system

Abstract

A system is provided.SOLUTION: A system comprising: means for analyzing input requirements using a production AI; means for selecting appropriate data marts based on the analyzed requirements; means for automatically generating SQL queries based on the selected data marts; means for presenting the generated SQL queries to a user and accepting user confirmations and modifications; and means for executing the confirmed or modified SQL queries and presenting the results 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] This solution solves the problem that in-house data warehouse (DWH) users spend a lot of time and effort in the process of selecting the appropriate data mart and creating the necessary SQL queries. In particular, it is difficult to understand the different references for each case and create accurate queries, so there is a need to speed up and streamline the process. [Means for solving the problem]

[0005] This invention provides a means for analyzing requirements using a generation AI and selecting an optimal data mart, and a means for automatically generating SQL queries based on the selected data mart. Furthermore, we propose a system that includes a means for presenting the generated SQL query to the user and accepting user confirmation and correction, and a means for executing the confirmed or corrected SQL query and presenting the execution results to the user. Specifically, the generation AI analyzes the user's requirements using natural language processing technology and selects an appropriate data mart by referring to past consultation history. As a result, DWH users can effectively extract data and perform analysis tasks quickly, even without SQL skills.

[0006] "Generative AI" is an artificial intelligence system that uses natural language processing technology to understand user requirements, select data marts, and generate SQL queries.

[0007] "Means for analyzing requirements" is a function that understands the requirements entered by the user and performs processing to convert them into appropriate data marts and queries.

[0008] The "means for selecting a data mart" is a function that determines which data mart is most suitable based on the analyzed requirements and recommends it to the user.

[0009] "Means for automatically generating SQL queries" is a function that automatically generates the required SQL queries based on the selected data mart.

[0010] "Means for accepting review and correction" is a function that presents the generated SQL query to the user and allows the user to review the query and correct it if necessary.

[0011] "Means for executing SQL queries" refers to the function of executing SQL queries confirmed or modified by a user on the data warehouse.

[0012] The "means for presenting the execution results to the user" is a function for obtaining the results of the executed SQL query and displaying them to the user.

[0013] "Natural language processing technology" is a technology that enables computers to understand and appropriately process human language.

[0014] "Past consultation history" refers to data on the requirements and solutions users have provided to the generated AI. [Brief explanation of the drawings]

[0015] [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

[0016] 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.

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

[0018] 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).

[0019] 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.

[0020] 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.

[0021] 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.

[0022] 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."

[0023] [First embodiment]

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

[0025] 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.

[0026] 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).

[0027] 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.

[0028] 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.

[0029] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form 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.

[0030] 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.

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

[0032] 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.

[0033] 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.

[0034] 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.

[0035] 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."

[0036] This invention is a system that streamlines the process for users of an in-house data warehouse (DWH) to select an appropriate data mart and create the necessary SQL queries. This system uses generative AI to analyze the user's requirements, select a data mart based on those requirements, and automatically generate SQL queries.

[0037] Overall system configuration

[0038] This system is centered around a generation AI running on a server and the user's device.

[0039] 1. Obtaining user requirements:

[0040] The user inputs the data analysis requirements from the terminal. For example, the requirement is "I want to obtain three months' worth of purchase history by customer."

[0041] 2. Requirements Analysis:

[0042] The server sends the user's requirements to the generation AI. The generation AI uses natural language processing technology to analyze the requirements and identify keywords and objectives. For example, it extracts "by customer," "purchase history," and "three months' worth" from the requirements.

[0043] 3. Data mart selection:

[0044] The generation AI refers to past consultation history and recommends the most appropriate data mart. For example, it recommends a "customer data mart" and a "purchase history data mart."

[0045] 4. Generate SQL queries:

[0046] The AI ​​automatically generates SQL queries based on the selected data mart. For example, it generates the following query:

[0047] sql

[0048] SELECT

[0049] customer_id,

[0050] purchase_date,

[0051] product_id,

[0052] amount

[0053] FROM

[0054] customer_data_mart

[0055] WHERE

[0056] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[0057] (The code is not listed here)

[0058] 5. Check and correct the query:

[0059] The server presents the generated SQL query to the user, who can review it and modify it if necessary (for example, if the user wants to add specific conditions to the query's output).

[0060] 6. Execute the query:

[0061] The server sends the query approved by the user to the DWH, executes it, and retrieves the execution results and displays them to the user.

[0062] Specific examples

[0063] For example, if a user enters a requirement such as "I want to get the purchase history for each customer for the past three months," it will work as follows:

[0064] 1. Obtain requirements:

[0065] The user enters "I would like to obtain three months' worth of purchase history by customer" into the input form on the terminal.

[0066] 2. Requirements analysis:

[0067] The server sends the requirements to the generation AI, which then extracts and analyzes the keywords "by customer," "purchase history," and "three months' worth."

[0068] 3. Data mart selection:

[0069] The generative AI recommends a "customer data mart" and a "purchase history data mart" based on similar past requirements. The server presents this information to the user.

[0070] 4. SQL query generation:

[0071] The generative AI automatically generates the following SQL query:

[0072] sql

[0073] SELECT

[0074] customer_id,

[0075] purchase_date,

[0076] product_id,

[0077] amount

[0078] FROM

[0079] customer_data_mart

[0080] WHERE

[0081] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[0082] (The code is not listed here)

[0083] 5. Check and correct the query:

[0084] The server presents the generated SQL query to the user, who then reviews it and, if necessary, modifies it.

[0085] 6. Query execution:

[0086] The server sends the modified query to the DWH, executes the query, retrieves the execution results, and displays them to the user.

[0087] This allows users to quickly obtain the data they need without having SQL skills. This system improves the efficiency of data analysis and greatly simplifies the work of users.

[0088] The processing flow will be explained below.

[0089] Step 1:

[0090] The user enters the requirements into the input form on the terminal. For example, the user enters the requirement "I want to obtain three months' worth of purchase history by customer."

[0091] Step 2:

[0092] The terminal transmits the input requirement data to the server.

[0093] Step 3:

[0094] The server sends the requirements data to the generation AI, which analyzes the requirements.

[0095] Step 4:

[0096] The generative AI uses natural language processing technology to extract and structure requirement keywords. For example, it identifies keywords such as "by customer," "purchase history," and "three months' worth."

[0097] Step 5:

[0098] The generation AI refers to past consultation history and selects the optimal data mart based on similar requirements. For example, it recommends a "customer data mart" and a "purchase history data mart."

[0099] Step 6:

[0100] The server receives the recommended data mart information from the generation AI and presents it to the user.

[0101] Step 7:

[0102] The user confirms the presented data mart information and makes a selection to proceed to the next step.

[0103] Step 8:

[0104] The AI ​​automatically generates the necessary SQL queries based on the selected data mart. For example, it generates the following query:

[0105] sql

[0106] SELECT

[0107] customer_id,

[0108] purchase_date,

[0109] product_id,

[0110] amount

[0111] FROM

[0112] customer_data_mart

[0113] WHERE

[0114] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[0115] (The code is not listed here)

[0116] Step 9:

[0117] The server presents the generated SQL query to the user.

[0118] Step 10:

[0119] The user reviews the proposed SQL query and makes corrections if necessary.

[0120] Step 11:

[0121] The user submits the modified query to the server.

[0122] Step 12:

[0123] The server sends the confirmed or modified query to the DWH for execution.

[0124] Step 13:

[0125] The server retrieves the query results from the DWH.

[0126] Step 14:

[0127] The server presents the obtained results to the user.

[0128] These steps allow users to efficiently use the DWH to quickly obtain the data they need. Automating the entire system process improves the efficiency of data analysis work and ensures accurate data extraction.

[0129] Example 1

[0130] 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."

[0131] In today's business environment, companies are required to analyze data stored in their internal data warehouses (DWHs) appropriately and quickly obtain the necessary information. However, data processing and SQL query creation require highly specialized knowledge, which reduces business efficiency. In particular, it is difficult for non-expert users to effectively utilize data.

[0132] 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.

[0133] In this invention, the server includes means for analyzing input requirements using a generation AI, means for selecting an appropriate data structure based on the analyzed requirements, means for automatically generating a data extraction query based on the selected data structure, means for presenting the generated data extraction query to a user and accepting confirmation and correction by the user, and means for executing the confirmed or corrected data extraction query and presenting the execution result to the user. This enables users to efficiently and quickly obtain required data without requiring advanced specialized knowledge.

[0134] "Generative AI" is an artificial intelligence technology that analyzes input requirements and automatically generates appropriate data structures and queries.

[0135] "Means" refers to a component of a system or technology that accomplishes a specific function or operation.

[0136] "Data structure" refers to the format or schema of information stored in a database or data warehouse.

[0137] A "data extraction query" is program code, such as SQL, used to retrieve specific information from a database.

[0138] "User" refers to a person who uses this system to analyze data and create queries.

[0139] "Natural language processing technology" is a technology that enables computers to understand, analyze, and generate language that humans use on a daily basis.

[0140] "Past response history" refers to data that records requirements that have been previously processed and the results of those processing.

[0141] A "data warehouse (DWH)" is a type of database system that centrally stores large amounts of data and is used for analysis and reporting.

[0142] This invention is a system that streamlines the process by which users of an in-house data warehouse (DWH) can select an appropriate data structure and automatically generate the necessary data extraction queries. This system uses a generative AI model to analyze user requirements, select a data structure based on the analysis, and automatically generate data extraction queries.

[0143] This system is centered around a generative AI model running on a server and the user's device.

[0144] To obtain user requirements, the user inputs the requirements for data analysis into an input form on the terminal. For example, the user might input a requirement such as "I want to obtain three months' worth of purchase history by customer." The terminal then sends this requirement to the server.

[0145] Next, the requirements are analyzed. The server receives the requirements data sent by the user and sends it to the generative AI model as a prompt. This prompt takes the form, "I would like to obtain the past three months' worth of purchase history by customer." The generative AI model uses natural language processing technology to analyze the requirements and extract keywords and objectives. For example, it extracts keywords such as "by customer," "purchase history," and "three months' worth."

[0146] Next, the data structure is selected. The generative AI model refers to the history of similar requirements processing in the past and returns a list of optimal data structures to the server. For example, it selects a "customer data structure" and a "purchase history data structure." The server then presents this information to the user.

[0147] Next, SQL queries are generated. The server uses a generative AI model to automatically generate data extraction queries based on the selected data structure. The generated queries may have the following format:

[0148] sql

[0149] SELECT

[0150] customer_id,

[0151] purchase_date,

[0152] product_id,

[0153] amount

[0154] FROM

[0155] customer_data_structure

[0156] WHERE

[0157] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[0158] The generated data extraction query is presented to the user by the server.

[0159] The user can review the query on their device screen and modify it as needed, for example to narrow it down to a specific customer ID.

[0160] Finally, the server sends the query modified and approved by the user to the DWH and executes it. The server retrieves the execution results and displays them to the user. This allows users to efficiently and quickly obtain the data they need, even without specialized knowledge of SQL.

[0161] An example of a prompt would be something like, "The user has entered the requirement, 'I want to retrieve the purchase history for each customer for the past three months.' Please select the optimal data structure based on this requirement and generate the necessary data extraction query."

[0162] Through this system, users can automatically generate the SQL queries required for data analysis, streamlining the data extraction process, enabling fast and accurate data analysis in the field of business, significantly improving work efficiency.

[0163] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0164] Step 1: Obtain user requirements

[0165] Specific behavior:

[0166] The user enters the requirements for data analysis into the input form on the terminal. For example, the user might enter, "I want to obtain three months' worth of purchase history by customer."

[0167] The terminal receives this input data and sends it to the server.

[0168] Input: Requirements entered by the user into the terminal

[0169] Output: Requirement data sent to the server

[0170] Step 2: Analyze requirements

[0171] Specific behavior:

[0172] The server receives the requirements data sent by the user.

[0173] The server creates a prompt for the generative AI model and sends it the requirements. An example of a prompt would be "I would like to obtain the purchase history for each customer for the past three months."

[0174] The generative AI model uses natural language processing technology to analyze requirements and extract keywords and objectives, such as "by customer," "purchase history," and "three months' worth."

[0175] Input: Requirement data submitted by the user

[0176] Output: Keywords and analysis results

[0177] Step 3: Choose a data structure

[0178] Specific behavior:

[0179] Based on the analysis results, the generative AI model refers to the history of similar requirements processing in the past.

[0180] The server receives a list of optimal data structures (data marts) from the generative AI model. For example, it selects a "customer data structure" and a "purchase history data structure."

[0181] The server presents this information to the user.

[0182] Input: Analysis results, past requirements processing history

[0183] Output: A list of selected data structures

[0184] Step 4: Generate SQL queries

[0185] Specific behavior:

[0186] The server uses the generative AI model to automatically generate SQL queries based on the selected data structure.

[0187] The generated query may have the following format:

[0188] sql

[0189] SELECT

[0190] customer_id,

[0191] purchase_date,

[0192] product_id,

[0193] amount

[0194] FROM

[0195] customer_data_structure

[0196] WHERE

[0197] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[0198] The server presents the generated data extraction query to the user.

[0199] Input: A list of selected data structures

[0200] Output: Auto-generated SQL query

[0201] Step 5: Check and fix your queries

[0202] Specific behavior:

[0203] The user sees the generated SQL query on the terminal screen.

[0204] The user modifies the query as needed, for example to narrow it down to a specific customer ID.

[0205] The modified SQL query is then sent to the server again.

[0206] Input: Generated SQL query

[0207] Output: SQL query confirmed or modified by the user

[0208] Step 6: Execute the query

[0209] Specific behavior:

[0210] The server sends the SQL query, which has been modified and approved by the user, to the DWH.

[0211] The server receives the execution results from the DWH and displays them to the user.

[0212] As a result of the execution, for example, a list of the customer's purchase history is displayed on the terminal.

[0213] Input: Confirmed or revised SQL query

[0214] Output: Execution results obtained from the DWH

[0215] (Application example 1)

[0216] 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."

[0217] Modern logistics centers are required to efficiently manage large volumes of shipping records and inventory data and quickly retrieve the necessary data. In particular, there is a growing need for systems that allow users to easily analyze data, even if they do not have specific SQL skills. However, conventional systems often require users to be familiar with complex database structures, which takes time to learn. It is also difficult to quickly select the appropriate data source. Therefore, there is a need to develop a system that can perform data analysis efficiently.

[0218] 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.

[0219] In this invention, the server includes means for analyzing input requirements using a generation AI, means for selecting an appropriate data mart based on the analyzed requirements, means for automatically generating an SQL query based on the selected data mart, means for presenting the generated SQL query to a user and accepting confirmation and correction by the user, means for executing the confirmed or corrected SQL query and presenting the execution result to the user, and means for automatically acquiring past shipping records and inventory data, specialized for data analysis of a logistics center. This enables users to efficiently analyze data and quickly acquire appropriate data even if they do not have specific SQL skills.

[0220] "Generative AI" is an artificial intelligence technology that analyzes user requirements, selects appropriate data marts, and automatically generates SQL queries.

[0221] A "data mart" is an aggregation of data sets extracted from a large data warehouse to address a specific business need.

[0222] An "SQL query" is a command in the Structured Query Language that describes an instruction to a database.

[0223] A "logistics center" is a facility that receives, stores, organizes, and ships goods and materials.

[0224] "Past shipping records" are historical information on products that have been shipped from the logistics center in the past.

[0225] "Inventory data" is information about the quantity and condition of products currently stored in the logistics center.

[0226] "Natural language processing technology" is a technology that enables computers to understand, analyze, and generate human language.

[0227] "Verify and fix" is the process whereby the user reviews the auto-generated SQL queries and makes changes if necessary.

[0228] This invention relates to a system that utilizes generative AI to improve the efficiency of data analysis in logistics centers. This system is centered around a server and user terminals.

[0229] First, the user inputs the requirements for data analysis from a terminal. For example, they might input a requirement such as "I want to obtain shipping records for each warehouse for the last three months." The server receives the user's requirements and sends them to the generation AI. This generation AI uses natural language processing technology to analyze the requirements and extract the keywords "warehouse," "shipping records," and "three months."

[0230] The generative AI then selects appropriate data marts based on these keywords, recommending, for example, "warehouse data mart" and "shipping record data mart," and the server presents this information to the user.

[0231] The Generative AI then automatically generates SQL queries based on the selected data mart. For example, the following SQL queries are generated:

[0232] SELECT

[0233] warehouse_id,

[0234] shipment_date,

[0235] product_id,

[0236] quantity

[0237] FROM

[0238] warehouse_data_mart

[0239] WHERE

[0240] shipment_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[0241] The server presents the generated SQL query to the user and accepts confirmation and modification by the user. The user confirms the query and makes modifications as necessary. The server then executes the confirmed or modified SQL query and presents the execution results to the user.

[0242] Specifically, it is implemented as shown in the following Python code, which uses an SQLite database and executes queries according to the user's requirements. The generative AI uses OpenAI's GPT model.

[0243] This allows the server to efficiently analyze data and quickly retrieve appropriate data, even if the user does not have specific SQL skills. This makes it easier for distribution centers to quickly check shipping records and manage inventory.

[0244] Examples of prompt statements

[0245] Prompt for the generative AI model:

[0246] Analyze the following requirements: Get shipping records for each warehouse for the last three months

[0247] As described above, the present invention provides a concrete example of a system that simplifies data management at a logistics center and enables efficient data analysis. This system allows users to quickly obtain the data they need without having to worry about complex database operations.

[0248] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0249] Step 1:

[0250] The user inputs the requirements for data analysis from the terminal. For example, they input a requirement such as "I want to obtain the shipping records for each warehouse for the last three months." This requirement is sent to the server.

[0251] Input: The requirement text entered by the user

[0252] Output: The text of the requirement as received by the server

[0253] Step 2:

[0254] The server sends the received requirements to the generation AI, which then analyzes them using natural language processing technology. During the analysis, keywords such as "warehouse," "shipping record," and "3 months" are extracted from the requirements.

[0255] Input: Text of the requirement received from the user

[0256] Output: Generated keyword list (e.g. "warehouse", "shipping record", "3 months")

[0257] Step 3:

[0258] Based on the extracted keywords, the generation AI refers to past consultation history and selects the appropriate data mart. For example, it might recommend a "warehouse data mart" and a "shipping record data mart."

[0259] Input: Keyword list, past consultation history data

[0260] Output: A list of recommended data marts (e.g., "Warehouse Data Mart" and "Shipping Record Data Mart")

[0261] Step 4:

[0262] The server automatically generates SQL queries based on the selected data mart using the AI. The generated SQL queries are, for example, as follows:

[0263] SELECT

[0264] warehouse_id,

[0265] shipment_date,

[0266] product_id,

[0267] quantity

[0268] FROM

[0269] warehouse_data_mart

[0270] WHERE

[0271] shipment_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[0272] Input: Recommended data mart list, keyword list

[0273] Output: Auto-generated SQL query

[0274] Step 5:

[0275] The server presents the generated SQL query to the user, who can review it and make modifications if necessary. The user can review the query and add or modify specific conditions.

[0276] Input: Auto-generated SQL query

[0277] Output: SQL query reviewed and corrected by the user

[0278] Step 6:

[0279] The server executes the verified or modified SQL query and retrieves the results, for example, by querying an SQLite database and retrieving the results.

[0280] Input: Confirmed or revised SQL query

[0281] Output: Query results (data retrieved from the database)

[0282] Step 7:

[0283] The server then presents the results of the query to the user, allowing the user to quickly obtain the data they need.

[0284] Input: Query results

[0285] Output: Execution results presented to the user (e.g., shipping records by warehouse for the past three months)

[0286] 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.

[0287] This invention is a system that enables users of an in-house data warehouse (DWH) to efficiently extract data and perform analytical tasks, and aims to improve the user experience by combining it with an emotion engine that recognizes the user's emotions.

