to the server, which then analyzes the text data using a natural language processing engine (for example, a natural language processing API). The input is the received text data, and the output is the extracted keyword "sofa" and attribute information "new." The analysis results are saved in internal memory. A specific example of operation is when the server sends text to a natural language processing API and receives the analysis results.

[0762] Step 3:

[0763] The server passes the text data to an emotion recognition engine (e.g., emotion recognition software) to recognize the user's emotional state (e.g., "joy"). The input is text data, and the output is the emotion recognition result (type of emotion). A specific example of operation is sending text data to an emotion recognition engine to obtain the emotional state.

[0764] Step 4:

[0765] The server retrieves product data using the APIs of multiple e-commerce sites (e.g., large online marketplaces) based on the extracted keyword "sofa." The input is the keyword "sofa," and the output is the retrieved product data (product images, price information, and links). A specific example of the operation is to send a keyword to the API of an e-commerce site and receive product data.

[0766] Step 5:

[0767] The server inputs the acquired product data images into a multimodal generative AI model (e.g., an AI model) and analyzes the suitability of the product. The input is the product data image, and the output is the set product feature vector. The user's emotional state is also taken into consideration here. A specific example of operation is inputting a product image into the AI ​​model and obtaining a feature vector.

[0768] Step 6:

[0769] Based on the analysis results, the server selects product information appropriate for the user and generates a message according to their emotional state. The input is the user's emotional state and the product's feature vector, and the output is the suggested product information and the generated message. As a specific example of operation, if the emotional state is "dissatisfied," a message containing highly rated products and discount information is generated.

[0770] Step 7:

[0771] The server sends a proposal message to the terminal, and the terminal displays the message to the user. The input is the proposal message sent from the server, and the output is the proposal message displayed in the user's chat window. A specific example of the operation is to display the proposal message in the chat window.

[0772] Step 8:

[0773] The user inputs (or clicks) "I want to buy this" for the selected product, and that information is sent to the server by the terminal. The input is the user's intention to purchase, and the output is a purchase request sent to the server. A specific example of this behavior is when the user clicks the product purchase button.

[0774] Step 9:

[0775] The server sends the user's purchase information to the e-commerce site's purchase interface and begins the purchase process. The input is the user's purchase information, and the output is a notification that the purchase process has been completed. A specific example of this operation is sending the user's account information and product information to the purchase API.

[0776] Step 10:

[0777] The e-commerce site completes the purchase process and sends a confirmation message to the server. The server generates a purchase completion message and sends it to the terminal, which then notifies the user. The input is the purchase completion notification and the output is the purchase completion message. A specific example of this behavior is when the terminal displays "Purchase completed!" in a chat window.

[0778] In this way, the system can suggest optimal products based on the user's desires and emotional state, and efficiently process the purchase.

[0779] (Application example 2)

[0780] 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."

[0781] Conventional e-commerce systems provide insufficient support for users to find products quickly and appropriately, and in particular do not provide product suggestions that take into account the user's emotional state. Furthermore, they are unable to respond flexibly to the user's emotions, which can lead to a decrease in satisfaction. Furthermore, when a user makes additional input to a product suggestion, a fast and accurate response to that request is required, but current systems do not adequately achieve this.

[0782] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0783] In this invention, the server includes: means for receiving text data entered by a user; means for analyzing the text data using a natural language processing engine to extract keywords and attribute information; means for acquiring product data from multiple e-commerce sites based on the extracted keywords; means for analyzing images of the acquired product data using a multimodal AI model to identify products that meet the user's requirements; means for recognizing the user's emotional state and adjusting product suggestions based on the recognized emotional state; and means for suggesting the identified product information to the user using a message with wording and tone that matches the user's emotional state. This allows the server to appropriately determine the user's emotional state and make optimal product suggestions based on that state, thereby improving the user's purchasing experience. Furthermore, the server can quickly respond to additional user input and re-suggest suitable products, thereby flexibly meeting user requests.

[0784] A "user" is a person who uses the system to search for products and consider purchasing them.

[0785] "Text data" refers to character information entered by the user.

[0786] A "natural language processing engine" is a computer program that analyzes and interprets text data and extracts keywords and attribute information.

[0787] "Keywords" are important words that represent products or requirements identified from the text entered by the user.

[0788] "Attribute information" refers to additional information that accompanies a keyword, such as adjectives like "new" or "cheap."

[0789] An "e-commerce site" is a website that sells and buys goods over the Internet.

[0790] "Product data" is information about a product, specifically, product name, price, image, description, etc.

[0791] A "multimodal AI model" is an artificial intelligence model that can simultaneously analyze multiple forms of data, such as text and images.

[0792] "User requirements" refer to the product characteristics and conditions desired by the user.

[0793] "Emotional state" refers to the psychological state (e.g., joy, dissatisfaction, etc.) that a user experiences when searching for a product.

[0794] "Product suggestion" refers to the system recommending products selected by the system to the user.

[0795] "Message wording and tone" refers to the content of the product proposal message, which is adjusted according to the emotional state, and how it is delivered.

[0796] "Additional input" refers to new text information added by the user after the initial input.

[0797] The "purchase procedure" refers to a series of steps required for the user to actually purchase the product selected by the user.

[0798] A "purchase API" is an application programming interface that works with e-commerce sites to automate the product purchasing process.

[0799] The present invention is a system that allows users to easily and quickly find and purchase the products they desire, and is combined with an emotion engine. This system operates via a chat interface, and when users input their product requests, it suggests products that match those requests.

[0800] System Configuration

[0801] The system mainly includes the following elements:

[0802] Terminal: A device that allows users to input text data. This includes smartphones and PCs.

[0803] Server: A central system that analyzes text data, acquires product data, recognizes emotions, and generates product suggestions.

[0804] Natural language processing engine: Software for analyzing text data and extracting keywords and attribute information.

[0805] Multimodal AI model: An artificial intelligence model for analyzing data in multiple formats, such as text data and image data.

[0806] Emotion Engine: An engine for recognizing and analyzing the user's emotional state.

[0807] Process Flow

[0808] 1. Accept user input:

[0809] The user uses the terminal to input text data into the chat window, for example, "I want a new sofa."

[0810] 2. Text data analysis:

[0811] The server passes the received text data to a natural language processing engine, which analyzes the text and extracts the important keyword "sofa" and attribute information "new" from the text.

[0812] 3. Emotion recognition:

[0813] The server uses an emotion engine to analyze the user's emotional state, for example classifying the user's text as an emotion such as "happy" or "frustrated."

[0814] 4. Product data acquisition:

[0815] The server retrieves product data from APIs of multiple e-commerce sites based on the extracted keywords, including product images, price information, and detailed descriptions.

[0816] 5. Multimodal AI analysis:

[0817] The server inputs the acquired product data into a multimodal AI model for analysis, and the analysis results take into account the user's emotional state.

[0818] 6. Product proposal generation:

[0819] The server then lists the identified product information and generates product suggestions with message wording and tone that correspond to the user's emotional state. For example, it could provide a message such as "This sofa is recommended!" along with product images, prices, and links via chat.

[0820] Hardware and software used

[0821] Hardware: smartphones, PCs, servers

[0822] software:

[0823] OpenAI API: An API for emotion recognition.

[0824] TextBlob: A tool that performs natural language processing and extracts keywords from text.

[0825] requests library: A library for obtaining product data using the API of an e-commerce site.

[0826] Specific examples

[0827] For example, if the user again types "Show me more gray sofas," the device sends this new text to the server, which again uses the NLP engine to extract the new keyword "gray" and performs similar search and image analysis to generate new suggestions for gray sofas, again using the emotion engine to adjust the suggestions based on the user's latest emotional state.

[0828] Prompt Sentence Examples

[0829] "I want a new sofa"

[0830] "Show me a cheaper sofa."

[0831] "I want to see a gray sofa."

[0832] As described above, this system realizes flexible product suggestions that take into account the user's emotional state, thereby significantly improving the user's purchasing experience.

[0833] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0834] Step 1:

[0835] Accepting user input

[0836] Operation: A user inputs a request for a product into a chat window on the terminal.

[0837] Input: Text data entered by the user: "I want a new sofa."

[0838] Output: The user's input text is sent from the terminal to the server.

[0839] Step 2:

[0840] Text data analysis

[0841] How it works: The server passes the text data to a natural language processing engine for analysis, extracting important keywords and attribute information.

[0842] Input: User input text "I want a new sofa."

[0843] Data processing: A natural language processing engine is used to analyze the text and extract the attribute information "new" and the keyword "sofa."

[0844] Output: The extracted keyword information of "new" and "sofa" is obtained.

[0845] Step 3:

[0846] emotion recognition

[0847] Operation: The server uses the emotion engine to analyze the user's emotional state.

[0848] Input: User input text "I want a new sofa."

[0849] Data processing: Using an emotion engine to classify emotions such as "happy" or "frustrated" from user text.

[0850] Output: The user's emotional state (e.g., "joy" is recognized).

[0851] Step 4:

[0852] Get product data

[0853] Operation: The server retrieves product data from the APIs of multiple e-commerce sites based on the extracted keywords.

[0854] Input: The extracted keyword "sofa" and attribute information "new".

[0855] Data processing: Call the API of the e-commerce site and obtain the relevant product information (product images, price information, detailed descriptions, etc.).

[0856] Output: A list of retrieved product data.

[0857] Step 5:

[0858] Multimodal AI analysis

[0859] Operation: The server inputs the acquired product data images into a multimodal AI model to identify products that meet the user's requirements.

[0860] Input: A list of retrieved product data and the user's emotional state.

[0861] Data processing: Using a multimodal AI model to analyze images and evaluate products based on emotional states.

[0862] Output: A list of products that match the user requirements.

[0863] Step 6:

[0864] Generate product suggestions

[0865] Operation: The server generates a suggestion message based on the identified product information with a message wording and tone that corresponds to the emotional state.

[0866] Input: A list of products that match the user requirements and the user's emotional state.

[0867] Data processing: Generate appropriate messages taking into account the emotional state and list the identified product information.

[0868] Output: Product suggestion message to be presented to the user.

[0869] Step 7:

[0870] Presenting product suggestions to users

[0871] Operation: The server sends the generated proposal message to the terminal and presents it to the user.

[0872] Input: The generated product suggestion message.

[0873] Output: Product suggestion message displayed in the device's chat window.

[0874] Example: "Show me more gray sofas."

[0875] When the user makes this input, "gray" is extracted as a new keyword, and data is acquired and analyzed again, generating new product proposals that match the user's requirements.

[0876] 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.

[0877] 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.

[0878] 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.

[0879] [Third embodiment]

[0880] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0881] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0882] 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).

[0883] 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.

[0884] 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.

[0885] 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).

[0886] 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.

[0887] 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.

[0888] 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.

[0889] 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.

[0890] 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.

[0891] 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."

[0892] The present invention is a system that allows users to easily and quickly find and purchase the products they desire. The system operates via a chat interface, allowing users to input requirements for the products they want, and then suggests products that match those requirements.

[0893] Program processing and specific examples

[0894] 1. Accepting input from the user

[0895] The terminal receives text data in which the user enters "I want a new sofa" in the chat window.

[0896] 2. Text interpretation and requirement extraction

[0897] The server passes the received text data to a natural language processing engine for analysis, which extracts the keyword "sofa" and the attribute information "new."

[0898] 3. Providing product images

[0899] The server retrieves product data (product images, price information, links, etc.) from APIs of multiple e-commerce sites based on the extracted keyword "sofa."

[0900] 4. Image Classification and Requirements Compliance Assessment

[0901] The server inputs the captured product images into a multimodal AI model to identify the product that best suits the user's requirements, for example, evaluating color and design features to select a modern sofa that meets the "new" requirements.

[0902] 5. Generate product suggestions and present them to the user

[0903] The server formats the identified product information and provides the user with a message via chat saying, "How about this sofa?" along with an image, price, and link to the selected product.

[0904] Specific examples

[0905] For example, if the user types "Show me more gray sofas," the device sends this new text to the server, which again uses its NLP engine to extract the new keyword "gray" and performs similar search and image analysis to generate new suggestions for gray sofas, allowing the user to find products that match their more specific requirements.

[0906] Purchase procedure

[0907] When the user selects a product and enters "I want to purchase this," the device sends this purchase intention to the server. The server then sends the user's account information to the e-commerce site's purchase API, automatically starting the purchase process. The e-commerce site then completes the purchase process and returns a confirmation message to the server. The server then notifies the user of the purchase completion message.

[0908] This system allows users to efficiently search for and purchase products, providing a superior purchasing experience.

[0909] The processing flow will be explained below.

[0910] Step 1:

[0911] The user types "I want a new sofa" into the chat window.

[0912] Step 2:

[0913] The terminal receives this text input and transmits it to the server in real time.

[0914] Step 3:

[0915] The server passes the received text data to a natural language processing engine for analysis.

[0916] Step 4:

[0917] The server uses a natural language processing engine to extract the keyword "sofa" and the attribute information "new" from the input text.

[0918] Step 5:

[0919] Based on the extracted keyword "sofa," the server sends requests to APIs of multiple e-commerce sites to retrieve related product data (product images, price information, links, etc.).

[0920] Step 6:

[0921] In response to the request, the server of the e-commerce site returns product data that matches the conditions to the server.

[0922] Step 7:

[0923] The server stores the acquired product data in a local database or memory.

[0924] Step 8:

[0925] The server inputs the saved product image data into a multimodal AI model for analysis.

[0926] Step 9:

[0927] The server uses a multimodal AI model to identify the product that best suits the user's requirements, evaluating color and design features to determine whether the product meets the user's requirements.

[0928] Step 10:

[0929] The server lists the identified product information and generates a suggestion message for the user, including product images, price information, and product links.

[0930] Step 11:

[0931] The server transmits the generated proposal message to the user through the terminal.

[0932] Step 12:

[0933] The user reviews the list of suggested products and provides additional input if more information is desired.

[0934] Step 13:

[0935] The terminal receives new text input and sends it back to the server.

[0936] Step 14:

[0937] The server passes the new text data to the natural language processing engine again to extract keywords and attribute information.

[0938] Step 15:

[0939] The server retrieves the product data again based on the new information extracted and analyzes it using a multimodal AI model.

[0940] Step 16:

[0941] The server generates an updated product list and again sends a suggestion message to the user.

[0942] Step 17:

[0943] The user selects a particular product and enters "I would like to purchase this."

[0944] Step 18:

[0945] The terminal transmits text data of the purchase intention to the server.

[0946] Step 19:

[0947] The server calls the purchase API of the e-commerce site and initiates the purchase process using the user's account information.

[0948] Step 20:

[0949] The e-commerce site's server processes the purchase and sends a confirmation message back to the server.

[0950] Step 21:

[0951] The server generates a purchase completion message and notifies the user through the terminal.

[0952] By going through each step, users can efficiently find products that meet their requirements and complete the purchasing process smoothly.

[0953] Example 1

[0954] 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."

[0955] Conventional online shopping systems have made it difficult for users to quickly and accurately find the products they are looking for. In particular, text-based searches have been problematic because they are unable to efficiently suggest products due to their inability to accommodate vague keywords and diverse user needs. Furthermore, there has been no system that can retrieve product information from multiple online markets and present products that best meet the user's requirements. As a result, users are forced to spend a long time searching for products, which often discourages them from making a purchase.

[0956] 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.

[0957] In this invention, the server includes means for receiving text data input by a user, means for analyzing the text data using a natural language processing engine to extract keywords and attribute information, means for acquiring product data from multiple online markets based on the extracted keywords, means for analyzing images of the acquired product data using a multi-function AI model to identify products that meet the user's requirements, and means for suggesting the identified product information to the user, thereby enabling the user to quickly and accurately find the product they are looking for.

[0958] "User" refers to an individual or corporation that uses the system to search for and purchase products.

[0959] "Text data" refers to character string information entered by the user, and refers to sentences used to explain searches and requirements.

[0960] A "natural language processing engine" refers to software or algorithms that analyze text data and extract keywords and attribute information.

[0961] "Keywords" refer to the main words or phrases that describe the product or its characteristics that a user is searching for.

[0962] "Attribute information" refers to information that indicates specific characteristics or requirements related to a keyword.

[0963] "Online marketplace" refers to platforms and websites that sell products over the internet.

[0964] "Product Data" means product information obtained from an online marketplace, including images, prices, links, etc.

[0965] A "multi-functional AI model" refers to an artificial intelligence model that integrates and analyzes data in multiple different formats to identify the products that best suit the user's requirements.

[0966] "Product information" is detailed information about a product suggested to a user, including an image, price, link, and the like.

[0967] The present invention is a system that helps users quickly and accurately find and purchase the products they are looking for. The system operates via a chat interface and suggests products based on product requirements entered by the user.

[0968] Hardware and Software

[0969] This system consists of a user terminal, a server, and multiple APIs. The main hardware and software used are as follows:

[0970] User device: Smartphone, tablet, PC, etc. Used by users to input text data through the chat window.

[0971] Server: A virtual server provided by a cloud service provider. For example, commonly used cloud platforms include Amazon Web Services (AWS) and Google Cloud Platform (GCP).

[0972] Natural language processing engine: Software used to analyze text data. Google Cloud Natural Language API is one such example.

[0973] Multimodal AI model: An AI model that analyzes image and text data to recommend optimal products. Typically, OpenAI's CLIP (Contrastive Language-Image Pretraining) model is used.

[0974] Online marketplace API: An interface for obtaining product data from multiple online marketplaces. For example, Amazon Marketplace API or Rakuten Marketplace API are used.

[0975] Specific operation of the system

[0976] When a user opens a chat window and types "I want a new sofa," the device receives this text data, stores it in its internal memory, and then sends it to the server.

[0977] The server passes the received text data to a natural language processing engine for analysis, which extracts the keyword "sofa" and the attribute information "new."

[0978] The server then sends requests to multiple online marketplace APIs based on these extracted keywords. For example, it uses the APIs of Amazon and Rakuten to perform a product search using the keyword "sofa." The product data retrieved from these online marketplaces includes images, prices, links, etc.

[0979] The acquired product data is input by the server into a multi-function AI model, which identifies the most suitable product based on the product image and extracted attribute information. For example, using the CLIP model, it matches the image with the keyword "new sofa" and ranks the most relevant products.

[0980] The identified product information is converted into JSON format and then reformatted to fit the chat interface. The user is then provided with an image, price, and link to the suggested product along with the message, "How about this sofa?"

[0981] Specific examples

[0982] For example, if the user types "Show me more gray sofas," the device receives this new text data and sends it to the server. The server then uses its NLP engine to extract the new keyword "gray" and requests a new search from the online marketplace API. It then performs image analysis on the new product data and suggests the gray sofa that best suits the user's requirements.

[0983] Prompt Sentence Examples

[0984] A user types the following into a chat window:

[0985] "I want a new sofa"

[0986] "Show me more gray sofas."

[0987] This system allows users to efficiently search for and purchase products, providing a superior purchasing experience.

[0988] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0989] Step 1:

[0990] The terminal receives text data entered into the chat window by the user.

[0991] Input: A user types "I want a new sofa" into a chat window.

[0992] Operation: The terminal stores this text data in its internal memory and generates a data packet.

[0993] Output: Send the generated data packet to the server.

[0994] Step 2:

[0995] The server receives the text data sent from the terminal.

[0996] Input: Data packets from the terminal.

[0997] How it works: The server takes this text data and prepares it to be passed to a natural language processing (NLP) engine.

[0998] Output: Text data ready for analysis.

[0999] Step 3:

[1000] The server analyzes the text data using a natural language processing engine to extract keywords and attribute information.

[1001] Input: Text data ready for analysis.

[1002] How it works: The server parses the text data using the Google Cloud Natural Language API.

[1003] Output: Keyword "sofa" and attribute information "new".

[1004] Step 4:

[1005] The server retrieves product data from multiple online marketplaces based on the extracted keywords.

[1006] Input: keyword "sofa".

[1007] Operation: The server sends an HTTP request to the Amazon Marketplace API and the Rakuten Marketplace API to search for products using the keyword "sofa."

[1008] Output: Retrieved product data (images, prices, links, etc.).

[1009] Step 5:

[1010] The server analyzes the acquired product data images using a multi-function AI model and identifies products that meet the user's requirements.

[1011] Input: Retrieved product data.

[1012] How it works: The server uses OpenAI's CLIP model to match product images with the attribute "new" and rank the products that best match the user's requirements.

[1013] Output: A ranked list of products.

[1014] Step 6:

[1015] The server proposes the identified product information to the user.

[1016] Input: A ranked list of products.

[1017] What it does: The server converts this information into JSON format and reformats it into a format suitable for a chat interface with the message "How would you like this sofa?"

[1018] Output: Send the formatted product suggestion message to the terminal.

[1019] Step 7:

[1020] The terminal displays the product suggestion message sent from the server to the user.

[1021] Input: A formatted product suggestion message from the server.

[1022] What it does: The device displays this message in a chat window, providing the user with a picture of the product, its price, and a link.

[1023] Output: Product suggestions to the user.

[1024] Step 8:

[1025] The user enters additional requests for the suggested products, for example, "Show me more gray sofas."

[1026] Input: Additional input from the user.

[1027] Operation: The user enters a new request into the chat window of the terminal, which receives it and prepares to retransmit it to the server.

[1028] Output: New data packet.

[1029] Step 9:

[1030] The server receives the new text data again and again uses the NLP engine to extract keywords and attribute information.

[1031] Input: A new data packet.

[1032] How it works: The server uses the Google Cloud Natural Language API to parse newly received text data and extract the keyword "gray."

[1033] Output: New keyword and attribute information.

[1034] Step 10:

[1035] The server retrieves product data again based on the new keywords, and identifies and suggests more suitable products.

[1036] Input: new keyword "gray".

[1037] Operation: The server requests a re-search from the Amazon Marketplace API and Rakuten Market API to obtain product data for "gray sofas." It then analyzes the product information using a multi-function AI model.

[1038] Output: Send new product suggestion message to the terminal.

[1039] (Application example 1)

[1040] 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."

[1041] Conventional e-commerce systems have faced challenges in making it difficult for users to efficiently search for products and quickly find those that meet their requirements. Furthermore, product search and suggestion functions using voice input and augmented reality (AR) technology interfaces were inadequate, leaving a need for an improved user experience.

[1042] 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.

[1043] In this invention, the server includes means for receiving text data input by a user, means for analyzing the text data using a natural language processing engine to extract keywords and attribute information, means for acquiring product data from multiple e-commerce sites based on the extracted keywords, means for analyzing images of the acquired product data using a multimodal AI model to identify products that meet the user's requirements, means for suggesting the identified product information to the user, means for converting voice data into text data, and means for presenting product information in augmented reality on a visual display device, thereby enabling users to easily search for products through voice input, visually confirm product information using AR technology, and complete the purchase process efficiently and intuitively.

[1044] "User" refers to a consumer who searches for and purchases products using this system.