[0288] Overall system configuration

[0289] The system includes the following main components:

[0290] 1. Generation AI

[0291] 2. Natural Language Processing Technology

[0292] 3. Emotion Engine

[0293] 4. Server

[0294] 5. User Device

[0295] Acquiring and analyzing user requirements

[0296] 1. Obtaining user requirements

[0297] The user enters requirements into an input form on the terminal. For example, the user enters a requirement such as "I want to obtain three months' worth of purchase history by customer."

[0298] 2. Emotion recognition

[0299] While the terminal is inputting user requirements, the emotion engine monitors the user's emotional state. For example, if the user is feeling stressed, the emotion engine will recognize this.

[0300] 3. Submit your requirements

[0301] The terminal transmits the input requirement data and the user's emotion data to the server.

[0302] Requirements and sentiment data analysis

[0303] 4. Requirements Analysis

[0304] The server sends the requirements data to the generation AI, which then uses natural language processing technology to extract and structure keywords from the requirements. For example, it identifies keywords such as "by customer," "purchase history," and "three months' worth."

[0305] 5. Emotion Data Analysis

[0306] The server analyzes the user's emotional state based on the data from the emotion engine and determines whether the user is feeling stressed.

[0307] Data mart selection and query generation

[0308] 6. Data Mart Selection

[0309] The generative AI refers to past consultation history and selects the optimal data mart based on the analyzed requirements and emotional state. For example, it recommends a "customer data mart" and a "purchase history data mart."

[0310] 7. Presentation to the User

[0311] The server receives the recommended data mart information from the generation AI and displays it on the device. It also adjusts the interface and provides additional support content according to the user's emotional state.

[0312] 8. Generating SQL Queries

[0313] The AI ​​automatically generates the necessary SQL queries based on the selected data mart. For example, it generates the following query:

[0314] sql

[0315] SELECT

[0316] customer_id,

[0317] purchase_date,

[0318] product_id,

[0319] amount

[0320] FROM

[0321] customer_data_mart

[0322] WHERE

[0323] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[0324] (The code is not listed here)

[0325] Review and run the query

[0326] 9. Check and correct your query

[0327] The server presents the generated SQL query to the user, who confirms it and modifies it if necessary. If the user is feeling nervous or stressed, the emotion engine displays an appropriate support message.

[0328] 10. Executing a Query

[0329] The server sends the confirmed or modified query to the DWH for execution.

[0330] Displaying execution results

[0331] 11. Obtaining and Presenting Results

[0332] The server retrieves the query results from the DWH and displays them on the device. It monitors the user's emotional state and interprets the results and suggests the next action if necessary.

[0333] Specific examples

[0334] For example, if a user enters the requirement "I want to get the purchase history for each customer for the past three months":

[0335] 1. Requirements acquisition: The user enters the requirements into an input form on the terminal.

[0336] 2. Emotion Recognition: The emotion engine recognizes the user's emotions (e.g., stress state).

[0337] 3. Sending requirements and emotion data: The device sends the data to the server.

[0338] 4. Requirements analysis: The server sends the requirements data to the generation AI for analysis.

[0339] 5. Emotional data analysis: Analyze user emotions.

[0340] 6. Data mart selection: Generative AI recommends the optimal data mart.

[0341] 7. Presentation to the user: The server presents emotional support along with data mart information.

[0342] 8. SQL Query Generation: Generative AI generates SQL queries.

[0343] 9. Review and modify the query: The user reviews the query and modifies it if necessary.

[0344] 10. Query execution: The server sends the query to the DWH and executes it.

[0345] 11. Retrieving and presenting results: The server retrieves the results and presents them to the user.

[0346] This system allows users to efficiently retrieve data and perform analysis work without stress, even if they do not have SQL skills.The emotion engine monitors the user's emotional state and provides appropriate support, resulting in a better user experience.

[0347] The processing flow will be explained below.

[0348] Step 1:

[0349] The user enters the requirements into the input form on the terminal. For example, the user enters "I want to obtain three months' worth of purchase history by customer."

[0350] Step 2:

[0351] The emotion engine monitors the user's emotional state, for example, analyzing the user's stress level from facial expressions and tone of voice via the camera and microphone while the user is typing.

[0352] Step 3:

[0353] The terminal transmits the input requirement data and emotion data to the server.

[0354] Step 4:

[0355] The server sends the requirement data and emotion data to the generation AI, which receives the requirement data and begins analyzing it.

[0356] Step 5:

[0357] Generative AI uses natural language processing technology to analyze requirement data and identify and extract keywords and objectives. For example, it identifies "by customer," "purchase history," and "three months' worth."

[0358] Step 6:

[0359] The server receives the analysis results from the generation AI and stores them along with the emotion data.

[0360] Step 7:

[0361] The generation AI refers to past consultation history and selects the appropriate data mart based on the analyzed requirements. For example, it recommends a "customer data mart" and a "purchase history data mart."

[0362] Step 8:

[0363] The server receives the recommendation data mart information from the generation AI and displays it on the device, and also adjusts the interface based on the user's emotional state and displays necessary support messages.

[0364] Step 9:

[0365] The user confirms the presented data mart information and makes a selection to proceed to the next step.

[0366] Step 10:

[0367] The AI ​​automatically generates the necessary SQL queries based on the selected data mart. For example, it generates the following query:

[0368] sql

[0369] SELECT

[0370] customer_id,

[0371] purchase_date,

[0372] product_id,

[0373] amount

[0374] FROM

[0375] customer_data_mart

[0376] WHERE

[0377] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[0378] (The code is not listed here)

[0379] Step 11:

[0380] The server presents the generated SQL query to the user.

[0381] Step 12:

[0382] The user checks the presented SQL query and modifies it as necessary. For example, if they want to extract only a specific product category, they can add that condition.

[0383] Step 13:

[0384] The user submits the modified query to the server.

[0385] Step 14:

[0386] The server sends the confirmed or modified query to the DWH and executes the query.

[0387] Step 15:

[0388] The server retrieves the query results from the DWH.

[0389] Step 16:

[0390] The server sends the results to the device and displays them to the user. It also monitors the user's emotional state again and displays a message to guide the user on the next action based on the results.

[0391] These steps allow users to efficiently extract data and perform data analysis.The emotion engine monitors the user's emotional state in real time and takes appropriate action to provide a better user experience.

[0392] Example 2

[0393] 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."

[0394] Conventional in-house data warehouse (DWH) systems have the drawback of making it difficult for users to efficiently extract and analyze data. Furthermore, simple data processing is often performed without considering the user's emotional state, resulting in a poor user experience. The present invention aims to solve these problems and provide a system that enables users to efficiently extract and analyze data without feeling stressed.

[0395] 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.

[0396] In this invention, the server includes: means for a user to input requirements into an input form on a terminal; means for an emotion engine to monitor the user's emotional state while the user is inputting the user's requirements; means for transmitting the input requirement data and the user's emotional data to the server; means for transmitting the requirement data to a generation AI and extracting and structuring keywords of the requirements using natural language processing technology; means for analyzing the user's emotional state based on the data from the emotion engine and determining whether the user is feeling stressed; means for selecting an optimal data mart based on the analyzed requirements and emotional state by referring to past consultation history; means for receiving recommended data mart information from the generation AI and presenting interface adjustments or additional support content according to the user's emotional state; means for automatically generating a required SQL query based on the selected data mart; means for presenting the generated SQL query to the user and accepting user confirmation and correction; and means for executing the confirmed or corrected SQL query and presenting the execution result to the user. This allows the user to efficiently extract and analyze data, and further enables the emotion engine to improve the user experience.

[0397] A "terminal" is an input / output device that a user uses to input requirements.

[0398] An "emotion engine" is a software and hardware system that monitors and analyzes a user's emotional state in real time.

[0399] "Requirement data" refers to specific requests for data extraction and analysis that are input by the user into the terminal.

[0400] "Emotion data" is data that represents the user's emotional state and is acquired by the emotion engine.

[0401] A "server" is a computer system that receives user requirement data and emotion data and performs various processes based on them.

[0402] "Generative AI" is an artificial intelligence that analyzes requirements data and extracts and structures keywords using natural language processing technology.

[0403] "Natural language processing technology" is information technology used to interpret and understand the meaning of human language.

[0404] A "data mart" is a portion of a database that stores a subset of data related to a particular business process or area.

[0405] An "SQL query" is a command statement written in the Structured Query Language (SQL) for searching and manipulating data in a database management system (DBMS).

[0406] The "past consultation history" is a record of requirements that the user has exchanged with the server in the past and the results of the processing.

[0407] "Adjusting the interface" refers to changing the screen display or user interface configuration in response to the user's emotional state.

[0408] "Additional support content" refers to supplementary information or messages provided depending on the user's emotional state.

[0409] This invention is a system that aims to enable users of in-house data warehouses (DWHs) to efficiently extract data and perform analysis work. Furthermore, by combining it with an emotion engine that recognizes user emotions, we aim to improve the user experience.

[0410] Overall system configuration

[0411] The system includes the following major components:

[0412] 1. Generation AI

[0413] 2. Natural Language Processing Technology

[0414] 3. Emotion Engine

[0415] 4. Server

[0416] 5. User Device

[0417] Hardware and Software

[0418] Terminal: An input device for users to enter requirements. It uses a web browser or GUI form.

[0419] Emotion engine: Uses sensors and software to monitor the user's facial expressions and typing speed in real time and analyze their emotional state. Specific examples include camera sensors and keyboard typing speed analysis software.

[0420] Server: A computer system that receives requirements data and emotion data and analyzes and structures the data using generative AI models and natural language processing techniques.

[0421] Generative AI: An artificial intelligence that analyzes user requirements data and extracts and structures keywords using natural language processing technology.

[0422] Data mart: A portion of a database that stores a subset of data related to a particular business process or area.

[0423] System Operation

[0424] When a user enters requirements into an input form on a terminal, for example, "I would like to obtain the past three months' worth of purchase history by customer," the emotion engine monitors the user's facial expressions and input speed and analyzes their emotional state. The terminal then sends the entered requirement data and emotion data to the server.

[0425] The server sends the received requirements data to the generation AI, which uses natural language processing technology to extract and structure keywords. Based on the extracted keywords and taking into account data from the emotion engine, the generation AI determines whether the user is feeling stressed. The generation AI refers to past consultation history and selects the optimal data mart based on the analyzed requirements and emotional state. The server receives the recommended data mart information and displays it on the device. At this time, it adjusts the interface and presents additional support content according to the user's emotional state.

[0426] The generation AI automatically generates the necessary SQL queries based on the selected data mart. The generated SQL queries are presented to the user, who can confirm and modify them. Finally, the confirmed or modified SQL queries are sent from the server to the DWH, and data is retrieved. The retrieved data is displayed on the terminal, and the user can confirm the results. At this time, the emotion engine displays support messages as needed.

[0427] Specific examples

[0428] 1. Obtaining requirements: The user enters the requirement "I want to obtain the purchase history for each customer for the past three months" into the input form on the terminal.

[0429] 2. Emotion Recognition: The emotion engine monitors the user's facial expressions and input speed to recognize stress levels.

[0430] 3. Sending requirements and emotion data: The terminal sends the requirements data and emotion data to the server.

[0431] 4. Requirements analysis: The generative AI extracts the keywords "by customer," "purchase history," and "three months' worth."

[0432] 5. Emotional data analysis: Determine whether the user is feeling stressed.

[0433] 6. Data mart selection: The generative AI selects the optimal data mart (e.g., "customer data mart" or "purchase history data mart").

[0434] 7. Presentation to the user: The recommended data mart and support content according to the emotional state are displayed on the terminal.

[0435] 8. SQL Query Generation: The generative AI automatically generates SQL queries like the following:

[0436] sql

[0437] SELECT

[0438] customer_id,

[0439] purchase_date,

[0440] product_id,

[0441] amount

[0442] FROM

[0443] customer_data_mart

[0444] WHERE

[0445] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[0446] 9. Review and modify the query: The user reviews the query and modifies it if necessary.

[0447] 10. Query execution: The server sends the confirmed or modified query to the DWH and executes it.

[0448] 11. Obtaining and presenting results: The server displays the obtained data on the terminal, and the user confirms the results.

[0449] Prompt Sentence Examples

[0450] Generate a SQL query to retrieve customer purchase history for the past three months.

[0451] This system allows users to efficiently retrieve data and perform analysis work without stress, even if they do not have SQL skills.The emotion engine monitors the user's emotional state and provides appropriate support, resulting in a better user experience.

[0452] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0453] Step 1:

[0454] The user inputs requirements into an input form on the terminal. Specifically, the user inputs the requirement "I want to obtain the purchase history for each customer for the past three months" into the input form. At this time, the terminal receives the information entered by the user as input data. The input data is requirement information in text format.

[0455] Step 2:

[0456] The emotion engine monitors the user's emotional state while they are entering their requirements. Specifically, a camera sensor analyzes the user's facial expressions and keyboard input speed in real time to generate emotion data. The input is the user's facial image and input speed data, and the output is emotional state data in text format (e.g., stress level).

[0457] Step 3:

[0458] The device sends the input requirement data and user emotion data to the server. The input is the user requirement data and emotion data, and the output is the requirement data and emotion data sent to the server. Specifically, it is sent to the server as JSON format data.

[0459] Step 4:

[0460] The server sends the requirements data to the generation AI, which uses natural language processing technology to extract and structure requirement keywords. The input is the requirements data (e.g., "I want to obtain the purchase history by customer for the past three months"), and the output is the extracted keywords (e.g., "by customer," "purchase history," "three months' worth"). Specifically, the generation AI uses a natural language processing engine to analyze and structure the keywords.

[0461] Step 5:

[0462] The server analyzes the user's emotional state based on data from the emotion engine and determines whether the user is feeling stressed. The input is emotional data (e.g., stress state), and the output is a judgment result of the user's emotional state (e.g., the user is feeling stressed). Specifically, the server runs an emotion analysis algorithm and quantifies the emotional state.

[0463] Step 6:

[0464] The generation AI refers to past consultation history and selects the optimal data mart based on the analyzed requirements and emotional state. The input is requirement keywords and emotional state data, and the output is a recommended data mart (e.g., "customer data mart" or "purchase history data mart"). Specifically, the generation AI runs a data mart selection algorithm to select the optimal data mart.

[0465] Step 7:

[0466] The server receives the recommendation data mart information from the generation AI and displays it on the device. Furthermore, it adjusts the interface and presents additional support content according to the user's emotional state. The input is the recommendation data mart and the user's emotional state, and the output is the information displayed on the device. Specifically, the server receives the recommendation data mart information and adjusts the UI according to the emotional state.

[0467] Step 8:

[0468] The generation AI automatically generates the required SQL queries based on the selected data mart. The input is the selected data mart information and requirement keywords, and the output is the generated SQL query. Specifically, the generation AI automatically generates the following SQL queries:

[0469] sql

[0470] SELECT

[0471] customer_id,

[0472] purchase_date,

[0473] product_id,

[0474] amount

[0475] FROM

[0476] customer_data_mart

[0477] WHERE

[0478] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[0479] Step 9:

[0480] The server presents the generated SQL query to the user, who confirms it and modifies it if necessary. The input is the generated SQL query, and the output is the SQL query that has been confirmed and modified by the user. Specifically, the server displays the SQL query on the terminal, and the user confirms and modifies the query content.

[0481] Step 10:

[0482] The server sends the confirmed or modified SQL query to the DWH and executes it. The input is the confirmed or modified SQL query, and the output is the query execution result from the DWH. In concrete terms, the server sends the SQL query to the data warehouse, executes the query, and retrieves the data.

[0483] Step 11:

[0484] The server retrieves the query execution results from the DWH and displays them on the terminal. It monitors the user's emotional state and, if necessary, interprets the results and suggests the next action. The input is the query execution results and the user's emotional state, and the output is the execution results and a support message displayed on the terminal. In concrete terms, the server retrieves the query results, and if the user is feeling stressed, it displays an interpretation support message.

[0485] (Application example 2)

[0486] 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."

[0487] With conventional data analysis systems, if users lacked SQL skills or database knowledge, it was difficult to efficiently extract and analyze data. Furthermore, because data was provided without taking into account the user's emotional state, the user interface and support were often inappropriate. This resulted in a poor user experience and reduced productivity, especially when tasks were complex or under stress.

[0488] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for analyzing input requirements using a generation AI, means for selecting an appropriate data mart based on the analyzed requirements, means for automatically generating an SQL query based on the selected data mart, means for presenting the generated SQL query to a user and accepting confirmation and correction by the user, means for executing the confirmed or corrected SQL query and presenting the execution result to the user, an emotion recognition engine for monitoring the user's emotional state, and means for generating and presenting a response method based on the emotional state. This enables users to efficiently extract and analyze data without advanced technical knowledge, and furthermore, by providing support appropriate to the user's emotional state, it is possible to reduce stress and improve the user experience.

[0489] "Generative AI" is an artificial intelligence technique used to analyze data and automate tasks.

[0490] The "means for analyzing requirements" is a method for analyzing the requirements data entered by the user and extracting important keywords and intentions.

[0491] The "means for selecting an appropriate data mart" is a method for selecting the most relevant data mart based on the analyzed requirements.

[0492] "Means for automatically generating SQL queries" refers to a method for automatically creating the necessary SQL queries for the selected data mart.

[0493] The "means for presenting the generated SQL query to the user and accepting confirmation and correction by the user" refers to a method for displaying the generated SQL query to the user so that the user can confirm or correct it.

[0494] "Means for executing a confirmed or modified SQL query and presenting the execution results to a user" refers to a method for executing an SQL query confirmed or modified by a user and displaying the results to the user.

[0495] The "emotion recognition engine that monitors the user's emotional state" is a system that analyzes the user's emotions in real time and determines their state.

[0496] The "means for generating and presenting a response method based on an emotional state" is a method for generating an optimal response method according to the emotional state of the user determined by an emotion recognition engine and presenting it to the user.

[0497] As an embodiment of the present invention, a customer service support system using smart glasses for store clerks in a brick-and-mortar store can be adopted. The following describes the specific roles of the server, terminal, and user.

[0498] server

[0499] The server performs the following operations using the following specific hardware and software:

[0500] 1. Generative AI and Natural Language Processing Technology

[0501] Software used: Python language, natural language processing library (e.g., SpaCy)

[0502] The server acquires the user's requirements and uses generation AI to extract and analyze keywords from the requirements.

[0503] For example, from the requirement "Please tell me your purchase history for the past three months," extract keywords such as "past three months" and "purchase history."

[0504] 2. Emotion Recognition Engine

[0505] Hardware used: Smart glasses camera

[0506] Software used: Emotion-FerPlus

[0507] The server analyzes the video data acquired through the glasses' camera and monitors the customer's emotional state in real time.

[0508] For example, determining whether a customer is stressed.

[0509] 3. Data mart selection and SQL query generation

[0510] Software used: SQLAlchemy

[0511] The server selects a data mart based on the analyzed requirements and automatically generates the necessary SQL queries.

[0512] For example, generate a query to extract data from a purchase history data mart within a specific period.

[0513] Terminal (smart glasses)

[0514] The smart glasses worn by the store clerk function as a terminal and support the following processes:

[0515] 1. Input customer requirements

[0516] Enter customer requirements using voice input or touch operation.

[0517] For example, a request such as "I want to know the purchase history for the past three months" is input.

[0518] 2. Displaying emotional states

[0519] The analysis results of the emotion recognition engine are displayed in real time.

[0520] For example, if a customer is feeling stressed, the glasses' display will say "Customer is feeling stressed."

[0521] 3. Displaying query results and suggesting how to respond

[0522] The data returned from the server and the response method based on the emotional state are displayed on the glasses' display.

[0523] For example, along with the purchase history as a query result, the message "The customer is stressed. Please respond calmly" may be displayed.

[0524] User (store clerk)

[0525] Store staff use the above system to efficiently handle customer service.

[0526] 1. Obtaining and inputting user requirements

[0527] Customer requirements are registered into the smart glasses using voice or touch input.

[0528] An example of a specific prompt might be: "The customer wants purchase history data for the past three months. Please generate a query."

[0529] 2. Checking and responding to emotional states

[0530] Check the customer's emotional state in real time and respond accordingly.

[0531] Specific example of how to respond: "The customer is stressed. Please respond calmly."

[0532] This system allows store clerks to efficiently obtain customer information without having SQL skills and to respond optimally according to the customer's emotional state, significantly improving customer satisfaction and store clerk work efficiency.

[0533] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0534] Step 1:

[0535] The user uses the smart glasses to input customer requirements. At this time, the user can use voice input or touch operation. Specifically, the user can say, "I would like to know my purchase history for the past three months." The input data is acquired through the smart glasses' built-in microphone or touchpad. The input content is converted into text format and sent to the server.

[0536] Step 2:

[0537] The server analyzes the received user requirements using a generative AI model and natural language processing technology. The input data is in text format, and the generative AI model extracts keywords. In this process, keywords such as "past 3 months" and "purchase history" are identified.

[0538] Step 3:

[0539] The server selects an appropriate data mart based on the analysis results. The server refers to past consultation history and recommends a data mart that matches the analyzed keywords. For example, the "purchase history data mart" may be selected. This information is also returned to the user in text format.

[0540] Step 4:

[0541] The server automatically generates SQL queries using the selected data mart. The generative AI model creates appropriate queries based on keywords, for example, the following SQL queries are generated:

[0542] sql

[0543] SELECT customer_id, purchase_date, product_id, amount FROM purchase_history WHERE purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[0544] The generated query is presented to the user in text form.

[0545] Step 5:

[0546] The user checks the generated SQL query through the smart glasses and makes corrections if necessary. The user visually checks the query displayed on the screen, accepts it as is if there are no problems, and makes corrections by voice or touch if necessary. The corrected query is then sent back to the server.

[0547] Step 6:

[0548] The server executes the verified or modified SQL query and retrieves the results from the data warehouse. The results are then retrieved from the database and converted to text format, which is then sent back to the user.

[0549] Step 7:

[0550] At the same time, the server monitors the customer's emotional state through the smart glasses' camera, and an emotion recognition engine analyzes the image data in real time to determine the customer's state, such as "feeling stressed."

[0551] Step 8:

[0552] The server generates the optimal response based on the customer's emotional state and displays it on the smart glasses. The generative AI model then uses the emotional data to devise a response method. For example, a message such as "The customer is feeling stressed. Please respond calmly" is displayed.

[0553] Step 9:

[0554] The user checks the execution results and response methods displayed on the smart glasses display and responds appropriately to the customer. Based on the results, the necessary information is conveyed to the customer, and appropriate communication is carried out based on emotional data. This optimizes customer response.

[0555] 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.

[0556] 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.

[0557] 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.

[0558] [Second embodiment]

[0559] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0560] 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.

[0561] 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).

[0562] 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.

[0563] 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.

[0564] 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).

[0565] 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.

[0566] 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.

[0567] 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.

[0568] 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.

[0569] 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.

[0570] 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."

[0571] This invention is a system that streamlines the process for users of an in-house data warehouse (DWH) to select an appropriate data mart and create the necessary SQL queries. This system uses generative AI to analyze the user's requirements, select a data mart based on those requirements, and automatically generate SQL queries.

[0572] Overall system configuration

[0573] This system is centered around a generation AI running on a server and the user's device.

[0574] 1. Obtaining user requirements:

[0575] The user inputs the data analysis requirements from the terminal. For example, the requirement is "I want to obtain three months' worth of purchase history by customer."

[0576] 2. Requirements Analysis:

[0577] The server sends the user's requirements to the generation AI. The generation AI uses natural language processing technology to analyze the requirements and identify keywords and objectives. For example, it extracts "by customer," "purchase history," and "three months' worth" from the requirements.

[0578] 3. Data mart selection:

[0579] The generation AI refers to past consultation history and recommends the most appropriate data mart. For example, it recommends a "customer data mart" and a "purchase history data mart."

[0580] 4. Generate SQL queries:

[0581] The AI ​​automatically generates SQL queries based on the selected data mart. For example, it generates the following query:

[0582] sql

[0583] SELECT

[0584] customer_id,

[0585] purchase_date,

[0586] product_id,

[0587] amount

[0588] FROM

[0589] customer_data_mart

[0590] WHERE

[0591] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[0592] (The code is not listed here)

[0593] 5. Check and correct the query:

[0594] The server presents the generated SQL query to the user, who can review it and modify it if necessary (for example, if the user wants to add specific conditions to the query's output).

[0595] 6. Execute the query:

[0596] The server sends the query approved by the user to the DWH, executes it, and retrieves the execution results and displays them to the user.

[0597] Specific examples

[0598] For example, if a user enters a requirement such as "I want to get the purchase history for each customer for the past three months," it will work as follows:

[0599] 1. Obtain requirements:

[0600] The user enters "I would like to obtain three months' worth of purchase history by customer" into the input form on the terminal.

[0601] 2. Requirements analysis:

[0602] The server sends the requirements to the generation AI, which then extracts and analyzes the keywords "by customer," "purchase history," and "three months' worth."

[0603] 3. Data mart selection:

[0604] The generative AI recommends a "customer data mart" and a "purchase history data mart" based on similar past requirements. The server presents this information to the user.

[0605] 4. SQL query generation:

[0606] The generative AI automatically generates the following SQL query:

[0607] sql

[0608] SELECT

[0609] customer_id,

[0610] purchase_date,

[0611] product_id,

[0612] amount

[0613] FROM

[0614] customer_data_mart

[0615] WHERE

[0616] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[0617] (The code is not listed here)

[0618] 5. Check and correct the query:

[0619] The server presents the generated SQL query to the user, who then reviews it and, if necessary, modifies it.

[0620] 6. Query execution:

[0621] The server sends the modified query to the DWH, executes the query, retrieves the execution results, and displays them to the user.

[0622] This allows users to quickly obtain the data they need without having SQL skills. This system improves the efficiency of data analysis and greatly simplifies the work of users.

[0623] The processing flow will be explained below.

[0624] Step 1:

[0625] The user enters the requirements into the input form on the terminal. For example, the user enters the requirement "I want to obtain three months' worth of purchase history by customer."

[0626] Step 2:

[0627] The terminal transmits the input requirement data to the server.

[0628] Step 3:

[0629] The server sends the requirements data to the generation AI, which analyzes the requirements.

[0630] Step 4:

[0631] The generative AI uses natural language processing technology to extract and structure requirement keywords. For example, it identifies keywords such as "by customer," "purchase history," and "three months' worth."

[0632] Step 5:

[0633] The generation AI refers to past consultation history and selects the optimal data mart based on similar requirements. For example, it recommends a "customer data mart" and a "purchase history data mart."

[0634] Step 6:

[0635] The server receives the recommended data mart information from the generation AI and presents it to the user.

[0636] Step 7:

[0637] The user confirms the presented data mart information and makes a selection to proceed to the next step.

[0638] Step 8:

[0639] The AI ​​automatically generates the necessary SQL queries based on the selected data mart. For example, it generates the following query:

[0640] sql

[0641] SELECT

[0642] customer_id,

[0643] purchase_date,

[0644] product_id,

[0645] amount

[0646] FROM

[0647] customer_data_mart

[0648] WHERE

[0649] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[0650] (The code is not listed here)

[0651] Step 9:

[0652] The server presents the generated SQL query to the user.

[0653] Step 10:

[0654] The user reviews the proposed SQL query and makes corrections if necessary.

[0655] Step 11:

[0656] The user submits the modified query to the server.

[0657] Step 12:

[0658] The server sends the confirmed or modified query to the DWH for execution.

[0659] Step 13:

[0660] The server retrieves the query results from the DWH.

[0661] Step 14:

[0662] The server presents the obtained results to the user.

[0663] These steps allow users to efficiently use the DWH to quickly obtain the data they need. Automating the entire system process improves the efficiency of data analysis work and ensures accurate data extraction.

[0664] Example 1

[0665] 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."

[0666] In today's business environment, companies are required to analyze data stored in their internal data warehouses (DWHs) appropriately and quickly obtain the necessary information. However, data processing and SQL query creation require highly specialized knowledge, which reduces business efficiency. In particular, it is difficult for non-expert users to effectively utilize data.

[0667] 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.

[0668] In this invention, the server includes means for analyzing input requirements using a generation AI, means for selecting an appropriate data structure based on the analyzed requirements, means for automatically generating a data extraction query based on the selected data structure, means for presenting the generated data extraction query to a user and accepting confirmation and correction by the user, and means for executing the confirmed or corrected data extraction query and presenting the execution result to the user. This enables users to efficiently and quickly obtain required data without requiring advanced specialized knowledge.

[0669] "Generative AI" is an artificial intelligence technology that analyzes input requirements and automatically generates appropriate data structures and queries.

[0670] "Means" refers to a component of a system or technology that accomplishes a specific function or operation.

[0671] "Data structure" refers to the format or schema of information stored in a database or data warehouse.

[0672] A "data extraction query" is program code, such as SQL, used to retrieve specific information from a database.

[0673] "User" refers to a person who uses this system to analyze data and create queries.

[0674] "Natural language processing technology" is a technology that enables computers to understand, analyze, and generate language that humans use on a daily basis.

[0675] "Past response history" refers to data that records requirements that have been previously processed and the results of those processing.

[0676] A "data warehouse (DWH)" is a type of database system that centrally stores large amounts of data and is used for analysis and reporting.

[0677] This invention is a system that streamlines the process by which users of an in-house data warehouse (DWH) can select an appropriate data structure and automatically generate the necessary data extraction queries. This system uses a generative AI model to analyze user requirements, select a data structure based on the analysis, and automatically generate data extraction queries.

[0678] This system is centered around a generative AI model running on a server and the user's device.

[0679] To obtain user requirements, the user inputs the requirements for data analysis into an input form on the terminal. For example, the user might input a requirement such as "I want to obtain three months' worth of purchase history by customer." The terminal then sends this requirement to the server.

[0680] Next, the requirements are analyzed. The server receives the requirements data sent by the user and sends it to the generative AI model as a prompt. This prompt takes the form, "I would like to obtain the past three months' worth of purchase history by customer." The generative AI model uses natural language processing technology to analyze the requirements and extract keywords and objectives. For example, it extracts keywords such as "by customer," "purchase history," and "three months' worth."

[0681] Next, the data structure is selected. The generative AI model refers to the history of similar requirements processing in the past and returns a list of optimal data structures to the server. For example, it selects a "customer data structure" and a "purchase history data structure." The server then presents this information to the user.

[0682] Next, SQL queries are generated. The server uses a generative AI model to automatically generate data extraction queries based on the selected data structure. The generated queries may have the following format:

[0683] sql

[0684] SELECT

[0685] customer_id,

[0686] purchase_date,

[0687] product_id,

[0688] amount

[0689] FROM

[0690] customer_data_structure

[0691] WHERE

[0692] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[0693] The generated data extraction query is presented to the user by the server.

[0694] The user can review the query on their device screen and modify it as needed, for example to narrow it down to a specific customer ID.

[0695] Finally, the server sends the query modified and approved by the user to the DWH and executes it. The server retrieves the execution results and displays them to the user. This allows users to efficiently and quickly obtain the data they need, even without specialized knowledge of SQL.

[0696] An example of a prompt would be something like, "The user has entered the requirement, 'I want to retrieve the purchase history for each customer for the past three months.' Please select the optimal data structure based on this requirement and generate the necessary data extraction query."

[0697] Through this system, users can automatically generate the SQL queries required for data analysis, streamlining the data extraction process, enabling fast and accurate data analysis in the field of business, significantly improving work efficiency.

[0698] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0699] Step 1: Obtain user requirements

[0700] Specific behavior:

[0701] The user enters the requirements for data analysis into the input form on the terminal. For example, the user might enter, "I want to obtain three months' worth of purchase history by customer."

[0702] The terminal receives this input data and sends it to the server.

[0703] Input: Requirements entered by the user into the terminal

[0704] Output: Requirement data sent to the server

[0705] Step 2: Analyze requirements

[0706] Specific behavior:

[0707] The server receives the requirements data sent by the user.

[0708] The server creates a prompt for the generative AI model and sends it the requirements. An example of a prompt would be "I would like to obtain the purchase history for each customer for the past three months."

[0709] The generative AI model uses natural language processing technology to analyze requirements and extract keywords and objectives, such as "by customer," "purchase history," and "three months' worth."

[0710] Input: Requirement data submitted by the user

[0711] Output: Keywords and analysis results

[0712] Step 3: Choose a data structure

[0713] Specific behavior:

[0714] Based on the analysis results, the generative AI model refers to the history of similar requirements processing in the past.

[0715] The server receives a list of optimal data structures (data marts) from the generative AI model. For example, it selects a "customer data structure" and a "purchase history data structure."

[0716] The server presents this information to the user.

[0717] Input: Analysis results, past requirements processing history

[0718] Output: A list of selected data structures

[0719] Step 4: Generate SQL queries

[0720] Specific behavior:

[0721] The server uses the generative AI model to automatically generate SQL queries based on the selected data structure.

[0722] The generated query may have the following format:

[0723] sql

[0724] SELECT

[0725] customer_id,

[0726] purchase_date,

[0727] product_id,

[0728] amount

[0729] FROM

[0730] customer_data_structure

[0731] WHERE

[0732] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[0733] The server presents the generated data extraction query to the user.

[0734] Input: A list of selected data structures

[0735] Output: Auto-generated SQL query

[0736] Step 5: Check and fix your queries

[0737] Specific behavior:

[0738] The user sees the generated SQL query on the terminal screen.

[0739] The user modifies the query as needed, for example to narrow it down to a specific customer ID.

[0740] The modified SQL query is then sent to the server again.

[0741] Input: Generated SQL query

[0742] Output: SQL query confirmed or modified by the user

[0743] Step 6: Execute the query

[0744] Specific behavior:

[0745] The server sends the SQL query, which has been modified and approved by the user, to the DWH.

[0746] The server receives the execution results from the DWH and displays them to the user.

[0747] As a result of the execution, for example, a list of the customer's purchase history is displayed on the terminal.

[0748] Input: Confirmed or revised SQL query

[0749] Output: Execution results obtained from the DWH

[0750] (Application example 1)

[0751] 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."

[0752] Modern logistics centers are required to efficiently manage large volumes of shipping records and inventory data and quickly retrieve the necessary data. In particular, there is a growing need for systems that allow users to easily analyze data, even if they do not have specific SQL skills. However, conventional systems often require users to be familiar with complex database structures, which takes time to learn. It is also difficult to quickly select the appropriate data source. Therefore, there is a need to develop a system that can perform data analysis efficiently.

[0753] 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.

[0754] In this invention, the server includes means for analyzing input requirements using a generation AI, means for selecting an appropriate data mart based on the analyzed requirements, means for automatically generating an SQL query based on the selected data mart, means for presenting the generated SQL query to a user and accepting confirmation and correction by the user, means for executing the confirmed or corrected SQL query and presenting the execution result to the user, and means for automatically acquiring past shipping records and inventory data, specialized for data analysis of a logistics center. This enables users to efficiently analyze data and quickly acquire appropriate data even if they do not have specific SQL skills.

[0755] "Generative AI" is an artificial intelligence technology that analyzes user requirements, selects appropriate data marts, and automatically generates SQL queries.

[0756] A "data mart" is an aggregation of data sets extracted from a large data warehouse to address a specific business need.

[0757] An "SQL query" is a command in the Structured Query Language that describes an instruction to a database.

[0758] A "logistics center" is a facility that receives, stores, organizes, and ships goods and materials.

[0759] "Past shipping records" are historical information on products that have been shipped from the logistics center in the past.

[0760] "Inventory data" is information about the quantity and condition of products currently stored in the logistics center.

[0761] "Natural language processing technology" is a technology that enables computers to understand, analyze, and generate human language.

[0762] "Verify and fix" is the process whereby the user reviews the auto-generated SQL queries and makes changes if necessary.

[0763] This invention relates to a system that utilizes generative AI to improve the efficiency of data analysis in logistics centers. This system is centered around a server and user terminals.

[0764] First, the user inputs the requirements for data analysis from a terminal. For example, they might input a requirement such as "I want to obtain shipping records for each warehouse for the last three months." The server receives the user's requirements and sends them to the generation AI. This generation AI uses natural language processing technology to analyze the requirements and extract the keywords "warehouse," "shipping records," and "three months."

[0765] The generative AI then selects appropriate data marts based on these keywords, recommending, for example, "warehouse data mart" and "shipping record data mart," and the server presents this information to the user.

[0766] The Generative AI then automatically generates SQL queries based on the selected data mart. For example, the following SQL queries are generated:

[0767] SELECT

[0768] warehouse_id,

[0769] shipment_date,

[0770] product_id,

[0771] quantity

[0772] FROM

[0773] warehouse_data_mart

[0774] WHERE

[0775] shipment_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[0776] The server presents the generated SQL query to the user and accepts confirmation and modification by the user. The user confirms the query and makes modifications as necessary. The server then executes the confirmed or modified SQL query and presents the execution results to the user.

[0777] Specifically, it is implemented as shown in the following Python code, which uses an SQLite database and executes queries according to the user's requirements. The generative AI uses OpenAI's GPT model.

[0778] This allows the server to efficiently analyze data and quickly retrieve appropriate data, even if the user does not have specific SQL skills. This makes it easier for distribution centers to quickly check shipping records and manage inventory.

[0779] Examples of prompt statements

[0780] Prompt for the generative AI model:

[0781] Analyze the following requirements: Get shipping records for each warehouse for the last three months

[0782] As described above, the present invention provides a concrete example of a system that simplifies data management at a logistics center and enables efficient data analysis. This system allows users to quickly obtain the data they need without having to worry about complex database operations.

[0783] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0784] Step 1:

[0785] The user inputs the requirements for data analysis from the terminal. For example, they input a requirement such as "I want to obtain the shipping records for each warehouse for the last three months." This requirement is sent to the server.

[0786] Input: The requirement text entered by the user

[0787] Output: The text of the requirement as received by the server

[0788] Step 2:

[0789] The server sends the received requirements to the generation AI, which then analyzes them using natural language processing technology. During the analysis, keywords such as "warehouse," "shipping record," and "3 months" are extracted from the requirements.

[0790] Input: Text of the requirement received from the user

[0791] Output: Generated keyword list (e.g. "warehouse", "shipping record", "3 months")

[0792] Step 3:

[0793] Based on the extracted keywords, the generation AI refers to past consultation history and selects the appropriate data mart. For example, it might recommend a "warehouse data mart" and a "shipping record data mart."

[0794] Input: Keyword list, past consultation history data

[0795] Output: A list of recommended data marts (e.g., "Warehouse Data Mart" and "Shipping Record Data Mart")

[0796] Step 4:

[0797] The server automatically generates SQL queries based on the selected data mart using the AI. The generated SQL queries are, for example, as follows:

[0798] SELECT

[0799] warehouse_id,

[0800] shipment_date,

[0801] product_id,

[0802] quantity

[0803] FROM

[0804] warehouse_data_mart

[0805] WHERE

[0806] shipment_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[0807] Input: Recommended data mart list, keyword list

[0808] Output: Auto-generated SQL query

[0809] Step 5:

[0810] The server presents the generated SQL query to the user, who can review it and make modifications if necessary. The user can review the query and add or modify specific conditions.

[0811] Input: Auto-generated SQL query

[0812] Output: SQL query reviewed and corrected by the user

[0813] Step 6:

[0814] The server executes the verified or modified SQL query and retrieves the results, for example, by querying an SQLite database and retrieving the results.