[1045] "Text data" refers to character information including requirements and conditions regarding products entered by a user.

[1046] "Natural language processing engine" refers to software and algorithms for understanding and analyzing human language.

[1047] "Keywords" refer to the main words necessary for searching for products, extracted from the user's input text.

[1048] "Attribute information" refers to specific characteristics or conditions related to a keyword.

[1049] "E-commerce site" refers to a website that sells goods and services over the Internet.

[1050] "Product Data" refers to information about a product obtained from an e-commerce site, primarily including the name, image, price, and link.

[1051] A "multimodal AI model" refers to an artificial intelligence model that integrates and analyzes multiple types of data (e.g., text data, image data).

[1052] A "visual display device" is a device that presents information to a user's field of vision, and primarily refers to smart glasses and augmented reality devices.

[1053] "Augmented reality (AR)" refers to the technology of overlaying digital information onto a real-world environment.

[1054] "Voice data" refers to sound information obtained from the user's speech.

[1055] "Product suggestion" refers to the act of presenting the most suitable product to the user based on analyzed product data.

[1056] The present invention is a system for enabling users to easily and quickly find and purchase desired products, and is particularly directed to a shopping assistant application that uses smart glasses. The system is configured as follows.

[1057] The server receives the text data entered by the user and analyzes it using a natural language processing engine. When the user enters product requirements by voice, the smart glasses' microphone picks up the voice data and converts it into text data. The converted text data is then analyzed by the natural language processing engine to extract keywords and attribute information.

[1058] Once the keywords are extracted, the server accesses the APIs of multiple e-commerce sites to retrieve the corresponding product data. An e-commerce site is a website that sells products and services over the Internet, and the product data includes the product name, image, price, and purchase link.

[1059] The acquired product data images are analyzed using a multimodal AI model, which is an artificial intelligence model that integrates and analyzes multiple types of data (e.g., text data, image data, etc.). This allows the system to identify the product that best suits the user's requirements.

[1060] The identified product information is then formatted and suggested to the user. The suggested product information is displayed as augmented reality (AR) on the smart glasses display. Augmented reality is a technology that overlays digital information on the real-world environment, allowing the user to intuitively check the product information superimposed on their field of vision.

[1061] For example, if a user says "I'm looking for red sneakers," the voice data is picked up through the smart glasses' microphone and converted into text data. The server then uses a natural language processing engine to analyze the converted text data and extract the keyword "red sneakers." Product data is then retrieved using the API of the e-commerce site, and analyzed using a multimodal AI model to identify the most suitable product.

[1062] As a result, product information for "red sneakers" is displayed in AR within the user's field of vision. When the user finds a specific product and says "I want to buy this," the server automatically initiates the purchase process through the e-commerce site's purchase API and sends a confirmation message to the user.

[1063] This system allows users to search for and purchase products efficiently and intuitively, providing an excellent purchasing experience.

[1064] Example prompt sentence:

[1065] The user speaks, "I'm looking for red sneakers."

[1066] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1067] Step 1:

[1068] The user speaks into the microphone of the smart glasses to input information about a product. For example, they might say, "I'm looking for red sneakers." This voice data is picked up by the microphone of the smart glasses. The input is voice data, and the output is voice data.

[1069] Step 2:

[1070] The device converts the acquired voice data into text data. This is done using a voice recognition library (e.g., Google's voice recognition API). This converts the voice data into text data. The input is voice data, and the output is text data.

[1071] Step 3:

[1072] Text data is sent to the server. The server analyzes the text data using a natural language processing engine (e.g., the BERT model) and extracts keywords and attribute information. This yields a keyword, for example, "red sneakers." The input is text data, and the output is keywords and attribute information.

[1073] Step 4:

[1074] The server accesses the APIs of multiple e-commerce sites based on the extracted keywords to retrieve related product data, including product names, images, prices, and purchase links. The input is the keywords, and the output is the product data.

[1075] Step 5:

[1076] The server inputs the acquired product data images into a multimodal AI model (e.g., CLIP model) and analyzes them. This identifies products that meet the user's requirements. The input is the product image, and the output is the identified product information.

[1077] Step 6:

[1078] The server formats the identified product information and presents it as augmented reality (AR) on the smart glasses display, allowing the user to view the product information directly in their field of vision. The input is the identified product information, and the output is the AR-displayed product information.

[1079] Step 7:

[1080] The user checks the display on the smart glasses, selects a specific product, and then voice-inputs, "I would like to purchase this." The input is voice data, and the output is text data indicating the purchase intention.

[1081] Step 8:

[1082] The terminal receives the voice data again, converts it into text data, and sends it to the server. The input is voice data, and the output is text data.

[1083] Step 9:

[1084] The server analyzes the received text data, confirms the user's purchase intention, and executes the purchase procedure via the e-commerce site's purchase API. The input is the text data of the purchase intention, and the output is purchase confirmation information.

[1085] Step 10:

[1086] The server confirms that the purchase has been completed and notifies the user of the result. The user receives a confirmation message on their smart glasses. The input is the purchase confirmation information, and the output is a user notification.

[1087] 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.

[1088] The present invention combines an emotion engine with a system that allows users to easily and quickly find and purchase the products they desire. The system operates via a chat interface, allowing users to input their desired product requirements and then suggests products that match those requirements. The system can also recognize the user's emotional state and adjust product suggestions based on the results.

[1089] Program processing and specific examples

[1090] 1. Accepting input from the user

[1091] The terminal receives text data in which the user enters "I want a new sofa" in the chat window.

[1092] 2. Text Interpretation and Emotion Recognition

[1093] The server passes the received text data to a natural language processing engine, analyzes the text, and uses an emotion engine to recognize the user's emotional state. For example, the emotion engine classifies the user's text as an emotion such as "joy" or "dissatisfaction."

[1094] 3. Extracting keywords and attribute information

[1095] The server uses a natural language processing engine to extract the keyword "sofa" and the attribute information "new" from the input text.

[1096] 4. Providing product images

[1097] The server retrieves product data (product images, price information, links, etc.) from APIs of multiple e-commerce sites based on the extracted keyword "sofa."

[1098] 5. Image Classification and Requirements Compliance Assessment

[1099] The server inputs the acquired product image data into a multimodal AI model for analysis. This analysis also takes into account the user's emotional state. For example, if a user's emotional state is "dissatisfied," the server will prioritize suggesting highly rated products and discount information.

[1100] 6. Generate product suggestions and present them to the user

[1101] The server then lists the identified product information and generates a suggestion message by adjusting the wording and tone of the message according to the user's emotional state. For example, it might provide a message such as "This sofa is recommended!" along with an image, price, and link to the selected product via chat.

[1102] Specific examples

[1103] For example, if the user again types "Show me more gray sofas," the device sends this new text to the server, which again uses the NLP engine to extract the new keyword "gray" and performs similar search and image analysis to generate new suggestions for gray sofas, again using the emotion engine to adjust the suggestions based on the user's latest emotional state.

[1104] Purchase procedure

[1105] When the user selects a product and enters "I want to purchase this," the device sends this purchase intention to the server. The server then sends the user's account information to the e-commerce site's purchase API, automatically starting the purchase process. The e-commerce site then completes the purchase process and returns a confirmation message to the server. The server then generates a purchase completion message and notifies the user via the device.

[1106] This system allows users to efficiently search for and purchase products, and provides an excellent purchasing experience with appropriate suggestions based on their emotional state.

[1107] The processing flow will be explained below.

[1108] Step 1:

[1109] The user types "I want a new sofa" into the chat window.

[1110] Step 2:

[1111] The terminal receives this text input and transmits it to the server in real time.

[1112] Step 3:

[1113] The server passes the received text data to a natural language processing engine and analyzes the text.

[1114] Step 4:

[1115] The server uses a natural language processing engine to extract the keyword "sofa" and the attribute information "new" from the input text.

[1116] Step 5:

[1117] The server inputs text data into an emotion engine to recognize the user's emotional state (e.g., "joy," "dissatisfaction," "excitement," etc.).

[1118] Step 6:

[1119] Based on the extracted keyword "sofa," the server sends requests to APIs of multiple e-commerce sites to retrieve related product data (product images, price information, links, etc.).

[1120] Step 7:

[1121] In response to the request, the server of the e-commerce site returns product data that matches the conditions to the server.

[1122] Step 8:

[1123] The server stores the acquired product data in a local database or memory.

[1124] Step 9:

[1125] The server inputs the saved product image data into a multimodal AI model for analysis.

[1126] Step 10:

[1127] The server uses a multimodal AI model to identify the products that best fit the user's requirements, taking into account not only color and design features but also the user's emotional state as recognized by an emotion engine.

[1128] Step 11:

[1129] The server then lists the identified product information and generates a suggested message by adjusting the wording and tone of the message according to the user's emotional state. For example, if the user is in a "dissatisfied" state, it adds an encouraging message such as "This sofa has been highly rated by many users!"

[1130] Step 12:

[1131] The server transmits the generated proposal message to the user through the terminal.

[1132] Step 13:

[1133] The user reviews the list of suggested products and provides additional input if they require more information (e.g., "Show me more gray sofas").

[1134] Step 14:

[1135] The terminal receives new text input and sends it back to the server.

[1136] Step 15:

[1137] The server again uses the NLP engine to extract the new keyword "gray" and similarly performs emotion recognition using the emotion engine.

[1138] Step 16:

[1139] The server retrieves product data again based on the new keywords and analyzes it using a multimodal AI model.

[1140] Step 17:

[1141] The server generates an updated product list and again creates and sends suggestion messages based on the user's latest emotional state.

[1142] Step 18:

[1143] The user selects a particular product and enters "I would like to purchase this."

[1144] Step 19:

[1145] The terminal transmits text data of the purchase intention to the server.

[1146] Step 20:

[1147] The server calls the purchase API of the e-commerce site and initiates the purchase process using the user's account information.

[1148] Step 21:

[1149] The e-commerce site's server processes the purchase and sends a confirmation message back to the server.

[1150] Step 22:

[1151] The server generates a purchase completion message and notifies the user through the terminal.

[1152] Through these specific processing steps, users can efficiently find products that meet their requirements and smoothly complete the purchasing process.The use of the emotion engine allows users to receive appropriate product suggestions based on their emotional state, significantly improving the purchasing experience.

[1153] Example 2

[1154] 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."

[1155] Conventional e-commerce systems have had difficulty in quickly and accurately searching for and suggesting products that users want. Furthermore, they were unable to suggest products that corresponded to the user's emotional state, resulting in low satisfaction. Furthermore, the purchasing process was cumbersome, making it difficult to improve the user experience.

[1156] 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.

[1157] In this invention, the server includes: means for receiving text data entered by a user; means for analyzing the text data using a natural language processing engine to extract keywords and attribute information; means for recognizing the user's emotional state from the text data; means for acquiring product data from multiple e-commerce sites based on the extracted keywords; means for analyzing images of the acquired product data using a multimodal generative AI model to identify products that meet the user's requirements and emotional state; and means for suggesting the identified product information to the user in a message corresponding to the user's emotional state. This allows for quick and accurate suggestions of products desired by the user, and further enables optimal product suggestions based on the user's emotional state. In addition, a smoother purchase process is expected to improve the user experience.

[1158] "User" means an individual or corporation that uses the system.

[1159] "Text data" is text information entered by the user.

[1160] A "natural language processing engine" is a software tool or algorithm for analyzing text data and extracting keywords and attribute information.

[1161] A "keyword" is an important word or phrase that identifies the product or service a user is looking for.

[1162] "Attribute information" is data that indicates characteristics and conditions that embody a user's request.

[1163] The "emotional state" is a recognition of the psychological state of the user when he or she inputs text data.

[1164] An "e-commerce site" is a website that sells and buys goods and services online.

[1165] "Product data" is information about a product, including images, prices, links, etc.

[1166] A "multimodal generative AI model" is an artificial intelligence model that integrates and analyzes multiple data formats (text, images, etc.).

[1167] "Analysis" is the processing of collected data to extract meaning and information.

[1168] "Proposal" is the act of providing appropriate product or service information to a user.

[1169] A "message" is text containing notifications or information sent from the system to a user.

[1170] A "purchase interface" is a software component that executes a purchase transaction on an e-commerce site.

[1171] The "purchase procedure" is a series of steps for purchasing the product or service selected by the user.

[1172] The present invention combines emotion recognition with a system that helps users quickly find and purchase the products they desire. The system operates via a chat interface and makes product suggestions based on text data entered by the user.

[1173] The system mainly consists of a server and a terminal. The terminal receives text data entered by the user in the chat window, and the server analyzes the data. The specific operation of the system is explained below.

[1174] First, the user enters the product requirements into the chat window on the device. For example, they might type, "I want a new sofa." This text data is received by the device and sent to the server.

[1175] The server passes the received text data to a natural language processing engine (e.g., Google Cloud Natural Language API) to extract keywords and attribute information. At the same time, it uses an emotion recognition engine (e.g., open-source emotion recognition software) to analyze the user's emotional state (e.g., joy, dissatisfaction). The analysis results allow the user's psychological state to be understood.

[1176] Next, the server retrieves product data (e.g., product images, price information, links, etc.) from multiple e-commerce sites (e.g., large online marketplaces) based on the extracted keywords, using the APIs of the e-commerce sites (e.g., widely used e-commerce APIs).

[1177] The server inputs the acquired product data into a multimodal generative AI model (e.g., CLIP model) and performs an integrated analysis of image and text elements. The system also takes the user's emotional state into consideration. For example, if the emotional state is "dissatisfied," it will prioritize suggestions of highly rated products and discount information.

[1178] The server generates product suggestions appropriate for the user based on the analysis results. The message is generated by adjusting the wording and tone of the message according to the user's emotional state. The message includes product images, prices, and links, and displays a message such as "We recommend this sofa!"

[1179] This allows users to view product details in the chat window and receive highly satisfying recommendations. For example, if the user types "Show me more gray sofas," the device sends this new text to the server, which again uses its natural language processing engine to extract the keyword "gray" and performs the same search and analysis. The emotion engine is also utilized again to adjust the recommendations based on the user's latest emotional state.

[1180] Finally, if the user selects the product and enters "I want to purchase this," the terminal sends this purchase intention to the server. The server starts the purchase procedure in cooperation with the purchase interface of the e-commerce site, and receives a confirmation message from the e-commerce site. The server then generates a purchase completion message and notifies the user through the terminal.

[1181] This system allows users to efficiently search for and purchase products, and provides an excellent purchasing experience by receiving appropriate suggestions based on their emotional state.

[1182] Prompt Sentence Examples

[1183] For example, if a user types "I'm looking for a red dress," the device sends the text data to the server, which then retrieves and suggests product data in a similar manner, using a search query based on the keyword "red dress" and further adjusting it according to the user's emotional state.

[1184] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1185] Step 1:

[1186] The user types "I want a new sofa" into the chat window. The device receives this text data and stores it in its internal memory. The input is text data based on the user's request, and the output is the stored text data. A specific example of this behavior is when the user types "I want a new sofa" and presses the SEND button.

[1187] Step 2:

[1188] The device sends the received text data to the server, which then analyzes the text data using a natural language processing engine (for example, a natural language processing API). The input is the received text data, and the output is the extracted keyword "sofa" and attribute information "new." The analysis results are saved in internal memory. A specific example of operation is when the server sends text to a natural language processing API and receives the analysis results.

[1189] Step 3:

[1190] The server passes the text data to an emotion recognition engine (e.g., emotion recognition software) to recognize the user's emotional state (e.g., "joy"). The input is text data, and the output is the emotion recognition result (type of emotion). A specific example of operation is sending text data to an emotion recognition engine to obtain the emotional state.

[1191] Step 4:

[1192] The server retrieves product data using the APIs of multiple e-commerce sites (e.g., large online marketplaces) based on the extracted keyword "sofa." The input is the keyword "sofa," and the output is the retrieved product data (product images, price information, and links). A specific example of the operation is to send a keyword to the API of an e-commerce site and receive product data.

[1193] Step 5:

[1194] The server inputs the acquired product data images into a multimodal generative AI model (e.g., an AI model) and analyzes the suitability of the product. The input is the product data image, and the output is the set product feature vector. The user's emotional state is also taken into consideration here. A specific example of operation is inputting a product image into the AI ​​model and obtaining a feature vector.

[1195] Step 6:

[1196] Based on the analysis results, the server selects product information appropriate for the user and generates a message according to their emotional state. The input is the user's emotional state and the product's feature vector, and the output is the suggested product information and the generated message. As a specific example of operation, if the emotional state is "dissatisfied," a message containing highly rated products and discount information is generated.

[1197] Step 7:

[1198] The server sends a proposal message to the terminal, and the terminal displays the message to the user. The input is the proposal message sent from the server, and the output is the proposal message displayed in the user's chat window. A specific example of the operation is to display the proposal message in the chat window.

[1199] Step 8:

[1200] The user inputs (or clicks) "I want to buy this" for the selected product, and that information is sent to the server by the terminal. The input is the user's intention to purchase, and the output is a purchase request sent to the server. A specific example of this behavior is when the user clicks the product purchase button.

[1201] Step 9:

[1202] The server sends the user's purchase information to the e-commerce site's purchase interface and begins the purchase process. The input is the user's purchase information, and the output is a notification that the purchase process has been completed. A specific example of this operation is sending the user's account information and product information to the purchase API.

[1203] Step 10:

[1204] The e-commerce site completes the purchase process and sends a confirmation message to the server. The server generates a purchase completion message and sends it to the terminal, which then notifies the user. The input is the purchase completion notification and the output is the purchase completion message. A specific example of this behavior is when the terminal displays "Purchase completed!" in a chat window.

[1205] In this way, the system can suggest optimal products based on the user's desires and emotional state, and efficiently process the purchase.

[1206] (Application example 2)

[1207] 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."

[1208] Conventional e-commerce systems provide insufficient support for users to find products quickly and appropriately, and in particular do not provide product suggestions that take into account the user's emotional state. Furthermore, they are unable to respond flexibly to the user's emotions, which can lead to a decrease in satisfaction. Furthermore, when a user makes additional input to a product suggestion, a fast and accurate response to that request is required, but current systems do not adequately achieve this.

[1209] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1210] In this invention, the server includes: means for receiving text data entered by a user; means for analyzing the text data using a natural language processing engine to extract keywords and attribute information; means for acquiring product data from multiple e-commerce sites based on the extracted keywords; means for analyzing images of the acquired product data using a multimodal AI model to identify products that meet the user's requirements; means for recognizing the user's emotional state and adjusting product suggestions based on the recognized emotional state; and means for suggesting the identified product information to the user using a message with wording and tone that matches the user's emotional state. This allows the server to appropriately determine the user's emotional state and make optimal product suggestions based on that state, thereby improving the user's purchasing experience. Furthermore, the server can quickly respond to additional user input and re-suggest suitable products, thereby flexibly meeting user requests.

[1211] A "user" is a person who uses the system to search for products and consider purchasing them.

[1212] "Text data" refers to character information entered by the user.

[1213] A "natural language processing engine" is a computer program that analyzes and interprets text data and extracts keywords and attribute information.

[1214] "Keywords" are important words that represent products or requirements identified from the text entered by the user.

[1215] "Attribute information" refers to additional information that accompanies a keyword, such as adjectives like "new" or "cheap."

[1216] An "e-commerce site" is a website that sells and buys goods over the Internet.

[1217] "Product data" is information about a product, specifically, product name, price, image, description, etc.

[1218] A "multimodal AI model" is an artificial intelligence model that can simultaneously analyze multiple forms of data, such as text and images.

[1219] "User requirements" refer to the product characteristics and conditions desired by the user.

[1220] "Emotional state" refers to the psychological state (e.g., joy, dissatisfaction, etc.) that a user experiences when searching for a product.

[1221] "Product suggestion" refers to the system recommending products selected by the system to the user.

[1222] "Message wording and tone" refers to the content of the product proposal message, which is adjusted according to the emotional state, and how it is delivered.

[1223] "Additional input" refers to new text information added by the user after the initial input.

[1224] The "purchase procedure" refers to a series of steps required for the user to actually purchase the product selected by the user.

[1225] A "purchase API" is an application programming interface that works with e-commerce sites to automate the product purchasing process.

[1226] The present invention is a system that allows users to easily and quickly find and purchase the products they desire, and is combined with an emotion engine. This system operates via a chat interface, and when users input their product requests, it suggests products that match those requests.

[1227] System Configuration

[1228] The system mainly includes the following elements:

[1229] Terminal: A device that allows users to input text data. This includes smartphones and PCs.

[1230] Server: A central system that analyzes text data, acquires product data, recognizes emotions, and generates product suggestions.

[1231] Natural language processing engine: Software for analyzing text data and extracting keywords and attribute information.

[1232] Multimodal AI model: An artificial intelligence model for analyzing data in multiple formats, such as text data and image data.

[1233] Emotion Engine: An engine for recognizing and analyzing the user's emotional state.

[1234] Process Flow

[1235] 1. Accept user input:

[1236] The user uses the terminal to input text data into the chat window, for example, "I want a new sofa."

[1237] 2. Text data analysis:

[1238] The server passes the received text data to a natural language processing engine, which analyzes the text and extracts the important keyword "sofa" and attribute information "new" from the text.

[1239] 3. Emotion recognition:

[1240] The server uses an emotion engine to analyze the user's emotional state, for example classifying the user's text as an emotion such as "happy" or "frustrated."

[1241] 4. Product data acquisition:

[1242] The server retrieves product data from APIs of multiple e-commerce sites based on the extracted keywords, including product images, price information, and detailed descriptions.

[1243] 5. Multimodal AI analysis:

[1244] The server inputs the acquired product data into a multimodal AI model for analysis, and the analysis results take into account the user's emotional state.

[1245] 6. Product proposal generation:

[1246] The server then lists the identified product information and generates product suggestions with message wording and tone that correspond to the user's emotional state. For example, it could provide a message such as "This sofa is recommended!" along with product images, prices, and links via chat.

[1247] Hardware and software used

[1248] Hardware: smartphones, PCs, servers

[1249] software:

[1250] OpenAI API: An API for emotion recognition.

[1251] TextBlob: A tool that performs natural language processing and extracts keywords from text.

[1252] requests library: A library for obtaining product data using the API of an e-commerce site.

[1253] Specific examples

[1254] For example, if the user again types "Show me more gray sofas," the device sends this new text to the server, which again uses the NLP engine to extract the new keyword "gray" and performs similar search and image analysis to generate new suggestions for gray sofas, again using the emotion engine to adjust the suggestions based on the user's latest emotional state.

[1255] Prompt Sentence Examples

[1256] "I want a new sofa"

[1257] "Show me a cheaper sofa."

[1258] "I want to see a gray sofa."

[1259] As described above, this system realizes flexible product suggestions that take into account the user's emotional state, thereby significantly improving the user's purchasing experience.

[1260] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1261] Step 1:

[1262] Accepting user input

[1263] Operation: A user inputs a request for a product into a chat window on the terminal.