[0815] Input: Confirmed or revised SQL query

[0816] Output: Query results (data retrieved from the database)

[0817] Step 7:

[0818] The server then presents the results of the query to the user, allowing the user to quickly obtain the data they need.

[0819] Input: Query results

[0820] Output: Execution results presented to the user (e.g., shipping records by warehouse for the past three months)

[0821] 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.

[0822] This invention is a system that enables users of an in-house data warehouse (DWH) to efficiently extract data and perform analytical tasks, and aims to improve the user experience by combining it with an emotion engine that recognizes the user's emotions.

[0823] Overall system configuration

[0824] The system includes the following main components:

[0825] 1. Generation AI

[0826] 2. Natural Language Processing Technology

[0827] 3. Emotion Engine

[0828] 4. Server

[0829] 5. User Device

[0830] Acquiring and analyzing user requirements

[0831] 1. Obtaining user requirements

[0832] The user enters requirements into an input form on the terminal. For example, the user enters a requirement such as "I want to obtain three months' worth of purchase history by customer."

[0833] 2. Emotion recognition

[0834] While the terminal is inputting user requirements, the emotion engine monitors the user's emotional state. For example, if the user is feeling stressed, the emotion engine will recognize this.

[0835] 3. Submit your requirements

[0836] The terminal transmits the input requirement data and the user's emotion data to the server.

[0837] Requirements and sentiment data analysis

[0838] 4. Requirements Analysis

[0839] The server sends the requirements data to the generation AI, which then uses natural language processing technology to extract and structure keywords from the requirements. For example, it identifies keywords such as "by customer," "purchase history," and "three months' worth."

[0840] 5. Emotion Data Analysis

[0841] The server analyzes the user's emotional state based on the data from the emotion engine and determines whether the user is feeling stressed.

[0842] Data mart selection and query generation

[0843] 6. Data Mart Selection

[0844] The generative AI refers to past consultation history and selects the optimal data mart based on the analyzed requirements and emotional state. For example, it recommends a "customer data mart" and a "purchase history data mart."

[0845] 7. Presentation to the User

[0846] The server receives the recommended data mart information from the generation AI and displays it on the device. It also adjusts the interface and provides additional support content according to the user's emotional state.

[0847] 8. Generating SQL Queries

[0848] The AI ​​automatically generates the necessary SQL queries based on the selected data mart. For example, it generates the following query:

[0849] sql

[0850] SELECT

[0851] customer_id,

[0852] purchase_date,

[0853] product_id,

[0854] amount

[0855] FROM

[0856] customer_data_mart

[0857] WHERE

[0858] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[0859] (The code is not listed here)

[0860] Review and run the query

[0861] 9. Check and correct your query

[0862] The server presents the generated SQL query to the user, who confirms it and modifies it if necessary. If the user is feeling nervous or stressed, the emotion engine displays an appropriate support message.

[0863] 10. Executing a Query

[0864] The server sends the confirmed or modified query to the DWH for execution.

[0865] Displaying execution results

[0866] 11. Obtaining and Presenting Results

[0867] The server retrieves the query results from the DWH and displays them on the device. It monitors the user's emotional state and interprets the results and suggests the next action if necessary.

[0868] Specific examples

[0869] For example, if a user enters the requirement "I want to get the purchase history for each customer for the past three months":

[0870] 1. Requirements acquisition: The user enters the requirements into an input form on the terminal.

[0871] 2. Emotion Recognition: The emotion engine recognizes the user's emotions (e.g., stress state).

[0872] 3. Sending requirements and emotion data: The device sends the data to the server.

[0873] 4. Requirements analysis: The server sends the requirements data to the generation AI for analysis.

[0874] 5. Emotional data analysis: Analyze user emotions.

[0875] 6. Data mart selection: Generative AI recommends the optimal data mart.

[0876] 7. Presentation to the user: The server presents emotional support along with data mart information.

[0877] 8. SQL Query Generation: Generative AI generates SQL queries.

[0878] 9. Review and modify the query: The user reviews the query and modifies it if necessary.

[0879] 10. Query execution: The server sends the query to the DWH and executes it.

[0880] 11. Retrieving and presenting results: The server retrieves the results and presents them to the user.

[0881] This system allows users to efficiently retrieve data and perform analysis work without stress, even if they do not have SQL skills.The emotion engine monitors the user's emotional state and provides appropriate support, resulting in a better user experience.

[0882] The processing flow will be explained below.

[0883] Step 1:

[0884] The user enters the requirements into the input form on the terminal. For example, the user enters "I want to obtain three months' worth of purchase history by customer."

[0885] Step 2:

[0886] The emotion engine monitors the user's emotional state, for example, analyzing the user's stress level from facial expressions and tone of voice via the camera and microphone while the user is typing.

[0887] Step 3:

[0888] The terminal transmits the input requirement data and emotion data to the server.

[0889] Step 4:

[0890] The server sends the requirement data and emotion data to the generation AI, which receives the requirement data and begins analyzing it.

[0891] Step 5:

[0892] Generative AI uses natural language processing technology to analyze requirement data and identify and extract keywords and objectives. For example, it identifies "by customer," "purchase history," and "three months' worth."

[0893] Step 6:

[0894] The server receives the analysis results from the generation AI and stores them along with the emotion data.

[0895] Step 7:

[0896] The generation AI refers to past consultation history and selects the appropriate data mart based on the analyzed requirements. For example, it recommends a "customer data mart" and a "purchase history data mart."

[0897] Step 8:

[0898] The server receives the recommendation data mart information from the generation AI and displays it on the device, and also adjusts the interface based on the user's emotional state and displays necessary support messages.

[0899] Step 9:

[0900] The user confirms the presented data mart information and makes a selection to proceed to the next step.

[0901] Step 10:

[0902] The AI ​​automatically generates the necessary SQL queries based on the selected data mart. For example, it generates the following query:

[0903] sql

[0904] SELECT

[0905] customer_id,

[0906] purchase_date,

[0907] product_id,

[0908] amount

[0909] FROM

[0910] customer_data_mart

[0911] WHERE

[0912] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[0913] (The code is not listed here)

[0914] Step 11:

[0915] The server presents the generated SQL query to the user.

[0916] Step 12:

[0917] The user checks the presented SQL query and modifies it as necessary. For example, if they want to extract only a specific product category, they can add that condition.

[0918] Step 13:

[0919] The user submits the modified query to the server.

[0920] Step 14:

[0921] The server sends the confirmed or modified query to the DWH and executes the query.

[0922] Step 15:

[0923] The server retrieves the query results from the DWH.

[0924] Step 16:

[0925] The server sends the results to the device and displays them to the user. It also monitors the user's emotional state again and displays a message to guide the user on the next action based on the results.

[0926] These steps allow users to efficiently extract data and perform data analysis.The emotion engine monitors the user's emotional state in real time and takes appropriate action to provide a better user experience.

[0927] Example 2

[0928] 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."

[0929] Conventional in-house data warehouse (DWH) systems have the drawback of making it difficult for users to efficiently extract and analyze data. Furthermore, simple data processing is often performed without considering the user's emotional state, resulting in a poor user experience. The present invention aims to solve these problems and provide a system that enables users to efficiently extract and analyze data without feeling stressed.

[0930] 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.

[0931] In this invention, the server includes: means for a user to input requirements into an input form on a terminal; means for an emotion engine to monitor the user's emotional state while the user is inputting the user's requirements; means for transmitting the input requirement data and the user's emotional data to the server; means for transmitting the requirement data to a generation AI and extracting and structuring keywords of the requirements using natural language processing technology; means for analyzing the user's emotional state based on the data from the emotion engine and determining whether the user is feeling stressed; means for selecting an optimal data mart based on the analyzed requirements and emotional state by referring to past consultation history; means for receiving recommended data mart information from the generation AI and presenting interface adjustments or additional support content according to the user's emotional state; means for automatically generating a required SQL query based on the selected data mart; means for presenting the generated SQL query to the user and accepting user confirmation and correction; and means for executing the confirmed or corrected SQL query and presenting the execution result to the user. This allows the user to efficiently extract and analyze data, and further enables the emotion engine to improve the user experience.

[0932] A "terminal" is an input / output device that a user uses to input requirements.

[0933] An "emotion engine" is a software and hardware system that monitors and analyzes a user's emotional state in real time.

[0934] "Requirement data" refers to specific requests for data extraction and analysis that are input by the user into the terminal.

[0935] "Emotion data" is data that represents the user's emotional state and is acquired by the emotion engine.

[0936] A "server" is a computer system that receives user requirement data and emotion data and performs various processes based on them.

[0937] "Generative AI" is an artificial intelligence that analyzes requirements data and extracts and structures keywords using natural language processing technology.

[0938] "Natural language processing technology" is information technology used to interpret and understand the meaning of human language.

[0939] A "data mart" is a portion of a database that stores a subset of data related to a particular business process or area.

[0940] An "SQL query" is a command statement written in the Structured Query Language (SQL) for searching and manipulating data in a database management system (DBMS).

[0941] The "past consultation history" is a record of requirements that the user has exchanged with the server in the past and the results of the processing.

[0942] "Adjusting the interface" refers to changing the screen display or user interface configuration in response to the user's emotional state.

[0943] "Additional support content" refers to supplementary information or messages provided depending on the user's emotional state.

[0944] This invention is a system that aims to enable users of in-house data warehouses (DWHs) to efficiently extract data and perform analysis work. Furthermore, by combining it with an emotion engine that recognizes user emotions, we aim to improve the user experience.

[0945] Overall system configuration

[0946] The system includes the following major components:

[0947] 1. Generation AI

[0948] 2. Natural Language Processing Technology

[0949] 3. Emotion Engine

[0950] 4. Server

[0951] 5. User Device

[0952] Hardware and Software

[0953] Terminal: An input device for users to enter requirements. It uses a web browser or GUI form.

[0954] Emotion engine: Uses sensors and software to monitor the user's facial expressions and typing speed in real time and analyze their emotional state. Specific examples include camera sensors and keyboard typing speed analysis software.

[0955] Server: A computer system that receives requirements data and emotion data and analyzes and structures the data using generative AI models and natural language processing techniques.

[0956] Generative AI: An artificial intelligence that analyzes user requirements data and extracts and structures keywords using natural language processing technology.

[0957] Data mart: A portion of a database that stores a subset of data related to a particular business process or area.

[0958] System Operation

[0959] When a user enters requirements into an input form on a terminal, for example, "I would like to obtain the past three months' worth of purchase history by customer," the emotion engine monitors the user's facial expressions and input speed and analyzes their emotional state. The terminal then sends the entered requirement data and emotion data to the server.

[0960] The server sends the received requirements data to the generation AI, which uses natural language processing technology to extract and structure keywords. Based on the extracted keywords and taking into account data from the emotion engine, the generation AI determines whether the user is feeling stressed. The generation AI refers to past consultation history and selects the optimal data mart based on the analyzed requirements and emotional state. The server receives the recommended data mart information and displays it on the device. At this time, it adjusts the interface and presents additional support content according to the user's emotional state.

[0961] The generation AI automatically generates the necessary SQL queries based on the selected data mart. The generated SQL queries are presented to the user, who can confirm and modify them. Finally, the confirmed or modified SQL queries are sent from the server to the DWH, and data is retrieved. The retrieved data is displayed on the terminal, and the user can confirm the results. At this time, the emotion engine displays support messages as needed.

[0962] Specific examples

[0963] 1. Obtaining requirements: The user enters the requirement "I want to obtain the purchase history for each customer for the past three months" into the input form on the terminal.

[0964] 2. Emotion Recognition: The emotion engine monitors the user's facial expressions and input speed to recognize stress levels.

[0965] 3. Sending requirements and emotion data: The terminal sends the requirements data and emotion data to the server.

[0966] 4. Requirements analysis: The generative AI extracts the keywords "by customer," "purchase history," and "three months' worth."

[0967] 5. Emotional data analysis: Determine whether the user is feeling stressed.

[0968] 6. Data mart selection: The generative AI selects the optimal data mart (e.g., "customer data mart" or "purchase history data mart").

[0969] 7. Presentation to the user: The recommended data mart and support content according to the emotional state are displayed on the terminal.

[0970] 8. SQL Query Generation: The generative AI automatically generates SQL queries like the following:

[0971] sql

[0972] SELECT

[0973] customer_id,

[0974] purchase_date,

[0975] product_id,

[0976] amount

[0977] FROM

[0978] customer_data_mart

[0979] WHERE

[0980] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[0981] 9. Review and modify the query: The user reviews the query and modifies it if necessary.

[0982] 10. Query execution: The server sends the confirmed or modified query to the DWH and executes it.

[0983] 11. Obtaining and presenting results: The server displays the obtained data on the terminal, and the user confirms the results.

[0984] Prompt Sentence Examples

[0985] Generate a SQL query to retrieve customer purchase history for the past three months.

[0986] This system allows users to efficiently retrieve data and perform analysis work without stress, even if they do not have SQL skills.The emotion engine monitors the user's emotional state and provides appropriate support, resulting in a better user experience.

[0987] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0988] Step 1:

[0989] The user inputs requirements into an input form on the terminal. Specifically, the user inputs the requirement "I want to obtain the purchase history for each customer for the past three months" into the input form. At this time, the terminal receives the information entered by the user as input data. The input data is requirement information in text format.

[0990] Step 2:

[0991] The emotion engine monitors the user's emotional state while they are entering their requirements. Specifically, a camera sensor analyzes the user's facial expressions and keyboard input speed in real time to generate emotion data. The input is the user's facial image and input speed data, and the output is emotional state data in text format (e.g., stress level).

[0992] Step 3:

[0993] The device sends the input requirement data and user emotion data to the server. The input is the user requirement data and emotion data, and the output is the requirement data and emotion data sent to the server. Specifically, it is sent to the server as JSON format data.

[0994] Step 4:

[0995] The server sends the requirements data to the generation AI, which uses natural language processing technology to extract and structure requirement keywords. The input is the requirements data (e.g., "I want to obtain the purchase history by customer for the past three months"), and the output is the extracted keywords (e.g., "by customer," "purchase history," "three months' worth"). Specifically, the generation AI uses a natural language processing engine to analyze and structure the keywords.

[0996] Step 5:

[0997] The server analyzes the user's emotional state based on data from the emotion engine and determines whether the user is feeling stressed. The input is emotional data (e.g., stress state), and the output is a judgment result of the user's emotional state (e.g., the user is feeling stressed). Specifically, the server runs an emotion analysis algorithm and quantifies the emotional state.

[0998] Step 6:

[0999] The generation AI refers to past consultation history and selects the optimal data mart based on the analyzed requirements and emotional state. The input is requirement keywords and emotional state data, and the output is a recommended data mart (e.g., "customer data mart" or "purchase history data mart"). Specifically, the generation AI runs a data mart selection algorithm to select the optimal data mart.

[1000] Step 7:

[1001] The server receives the recommendation data mart information from the generation AI and displays it on the device. Furthermore, it adjusts the interface and presents additional support content according to the user's emotional state. The input is the recommendation data mart and the user's emotional state, and the output is the information displayed on the device. Specifically, the server receives the recommendation data mart information and adjusts the UI according to the emotional state.

[1002] Step 8:

[1003] The generation AI automatically generates the required SQL queries based on the selected data mart. The input is the selected data mart information and requirement keywords, and the output is the generated SQL query. Specifically, the generation AI automatically generates the following SQL queries:

[1004] sql

[1005] SELECT

[1006] customer_id,

[1007] purchase_date,

[1008] product_id,

[1009] amount

[1010] FROM

[1011] customer_data_mart

[1012] WHERE

[1013] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[1014] Step 9:

[1015] The server presents the generated SQL query to the user, who confirms it and modifies it if necessary. The input is the generated SQL query, and the output is the SQL query that has been confirmed and modified by the user. Specifically, the server displays the SQL query on the terminal, and the user confirms and modifies the query content.

[1016] Step 10:

[1017] The server sends the confirmed or modified SQL query to the DWH and executes it. The input is the confirmed or modified SQL query, and the output is the query execution result from the DWH. In concrete terms, the server sends the SQL query to the data warehouse, executes the query, and retrieves the data.

[1018] Step 11:

[1019] The server retrieves the query execution results from the DWH and displays them on the terminal. It monitors the user's emotional state and, if necessary, interprets the results and suggests the next action. The input is the query execution results and the user's emotional state, and the output is the execution results and a support message displayed on the terminal. In concrete terms, the server retrieves the query results, and if the user is feeling stressed, it displays an interpretation support message.

[1020] (Application example 2)

[1021] 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."

[1022] With conventional data analysis systems, if users lacked SQL skills or database knowledge, it was difficult to efficiently extract and analyze data. Furthermore, because data was provided without taking into account the user's emotional state, the user interface and support were often inappropriate. This resulted in a poor user experience and reduced productivity, especially when tasks were complex or under stress.

[1023] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for analyzing input requirements using a generation AI, means for selecting an appropriate data mart based on the analyzed requirements, means for automatically generating an SQL query based on the selected data mart, means for presenting the generated SQL query to a user and accepting confirmation and correction by the user, means for executing the confirmed or corrected SQL query and presenting the execution result to the user, an emotion recognition engine for monitoring the user's emotional state, and means for generating and presenting a response method based on the emotional state. This enables users to efficiently extract and analyze data without advanced technical knowledge, and furthermore, by providing support appropriate to the user's emotional state, it is possible to reduce stress and improve the user experience.

[1024] "Generative AI" is an artificial intelligence technique used to analyze data and automate tasks.

[1025] The "means for analyzing requirements" is a method for analyzing the requirements data entered by the user and extracting important keywords and intentions.

[1026] The "means for selecting an appropriate data mart" is a method for selecting the most relevant data mart based on the analyzed requirements.

[1027] "Means for automatically generating SQL queries" refers to a method for automatically creating the necessary SQL queries for the selected data mart.

[1028] The "means for presenting the generated SQL query to the user and accepting confirmation and correction by the user" refers to a method for displaying the generated SQL query to the user so that the user can confirm or correct it.

[1029] "Means for executing a confirmed or modified SQL query and presenting the execution results to a user" refers to a method for executing an SQL query confirmed or modified by a user and displaying the results to the user.

[1030] The "emotion recognition engine that monitors the user's emotional state" is a system that analyzes the user's emotions in real time and determines their state.

[1031] The "means for generating and presenting a response method based on an emotional state" is a method for generating an optimal response method according to the emotional state of the user determined by an emotion recognition engine and presenting it to the user.

[1032] As an embodiment of the present invention, a customer service support system using smart glasses for store clerks in a brick-and-mortar store can be adopted. The following describes the specific roles of the server, terminal, and user.

[1033] server

[1034] The server performs the following operations using the following specific hardware and software:

[1035] 1. Generative AI and Natural Language Processing Technology

[1036] Software used: Python language, natural language processing library (e.g., SpaCy)

[1037] The server acquires the user's requirements and uses generation AI to extract and analyze keywords from the requirements.

[1038] For example, from the requirement "Please tell me your purchase history for the past three months," extract keywords such as "past three months" and "purchase history."

[1039] 2. Emotion Recognition Engine

[1040] Hardware used: Smart glasses camera

[1041] Software used: Emotion-FerPlus

[1042] The server analyzes the video data acquired through the glasses' camera and monitors the customer's emotional state in real time.

[1043] For example, determining whether a customer is stressed.

[1044] 3. Data mart selection and SQL query generation

[1045] Software used: SQLAlchemy

[1046] The server selects a data mart based on the analyzed requirements and automatically generates the necessary SQL queries.

[1047] For example, generate a query to extract data from a purchase history data mart within a specific period.

[1048] Terminal (smart glasses)

[1049] The smart glasses worn by the store clerk function as a terminal and support the following processes:

[1050] 1. Input customer requirements

[1051] Enter customer requirements using voice input or touch operation.