[1264] Input: Text data entered by the user: "I want a new sofa."

[1265] Output: The user's input text is sent from the terminal to the server.

[1266] Step 2:

[1267] Text data analysis

[1268] How it works: The server passes the text data to a natural language processing engine for analysis, extracting important keywords and attribute information.

[1269] Input: User input text "I want a new sofa."

[1270] Data processing: A natural language processing engine is used to analyze the text and extract the attribute information "new" and the keyword "sofa."

[1271] Output: The extracted keyword information of "new" and "sofa" is obtained.

[1272] Step 3:

[1273] emotion recognition

[1274] Operation: The server uses the emotion engine to analyze the user's emotional state.

[1275] Input: User input text "I want a new sofa."

[1276] Data processing: Using an emotion engine to classify emotions such as "happy" or "frustrated" from user text.

[1277] Output: The user's emotional state (e.g., "joy" is recognized).

[1278] Step 4:

[1279] Get product data

[1280] Operation: The server retrieves product data from the APIs of multiple e-commerce sites based on the extracted keywords.

[1281] Input: The extracted keyword "sofa" and attribute information "new".

[1282] Data processing: Call the API of the e-commerce site and obtain the relevant product information (product images, price information, detailed descriptions, etc.).

[1283] Output: A list of retrieved product data.

[1284] Step 5:

[1285] Multimodal AI analysis

[1286] Operation: The server inputs the acquired product data images into a multimodal AI model to identify products that meet the user's requirements.

[1287] Input: A list of retrieved product data and the user's emotional state.

[1288] Data processing: Using a multimodal AI model to analyze images and evaluate products based on emotional states.

[1289] Output: A list of products that match the user requirements.

[1290] Step 6:

[1291] Generate product suggestions

[1292] Operation: The server generates a suggestion message based on the identified product information with a message wording and tone that corresponds to the emotional state.

[1293] Input: A list of products that match the user requirements and the user's emotional state.

[1294] Data processing: Generate appropriate messages taking into account the emotional state and list the identified product information.

[1295] Output: Product suggestion message to be presented to the user.

[1296] Step 7:

[1297] Presenting product suggestions to users

[1298] Operation: The server sends the generated proposal message to the terminal and presents it to the user.

[1299] Input: The generated product suggestion message.

[1300] Output: Product suggestion message displayed in the device's chat window.

[1301] Example: "Show me more gray sofas."

[1302] When the user makes this input, "gray" is extracted as a new keyword, and data is acquired and analyzed again, generating new product proposals that match the user's requirements.

[1303] 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.

[1304] 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.

[1305] 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.

[1306] [Fourth embodiment]

[1307] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1308] 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.

[1309] 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).

[1310] 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.

[1311] 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.

[1312] 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).

[1313] 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.

[1314] 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.

[1315] 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.

[1316] 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.

[1317] 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.

[1318] 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.

[1319] 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."

[1320] The present invention is a system that allows users to easily and quickly find and purchase the products they desire. The system operates via a chat interface, allowing users to input requirements for the products they want, and then suggests products that match those requirements.

[1321] Program processing and specific examples

[1322] 1. Accepting input from the user

[1323] The terminal receives text data in which the user enters "I want a new sofa" in the chat window.

[1324] 2. Text interpretation and requirement extraction

[1325] The server passes the received text data to a natural language processing engine for analysis, which extracts the keyword "sofa" and the attribute information "new."

[1326] 3. Providing product images

[1327] The server retrieves product data (product images, price information, links, etc.) from APIs of multiple e-commerce sites based on the extracted keyword "sofa."

[1328] 4. Image Classification and Requirements Compliance Assessment

[1329] The server inputs the captured product images into a multimodal AI model to identify the product that best suits the user's requirements, for example, evaluating color and design features to select a modern sofa that meets the "new" requirements.

[1330] 5. Generate product suggestions and present them to the user

[1331] The server formats the identified product information and provides the user with a message via chat saying, "How about this sofa?" along with an image, price, and link to the selected product.

[1332] Specific examples

[1333] For example, if the user types "Show me more gray sofas," the device sends this new text to the server, which again uses its NLP engine to extract the new keyword "gray" and performs similar search and image analysis to generate new suggestions for gray sofas, allowing the user to find products that match their more specific requirements.

[1334] Purchase procedure

[1335] When the user selects a product and enters "I want to purchase this," the device sends this purchase intention to the server. The server then sends the user's account information to the e-commerce site's purchase API, automatically starting the purchase process. The e-commerce site then completes the purchase process and returns a confirmation message to the server. The server then notifies the user of the purchase completion message.

[1336] This system allows users to efficiently search for and purchase products, providing a superior purchasing experience.

[1337] The processing flow will be explained below.

[1338] Step 1:

[1339] The user types "I want a new sofa" into the chat window.

[1340] Step 2:

[1341] The terminal receives this text input and transmits it to the server in real time.

[1342] Step 3:

[1343] The server passes the received text data to a natural language processing engine for analysis.

[1344] Step 4:

[1345] The server uses a natural language processing engine to extract the keyword "sofa" and the attribute information "new" from the input text.

[1346] Step 5:

[1347] Based on the extracted keyword "sofa," the server sends requests to APIs of multiple e-commerce sites to retrieve related product data (product images, price information, links, etc.).

[1348] Step 6:

[1349] In response to the request, the server of the e-commerce site returns product data that matches the conditions to the server.

[1350] Step 7:

[1351] The server stores the acquired product data in a local database or memory.

[1352] Step 8:

[1353] The server inputs the saved product image data into a multimodal AI model for analysis.

[1354] Step 9:

[1355] The server uses a multimodal AI model to identify the product that best suits the user's requirements, evaluating color and design features to determine whether the product meets the user's requirements.

[1356] Step 10:

[1357] The server lists the identified product information and generates a suggestion message for the user, including product images, price information, and product links.

[1358] Step 11:

[1359] The server transmits the generated proposal message to the user through the terminal.

[1360] Step 12:

[1361] The user reviews the list of suggested products and provides additional input if more information is desired.

[1362] Step 13:

[1363] The terminal receives new text input and sends it back to the server.

[1364] Step 14:

[1365] The server passes the new text data to the natural language processing engine again to extract keywords and attribute information.

[1366] Step 15:

[1367] The server retrieves the product data again based on the new information extracted and analyzes it using a multimodal AI model.

[1368] Step 16:

[1369] The server generates an updated product list and again sends a suggestion message to the user.

[1370] Step 17:

[1371] The user selects a particular product and enters "I would like to purchase this."

[1372] Step 18:

[1373] The terminal transmits text data of the purchase intention to the server.

[1374] Step 19:

[1375] The server calls the purchase API of the e-commerce site and initiates the purchase process using the user's account information.

[1376] Step 20:

[1377] The e-commerce site's server processes the purchase and sends a confirmation message back to the server.

[1378] Step 21:

[1379] The server generates a purchase completion message and notifies the user through the terminal.

[1380] By going through each step, users can efficiently find products that meet their requirements and complete the purchasing process smoothly.

[1381] Example 1

[1382] 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."

[1383] Conventional online shopping systems have made it difficult for users to quickly and accurately find the products they are looking for. In particular, text-based searches have been problematic because they are unable to efficiently suggest products due to their inability to accommodate vague keywords and diverse user needs. Furthermore, there has been no system that can retrieve product information from multiple online markets and present products that best meet the user's requirements. As a result, users are forced to spend a long time searching for products, which often discourages them from making a purchase.

[1384] 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.

[1385] In this invention, the server includes means for receiving text data input by a user, means for analyzing the text data using a natural language processing engine to extract keywords and attribute information, means for acquiring product data from multiple online markets based on the extracted keywords, means for analyzing images of the acquired product data using a multi-function AI model to identify products that meet the user's requirements, and means for suggesting the identified product information to the user, thereby enabling the user to quickly and accurately find the product they are looking for.

[1386] "User" refers to an individual or corporation that uses the system to search for and purchase products.

[1387] "Text data" refers to character string information entered by the user, and refers to sentences used to explain searches and requirements.

[1388] A "natural language processing engine" refers to software or algorithms that analyze text data and extract keywords and attribute information.

[1389] "Keywords" refer to the main words or phrases that describe the product or its characteristics that a user is searching for.

[1390] "Attribute information" refers to information that indicates specific characteristics or requirements related to a keyword.

[1391] "Online marketplace" refers to platforms and websites that sell products over the internet.

[1392] "Product Data" means product information obtained from an online marketplace, including images, prices, links, etc.

[1393] A "multi-functional AI model" refers to an artificial intelligence model that integrates and analyzes data in multiple different formats to identify the products that best suit the user's requirements.

[1394] "Product information" is detailed information about a product suggested to a user, including an image, price, link, and the like.

[1395] The present invention is a system that helps users quickly and accurately find and purchase the products they are looking for. The system operates via a chat interface and suggests products based on product requirements entered by the user.

[1396] Hardware and Software

[1397] This system consists of a user terminal, a server, and multiple APIs. The main hardware and software used are as follows:

[1398] User device: Smartphone, tablet, PC, etc. Used by users to input text data through the chat window.

[1399] Server: A virtual server provided by a cloud service provider. For example, commonly used cloud platforms include Amazon Web Services (AWS) and Google Cloud Platform (GCP).

[1400] Natural language processing engine: Software used to analyze text data. Google Cloud Natural Language API is one such example.

[1401] Multimodal AI model: An AI model that analyzes image and text data to recommend optimal products. Typically, OpenAI's CLIP (Contrastive Language-Image Pretraining) model is used.

[1402] Online marketplace API: An interface for obtaining product data from multiple online marketplaces. For example, Amazon Marketplace API or Rakuten Marketplace API are used.

[1403] Specific operation of the system

[1404] When a user opens a chat window and types "I want a new sofa," the device receives this text data, stores it in its internal memory, and then sends it to the server.

[1405] The server passes the received text data to a natural language processing engine for analysis, which extracts the keyword "sofa" and the attribute information "new."

[1406] The server then sends requests to multiple online marketplace APIs based on these extracted keywords. For example, it uses the APIs of Amazon and Rakuten to perform a product search using the keyword "sofa." The product data retrieved from these online marketplaces includes images, prices, links, etc.

[1407] The acquired product data is input by the server into a multi-function AI model, which identifies the most suitable product based on the product image and extracted attribute information. For example, using the CLIP model, it matches the image with the keyword "new sofa" and ranks the most relevant products.

[1408] The identified product information is converted into JSON format and then reformatted to fit the chat interface. The user is then provided with an image, price, and link to the suggested product along with the message, "How about this sofa?"

[1409] Specific examples

[1410] For example, if the user types "Show me more gray sofas," the device receives this new text data and sends it to the server. The server then uses its NLP engine to extract the new keyword "gray" and requests a new search from the online marketplace API. It then performs image analysis on the new product data and suggests the gray sofa that best suits the user's requirements.

[1411] Prompt Sentence Examples

[1412] A user types the following into a chat window:

[1413] "I want a new sofa"

[1414] "Show me more gray sofas."

[1415] This system allows users to efficiently search for and purchase products, providing a superior purchasing experience.

[1416] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1417] Step 1:

[1418] The terminal receives text data entered into the chat window by the user.

[1419] Input: A user types "I want a new sofa" into a chat window.

[1420] Operation: The terminal stores this text data in its internal memory and generates a data packet.

[1421] Output: Send the generated data packet to the server.

[1422] Step 2:

[1423] The server receives the text data sent from the terminal.

[1424] Input: Data packets from the terminal.