[1052] For example, a request such as "I want to know the purchase history for the past three months" is input.

[1053] 2. Displaying emotional states

[1054] The analysis results of the emotion recognition engine are displayed in real time.

[1055] For example, if a customer is feeling stressed, the glasses' display will say "Customer is feeling stressed."

[1056] 3. Displaying query results and suggesting how to respond

[1057] The data returned from the server and the response method based on the emotional state are displayed on the glasses' display.

[1058] For example, along with the purchase history as a query result, the message "The customer is stressed. Please respond calmly" may be displayed.

[1059] User (store clerk)

[1060] Store staff use the above system to efficiently handle customer service.

[1061] 1. Obtaining and inputting user requirements

[1062] Customer requirements are registered into the smart glasses using voice or touch input.

[1063] An example of a specific prompt might be: "The customer wants purchase history data for the past three months. Please generate a query."

[1064] 2. Checking and responding to emotional states

[1065] Check the customer's emotional state in real time and respond accordingly.

[1066] Specific example of how to respond: "The customer is stressed. Please respond calmly."

[1067] This system allows store clerks to efficiently obtain customer information without having SQL skills and to respond optimally according to the customer's emotional state, significantly improving customer satisfaction and store clerk work efficiency.

[1068] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1069] Step 1:

[1070] The user uses the smart glasses to input customer requirements. At this time, the user can use voice input or touch operation. Specifically, the user can say, "I would like to know my purchase history for the past three months." The input data is acquired through the smart glasses' built-in microphone or touchpad. The input content is converted into text format and sent to the server.

[1071] Step 2:

[1072] The server analyzes the received user requirements using a generative AI model and natural language processing technology. The input data is in text format, and the generative AI model extracts keywords. In this process, keywords such as "past 3 months" and "purchase history" are identified.

[1073] Step 3:

[1074] The server selects an appropriate data mart based on the analysis results. The server refers to past consultation history and recommends a data mart that matches the analyzed keywords. For example, the "purchase history data mart" may be selected. This information is also returned to the user in text format.

[1075] Step 4:

[1076] The server automatically generates SQL queries using the selected data mart. The generative AI model creates appropriate queries based on keywords, for example, the following SQL queries are generated:

[1077] sql

[1078] SELECT customer_id, purchase_date, product_id, amount FROM purchase_history WHERE purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[1079] The generated query is presented to the user in text form.

[1080] Step 5:

[1081] The user checks the generated SQL query through the smart glasses and makes corrections if necessary. The user visually checks the query displayed on the screen, accepts it as is if there are no problems, and makes corrections by voice or touch if necessary. The corrected query is then sent back to the server.

[1082] Step 6:

[1083] The server executes the verified or modified SQL query and retrieves the results from the data warehouse. The results are then retrieved from the database and converted to text format, which is then sent back to the user.

[1084] Step 7:

[1085] At the same time, the server monitors the customer's emotional state through the smart glasses' camera, and an emotion recognition engine analyzes the image data in real time to determine the customer's state, such as "feeling stressed."

[1086] Step 8:

[1087] The server generates the optimal response based on the customer's emotional state and displays it on the smart glasses. The generative AI model then uses the emotional data to devise a response method. For example, a message such as "The customer is feeling stressed. Please respond calmly" is displayed.

[1088] Step 9:

[1089] The user checks the execution results and response methods displayed on the smart glasses display and responds appropriately to the customer. Based on the results, the necessary information is conveyed to the customer, and appropriate communication is carried out based on emotional data. This optimizes customer response.

[1090] 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.

[1091] 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.

[1092] 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.

[1093] [Third embodiment]

[1094] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1095] 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.

[1096] 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).

[1097] 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.

[1098] 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.

[1099] 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).

[1100] 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.

[1101] 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.

[1102] 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.

[1103] 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.

[1104] 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.

[1105] 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."

[1106] This invention is a system that streamlines the process for users of an in-house data warehouse (DWH) to select an appropriate data mart and create the necessary SQL queries. This system uses generative AI to analyze the user's requirements, select a data mart based on those requirements, and automatically generate SQL queries.

[1107] Overall system configuration

[1108] This system is centered around a generation AI running on a server and the user's device.

[1109] 1. Obtaining user requirements:

[1110] The user inputs the data analysis requirements from the terminal. For example, the requirement is "I want to obtain three months' worth of purchase history by customer."

[1111] 2. Requirements Analysis:

[1112] The server sends the user's requirements to the generation AI. The generation AI uses natural language processing technology to analyze the requirements and identify keywords and objectives. For example, it extracts "by customer," "purchase history," and "three months' worth" from the requirements.

[1113] 3. Data mart selection:

[1114] The generation AI refers to past consultation history and recommends the most appropriate data mart. For example, it recommends a "customer data mart" and a "purchase history data mart."

[1115] 4. Generate SQL queries:

[1116] The AI ​​automatically generates SQL queries based on the selected data mart. For example, it generates the following query:

[1117] sql

[1118] SELECT

[1119] customer_id,

[1120] purchase_date,

[1121] product_id,

[1122] amount

[1123] FROM

[1124] customer_data_mart

[1125] WHERE

[1126] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[1127] (The code is not listed here)

[1128] 5. Check and correct the query:

[1129] The server presents the generated SQL query to the user, who can review it and modify it if necessary (for example, if the user wants to add specific conditions to the query's output).

[1130] 6. Execute the query:

[1131] The server sends the query approved by the user to the DWH, executes it, and retrieves the execution results and displays them to the user.

[1132] Specific examples

[1133] For example, if a user enters a requirement such as "I want to get the purchase history for each customer for the past three months," it will work as follows:

[1134] 1. Obtain requirements:

[1135] The user enters "I would like to obtain three months' worth of purchase history by customer" into the input form on the terminal.

[1136] 2. Requirements analysis:

[1137] The server sends the requirements to the generation AI, which then extracts and analyzes the keywords "by customer," "purchase history," and "three months' worth."

[1138] 3. Data mart selection:

[1139] The generative AI recommends a "customer data mart" and a "purchase history data mart" based on similar past requirements. The server presents this information to the user.

[1140] 4. SQL query generation:

[1141] The generative AI automatically generates the following SQL query:

[1142] sql

[1143] SELECT

[1144] customer_id,

[1145] purchase_date,

[1146] product_id,

[1147] amount

[1148] FROM

[1149] customer_data_mart

[1150] WHERE

[1151] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[1152] (The code is not listed here)

[1153] 5. Check and correct the query:

[1154] The server presents the generated SQL query to the user, who then reviews it and, if necessary, modifies it.

[1155] 6. Query execution:

[1156] The server sends the modified query to the DWH, executes the query, retrieves the execution results, and displays them to the user.

[1157] This allows users to quickly obtain the data they need without having SQL skills. This system improves the efficiency of data analysis and greatly simplifies the work of users.

[1158] The processing flow will be explained below.

[1159] Step 1:

[1160] The user enters the requirements into the input form on the terminal. For example, the user enters the requirement "I want to obtain three months' worth of purchase history by customer."

[1161] Step 2:

[1162] The terminal transmits the input requirement data to the server.

[1163] Step 3:

[1164] The server sends the requirements data to the generation AI, which analyzes the requirements.

[1165] Step 4:

[1166] The generative AI uses natural language processing technology to extract and structure requirement keywords. For example, it identifies keywords such as "by customer," "purchase history," and "three months' worth."

[1167] Step 5:

[1168] The generation AI refers to past consultation history and selects the optimal data mart based on similar requirements. For example, it recommends a "customer data mart" and a "purchase history data mart."

[1169] Step 6:

[1170] The server receives the recommended data mart information from the generation AI and presents it to the user.

[1171] Step 7:

[1172] The user confirms the presented data mart information and makes a selection to proceed to the next step.

[1173] Step 8:

[1174] The AI ​​automatically generates the necessary SQL queries based on the selected data mart. For example, it generates the following query:

[1175] sql

[1176] SELECT

[1177] customer_id,

[1178] purchase_date,

[1179] product_id,

[1180] amount

[1181] FROM

[1182] customer_data_mart

[1183] WHERE

[1184] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[1185] (The code is not listed here)

[1186] Step 9:

[1187] The server presents the generated SQL query to the user.

[1188] Step 10:

[1189] The user reviews the proposed SQL query and makes corrections if necessary.

[1190] Step 11:

[1191] The user submits the modified query to the server.

[1192] Step 12:

[1193] The server sends the confirmed or modified query to the DWH for execution.

[1194] Step 13:

[1195] The server retrieves the query results from the DWH.

[1196] Step 14:

[1197] The server presents the obtained results to the user.

[1198] These steps allow users to efficiently use the DWH to quickly obtain the data they need. Automating the entire system process improves the efficiency of data analysis work and ensures accurate data extraction.

[1199] Example 1

[1200] 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."

[1201] In today's business environment, companies are required to analyze data stored in their internal data warehouses (DWHs) appropriately and quickly obtain the necessary information. However, data processing and SQL query creation require highly specialized knowledge, which reduces business efficiency. In particular, it is difficult for non-expert users to effectively utilize data.

[1202] 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.

[1203] In this invention, the server includes means for analyzing input requirements using a generation AI, means for selecting an appropriate data structure based on the analyzed requirements, means for automatically generating a data extraction query based on the selected data structure, means for presenting the generated data extraction query to a user and accepting confirmation and correction by the user, and means for executing the confirmed or corrected data extraction query and presenting the execution result to the user. This enables users to efficiently and quickly obtain required data without requiring advanced specialized knowledge.

[1204] "Generative AI" is an artificial intelligence technology that analyzes input requirements and automatically generates appropriate data structures and queries.

[1205] "Means" refers to a component of a system or technology that accomplishes a specific function or operation.

[1206] "Data structure" refers to the format or schema of information stored in a database or data warehouse.

[1207] A "data extraction query" is program code, such as SQL, used to retrieve specific information from a database.

[1208] "User" refers to a person who uses this system to analyze data and create queries.

[1209] "Natural language processing technology" is a technology that enables computers to understand, analyze, and generate language that humans use on a daily basis.

[1210] "Past response history" refers to data that records requirements that have been previously processed and the results of those processing.

[1211] A "data warehouse (DWH)" is a type of database system that centrally stores large amounts of data and is used for analysis and reporting.

[1212] This invention is a system that streamlines the process by which users of an in-house data warehouse (DWH) can select an appropriate data structure and automatically generate the necessary data extraction queries. This system uses a generative AI model to analyze user requirements, select a data structure based on the analysis, and automatically generate data extraction queries.

[1213] This system is centered around a generative AI model running on a server and the user's device.

[1214] To obtain user requirements, the user inputs the requirements for data analysis into an input form on the terminal. For example, the user might input a requirement such as "I want to obtain three months' worth of purchase history by customer." The terminal then sends this requirement to the server.

[1215] Next, the requirements are analyzed. The server receives the requirements data sent by the user and sends it to the generative AI model as a prompt. This prompt takes the form, "I would like to obtain the past three months' worth of purchase history by customer." The generative AI model uses natural language processing technology to analyze the requirements and extract keywords and objectives. For example, it extracts keywords such as "by customer," "purchase history," and "three months' worth."

[1216] Next, the data structure is selected. The generative AI model refers to the history of similar requirements processing in the past and returns a list of optimal data structures to the server. For example, it selects a "customer data structure" and a "purchase history data structure." The server then presents this information to the user.

[1217] Next, SQL queries are generated. The server uses a generative AI model to automatically generate data extraction queries based on the selected data structure. The generated queries may have the following format:

[1218] sql

[1219] SELECT

[1220] customer_id,

[1221] purchase_date,

[1222] product_id,

[1223] amount

[1224] FROM

[1225] customer_data_structure

[1226] WHERE

[1227] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[1228] The generated data extraction query is presented to the user by the server.

[1229] The user can review the query on their device screen and modify it as needed, for example to narrow it down to a specific customer ID.

[1230] Finally, the server sends the query modified and approved by the user to the DWH and executes it. The server retrieves the execution results and displays them to the user. This allows users to efficiently and quickly obtain the data they need, even without specialized knowledge of SQL.

[1231] An example of a prompt would be something like, "The user has entered the requirement, 'I want to retrieve the purchase history for each customer for the past three months.' Please select the optimal data structure based on this requirement and generate the necessary data extraction query."

[1232] Through this system, users can automatically generate the SQL queries required for data analysis, streamlining the data extraction process, enabling fast and accurate data analysis in the field of business, significantly improving work efficiency.

[1233] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1234] Step 1: Obtain user requirements

[1235] Specific behavior:

[1236] The user enters the requirements for data analysis into the input form on the terminal. For example, the user might enter, "I want to obtain three months' worth of purchase history by customer."

[1237] The terminal receives this input data and sends it to the server.

[1238] Input: Requirements entered by the user into the terminal

[1239] Output: Requirement data sent to the server

[1240] Step 2: Analyze requirements

[1241] Specific behavior:

[1242] The server receives the requirements data sent by the user.

[1243] The server creates a prompt for the generative AI model and sends it the requirements. An example of a prompt would be "I would like to obtain the purchase history for each customer for the past three months."

[1244] The generative AI model uses natural language processing technology to analyze requirements and extract keywords and objectives, such as "by customer," "purchase history," and "three months' worth."

[1245] Input: Requirement data submitted by the user

[1246] Output: Keywords and analysis results

[1247] Step 3: Choose a data structure

[1248] Specific behavior:

[1249] Based on the analysis results, the generative AI model refers to the history of similar requirements processing in the past.

[1250] The server receives a list of optimal data structures (data marts) from the generative AI model. For example, it selects a "customer data structure" and a "purchase history data structure."

[1251] The server presents this information to the user.

[1252] Input: Analysis results, past requirements processing history

[1253] Output: A list of selected data structures

[1254] Step 4: Generate SQL queries

[1255] Specific behavior:

[1256] The server uses the generative AI model to automatically generate SQL queries based on the selected data structure.

[1257] The generated query may have the following format:

[1258] sql

[1259] SELECT

[1260] customer_id,

[1261] purchase_date,

[1262] product_id,

[1263] amount

[1264] FROM

[1265] customer_data_structure

[1266] WHERE

[1267] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[1268] The server presents the generated data extraction query to the user.

[1269] Input: A list of selected data structures

[1270] Output: Auto-generated SQL query

[1271] Step 5: Check and fix your queries

[1272] Specific behavior:

[1273] The user sees the generated SQL query on the terminal screen.

[1274] The user modifies the query as needed, for example to narrow it down to a specific customer ID.

[1275] The modified SQL query is then sent to the server again.

[1276] Input: Generated SQL query

[1277] Output: SQL query confirmed or modified by the user

[1278] Step 6: Execute the query

[1279] Specific behavior:

[1280] The server sends the SQL query, which has been modified and approved by the user, to the DWH.

[1281] The server receives the execution results from the DWH and displays them to the user.

[1282] As a result of the execution, for example, a list of the customer's purchase history is displayed on the terminal.

[1283] Input: Confirmed or revised SQL query

[1284] Output: Execution results obtained from the DWH

[1285] (Application example 1)

[1286] 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."

[1287] Modern logistics centers are required to efficiently manage large volumes of shipping records and inventory data and quickly retrieve the necessary data. In particular, there is a growing need for systems that allow users to easily analyze data, even if they do not have specific SQL skills. However, conventional systems often require users to be familiar with complex database structures, which takes time to learn. It is also difficult to quickly select the appropriate data source. Therefore, there is a need to develop a system that can perform data analysis efficiently.

[1288] 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.

[1289] In this invention, the server includes means for analyzing input requirements using a generation AI, means for selecting an appropriate data mart based on the analyzed requirements, means for automatically generating an SQL query based on the selected data mart, means for presenting the generated SQL query to a user and accepting confirmation and correction by the user, means for executing the confirmed or corrected SQL query and presenting the execution result to the user, and means for automatically acquiring past shipping records and inventory data, specialized for data analysis of a logistics center. This enables users to efficiently analyze data and quickly acquire appropriate data even if they do not have specific SQL skills.

[1290] "Generative AI" is an artificial intelligence technology that analyzes user requirements, selects appropriate data marts, and automatically generates SQL queries.

[1291] A "data mart" is an aggregation of data sets extracted from a large data warehouse to address a specific business need.

[1292] An "SQL query" is a command in the Structured Query Language that describes an instruction to a database.

[1293] A "logistics center" is a facility that receives, stores, organizes, and ships goods and materials.

[1294] "Past shipping records" are historical information on products that have been shipped from the logistics center in the past.

[1295] "Inventory data" is information about the quantity and condition of products currently stored in the logistics center.

[1296] "Natural language processing technology" is a technology that enables computers to understand, analyze, and generate human language.

[1297] "Verify and fix" is the process whereby the user reviews the auto-generated SQL queries and makes changes if necessary.

[1298] This invention relates to a system that utilizes generative AI to improve the efficiency of data analysis in logistics centers. This system is centered around a server and user terminals.

[1299] First, the user inputs the requirements for data analysis from a terminal. For example, they might input a requirement such as "I want to obtain shipping records for each warehouse for the last three months." The server receives the user's requirements and sends them to the generation AI. This generation AI uses natural language processing technology to analyze the requirements and extract the keywords "warehouse," "shipping records," and "three months."

[1300] The generative AI then selects appropriate data marts based on these keywords, recommending, for example, "warehouse data mart" and "shipping record data mart," and the server presents this information to the user.

[1301] The Generative AI then automatically generates SQL queries based on the selected data mart. For example, the following SQL queries are generated:

[1302] SELECT

[1303] warehouse_id,

[1304] shipment_date,

[1305] product_id,

[1306] quantity

[1307] FROM

[1308] warehouse_data_mart

[1309] WHERE

[1310] shipment_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[1311] The server presents the generated SQL query to the user and accepts confirmation and modification by the user. The user confirms the query and makes modifications as necessary. The server then executes the confirmed or modified SQL query and presents the execution results to the user.

[1312] Specifically, it is implemented as shown in the following Python code, which uses an SQLite database and executes queries according to the user's requirements. The generative AI uses OpenAI's GPT model.

[1313] This allows the server to efficiently analyze data and quickly retrieve appropriate data, even if the user does not have specific SQL skills. This makes it easier for distribution centers to quickly check shipping records and manage inventory.

[1314] Examples of prompt statements

[1315] Prompt for the generative AI model:

[1316] Analyze the following requirements: Get shipping records for each warehouse for the last three months

[1317] As described above, the present invention provides a concrete example of a system that simplifies data management at a logistics center and enables efficient data analysis. This system allows users to quickly obtain the data they need without having to worry about complex database operations.

[1318] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1319] Step 1:

[1320] The user inputs the requirements for data analysis from the terminal. For example, they input a requirement such as "I want to obtain the shipping records for each warehouse for the last three months." This requirement is sent to the server.

[1321] Input: The requirement text entered by the user

[1322] Output: The text of the requirement as received by the server

[1323] Step 2:

[1324] The server sends the received requirements to the generation AI, which then analyzes them using natural language processing technology. During the analysis, keywords such as "warehouse," "shipping record," and "3 months" are extracted from the requirements.

[1325] Input: Text of the requirement received from the user

[1326] Output: Generated keyword list (e.g. "warehouse", "shipping record", "3 months")

[1327] Step 3:

[1328] Based on the extracted keywords, the generation AI refers to past consultation history and selects the appropriate data mart. For example, it might recommend a "warehouse data mart" and a "shipping record data mart."

[1329] Input: Keyword list, past consultation history data

[1330] Output: A list of recommended data marts (e.g., "Warehouse Data Mart" and "Shipping Record Data Mart")

[1331] Step 4:

[1332] The server automatically generates SQL queries based on the selected data mart using the AI. The generated SQL queries are, for example, as follows:

[1333] SELECT

[1334] warehouse_id,

[1335] shipment_date,

[1336] product_id,

[1337] quantity

[1338] FROM

[1339] warehouse_data_mart

[1340] WHERE

[1341] shipment_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[1342] Input: Recommended data mart list, keyword list

[1343] Output: Auto-generated SQL query

[1344] Step 5:

[1345] The server presents the generated SQL query to the user, who can review it and make modifications if necessary. The user can review the query and add or modify specific conditions.