[1425] How it works: The server takes this text data and prepares it to be passed to a natural language processing (NLP) engine.

[1426] Output: Text data ready for analysis.

[1427] Step 3:

[1428] The server analyzes the text data using a natural language processing engine to extract keywords and attribute information.

[1429] Input: Text data ready for analysis.

[1430] How it works: The server parses the text data using the Google Cloud Natural Language API.

[1431] Output: Keyword "sofa" and attribute information "new".

[1432] Step 4:

[1433] The server retrieves product data from multiple online marketplaces based on the extracted keywords.

[1434] Input: keyword "sofa".

[1435] Operation: The server sends an HTTP request to the Amazon Marketplace API and the Rakuten Marketplace API to search for products using the keyword "sofa."

[1436] Output: Retrieved product data (images, prices, links, etc.).

[1437] Step 5:

[1438] The server analyzes the acquired product data images using a multi-function AI model and identifies products that meet the user's requirements.

[1439] Input: Retrieved product data.

[1440] How it works: The server uses OpenAI's CLIP model to match product images with the attribute "new" and rank the products that best match the user's requirements.

[1441] Output: A ranked list of products.

[1442] Step 6:

[1443] The server proposes the identified product information to the user.

[1444] Input: A ranked list of products.

[1445] What it does: The server converts this information into JSON format and reformats it into a format suitable for a chat interface with the message "How would you like this sofa?"

[1446] Output: Send the formatted product suggestion message to the terminal.

[1447] Step 7:

[1448] The terminal displays the product suggestion message sent from the server to the user.

[1449] Input: A formatted product suggestion message from the server.

[1450] What it does: The device displays this message in a chat window, providing the user with a picture of the product, its price, and a link.

[1451] Output: Product suggestions to the user.

[1452] Step 8:

[1453] The user enters additional requests for the suggested products, for example, "Show me more gray sofas."

[1454] Input: Additional input from the user.

[1455] Operation: The user enters a new request into the chat window of the terminal, which receives it and prepares to retransmit it to the server.

[1456] Output: New data packet.

[1457] Step 9:

[1458] The server receives the new text data again and again uses the NLP engine to extract keywords and attribute information.

[1459] Input: A new data packet.

[1460] How it works: The server uses the Google Cloud Natural Language API to parse newly received text data and extract the keyword "gray."

[1461] Output: New keyword and attribute information.

[1462] Step 10:

[1463] The server retrieves product data again based on the new keywords, and identifies and suggests more suitable products.

[1464] Input: new keyword "gray".

[1465] Operation: The server requests a re-search from the Amazon Marketplace API and Rakuten Market API to obtain product data for "gray sofas." It then analyzes the product information using a multi-function AI model.

[1466] Output: Send new product suggestion message to the terminal.

[1467] (Application example 1)

[1468] 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."

[1469] Conventional e-commerce systems have faced challenges in making it difficult for users to efficiently search for products and quickly find those that meet their requirements. Furthermore, product search and suggestion functions using voice input and augmented reality (AR) technology interfaces were inadequate, leaving a need for an improved user experience.

[1470] 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.

[1471] In this invention, the server includes means for receiving text data input by a user, means for analyzing the text data using a natural language processing engine to extract keywords and attribute information, means for acquiring product data from multiple e-commerce sites based on the extracted keywords, means for analyzing images of the acquired product data using a multimodal AI model to identify products that meet the user's requirements, means for suggesting the identified product information to the user, means for converting voice data into text data, and means for presenting product information in augmented reality on a visual display device, thereby enabling users to easily search for products through voice input, visually confirm product information using AR technology, and complete the purchase process efficiently and intuitively.

[1472] "User" refers to a consumer who searches for and purchases products using this system.

[1473] "Text data" refers to character information including requirements and conditions regarding products entered by a user.

[1474] "Natural language processing engine" refers to software and algorithms for understanding and analyzing human language.

[1475] "Keywords" refer to the main words necessary for searching for products, extracted from the user's input text.

[1476] "Attribute information" refers to specific characteristics or conditions related to a keyword.

[1477] "E-commerce site" refers to a website that sells goods and services over the Internet.

[1478] "Product Data" refers to information about a product obtained from an e-commerce site, primarily including the name, image, price, and link.

[1479] A "multimodal AI model" refers to an artificial intelligence model that integrates and analyzes multiple types of data (e.g., text data, image data).

[1480] A "visual display device" is a device that presents information to a user's field of vision, and primarily refers to smart glasses and augmented reality devices.

[1481] "Augmented reality (AR)" refers to the technology of overlaying digital information onto a real-world environment.

[1482] "Voice data" refers to sound information obtained from the user's speech.

[1483] "Product suggestion" refers to the act of presenting the most suitable product to the user based on analyzed product data.

[1484] The present invention is a system for enabling users to easily and quickly find and purchase desired products, and is particularly directed to a shopping assistant application that uses smart glasses. The system is configured as follows.

[1485] The server receives the text data entered by the user and analyzes it using a natural language processing engine. When the user enters product requirements by voice, the smart glasses' microphone picks up the voice data and converts it into text data. The converted text data is then analyzed by the natural language processing engine to extract keywords and attribute information.

[1486] Once the keywords are extracted, the server accesses the APIs of multiple e-commerce sites to retrieve the corresponding product data. An e-commerce site is a website that sells products and services over the Internet, and the product data includes the product name, image, price, and purchase link.

[1487] The acquired product data images are analyzed using a multimodal AI model, which is an artificial intelligence model that integrates and analyzes multiple types of data (e.g., text data, image data, etc.). This allows the system to identify the product that best suits the user's requirements.

[1488] The identified product information is then formatted and suggested to the user. The suggested product information is displayed as augmented reality (AR) on the smart glasses display. Augmented reality is a technology that overlays digital information on the real-world environment, allowing the user to intuitively check the product information superimposed on their field of vision.

[1489] For example, if a user says "I'm looking for red sneakers," the voice data is picked up through the smart glasses' microphone and converted into text data. The server then uses a natural language processing engine to analyze the converted text data and extract the keyword "red sneakers." Product data is then retrieved using the API of the e-commerce site, and analyzed using a multimodal AI model to identify the most suitable product.

[1490] As a result, product information for "red sneakers" is displayed in AR within the user's field of vision. When the user finds a specific product and says "I want to buy this," the server automatically initiates the purchase process through the e-commerce site's purchase API and sends a confirmation message to the user.

[1491] This system allows users to search for and purchase products efficiently and intuitively, providing an excellent purchasing experience.

[1492] Example prompt sentence:

[1493] The user speaks, "I'm looking for red sneakers."

[1494] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1495] Step 1:

[1496] The user speaks into the microphone of the smart glasses to input information about a product. For example, they might say, "I'm looking for red sneakers." This voice data is picked up by the microphone of the smart glasses. The input is voice data, and the output is voice data.

[1497] Step 2:

[1498] The device converts the acquired voice data into text data. This is done using a voice recognition library (e.g., Google's voice recognition API). This converts the voice data into text data. The input is voice data, and the output is text data.

[1499] Step 3:

[1500] Text data is sent to the server. The server analyzes the text data using a natural language processing engine (e.g., the BERT model) and extracts keywords and attribute information. This yields a keyword, for example, "red sneakers." The input is text data, and the output is keywords and attribute information.

[1501] Step 4:

[1502] The server accesses the APIs of multiple e-commerce sites based on the extracted keywords to retrieve related product data, including product names, images, prices, and purchase links. The input is the keywords, and the output is the product data.

[1503] Step 5:

[1504] The server inputs the acquired product data images into a multimodal AI model (e.g., CLIP model) and analyzes them. This identifies products that meet the user's requirements. The input is the product image, and the output is the identified product information.

[1505] Step 6:

[1506] The server formats the identified product information and presents it as augmented reality (AR) on the smart glasses display, allowing the user to view the product information directly in their field of vision. The input is the identified product information, and the output is the AR-displayed product information.

[1507] Step 7:

[1508] The user checks the display on the smart glasses, selects a specific product, and then voice-inputs, "I would like to purchase this." The input is voice data, and the output is text data indicating the purchase intention.

[1509] Step 8:

[1510] The terminal receives the voice data again, converts it into text data, and sends it to the server. The input is voice data, and the output is text data.

[1511] Step 9:

[1512] The server analyzes the received text data, confirms the user's purchase intention, and executes the purchase procedure via the e-commerce site's purchase API. The input is the text data of the purchase intention, and the output is purchase confirmation information.

[1513] Step 10:

[1514] The server confirms that the purchase has been completed and notifies the user of the result. The user receives a confirmation message on their smart glasses. The input is the purchase confirmation information, and the output is a user notification.

[1515] 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.

[1516] The present invention combines an emotion engine with a system that allows users to easily and quickly find and purchase the products they desire. The system operates via a chat interface, allowing users to input their desired product requirements and then suggests products that match those requirements. The system can also recognize the user's emotional state and adjust product suggestions based on the results.

[1517] Program processing and specific examples

[1518] 1. Accepting input from the user

[1519] The terminal receives text data in which the user enters "I want a new sofa" in the chat window.

[1520] 2. Text Interpretation and Emotion Recognition

[1521] The server passes the received text data to a natural language processing engine, analyzes the text, and uses an emotion engine to recognize the user's emotional state. For example, the emotion engine classifies the user's text as an emotion such as "joy" or "dissatisfaction."

[1522] 3. Extracting keywords and attribute information

[1523] The server uses a natural language processing engine to extract the keyword "sofa" and the attribute information "new" from the input text.

[1524] 4. Providing product images

[1525] The server retrieves product data (product images, price information, links, etc.) from APIs of multiple e-commerce sites based on the extracted keyword "sofa."

[1526] 5. Image Classification and Requirements Compliance Assessment

[1527] The server inputs the acquired product image data into a multimodal AI model for analysis. This analysis also takes into account the user's emotional state. For example, if a user's emotional state is "dissatisfied," the server will prioritize suggesting highly rated products and discount information.

[1528] 6. Generate product suggestions and present them to the user

[1529] The server then lists the identified product information and generates a suggestion message by adjusting the wording and tone of the message according to the user's emotional state. For example, it might provide a message such as "This sofa is recommended!" along with an image, price, and link to the selected product via chat.

[1530] Specific examples

[1531] For example, if the user again types "Show me more gray sofas," the device sends this new text to the server, which again uses the NLP engine to extract the new keyword "gray" and performs similar search and image analysis to generate new suggestions for gray sofas, again using the emotion engine to adjust the suggestions based on the user's latest emotional state.

[1532] Purchase procedure

[1533] When the user selects a product and enters "I want to purchase this," the device sends this purchase intention to the server. The server then sends the user's account information to the e-commerce site's purchase API, automatically starting the purchase process. The e-commerce site then completes the purchase process and returns a confirmation message to the server. The server then generates a purchase completion message and notifies the user via the device.

[1534] This system allows users to efficiently search for and purchase products, and provides an excellent purchasing experience with appropriate suggestions based on their emotional state.

[1535] The processing flow will be explained below.

[1536] Step 1:

[1537] The user types "I want a new sofa" into the chat window.

[1538] Step 2:

[1539] The terminal receives this text input and transmits it to the server in real time.

[1540] Step 3:

[1541] The server passes the received text data to a natural language processing engine and analyzes the text.

[1542] Step 4:

[1543] The server uses a natural language processing engine to extract the keyword "sofa" and the attribute information "new" from the input text.

[1544] Step 5:

[1545] The server inputs text data into an emotion engine to recognize the user's emotional state (e.g., "joy," "dissatisfaction," "excitement," etc.).

[1546] Step 6:

[1547] Based on the extracted keyword "sofa," the server sends requests to APIs of multiple e-commerce sites to retrieve related product data (product images, price information, links, etc.).

[1548] Step 7:

[1549] In response to the request, the server of the e-commerce site returns product data that matches the conditions to the server.

[1550] Step 8:

[1551] The server stores the acquired product data in a local database or memory.

[1552] Step 9:

[1553] The server inputs the saved product image data into a multimodal AI model for analysis.

[1554] Step 10:

[1555] The server uses a multimodal AI model to identify the products that best fit the user's requirements, taking into account not only color and design features but also the user's emotional state as recognized by an emotion engine.

[1556] Step 11:

[1557] The server then lists the identified product information and generates a suggested message by adjusting the wording and tone of the message according to the user's emotional state. For example, if the user is in a "dissatisfied" state, it adds an encouraging message such as "This sofa has been highly rated by many users!"

[1558] Step 12:

[1559] The server transmits the generated proposal message to the user through the terminal.

[1560] Step 13:

[1561] The user reviews the list of suggested products and provides additional input if they require more information (e.g., "Show me more gray sofas").

[1562] Step 14:

[1563] The terminal receives new text input and sends it back to the server.

[1564] Step 15:

[1565] The server again uses the NLP engine to extract the new keyword "gray" and similarly performs emotion recognition using the emotion engine.

[1566] Step 16:

[1567] The server retrieves product data again based on the new keywords and analyzes it using a multimodal AI model.

[1568] Step 17:

[1569] The server generates an updated product list and again creates and sends suggestion messages based on the user's latest emotional state.

[1570] Step 18:

[1571] The user selects a particular product and enters "I would like to purchase this."

[1572] Step 19:

[1573] The terminal transmits text data of the purchase intention to the server.

[1574] Step 20:

[1575] The server calls the purchase API of the e-commerce site and initiates the purchase process using the user's account information.

[1576] Step 21:

[1577] The e-commerce site's server processes the purchase and sends a confirmation message back to the server.

[1578] Step 22:

[1579] The server generates a purchase completion message and notifies the user through the terminal.

[1580] Through these specific processing steps, users can efficiently find products that meet their requirements and smoothly complete the purchasing process.The use of the emotion engine allows users to receive appropriate product suggestions based on their emotional state, significantly improving the purchasing experience.

[1581] Example 2

[1582] 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."

[1583] Conventional e-commerce systems have had difficulty in quickly and accurately searching for and suggesting products that users want. Furthermore, they were unable to suggest products that corresponded to the user's emotional state, resulting in low satisfaction. Furthermore, the purchasing process was cumbersome, making it difficult to improve the user experience.

[1584] 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.

[1585] In this invention, the server includes: means for receiving text data entered by a user; means for analyzing the text data using a natural language processing engine to extract keywords and attribute information; means for recognizing the user's emotional state from the text data; means for acquiring product data from multiple e-commerce sites based on the extracted keywords; means for analyzing images of the acquired product data using a multimodal generative AI model to identify products that meet the user's requirements and emotional state; and means for suggesting the identified product information to the user in a message corresponding to the user's emotional state. This allows for quick and accurate suggestions of products desired by the user, and further enables optimal product suggestions based on the user's emotional state. In addition, a smoother purchase process is expected to improve the user experience.

[1586] "User" means an individual or corporation that uses the system.

[1587] "Text data" is text information entered by the user.

[1588] A "natural language processing engine" is a software tool or algorithm for analyzing text data and extracting keywords and attribute information.

[1589] A "keyword" is an important word or phrase that identifies the product or service a user is looking for.

[1590] "Attribute information" is data that indicates characteristics and conditions that embody a user's request.

[1591] The "emotional state" is a recognition of the psychological state of the user when he or she inputs text data.

[1592] An "e-commerce site" is a website that sells and buys goods and services online.

[1593] "Product data" is information about a product, including images, prices, links, etc.

[1594] A "multimodal generative AI model" is an artificial intelligence model that integrates and analyzes multiple data formats (text, images, etc.).

[1595] "Analysis" is the processing of collected data to extract meaning and information.

[1596] "Proposal" is the act of providing appropriate product or service information to a user.

[1597] A "message" is text containing notifications or information sent from the system to a user.

[1598] A "purchase interface" is a software component that executes a purchase transaction on an e-commerce site.

[1599] The "purchase procedure" is a series of steps for purchasing the product or service selected by the user.

[1600] The present invention combines emotion recognition with a system that helps users quickly find and purchase the products they desire. The system operates via a chat interface and makes product suggestions based on text data entered by the user.

[1601] The system mainly consists of a server and a terminal. The terminal receives text data entered by the user in the chat window, and the server analyzes the data. The specific operation of the system is explained below.

[1602] First, the user enters the product requirements into the chat window on the device. For example, they might type, "I want a new sofa." This text data is received by the device and sent to the server.

[1603] The server passes the received text data to a natural language processing engine (e.g., Google Cloud Natural Language API) to extract keywords and attribute information. At the same time, it uses an emotion recognition engine (e.g., open-source emotion recognition software) to analyze the user's emotional state (e.g., joy, dissatisfaction). The analysis results allow the user's psychological state to be understood.

[1604] Next, the server retrieves product data (e.g., product images, price information, links, etc.) from multiple e-commerce sites (e.g., large online marketplaces) based on the extracted keywords, using the APIs of the e-commerce sites (e.g., widely used e-commerce APIs).

[1605] The server inputs the acquired product data into a multimodal generative AI model (e.g., CLIP model) and performs an integrated analysis of image and text elements. The system also takes the user's emotional state into consideration. For example, if the emotional state is "dissatisfied," it will prioritize suggestions of highly rated products and discount information.

[1606] The server generates product suggestions appropriate for the user based on the analysis results. The message is generated by adjusting the wording and tone of the message according to the user's emotional state. The message includes product images, prices, and links, and displays a message such as "We recommend this sofa!"

[1607] This allows users to view product details in the chat window and receive highly satisfying recommendations. For example, if the user types "Show me more gray sofas," the device sends this new text to the server, which again uses its natural language processing engine to extract the keyword "gray" and performs the same search and analysis. The emotion engine is also utilized again to adjust the recommendations based on the user's latest emotional state.

[1608] Finally, if the user selects the product and enters "I want to purchase this," the terminal sends this purchase intention to the server. The server starts the purchase procedure in cooperation with the purchase interface of the e-commerce site, and receives a confirmation message from the e-commerce site. The server then generates a purchase completion message and notifies the user through the terminal.

[1609] This system allows users to efficiently search for and purchase products, and provides an excellent purchasing experience by receiving appropriate suggestions based on their emotional state.

[1610] Prompt Sentence Examples

[1611] For example, if a user types "I'm looking for a red dress," the device sends the text data to the server, which then retrieves and suggests product data in a similar manner, using a search query based on the keyword "red dress" and further adjusting it according to the user's emotional state.

[1612] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1613] Step 1:

[1614] The user types "I want a new sofa" into the chat window. The device receives this text data and stores it in its internal memory. The input is text data based on the user's request, and the output is the stored text data. A specific example of this behavior is when the user types "I want a new sofa" and presses the SEND button.

[1615] Step 2:

[1616] The device sends the received text data to the server, which then analyzes the text data using a natural language processing engine (for example, a natural language processing API). The input is the received text data, and the output is the extracted keyword "sofa" and attribute information "new." The analysis results are saved in internal memory. A specific example of operation is when the server sends text to a natural language processing API and receives the analysis results.

[1617] Step 3:

[1618] The server passes the text data to an emotion recognition engine (e.g., emotion recognition software) to recognize the user's emotional state (e.g., "joy"). The input is text data, and the output is the emotion recognition result (type of emotion). A specific example of operation is sending text data to an emotion recognition engine to obtain the emotional state.

[1619] Step 4:

[1620] The server retrieves product data using the APIs of multiple e-commerce sites (e.g., large online marketplaces) based on the extracted keyword "sofa." The input is the keyword "sofa," and the output is the retrieved product data (product images, price information, and links). A specific example of the operation is to send a keyword to the API of an e-commerce site and receive product data.

[1621] Step 5:

[1622] The server inputs the acquired product data images into a multimodal generative AI model (e.g., an AI model) and analyzes the suitability of the product. The input is the product data image, and the output is the set product feature vector. The user's emotional state is also taken into consideration here. A specific example of operation is inputting a product image into the AI ​​model and obtaining a feature vector.

[1623] Step 6:

[1624] Based on the analysis results, the server selects product information appropriate for the user and generates a message according to their emotional state. The input is the user's emotional state and the product's feature vector, and the output is the suggested product information and the generated message. As a specific example of operation, if the emotional state is "dissatisfied," a message containing highly rated products and discount information is generated.

[1625] Step 7:

[1626] The server sends a proposal message to the terminal, and the terminal displays the message to the user. The input is the proposal message sent from the server, and the output is the proposal message displayed in the user's chat window. A specific example of the operation is to display the proposal message in the chat window.

[1627] Step 8:

[1628] The user inputs (or clicks) "I want to buy this" for the selected product, and that information is sent to the server by the terminal. The input is the user's intention to purchase, and the output is a purchase request sent to the server. A specific example of this behavior is when the user clicks the product purchase button.

[1629] Step 9:

[1630] The server sends the user's purchase information to the e-commerce site's purchase interface and begins the purchase process. The input is the user's purchase information, and the output is a notification that the purchase process has been completed. A specific example of this operation is sending the user's account information and product information to the purchase API.

[1631] Step 10:

[1632] The e-commerce site completes the purchase process and sends a confirmation message to the server. The server generates a purchase completion message and sends it to the terminal, which then notifies the user. The input is the purchase completion notification and the output is the purchase completion message. A specific example of this behavior is when the terminal displays "Purchase completed!" in a chat window.

[1633] In this way, the system can suggest optimal products based on the user's desires and emotional state, and efficiently process the purchase.

[1634] (Application example 2)

[1635] 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."

[1636] Conventional e-commerce systems provide insufficient support for users to find products quickly and appropriately, and in particular do not provide product suggestions that take into account the user's emotional state. Furthermore, they are unable to respond flexibly to the user's emotions, which can lead to a decrease in satisfaction. Furthermore, when a user makes additional input to a product suggestion, a fast and accurate response to that request is required, but current systems do not adequately achieve this.

[1637] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1638] In this invention, the server includes: means for receiving text data entered by a user; means for analyzing the text data using a natural language processing engine to extract keywords and attribute information; means for acquiring product data from multiple e-commerce sites based on the extracted keywords; means for analyzing images of the acquired product data using a multimodal AI model to identify products that meet the user's requirements; means for recognizing the user's emotional state and adjusting product suggestions based on the recognized emotional state; and means for suggesting the identified product information to the user using a message with wording and tone that matches the user's emotional state. This allows the server to appropriately determine the user's emotional state and make optimal product suggestions based on that state, thereby improving the user's purchasing experience. Furthermore, the server can quickly respond to additional user input and re-suggest suitable products, thereby flexibly meeting user requests.

[1639] A "user" is a person who uses the system to search for products and consider purchasing them.

[1640] "Text data" refers to character information entered by the user.

[1641] A "natural language processing engine" is a computer program that analyzes and interprets text data and extracts keywords and attribute information.

[1642] "Keywords" are important words that represent products or requirements identified from the text entered by the user.

[1643] "Attribute information" refers to additional information that accompanies a keyword, such as adjectives like "new" or "cheap."

[1644] An "e-commerce site" is a website that sells and buys goods over the Internet.