[1346] Input: Auto-generated SQL query

[1347] Output: SQL query reviewed and corrected by the user

[1348] Step 6:

[1349] The server executes the verified or modified SQL query and retrieves the results, for example, by querying an SQLite database and retrieving the results.

[1350] Input: Confirmed or revised SQL query

[1351] Output: Query results (data retrieved from the database)

[1352] Step 7:

[1353] The server then presents the results of the query to the user, allowing the user to quickly obtain the data they need.

[1354] Input: Query results

[1355] Output: Execution results presented to the user (e.g., shipping records by warehouse for the past three months)

[1356] 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.

[1357] This invention is a system that enables users of an in-house data warehouse (DWH) to efficiently extract data and perform analytical tasks, and aims to improve the user experience by combining it with an emotion engine that recognizes the user's emotions.

[1358] Overall system configuration

[1359] The system includes the following main components:

[1360] 1. Generation AI

[1361] 2. Natural Language Processing Technology

[1362] 3. Emotion Engine

[1363] 4. Server

[1364] 5. User Device

[1365] Acquiring and analyzing user requirements

[1366] 1. Obtaining user requirements

[1367] The user enters requirements into an input form on the terminal. For example, the user enters a requirement such as "I want to obtain three months' worth of purchase history by customer."

[1368] 2. Emotion recognition

[1369] While the terminal is inputting user requirements, the emotion engine monitors the user's emotional state. For example, if the user is feeling stressed, the emotion engine will recognize this.

[1370] 3. Submit your requirements

[1371] The terminal transmits the input requirement data and the user's emotion data to the server.

[1372] Requirements and sentiment data analysis

[1373] 4. Requirements Analysis

[1374] The server sends the requirements data to the generation AI, which then uses natural language processing technology to extract and structure keywords from the requirements. For example, it identifies keywords such as "by customer," "purchase history," and "three months' worth."

[1375] 5. Emotion Data Analysis

[1376] The server analyzes the user's emotional state based on the data from the emotion engine and determines whether the user is feeling stressed.

[1377] Data mart selection and query generation

[1378] 6. Data Mart Selection

[1379] The generative AI refers to past consultation history and selects the optimal data mart based on the analyzed requirements and emotional state. For example, it recommends a "customer data mart" and a "purchase history data mart."

[1380] 7. Presentation to the User

[1381] The server receives the recommended data mart information from the generation AI and displays it on the device. It also adjusts the interface and provides additional support content according to the user's emotional state.

[1382] 8. Generating SQL Queries

[1383] The AI ​​automatically generates the necessary SQL queries based on the selected data mart. For example, it generates the following query:

[1384] sql

[1385] SELECT

[1386] customer_id,

[1387] purchase_date,

[1388] product_id,

[1389] amount

[1390] FROM

[1391] customer_data_mart

[1392] WHERE

[1393] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[1394] (The code is not listed here)

[1395] Review and run the query

[1396] 9. Check and correct your query

[1397] The server presents the generated SQL query to the user, who confirms it and modifies it if necessary. If the user is feeling nervous or stressed, the emotion engine displays an appropriate support message.

[1398] 10. Executing a Query

[1399] The server sends the confirmed or modified query to the DWH for execution.

[1400] Displaying execution results

[1401] 11. Obtaining and Presenting Results

[1402] The server retrieves the query results from the DWH and displays them on the device. It monitors the user's emotional state and interprets the results and suggests the next action if necessary.

[1403] Specific examples

[1404] For example, if a user enters the requirement "I want to get the purchase history for each customer for the past three months":

[1405] 1. Requirements acquisition: The user enters the requirements into an input form on the terminal.

[1406] 2. Emotion Recognition: The emotion engine recognizes the user's emotions (e.g., stress state).

[1407] 3. Sending requirements and emotion data: The device sends the data to the server.

[1408] 4. Requirements analysis: The server sends the requirements data to the generation AI for analysis.

[1409] 5. Emotional data analysis: Analyze user emotions.

[1410] 6. Data mart selection: Generative AI recommends the optimal data mart.

[1411] 7. Presentation to the user: The server presents emotional support along with data mart information.

[1412] 8. SQL Query Generation: Generative AI generates SQL queries.

[1413] 9. Review and modify the query: The user reviews the query and modifies it if necessary.

[1414] 10. Query execution: The server sends the query to the DWH and executes it.

[1415] 11. Retrieving and presenting results: The server retrieves the results and presents them to the user.

[1416] This system allows users to efficiently retrieve data and perform analysis work without stress, even if they do not have SQL skills.The emotion engine monitors the user's emotional state and provides appropriate support, resulting in a better user experience.

[1417] The processing flow will be explained below.

[1418] Step 1:

[1419] The user enters the requirements into the input form on the terminal. For example, the user enters "I want to obtain three months' worth of purchase history by customer."

[1420] Step 2:

[1421] The emotion engine monitors the user's emotional state, for example, analyzing the user's stress level from facial expressions and tone of voice via the camera and microphone while the user is typing.

[1422] Step 3:

[1423] The terminal transmits the input requirement data and emotion data to the server.

[1424] Step 4:

[1425] The server sends the requirement data and emotion data to the generation AI, which receives the requirement data and begins analyzing it.

[1426] Step 5:

[1427] Generative AI uses natural language processing technology to analyze requirement data and identify and extract keywords and objectives. For example, it identifies "by customer," "purchase history," and "three months' worth."

[1428] Step 6:

[1429] The server receives the analysis results from the generation AI and stores them along with the emotion data.

[1430] Step 7:

[1431] The generation AI refers to past consultation history and selects the appropriate data mart based on the analyzed requirements. For example, it recommends a "customer data mart" and a "purchase history data mart."

[1432] Step 8:

[1433] The server receives the recommendation data mart information from the generation AI and displays it on the device, and also adjusts the interface based on the user's emotional state and displays necessary support messages.

[1434] Step 9:

[1435] The user confirms the presented data mart information and makes a selection to proceed to the next step.

[1436] Step 10:

[1437] The AI ​​automatically generates the necessary SQL queries based on the selected data mart. For example, it generates the following query:

[1438] sql

[1439] SELECT

[1440] customer_id,

[1441] purchase_date,

[1442] product_id,

[1443] amount

[1444] FROM

[1445] customer_data_mart

[1446] WHERE

[1447] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[1448] (The code is not listed here)

[1449] Step 11:

[1450] The server presents the generated SQL query to the user.

[1451] Step 12:

[1452] The user checks the presented SQL query and modifies it as necessary. For example, if they want to extract only a specific product category, they can add that condition.

[1453] Step 13:

[1454] The user submits the modified query to the server.

[1455] Step 14:

[1456] The server sends the confirmed or modified query to the DWH and executes the query.

[1457] Step 15:

[1458] The server retrieves the query results from the DWH.

[1459] Step 16:

[1460] The server sends the results to the device and displays them to the user. It also monitors the user's emotional state again and displays a message to guide the user on the next action based on the results.

[1461] These steps allow users to efficiently extract data and perform data analysis.The emotion engine monitors the user's emotional state in real time and takes appropriate action to provide a better user experience.

[1462] Example 2

[1463] 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."

[1464] Conventional in-house data warehouse (DWH) systems have the drawback of making it difficult for users to efficiently extract and analyze data. Furthermore, simple data processing is often performed without considering the user's emotional state, resulting in a poor user experience. The present invention aims to solve these problems and provide a system that enables users to efficiently extract and analyze data without feeling stressed.

[1465] 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.

[1466] In this invention, the server includes: means for a user to input requirements into an input form on a terminal; means for an emotion engine to monitor the user's emotional state while the user is inputting the user's requirements; means for transmitting the input requirement data and the user's emotional data to the server; means for transmitting the requirement data to a generation AI and extracting and structuring keywords of the requirements using natural language processing technology; means for analyzing the user's emotional state based on the data from the emotion engine and determining whether the user is feeling stressed; means for selecting an optimal data mart based on the analyzed requirements and emotional state by referring to past consultation history; means for receiving recommended data mart information from the generation AI and presenting interface adjustments or additional support content according to the user's emotional state; means for automatically generating a required SQL query based on the selected data mart; means for presenting the generated SQL query to the user and accepting user confirmation and correction; and means for executing the confirmed or corrected SQL query and presenting the execution result to the user. This allows the user to efficiently extract and analyze data, and further enables the emotion engine to improve the user experience.

[1467] A "terminal" is an input / output device that a user uses to input requirements.

[1468] An "emotion engine" is a software and hardware system that monitors and analyzes a user's emotional state in real time.

[1469] "Requirement data" refers to specific requests for data extraction and analysis that are input by the user into the terminal.

[1470] "Emotion data" is data that represents the user's emotional state and is acquired by the emotion engine.

[1471] A "server" is a computer system that receives user requirement data and emotion data and performs various processes based on them.

[1472] "Generative AI" is an artificial intelligence that analyzes requirements data and extracts and structures keywords using natural language processing technology.

[1473] "Natural language processing technology" is information technology used to interpret and understand the meaning of human language.

[1474] A "data mart" is a portion of a database that stores a subset of data related to a particular business process or area.

[1475] An "SQL query" is a command statement written in the Structured Query Language (SQL) for searching and manipulating data in a database management system (DBMS).

[1476] The "past consultation history" is a record of requirements that the user has exchanged with the server in the past and the results of the processing.

[1477] "Adjusting the interface" refers to changing the screen display or user interface configuration in response to the user's emotional state.

[1478] "Additional support content" refers to supplementary information or messages provided depending on the user's emotional state.

[1479] This invention is a system that aims to enable users of in-house data warehouses (DWHs) to efficiently extract data and perform analysis work. Furthermore, by combining it with an emotion engine that recognizes user emotions, we aim to improve the user experience.

[1480] Overall system configuration

[1481] The system includes the following major components:

[1482] 1. Generation AI

[1483] 2. Natural Language Processing Technology

[1484] 3. Emotion Engine

[1485] 4. Server

[1486] 5. User Device

[1487] Hardware and Software

[1488] Terminal: An input device for users to enter requirements. It uses a web browser or GUI form.

[1489] Emotion engine: Uses sensors and software to monitor the user's facial expressions and typing speed in real time and analyze their emotional state. Specific examples include camera sensors and keyboard typing speed analysis software.

[1490] Server: A computer system that receives requirements data and emotion data and analyzes and structures the data using generative AI models and natural language processing techniques.

[1491] Generative AI: An artificial intelligence that analyzes user requirements data and extracts and structures keywords using natural language processing technology.

[1492] Data mart: A portion of a database that stores a subset of data related to a particular business process or area.

[1493] System Operation

[1494] When a user enters requirements into an input form on a terminal, for example, "I would like to obtain the past three months' worth of purchase history by customer," the emotion engine monitors the user's facial expressions and input speed and analyzes their emotional state. The terminal then sends the entered requirement data and emotion data to the server.

[1495] The server sends the received requirements data to the generation AI, which uses natural language processing technology to extract and structure keywords. Based on the extracted keywords and taking into account data from the emotion engine, the generation AI determines whether the user is feeling stressed. The generation AI refers to past consultation history and selects the optimal data mart based on the analyzed requirements and emotional state. The server receives the recommended data mart information and displays it on the device. At this time, it adjusts the interface and presents additional support content according to the user's emotional state.

[1496] The generation AI automatically generates the necessary SQL queries based on the selected data mart. The generated SQL queries are presented to the user, who can confirm and modify them. Finally, the confirmed or modified SQL queries are sent from the server to the DWH, and data is retrieved. The retrieved data is displayed on the terminal, and the user can confirm the results. At this time, the emotion engine displays support messages as needed.

[1497] Specific examples

[1498] 1. Obtaining requirements: The user enters the requirement "I want to obtain the purchase history for each customer for the past three months" into the input form on the terminal.

[1499] 2. Emotion Recognition: The emotion engine monitors the user's facial expressions and input speed to recognize stress levels.

[1500] 3. Sending requirements and emotion data: The terminal sends the requirements data and emotion data to the server.

[1501] 4. Requirements analysis: The generative AI extracts the keywords "by customer," "purchase history," and "three months' worth."

[1502] 5. Emotional data analysis: Determine whether the user is feeling stressed.

[1503] 6. Data mart selection: The generative AI selects the optimal data mart (e.g., "customer data mart" or "purchase history data mart").

[1504] 7. Presentation to the user: The recommended data mart and support content according to the emotional state are displayed on the terminal.

[1505] 8. SQL Query Generation: The generative AI automatically generates SQL queries like the following:

[1506] sql

[1507] SELECT

[1508] customer_id,

[1509] purchase_date,

[1510] product_id,

[1511] amount

[1512] FROM

[1513] customer_data_mart

[1514] WHERE

[1515] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[1516] 9. Review and modify the query: The user reviews the query and modifies it if necessary.

[1517] 10. Query execution: The server sends the confirmed or modified query to the DWH and executes it.

[1518] 11. Obtaining and presenting results: The server displays the obtained data on the terminal, and the user confirms the results.

[1519] Prompt Sentence Examples

[1520] Generate a SQL query to retrieve customer purchase history for the past three months.

[1521] This system allows users to efficiently retrieve data and perform analysis work without stress, even if they do not have SQL skills.The emotion engine monitors the user's emotional state and provides appropriate support, resulting in a better user experience.

[1522] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1523] Step 1:

[1524] The user inputs requirements into an input form on the terminal. Specifically, the user inputs the requirement "I want to obtain the purchase history for each customer for the past three months" into the input form. At this time, the terminal receives the information entered by the user as input data. The input data is requirement information in text format.

[1525] Step 2:

[1526] The emotion engine monitors the user's emotional state while they are entering their requirements. Specifically, a camera sensor analyzes the user's facial expressions and keyboard input speed in real time to generate emotion data. The input is the user's facial image and input speed data, and the output is emotional state data in text format (e.g., stress level).

[1527] Step 3:

[1528] The device sends the input requirement data and user emotion data to the server. The input is the user requirement data and emotion data, and the output is the requirement data and emotion data sent to the server. Specifically, it is sent to the server as JSON format data.

[1529] Step 4:

[1530] The server sends the requirements data to the generation AI, which uses natural language processing technology to extract and structure requirement keywords. The input is the requirements data (e.g., "I want to obtain the purchase history by customer for the past three months"), and the output is the extracted keywords (e.g., "by customer," "purchase history," "three months' worth"). Specifically, the generation AI uses a natural language processing engine to analyze and structure the keywords.

[1531] Step 5:

[1532] The server analyzes the user's emotional state based on data from the emotion engine and determines whether the user is feeling stressed. The input is emotional data (e.g., stress state), and the output is a judgment result of the user's emotional state (e.g., the user is feeling stressed). Specifically, the server runs an emotion analysis algorithm and quantifies the emotional state.

[1533] Step 6:

[1534] The generation AI refers to past consultation history and selects the optimal data mart based on the analyzed requirements and emotional state. The input is requirement keywords and emotional state data, and the output is a recommended data mart (e.g., "customer data mart" or "purchase history data mart"). Specifically, the generation AI runs a data mart selection algorithm to select the optimal data mart.

[1535] Step 7:

[1536] The server receives the recommendation data mart information from the generation AI and displays it on the device. Furthermore, it adjusts the interface and presents additional support content according to the user's emotional state. The input is the recommendation data mart and the user's emotional state, and the output is the information displayed on the device. Specifically, the server receives the recommendation data mart information and adjusts the UI according to the emotional state.

[1537] Step 8:

[1538] The generation AI automatically generates the required SQL queries based on the selected data mart. The input is the selected data mart information and requirement keywords, and the output is the generated SQL query. Specifically, the generation AI automatically generates the following SQL queries:

[1539] sql

[1540] SELECT

[1541] customer_id,

[1542] purchase_date,

[1543] product_id,

[1544] amount

[1545] FROM

[1546] customer_data_mart

[1547] WHERE

[1548] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[1549] Step 9:

[1550] The server presents the generated SQL query to the user, who confirms it and modifies it if necessary. The input is the generated SQL query, and the output is the SQL query that has been confirmed and modified by the user. Specifically, the server displays the SQL query on the terminal, and the user confirms and modifies the query content.

[1551] Step 10:

[1552] The server sends the confirmed or modified SQL query to the DWH and executes it. The input is the confirmed or modified SQL query, and the output is the query execution result from the DWH. In concrete terms, the server sends the SQL query to the data warehouse, executes the query, and retrieves the data.

[1553] Step 11:

[1554] The server retrieves the query execution results from the DWH and displays them on the terminal. It monitors the user's emotional state and, if necessary, interprets the results and suggests the next action. The input is the query execution results and the user's emotional state, and the output is the execution results and a support message displayed on the terminal. In concrete terms, the server retrieves the query results, and if the user is feeling stressed, it displays an interpretation support message.

[1555] (Application example 2)

[1556] 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."

[1557] With conventional data analysis systems, if users lacked SQL skills or database knowledge, it was difficult to efficiently extract and analyze data. Furthermore, because data was provided without taking into account the user's emotional state, the user interface and support were often inappropriate. This resulted in a poor user experience and reduced productivity, especially when tasks were complex or under stress.

[1558] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for analyzing input requirements using a generation AI, means for selecting an appropriate data mart based on the analyzed requirements, means for automatically generating an SQL query based on the selected data mart, means for presenting the generated SQL query to a user and accepting confirmation and correction by the user, means for executing the confirmed or corrected SQL query and presenting the execution result to the user, an emotion recognition engine for monitoring the user's emotional state, and means for generating and presenting a response method based on the emotional state. This enables users to efficiently extract and analyze data without advanced technical knowledge, and furthermore, by providing support appropriate to the user's emotional state, it is possible to reduce stress and improve the user experience.

[1559] "Generative AI" is an artificial intelligence technique used to analyze data and automate tasks.

[1560] The "means for analyzing requirements" is a method for analyzing the requirements data entered by the user and extracting important keywords and intentions.

[1561] The "means for selecting an appropriate data mart" is a method for selecting the most relevant data mart based on the analyzed requirements.

[1562] "Means for automatically generating SQL queries" refers to a method for automatically creating the necessary SQL queries for the selected data mart.

[1563] The "means for presenting the generated SQL query to the user and accepting confirmation and correction by the user" refers to a method for displaying the generated SQL query to the user so that the user can confirm or correct it.

[1564] "Means for executing a confirmed or modified SQL query and presenting the execution results to a user" refers to a method for executing an SQL query confirmed or modified by a user and displaying the results to the user.

[1565] The "emotion recognition engine that monitors the user's emotional state" is a system that analyzes the user's emotions in real time and determines their state.

[1566] The "means for generating and presenting a response method based on an emotional state" is a method for generating an optimal response method according to the emotional state of the user determined by an emotion recognition engine and presenting it to the user.

[1567] As an embodiment of the present invention, a customer service support system using smart glasses for store clerks in a brick-and-mortar store can be adopted. The following describes the specific roles of the server, terminal, and user.

[1568] server

[1569] The server performs the following operations using the following specific hardware and software:

[1570] 1. Generative AI and Natural Language Processing Technology

[1571] Software used: Python language, natural language processing library (e.g., SpaCy)

[1572] The server acquires the user's requirements and uses generation AI to extract and analyze keywords from the requirements.

[1573] For example, from the requirement "Please tell me your purchase history for the past three months," extract keywords such as "past three months" and "purchase history."

[1574] 2. Emotion Recognition Engine

[1575] Hardware used: Smart glasses camera

[1576] Software used: Emotion-FerPlus

[1577] The server analyzes the video data acquired through the glasses' camera and monitors the customer's emotional state in real time.