[1645] "Product data" is information about a product, specifically, product name, price, image, description, etc.

[1646] A "multimodal AI model" is an artificial intelligence model that can simultaneously analyze multiple forms of data, such as text and images.

[1647] "User requirements" refer to the product characteristics and conditions desired by the user.

[1648] "Emotional state" refers to the psychological state (e.g., joy, dissatisfaction, etc.) that a user experiences when searching for a product.

[1649] "Product suggestion" refers to the system recommending products selected by the system to the user.

[1650] "Message wording and tone" refers to the content of the product proposal message, which is adjusted according to the emotional state, and how it is delivered.

[1651] "Additional input" refers to new text information added by the user after the initial input.

[1652] The "purchase procedure" refers to a series of steps required for the user to actually purchase the product selected by the user.

[1653] A "purchase API" is an application programming interface that works with e-commerce sites to automate the product purchasing process.

[1654] The present invention is a system that allows users to easily and quickly find and purchase the products they desire, and is combined with an emotion engine. This system operates via a chat interface, and when users input their product requests, it suggests products that match those requests.

[1655] System Configuration

[1656] The system mainly includes the following elements:

[1657] Terminal: A device that allows users to input text data. This includes smartphones and PCs.

[1658] Server: A central system that analyzes text data, acquires product data, recognizes emotions, and generates product suggestions.

[1659] Natural language processing engine: Software for analyzing text data and extracting keywords and attribute information.

[1660] Multimodal AI model: An artificial intelligence model for analyzing data in multiple formats, such as text data and image data.

[1661] Emotion Engine: An engine for recognizing and analyzing the user's emotional state.

[1662] Process Flow

[1663] 1. Accept user input:

[1664] The user uses the terminal to input text data into the chat window, for example, "I want a new sofa."

[1665] 2. Text data analysis:

[1666] The server passes the received text data to a natural language processing engine, which analyzes the text and extracts the important keyword "sofa" and attribute information "new" from the text.

[1667] 3. Emotion recognition:

[1668] The server uses an emotion engine to analyze the user's emotional state, for example classifying the user's text as an emotion such as "happy" or "frustrated."

[1669] 4. Product data acquisition:

[1670] The server retrieves product data from APIs of multiple e-commerce sites based on the extracted keywords, including product images, price information, and detailed descriptions.

[1671] 5. Multimodal AI analysis:

[1672] The server inputs the acquired product data into a multimodal AI model for analysis, and the analysis results take into account the user's emotional state.

[1673] 6. Product proposal generation:

[1674] The server then lists the identified product information and generates product suggestions with message wording and tone that correspond to the user's emotional state. For example, it could provide a message such as "This sofa is recommended!" along with product images, prices, and links via chat.

[1675] Hardware and software used

[1676] Hardware: smartphones, PCs, servers

[1677] software:

[1678] OpenAI API: An API for emotion recognition.

[1679] TextBlob: A tool that performs natural language processing and extracts keywords from text.

[1680] requests library: A library for obtaining product data using the API of an e-commerce site.

[1681] Specific examples

[1682] For example, if the user again types "Show me more gray sofas," the device sends this new text to the server, which again uses the NLP engine to extract the new keyword "gray" and performs similar search and image analysis to generate new suggestions for gray sofas, again using the emotion engine to adjust the suggestions based on the user's latest emotional state.

[1683] Prompt Sentence Examples

[1684] "I want a new sofa"

[1685] "Show me a cheaper sofa."

[1686] "I want to see a gray sofa."

[1687] As described above, this system realizes flexible product suggestions that take into account the user's emotional state, thereby significantly improving the user's purchasing experience.

[1688] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1689] Step 1:

[1690] Accepting user input

[1691] Operation: A user inputs a request for a product into a chat window on the terminal.

[1692] Input: Text data entered by the user: "I want a new sofa."

[1693] Output: The user's input text is sent from the terminal to the server.

[1694] Step 2:

[1695] Text data analysis

[1696] How it works: The server passes the text data to a natural language processing engine for analysis, extracting important keywords and attribute information.

[1697] Input: User input text "I want a new sofa."

[1698] Data processing: A natural language processing engine is used to analyze the text and extract the attribute information "new" and the keyword "sofa."

[1699] Output: The extracted keyword information of "new" and "sofa" is obtained.

[1700] Step 3:

[1701] emotion recognition

[1702] Operation: The server uses the emotion engine to analyze the user's emotional state.

[1703] Input: User input text "I want a new sofa."

[1704] Data processing: Using an emotion engine to classify emotions such as "happy" or "frustrated" from user text.

[1705] Output: The user's emotional state (e.g., "joy" is recognized).

[1706] Step 4:

[1707] Get product data

[1708] Operation: The server retrieves product data from the APIs of multiple e-commerce sites based on the extracted keywords.

[1709] Input: The extracted keyword "sofa" and attribute information "new".

[1710] Data processing: Call the API of the e-commerce site and obtain the relevant product information (product images, price information, detailed descriptions, etc.).

[1711] Output: A list of retrieved product data.

[1712] Step 5:

[1713] Multimodal AI analysis

[1714] Operation: The server inputs the acquired product data images into a multimodal AI model to identify products that meet the user's requirements.

[1715] Input: A list of retrieved product data and the user's emotional state.

[1716] Data processing: Using a multimodal AI model to analyze images and evaluate products based on emotional states.

[1717] Output: A list of products that match the user requirements.

[1718] Step 6:

[1719] Generate product suggestions

[1720] Operation: The server generates a suggestion message based on the identified product information with a message wording and tone that corresponds to the emotional state.

[1721] Input: A list of products that match the user requirements and the user's emotional state.

[1722] Data processing: Generate appropriate messages taking into account the emotional state and list the identified product information.

[1723] Output: Product suggestion message to be presented to the user.

[1724] Step 7:

[1725] Presenting product suggestions to users

[1726] Operation: The server sends the generated proposal message to the terminal and presents it to the user.

[1727] Input: The generated product suggestion message.

[1728] Output: Product suggestion message displayed in the device's chat window.

[1729] Example: "Show me more gray sofas."

[1730] When the user makes this input, "gray" is extracted as a new keyword, and data is acquired and analyzed again, generating new product proposals that match the user's requirements.

[1731] 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.

[1732] 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.

[1733] 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.

[1734] 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.

[1735] 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.

[1736] 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.

[1737] 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).

[1738] 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.

[1739] 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."

[1740] 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.

[1741] 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).

[1742] 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.

[1743] 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.

[1744] 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.

[1745] 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.

[1746] 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.

[1747] 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.

[1748] 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.

[1749] 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.

[1750] 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.

[1751] 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.

[1752] The following is further disclosed regarding the above embodiment.

[1753] (Claim 1)

[1754] means for receiving text data input by a user;

[1755] means for analyzing the text data using a natural language processing engine to extract keywords and attribute information;

[1756] means for acquiring product data from a plurality of e-commerce sites based on the extracted keywords;

[1757] A means for analyzing the acquired product data images using a multimodal AI model and identifying products that meet the user's requirements;

[1758] means for suggesting the identified product information to a user;

[1759] A system including:

[1760] (Claim 2)

[1761] The system according to claim 1, further comprising means for acquiring product data again based on additional input from the user, and identifying and suggesting further suitable products.

[1762] (Claim 3)

[1763] 10. The system according to claim 1, further comprising means for cooperating with a purchase API of an e-commerce site and executing a purchase procedure to purchase the product selected by the user.

[1764] "Example 1"

[1765] (Claim 1)

[1766] means for receiving text data input by a user;

[1767] means for analyzing the text data using a natural language processing engine to extract keywords and attribute information;

[1768] means for acquiring product data from multiple online marketplaces based on the extracted keywords;

[1769] A means for analyzing the acquired product data images using a multi-function AI model and identifying products that meet the user's requirements;

[1770] means for suggesting the identified product information to a user;

[1771] A system including:

[1772] (Claim 2)

[1773] The system according to claim 1, further comprising means for acquiring product data again based on additional input from the user, and identifying and suggesting further suitable products.

[1774] (Claim 3)

[1775] 10. The system of claim 1, further comprising means for interacting with a purchasing interface of an online marketplace to execute a purchasing process to purchase the product selected by the user.

[1776] "Application Example 1"

[1777] (Claim 1)

[1778] means for receiving text data input by a user;

[1779] means for analyzing the text data using a natural language processing engine to extract keywords and attribute information;

[1780] means for acquiring product data from a plurality of e-commerce sites based on the extracted keywords;

[1781] A means for analyzing the acquired product data images using a multimodal AI model and identifying products that meet the user's requirements;

[1782] means for suggesting the identified product information to a user;

[1783] means for converting voice data into text data;

[1784] means for presenting product information in augmented reality on a visual display device;

[1785] A system including:

[1786] (Claim 2)

[1787] The system according to claim 1, further comprising means for acquiring product data again based on additional input from the user, and identifying and suggesting further suitable products.

[1788] (Claim 3)

[1789] 10. The system according to claim 1, further comprising means for cooperating with a purchase API of an e-commerce site and executing a purchase procedure to purchase the product selected by the user.

[1790] "Example 2: Combining Emotion Engines"

[1791] (Claim 1)

[1792] means for receiving text data input by a user;

[1793] means for analyzing the text data using a natural language processing engine to extract keywords and attribute information;

[1794] means for recognizing a user's emotional state from text data;

[1795] means for acquiring product data from a plurality of e-commerce sites based on the extracted keywords;

[1796] A means for analyzing the acquired product data images using a multimodal generative AI model to identify products that match the user's requirements and emotional state;

[1797] means for suggesting the identified product information to the user in a message according to the user's emotional state;

[1798] A system including:

[1799] (Claim 2)

[1800] 2. The system according to claim 1, further comprising means for acquiring product data again based on additional input from the user, and identifying and suggesting further suitable products.

[1801] (Claim 3)

[1802] 10. The system of claim 1, further comprising means for cooperating with a purchasing interface of an e-commerce site to carry out a purchasing process to purchase the product selected by the user.

[1803] "Application example 2 when combining emotion engines"

[1804] (Claim 1)

[1805] means for receiving text data input by a user;

[1806] means for analyzing the text data using a natural language processing engine to extract keywords and attribute information;

[1807] means for acquiring product data from a plurality of e-commerce sites based on the extracted keywords;

[1808] A means for analyzing the acquired product data images using a multimodal AI model and identifying products that meet the user's requirements;

[1809] means for recognizing an emotional state of a user and adjusting product suggestions in response to the recognized emotional state;

[1810] a means for suggesting the identified product information to the user in a message wording and tone according to the user's emotional state;

[1811] A system including:

[1812] (Claim 2)

[1813] 2. The system according to claim 1, further comprising means for acquiring product data again based on additional input from the user, and identifying and suggesting further suitable products.

[1814] (Claim 3)

[1815] 10. The system according to claim 1, further comprising means for cooperating with a purchase API of an e-commerce site and executing a purchase procedure to purchase the product selected by the user. [Explanation of symbols]

[1816] 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. means for receiving text data input by a user; means for analyzing the text data using a natural language processing engine to extract keywords and attribute information; means for acquiring product data from a plurality of e-commerce sites based on the extracted keywords; A means for analyzing the acquired product data images using a multimodal AI model and identifying products that meet the user's requirements; means for suggesting the identified product information to a user; A system including:

2. The system according to claim 1 , further comprising means for acquiring product data again based on the additional input from the user, and identifying and suggesting further suitable products.

3. 2. The system according to claim 1, further comprising means for cooperating with a purchase API of an electronic commerce site and executing a purchase procedure in order to purchase the product selected by the user.

Images