[1578] For example, determining whether a customer is stressed.

[1579] 3. Data mart selection and SQL query generation

[1580] Software used: SQLAlchemy

[1581] The server selects a data mart based on the analyzed requirements and automatically generates the necessary SQL queries.

[1582] For example, generate a query to extract data from a purchase history data mart within a specific period.

[1583] Terminal (smart glasses)

[1584] The smart glasses worn by the store clerk function as a terminal and support the following processes:

[1585] 1. Input customer requirements

[1586] Enter customer requirements using voice input or touch operation.

[1587] For example, a request such as "I want to know the purchase history for the past three months" is input.

[1588] 2. Displaying emotional states

[1589] The analysis results of the emotion recognition engine are displayed in real time.

[1590] For example, if a customer is feeling stressed, the glasses' display will say "Customer is feeling stressed."

[1591] 3. Displaying query results and suggesting how to respond

[1592] The data returned from the server and the response method based on the emotional state are displayed on the glasses' display.

[1593] For example, along with the purchase history as a query result, the message "The customer is stressed. Please respond calmly" may be displayed.

[1594] User (store clerk)

[1595] Store staff use the above system to efficiently handle customer service.

[1596] 1. Obtaining and inputting user requirements

[1597] Customer requirements are registered into the smart glasses using voice or touch input.

[1598] An example of a specific prompt might be: "The customer wants purchase history data for the past three months. Please generate a query."

[1599] 2. Checking and responding to emotional states

[1600] Check the customer's emotional state in real time and respond accordingly.

[1601] Specific example of how to respond: "The customer is stressed. Please respond calmly."

[1602] This system allows store clerks to efficiently obtain customer information without having SQL skills and to respond optimally according to the customer's emotional state, significantly improving customer satisfaction and store clerk work efficiency.

[1603] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1604] Step 1:

[1605] The user uses the smart glasses to input customer requirements. At this time, the user can use voice input or touch operation. Specifically, the user can say, "I would like to know my purchase history for the past three months." The input data is acquired through the smart glasses' built-in microphone or touchpad. The input content is converted into text format and sent to the server.

[1606] Step 2:

[1607] The server analyzes the received user requirements using a generative AI model and natural language processing technology. The input data is in text format, and the generative AI model extracts keywords. In this process, keywords such as "past 3 months" and "purchase history" are identified.

[1608] Step 3:

[1609] The server selects an appropriate data mart based on the analysis results. The server refers to past consultation history and recommends a data mart that matches the analyzed keywords. For example, the "purchase history data mart" may be selected. This information is also returned to the user in text format.

[1610] Step 4:

[1611] The server automatically generates SQL queries using the selected data mart. The generative AI model creates appropriate queries based on keywords, for example, the following SQL queries are generated:

[1612] sql

[1613] SELECT customer_id, purchase_date, product_id, amount FROM purchase_history WHERE purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[1614] The generated query is presented to the user in text form.

[1615] Step 5:

[1616] The user checks the generated SQL query through the smart glasses and makes corrections if necessary. The user visually checks the query displayed on the screen, accepts it as is if there are no problems, and makes corrections by voice or touch if necessary. The corrected query is then sent back to the server.

[1617] Step 6:

[1618] The server executes the verified or modified SQL query and retrieves the results from the data warehouse. The results are then retrieved from the database and converted to text format, which is then sent back to the user.

[1619] Step 7:

[1620] At the same time, the server monitors the customer's emotional state through the smart glasses' camera, and an emotion recognition engine analyzes the image data in real time to determine the customer's state, such as "feeling stressed."

[1621] Step 8:

[1622] The server generates the optimal response based on the customer's emotional state and displays it on the smart glasses. The generative AI model then uses the emotional data to devise a response method. For example, a message such as "The customer is feeling stressed. Please respond calmly" is displayed.

[1623] Step 9:

[1624] The user checks the execution results and response methods displayed on the smart glasses display and responds appropriately to the customer. Based on the results, the necessary information is conveyed to the customer, and appropriate communication is carried out based on emotional data. This optimizes customer response.

[1625] 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.

[1626] 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.

[1627] 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.

[1628] [Fourth embodiment]

[1629] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1630] 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.

[1631] 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).

[1632] 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.

[1633] 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.

[1634] 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).

[1635] 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.

[1636] 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.

[1637] 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.

[1638] 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.

[1639] 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.

[1640] 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.

[1641] 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."

[1642] This invention is a system that streamlines the process for users of an in-house data warehouse (DWH) to select an appropriate data mart and create the necessary SQL queries. This system uses generative AI to analyze the user's requirements, select a data mart based on those requirements, and automatically generate SQL queries.

[1643] Overall system configuration

[1644] This system is centered around a generation AI running on a server and the user's device.

[1645] 1. Obtaining user requirements:

[1646] The user inputs the data analysis requirements from the terminal. For example, the requirement is "I want to obtain three months' worth of purchase history by customer."

[1647] 2. Requirements Analysis:

[1648] The server sends the user's requirements to the generation AI. The generation AI uses natural language processing technology to analyze the requirements and identify keywords and objectives. For example, it extracts "by customer," "purchase history," and "three months' worth" from the requirements.

[1649] 3. Data mart selection:

[1650] The generation AI refers to past consultation history and recommends the most appropriate data mart. For example, it recommends a "customer data mart" and a "purchase history data mart."

[1651] 4. Generate SQL queries:

[1652] The AI ​​automatically generates SQL queries based on the selected data mart. For example, it generates the following query:

[1653] sql

[1654] SELECT

[1655] customer_id,

[1656] purchase_date,

[1657] product_id,

[1658] amount

[1659] FROM

[1660] customer_data_mart

[1661] WHERE

[1662] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[1663] (The code is not listed here)

[1664] 5. Check and correct the query:

[1665] The server presents the generated SQL query to the user, who can review it and modify it if necessary (for example, if the user wants to add specific conditions to the query's output).

[1666] 6. Execute the query:

[1667] The server sends the query approved by the user to the DWH, executes it, and retrieves the execution results and displays them to the user.

[1668] Specific examples

[1669] For example, if a user enters a requirement such as "I want to get the purchase history for each customer for the past three months," it will work as follows:

[1670] 1. Obtain requirements:

[1671] The user enters "I would like to obtain three months' worth of purchase history by customer" into the input form on the terminal.

[1672] 2. Requirements analysis:

[1673] The server sends the requirements to the generation AI, which then extracts and analyzes the keywords "by customer," "purchase history," and "three months' worth."

[1674] 3. Data mart selection:

[1675] The generative AI recommends a "customer data mart" and a "purchase history data mart" based on similar past requirements. The server presents this information to the user.

[1676] 4. SQL query generation:

[1677] The generative AI automatically generates the following SQL query:

[1678] sql

[1679] SELECT

[1680] customer_id,

[1681] purchase_date,

[1682] product_id,

[1683] amount

[1684] FROM

[1685] customer_data_mart

[1686] WHERE

[1687] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[1688] (The code is not listed here)

[1689] 5. Check and correct the query:

[1690] The server presents the generated SQL query to the user, who then reviews it and, if necessary, modifies it.

[1691] 6. Query execution:

[1692] The server sends the modified query to the DWH, executes the query, retrieves the execution results, and displays them to the user.

[1693] This allows users to quickly obtain the data they need without having SQL skills. This system improves the efficiency of data analysis and greatly simplifies the work of users.

[1694] The processing flow will be explained below.

[1695] Step 1:

[1696] The user enters the requirements into the input form on the terminal. For example, the user enters the requirement "I want to obtain three months' worth of purchase history by customer."

[1697] Step 2:

[1698] The terminal transmits the input requirement data to the server.

[1699] Step 3:

[1700] The server sends the requirements data to the generation AI, which analyzes the requirements.

[1701] Step 4:

[1702] The generative AI uses natural language processing technology to extract and structure requirement keywords. For example, it identifies keywords such as "by customer," "purchase history," and "three months' worth."

[1703] Step 5:

[1704] The generation AI refers to past consultation history and selects the optimal data mart based on similar requirements. For example, it recommends a "customer data mart" and a "purchase history data mart."

[1705] Step 6:

[1706] The server receives the recommended data mart information from the generation AI and presents it to the user.

[1707] Step 7:

[1708] The user confirms the presented data mart information and makes a selection to proceed to the next step.

[1709] Step 8:

[1710] The AI ​​automatically generates the necessary SQL queries based on the selected data mart. For example, it generates the following query:

[1711] sql

[1712] SELECT

[1713] customer_id,

[1714] purchase_date,

[1715] product_id,

[1716] amount

[1717] FROM

[1718] customer_data_mart

[1719] WHERE

[1720] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[1721] (The code is not listed here)

[1722] Step 9:

[1723] The server presents the generated SQL query to the user.

[1724] Step 10:

[1725] The user reviews the proposed SQL query and makes corrections if necessary.

[1726] Step 11:

[1727] The user submits the modified query to the server.

[1728] Step 12:

[1729] The server sends the confirmed or modified query to the DWH for execution.

[1730] Step 13:

[1731] The server retrieves the query results from the DWH.

[1732] Step 14:

[1733] The server presents the obtained results to the user.

[1734] These steps allow users to efficiently use the DWH to quickly obtain the data they need. Automating the entire system process improves the efficiency of data analysis work and ensures accurate data extraction.

[1735] Example 1

[1736] 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."

[1737] In today's business environment, companies are required to analyze data stored in their internal data warehouses (DWHs) appropriately and quickly obtain the necessary information. However, data processing and SQL query creation require highly specialized knowledge, which reduces business efficiency. In particular, it is difficult for non-expert users to effectively utilize data.

[1738] 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.

[1739] In this invention, the server includes means for analyzing input requirements using a generation AI, means for selecting an appropriate data structure based on the analyzed requirements, means for automatically generating a data extraction query based on the selected data structure, means for presenting the generated data extraction query to a user and accepting confirmation and correction by the user, and means for executing the confirmed or corrected data extraction query and presenting the execution result to the user. This enables users to efficiently and quickly obtain required data without requiring advanced specialized knowledge.

[1740] "Generative AI" is an artificial intelligence technology that analyzes input requirements and automatically generates appropriate data structures and queries.

[1741] "Means" refers to a component of a system or technology that accomplishes a specific function or operation.

[1742] "Data structure" refers to the format or schema of information stored in a database or data warehouse.

[1743] A "data extraction query" is program code, such as SQL, used to retrieve specific information from a database.

[1744] "User" refers to a person who uses this system to analyze data and create queries.

[1745] "Natural language processing technology" is a technology that enables computers to understand, analyze, and generate language that humans use on a daily basis.

[1746] "Past response history" refers to data that records requirements that have been previously processed and the results of those processing.

[1747] A "data warehouse (DWH)" is a type of database system that centrally stores large amounts of data and is used for analysis and reporting.

[1748] This invention is a system that streamlines the process by which users of an in-house data warehouse (DWH) can select an appropriate data structure and automatically generate the necessary data extraction queries. This system uses a generative AI model to analyze user requirements, select a data structure based on the analysis, and automatically generate data extraction queries.

[1749] This system is centered around a generative AI model running on a server and the user's device.

[1750] To obtain user requirements, the user inputs the requirements for data analysis into an input form on the terminal. For example, the user might input a requirement such as "I want to obtain three months' worth of purchase history by customer." The terminal then sends this requirement to the server.

[1751] Next, the requirements are analyzed. The server receives the requirements data sent by the user and sends it to the generative AI model as a prompt. This prompt takes the form, "I would like to obtain the past three months' worth of purchase history by customer." The generative AI model uses natural language processing technology to analyze the requirements and extract keywords and objectives. For example, it extracts keywords such as "by customer," "purchase history," and "three months' worth."

[1752] Next, the data structure is selected. The generative AI model refers to the history of similar requirements processing in the past and returns a list of optimal data structures to the server. For example, it selects a "customer data structure" and a "purchase history data structure." The server then presents this information to the user.

[1753] Next, SQL queries are generated. The server uses a generative AI model to automatically generate data extraction queries based on the selected data structure. The generated queries may have the following format:

[1754] sql

[1755] SELECT

[1756] customer_id,

[1757] purchase_date,

[1758] product_id,

[1759] amount

[1760] FROM

[1761] customer_data_structure

[1762] WHERE

[1763] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[1764] The generated data extraction query is presented to the user by the server.

[1765] The user can review the query on their device screen and modify it as needed, for example to narrow it down to a specific customer ID.

[1766] Finally, the server sends the query modified and approved by the user to the DWH and executes it. The server retrieves the execution results and displays them to the user. This allows users to efficiently and quickly obtain the data they need, even without specialized knowledge of SQL.

[1767] An example of a prompt would be something like, "The user has entered the requirement, 'I want to retrieve the purchase history for each customer for the past three months.' Please select the optimal data structure based on this requirement and generate the necessary data extraction query."

[1768] Through this system, users can automatically generate the SQL queries required for data analysis, streamlining the data extraction process, enabling fast and accurate data analysis in the field of business, significantly improving work efficiency.

[1769] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1770] Step 1: Obtain user requirements

[1771] Specific behavior:

[1772] The user enters the requirements for data analysis into the input form on the terminal. For example, the user might enter, "I want to obtain three months' worth of purchase history by customer."

[1773] The terminal receives this input data and sends it to the server.

[1774] Input: Requirements entered by the user into the terminal

[1775] Output: Requirement data sent to the server

[1776] Step 2: Analyze requirements

[1777] Specific behavior:

[1778] The server receives the requirements data sent by the user.

[1779] The server creates a prompt for the generative AI model and sends it the requirements. An example of a prompt would be "I would like to obtain the purchase history for each customer for the past three months."

[1780] The generative AI model uses natural language processing technology to analyze requirements and extract keywords and objectives, such as "by customer," "purchase history," and "three months' worth."

[1781] Input: Requirement data submitted by the user

[1782] Output: Keywords and analysis results

[1783] Step 3: Choose a data structure

[1784] Specific behavior:

[1785] Based on the analysis results, the generative AI model refers to the history of similar requirements processing in the past.

[1786] The server receives a list of optimal data structures (data marts) from the generative AI model. For example, it selects a "customer data structure" and a "purchase history data structure."

[1787] The server presents this information to the user.

[1788] Input: Analysis results, past requirements processing history

[1789] Output: A list of selected data structures

[1790] Step 4: Generate SQL queries

[1791] Specific behavior:

[1792] The server uses the generative AI model to automatically generate SQL queries based on the selected data structure.

[1793] The generated query may have the following format:

[1794] sql

[1795] SELECT

[1796] customer_id,

[1797] purchase_date,

[1798] product_id,

[1799] amount

[1800] FROM

[1801] customer_data_structure

[1802] WHERE

[1803] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[1804] The server presents the generated data extraction query to the user.

[1805] Input: A list of selected data structures

[1806] Output: Auto-generated SQL query

[1807] Step 5: Check and fix your queries

[1808] Specific behavior:

[1809] The user sees the generated SQL query on the terminal screen.

[1810] The user modifies the query as needed, for example to narrow it down to a specific customer ID.

[1811] The modified SQL query is then sent to the server again.

[1812] Input: Generated SQL query

[1813] Output: SQL query confirmed or modified by the user

[1814] Step 6: Execute the query

[1815] Specific behavior:

[1816] The server sends the SQL query, which has been modified and approved by the user, to the DWH.

[1817] The server receives the execution results from the DWH and displays them to the user.

[1818] As a result of the execution, for example, a list of the customer's purchase history is displayed on the terminal.

[1819] Input: Confirmed or revised SQL query

[1820] Output: Execution results obtained from the DWH

[1821] (Application example 1)

[1822] 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."

[1823] Modern logistics centers are required to efficiently manage large volumes of shipping records and inventory data and quickly retrieve the necessary data. In particular, there is a growing need for systems that allow users to easily analyze data, even if they do not have specific SQL skills. However, conventional systems often require users to be familiar with complex database structures, which takes time to learn. It is also difficult to quickly select the appropriate data source. Therefore, there is a need to develop a system that can perform data analysis efficiently.

[1824] 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.

[1825] In this invention, the server includes means for analyzing input requirements using a generation AI, means for selecting an appropriate data mart based on the analyzed requirements, means for automatically generating an SQL query based on the selected data mart, means for presenting the generated SQL query to a user and accepting confirmation and correction by the user, means for executing the confirmed or corrected SQL query and presenting the execution result to the user, and means for automatically acquiring past shipping records and inventory data, specialized for data analysis of a logistics center. This enables users to efficiently analyze data and quickly acquire appropriate data even if they do not have specific SQL skills.

[1826] "Generative AI" is an artificial intelligence technology that analyzes user requirements, selects appropriate data marts, and automatically generates SQL queries.

[1827] A "data mart" is an aggregation of data sets extracted from a large data warehouse to address a specific business need.

[1828] An "SQL query" is a command in the Structured Query Language that describes an instruction to a database.

[1829] A "logistics center" is a facility that receives, stores, organizes, and ships goods and materials.

[1830] "Past shipping records" are historical information on products that have been shipped from the logistics center in the past.

[1831] "Inventory data" is information about the quantity and condition of products currently stored in the logistics center.

[1832] "Natural language processing technology" is a technology that enables computers to understand, analyze, and generate human language.

[1833] "Verify and fix" is the process whereby the user reviews the auto-generated SQL queries and makes changes if necessary.

[1834] This invention relates to a system that utilizes generative AI to improve the efficiency of data analysis in logistics centers. This system is centered around a server and user terminals.

[1835] First, the user inputs the requirements for data analysis from a terminal. For example, they might input a requirement such as "I want to obtain shipping records for each warehouse for the last three months." The server receives the user's requirements and sends them to the generation AI. This generation AI uses natural language processing technology to analyze the requirements and extract the keywords "warehouse," "shipping records," and "three months."

[1836] The generative AI then selects appropriate data marts based on these keywords, recommending, for example, "warehouse data mart" and "shipping record data mart," and the server presents this information to the user.

[1837] The Generative AI then automatically generates SQL queries based on the selected data mart. For example, the following SQL queries are generated:

[1838] SELECT

[1839] warehouse_id,

[1840] shipment_date,

[1841] product_id,

[1842] quantity

[1843] FROM

[1844] warehouse_data_mart

[1845] WHERE

[1846] shipment_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[1847] The server presents the generated SQL query to the user and accepts confirmation and modification by the user. The user confirms the query and makes modifications as necessary. The server then executes the confirmed or modified SQL query and presents the execution results to the user.

[1848] Specifically, it is implemented as shown in the following Python code, which uses an SQLite database and executes queries according to the user's requirements. The generative AI uses OpenAI's GPT model.

[1849] This allows the server to efficiently analyze data and quickly retrieve appropriate data, even if the user does not have specific SQL skills. This makes it easier for distribution centers to quickly check shipping records and manage inventory.

[1850] Examples of prompt statements

[1851] Prompt for the generative AI model:

[1852] Analyze the following requirements: Get shipping records for each warehouse for the last three months

[1853] As described above, the present invention provides a concrete example of a system that simplifies data management at a logistics center and enables efficient data analysis. This system allows users to quickly obtain the data they need without having to worry about complex database operations.

[1854] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1855] Step 1:

[1856] The user inputs the requirements for data analysis from the terminal. For example, they input a requirement such as "I want to obtain the shipping records for each warehouse for the last three months." This requirement is sent to the server.

[1857] Input: The requirement text entered by the user

[1858] Output: The text of the requirement as received by the server

[1859] Step 2:

[1860] The server sends the received requirements to the generation AI, which then analyzes them using natural language processing technology. During the analysis, keywords such as "warehouse," "shipping record," and "3 months" are extracted from the requirements.

[1861] Input: Text of the requirement received from the user

[1862] Output: Generated keyword list (e.g. "warehouse", "shipping record", "3 months")

[1863] Step 3:

[1864] Based on the extracted keywords, the generation AI refers to past consultation history and selects the appropriate data mart. For example, it might recommend a "warehouse data mart" and a "shipping record data mart."

[1865] Input: Keyword list, past consultation history data

[1866] Output: A list of recommended data marts (e.g., "Warehouse Data Mart" and "Shipping Record Data Mart")

[1867] Step 4:

[1868] The server automatically generates SQL queries based on the selected data mart using the AI. The generated SQL queries are, for example, as follows:

[1869] SELECT

[1870] warehouse_id,

[1871] shipment_date,

[1872] product_id,

[1873] quantity

[1874] FROM

[1875] warehouse_data_mart

[1876] WHERE

[1877] shipment_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[1878] Input: Recommended data mart list, keyword list

[1879] Output: Auto-generated SQL query

[1880] Step 5:

[1881] The server presents the generated SQL query to the user, who can review it and make modifications if necessary. The user can review the query and add or modify specific conditions.

[1882] Input: Auto-generated SQL query

[1883] Output: SQL query reviewed and corrected by the user

[1884] Step 6:

[1885] The server executes the verified or modified SQL query and retrieves the results, for example, by querying an SQLite database and retrieving the results.

[1886] Input: Confirmed or revised SQL query

[1887] Output: Query results (data retrieved from the database)

[1888] Step 7:

[1889] The server then presents the results of the query to the user, allowing the user to quickly obtain the data they need.

[1890] Input: Query results

[1891] Output: Execution results presented to the user (e.g., shipping records by warehouse for the past three months)

[1892] 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.

[1893] This invention is a system that enables users of an in-house data warehouse (DWH) to efficiently extract data and perform analytical tasks, and aims to improve the user experience by combining it with an emotion engine that recognizes the user's emotions.

[1894] Overall system configuration

[1895] The system includes the following main components:

[1896] 1. Generation AI

[1897] 2. Natural Language Processing Technology

[1898] 3. Emotion Engine

[1899] 4. Server

[1900] 5. User Device

[1901] Acquiring and analyzing user requirements

[1902] 1. Obtaining user requirements

[1903] The user enters requirements into an input form on the terminal. For example, the user enters a requirement such as "I want to obtain three months' worth of purchase history by customer."

[1904] 2. Emotion recognition

[1905] While the terminal is inputting user requirements, the emotion engine monitors the user's emotional state. For example, if the user is feeling stressed, the emotion engine will recognize this.

[1906] 3. Submit your requirements

[1907] The terminal transmits the input requirement data and the user's emotion data to the server.

[1908] Requirements and sentiment data analysis

[1909] 4. Requirements Analysis

[1910] The server sends the requirements data to the generation AI, which then uses natural language processing technology to extract and structure keywords from the requirements. For example, it identifies keywords such as "by customer," "purchase history," and "three months' worth."

[1911] 5. Emotion Data Analysis

[1912] The server analyzes the user's emotional state based on the data from the emotion engine and determines whether the user is feeling stressed.

[1913] Data mart selection and query generation

[1914] 6. Data Mart Selection

[1915] The generative AI refers to past consultation history and selects the optimal data mart based on the analyzed requirements and emotional state. For example, it recommends a "customer data mart" and a "purchase history data mart."

[1916] 7. Presentation to the User

[1917] The server receives the recommended data mart information from the generation AI and displays it on the device. It also adjusts the interface and provides additional support content according to the user's emotional state.

[1918] 8. Generating SQL Queries

[1919] The AI ​​automatically generates the necessary SQL queries based on the selected data mart. For example, it generates the following query:

[1920] sql

[1921] SELECT

[1922] customer_id,

[1923] purchase_date,

[1924] product_id,

[1925] amount

[1926] FROM

[1927] customer_data_mart

[1928] WHERE

[1929] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[1930] (The code is not listed here)

[1931] Review and run the query

[1932] 9. Check and correct your query

[1933] The server presents the generated SQL query to the user, who confirms it and modifies it if necessary. If the user is feeling nervous or stressed, the emotion engine displays an appropriate support message.

[1934] 10. Executing a Query

[1935] The server sends the confirmed or modified query to the DWH for execution.

[1936] Displaying execution results

[1937] 11. Obtaining and Presenting Results

[1938] The server retrieves the query results from the DWH and displays them on the device. It monitors the user's emotional state and interprets the results and suggests the next action if necessary.

[1939] Specific examples

[1940] For example, if a user enters the requirement "I want to get the purchase history for each customer for the past three months":

[1941] 1. Requirements acquisition: The user enters the requirements into an input form on the terminal.

[1942] 2. Emotion Recognition: The emotion engine recognizes the user's emotions (e.g., stress state).

[1943] 3. Sending requirements and emotion data: The device sends the data to the server.

[1944] 4. Requirements analysis: The server sends the requirements data to the generation AI for analysis.

[1945] 5. Emotional data analysis: Analyze user emotions.

[1946] 6. Data mart selection: Generative AI recommends the optimal data mart.

[1947] 7. Presentation to the user: The server presents emotional support along with data mart information.

[1948] 8. SQL Query Generation: Generative AI generates SQL queries.

[1949] 9. Review and modify the query: The user reviews the query and modifies it if necessary.

[1950] 10. Query execution: The server sends the query to the DWH and executes it.

[1951] 11. Retrieving and presenting results: The server retrieves the results and presents them to the user.

[1952] This system allows users to efficiently retrieve data and perform analysis work without stress, even if they do not have SQL skills.The emotion engine monitors the user's emotional state and provides appropriate support, resulting in a better user experience.

[1953] The processing flow will be explained below.

[1954] Step 1:

[1955] The user enters the requirements into the input form on the terminal. For example, the user enters "I want to obtain three months' worth of purchase history by customer."

[1956] Step 2:

[1957] The emotion engine monitors the user's emotional state, for example, analyzing the user's stress level from facial expressions and tone of voice via the camera and microphone while the user is typing.

[1958] Step 3:

[1959] The terminal transmits the input requirement data and emotion data to the server.

[1960] Step 4:

[1961] The server sends the requirement data and emotion data to the generation AI, which receives the requirement data and begins analyzing it.

[1962] Step 5:

[1963] Generative AI uses natural language processing technology to analyze requirement data and identify and extract keywords and objectives. For example, it identifies "by customer," "purchase history," and "three months' worth."

[1964] Step 6:

[1965] The server receives the analysis results from the generation AI and stores them along with the emotion data.

[1966] Step 7:

[1967] The generation AI refers to past consultation history and selects the appropriate data mart based on the analyzed requirements. For example, it recommends a "customer data mart" and a "purchase history data mart."

[1968] Step 8:

[1969] The server receives the recommendation data mart information from the generation AI and displays it on the device, and also adjusts the interface based on the user's emotional state and displays necessary support messages.

[1970] Step 9:

[1971] The user confirms the presented data mart information and makes a selection to proceed to the next step.

[1972] Step 10:

[1973] The AI ​​automatically generates the necessary SQL queries based on the selected data mart. For example, it generates the following query:

[1974] sql

[1975] SELECT

[1976] customer_id,

[1977] purchase_date,

[1978] product_id,

[1979] amount

[1980] FROM

[1981] customer_data_mart

[1982] WHERE

[1983] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[1984] (The code is not listed here)

[1985] Step 11:

[1986] The server presents the generated SQL query to the user.

[1987] Step 12:

[1988] The user checks the presented SQL query and modifies it as necessary. For example, if they want to extract only a specific product category, they can add that condition.

[1989] Step 13:

[1990] The user submits the modified query to the server.

[1991] Step 14:

[1992] The server sends the confirmed or modified query to the DWH and executes the query.

[1993] Step 15:

[1994] The server retrieves the query results from the DWH.

[1995] Step 16:

[1996] The server sends the results to the device and displays them to the user. It also monitors the user's emotional state again and displays a message to guide the user on the next action based on the results.

[1997] These steps allow users to efficiently extract data and perform data analysis.The emotion engine monitors the user's emotional state in real time and takes appropriate action to provide a better user experience.

[1998] Example 2

[1999] 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."

[2000] Conventional in-house data warehouse (DWH) systems have the drawback of making it difficult for users to efficiently extract and analyze data. Furthermore, simple data processing is often performed without considering the user's emotional state, resulting in a poor user experience. The present invention aims to solve these problems and provide a system that enables users to efficiently extract and analyze data without feeling stressed.

[2001] 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.

[2002] In this invention, the server includes: means for a user to input requirements into an input form on a terminal; means for an emotion engine to monitor the user's emotional state while the user is inputting the user's requirements; means for transmitting the input requirement data and the user's emotional data to the server; means for transmitting the requirement data to a generation AI and extracting and structuring keywords of the requirements using natural language processing technology; means for analyzing the user's emotional state based on the data from the emotion engine and determining whether the user is feeling stressed; means for selecting an optimal data mart based on the analyzed requirements and emotional state by referring to past consultation history; means for receiving recommended data mart information from the generation AI and presenting interface adjustments or additional support content according to the user's emotional state; means for automatically generating a required SQL query based on the selected data mart; means for presenting the generated SQL query to the user and accepting user confirmation and correction; and means for executing the confirmed or corrected SQL query and presenting the execution result to the user. This allows the user to efficiently extract and analyze data, and further enables the emotion engine to improve the user experience.

[2003] A "terminal" is an input / output device that a user uses to input requirements.

[2004] An "emotion engine" is a software and hardware system that monitors and analyzes a user's emotional state in real time.

[2005] "Requirement data" refers to specific requests for data extraction and analysis that are input by the user into the terminal.

[2006] "Emotion data" is data that represents the user's emotional state and is acquired by the emotion engine.

[2007] A "server" is a computer system that receives user requirement data and emotion data and performs various processes based on them.

[2008] "Generative AI" is an artificial intelligence that analyzes requirements data and extracts and structures keywords using natural language processing technology.

[2009] "Natural language processing technology" is information technology used to interpret and understand the meaning of human language.

[2010] A "data mart" is a portion of a database that stores a subset of data related to a particular business process or area.

[2011] An "SQL query" is a command statement written in the Structured Query Language (SQL) for searching and manipulating data in a database management system (DBMS).

[2012] The "past consultation history" is a record of requirements that the user has exchanged with the server in the past and the results of the processing.

[2013] "Adjusting the interface" refers to changing the screen display or user interface configuration in response to the user's emotional state.

[2014] "Additional support content" refers to supplementary information or messages provided depending on the user's emotional state.

[2015] This invention is a system that aims to enable users of in-house data warehouses (DWHs) to efficiently extract data and perform analysis work. Furthermore, by combining it with an emotion engine that recognizes user emotions, we aim to improve the user experience.

[2016] Overall system configuration

[2017] The system includes the following major components:

[2018] 1. Generation AI

[2019] 2. Natural Language Processing Technology

[2020] 3. Emotion Engine

[2021] 4. Server

[2022] 5. User Device

[2023] Hardware and Software

[2024] Terminal: An input device for users to enter requirements. It uses a web browser or GUI form.

[2025] Emotion engine: Uses sensors and software to monitor the user's facial expressions and typing speed in real time and analyze their emotional state. Specific examples include camera sensors and keyboard typing speed analysis software.

[2026] Server: A computer system that receives requirements data and emotion data and analyzes and structures the data using generative AI models and natural language processing techniques.

[2027] Generative AI: An artificial intelligence that analyzes user requirements data and extracts and structures keywords using natural language processing technology.

[2028] Data mart: A portion of a database that stores a subset of data related to a particular business process or area.

[2029] System Operation

[2030] When a user enters requirements into an input form on a terminal, for example, "I would like to obtain the past three months' worth of purchase history by customer," the emotion engine monitors the user's facial expressions and input speed and analyzes their emotional state. The terminal then sends the entered requirement data and emotion data to the server.

[2031] The server sends the received requirements data to the generation AI, which uses natural language processing technology to extract and structure keywords. Based on the extracted keywords and taking into account data from the emotion engine, the generation AI determines whether the user is feeling stressed. The generation AI refers to past consultation history and selects the optimal data mart based on the analyzed requirements and emotional state. The server receives the recommended data mart information and displays it on the device. At this time, it adjusts the interface and presents additional support content according to the user's emotional state.

[2032] The generation AI automatically generates the necessary SQL queries based on the selected data mart. The generated SQL queries are presented to the user, who can confirm and modify them. Finally, the confirmed or modified SQL queries are sent from the server to the DWH, and data is retrieved. The retrieved data is displayed on the terminal, and the user can confirm the results. At this time, the emotion engine displays support messages as needed.

[2033] Specific examples

[2034] 1. Obtaining requirements: The user enters the requirement "I want to obtain the purchase history for each customer for the past three months" into the input form on the terminal.

[2035] 2. Emotion Recognition: The emotion engine monitors the user's facial expressions and input speed to recognize stress levels.

[2036] 3. Sending requirements and emotion data: The terminal sends the requirements data and emotion data to the server.

[2037] 4. Requirements analysis: The generative AI extracts the keywords "by customer," "purchase history," and "three months' worth."

[2038] 5. Emotional data analysis: Determine whether the user is feeling stressed.

[2039] 6. Data mart selection: The generative AI selects the optimal data mart (e.g., "customer data mart" or "purchase history data mart").

[2040] 7. Presentation to the user: The recommended data mart and support content according to the emotional state are displayed on the terminal.

[2041] 8. SQL Query Generation: The generative AI automatically generates SQL queries like the following:

[2042] sql

[2043] SELECT

[2044] customer_id,

[2045] purchase_date,

[2046] product_id,

[2047] amount

[2048] FROM

[2049] customer_data_mart

[2050] WHERE

[2051] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[2052] 9. Review and modify the query: The user reviews the query and modifies it if necessary.

[2053] 10. Query execution: The server sends the confirmed or modified query to the DWH and executes it.

[2054] 11. Obtaining and presenting results: The server displays the obtained data on the terminal, and the user confirms the results.

[2055] Prompt Sentence Examples

[2056] Generate a SQL query to retrieve customer purchase history for the past three months.

[2057] This system allows users to efficiently retrieve data and perform analysis work without stress, even if they do not have SQL skills.The emotion engine monitors the user's emotional state and provides appropriate support, resulting in a better user experience.

[2058] The flow of the identification process in the second embodiment will be described with reference to FIG.

[2059] Step 1:

[2060] The user inputs requirements into an input form on the terminal. Specifically, the user inputs the requirement "I want to obtain the purchase history for each customer for the past three months" into the input form. At this time, the terminal receives the information entered by the user as input data. The input data is requirement information in text format.

[2061] Step 2:

[2062] The emotion engine monitors the user's emotional state while they are entering their requirements. Specifically, a camera sensor analyzes the user's facial expressions and keyboard input speed in real time to generate emotion data. The input is the user's facial image and input speed data, and the output is emotional state data in text format (e.g., stress level).

[2063] Step 3:

[2064] The device sends the input requirement data and user emotion data to the server. The input is the user requirement data and emotion data, and the output is the requirement data and emotion data sent to the server. Specifically, it is sent to the server as JSON format data.

[2065] Step 4:

[2066] The server sends the requirements data to the generation AI, which uses natural language processing technology to extract and structure requirement keywords. The input is the requirements data (e.g., "I want to obtain the purchase history by customer for the past three months"), and the output is the extracted keywords (e.g., "by customer," "purchase history," "three months' worth"). Specifically, the generation AI uses a natural language processing engine to analyze and structure the keywords.

[2067] Step 5:

[2068] The server analyzes the user's emotional state based on data from the emotion engine and determines whether the user is feeling stressed. The input is emotional data (e.g., stress state), and the output is a judgment result of the user's emotional state (e.g., the user is feeling stressed). Specifically, the server runs an emotion analysis algorithm and quantifies the emotional state.

[2069] Step 6:

[2070] The generation AI refers to past consultation history and selects the optimal data mart based on the analyzed requirements and emotional state. The input is requirement keywords and emotional state data, and the output is a recommended data mart (e.g., "customer data mart" or "purchase history data mart"). Specifically, the generation AI runs a data mart selection algorithm to select the optimal data mart.

[2071] Step 7:

[2072] The server receives the recommendation data mart information from the generation AI and displays it on the device. Furthermore, it adjusts the interface and presents additional support content according to the user's emotional state. The input is the recommendation data mart and the user's emotional state, and the output is the information displayed on the device. Specifically, the server receives the recommendation data mart information and adjusts the UI according to the emotional state.

[2073] Step 8:

[2074] The generation AI automatically generates the required SQL queries based on the selected data mart. The input is the selected data mart information and requirement keywords, and the output is the generated SQL query. Specifically, the generation AI automatically generates the following SQL queries:

[2075] sql

[2076] SELECT

[2077] customer_id,

[2078] purchase_date,

[2079] product_id,

[2080] amount

[2081] FROM

[2082] customer_data_mart

[2083] WHERE

[2084] purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[2085] Step 9:

[2086] The server presents the generated SQL query to the user, who confirms it and modifies it if necessary. The input is the generated SQL query, and the output is the SQL query that has been confirmed and modified by the user. Specifically, the server displays the SQL query on the terminal, and the user confirms and modifies the query content.

[2087] Step 10:

[2088] The server sends the confirmed or modified SQL query to the DWH and executes it. The input is the confirmed or modified SQL query, and the output is the query execution result from the DWH. In concrete terms, the server sends the SQL query to the data warehouse, executes the query, and retrieves the data.

[2089] Step 11:

[2090] The server retrieves the query execution results from the DWH and displays them on the terminal. It monitors the user's emotional state and, if necessary, interprets the results and suggests the next action. The input is the query execution results and the user's emotional state, and the output is the execution results and a support message displayed on the terminal. In concrete terms, the server retrieves the query results, and if the user is feeling stressed, it displays an interpretation support message.

[2091] (Application example 2)

[2092] 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."

[2093] With conventional data analysis systems, if users lacked SQL skills or database knowledge, it was difficult to efficiently extract and analyze data. Furthermore, because data was provided without taking into account the user's emotional state, the user interface and support were often inappropriate. This resulted in a poor user experience and reduced productivity, especially when tasks were complex or under stress.

[2094] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for analyzing input requirements using a generation AI, means for selecting an appropriate data mart based on the analyzed requirements, means for automatically generating an SQL query based on the selected data mart, means for presenting the generated SQL query to a user and accepting confirmation and correction by the user, means for executing the confirmed or corrected SQL query and presenting the execution result to the user, an emotion recognition engine for monitoring the user's emotional state, and means for generating and presenting a response method based on the emotional state. This enables users to efficiently extract and analyze data without advanced technical knowledge, and furthermore, by providing support appropriate to the user's emotional state, it is possible to reduce stress and improve the user experience.

[2095] "Generative AI" is an artificial intelligence technique used to analyze data and automate tasks.

[2096] The "means for analyzing requirements" is a method for analyzing the requirements data entered by the user and extracting important keywords and intentions.

[2097] The "means for selecting an appropriate data mart" is a method for selecting the most relevant data mart based on the analyzed requirements.

[2098] "Means for automatically generating SQL queries" refers to a method for automatically creating the necessary SQL queries for the selected data mart.

[2099] The "means for presenting the generated SQL query to the user and accepting confirmation and correction by the user" refers to a method for displaying the generated SQL query to the user so that the user can confirm or correct it.

[2100] "Means for executing a confirmed or modified SQL query and presenting the execution results to a user" refers to a method for executing an SQL query confirmed or modified by a user and displaying the results to the user.

[2101] The "emotion recognition engine that monitors the user's emotional state" is a system that analyzes the user's emotions in real time and determines their state.

[2102] The "means for generating and presenting a response method based on an emotional state" is a method for generating an optimal response method according to the emotional state of the user determined by an emotion recognition engine and presenting it to the user.

[2103] As an embodiment of the present invention, a customer service support system using smart glasses for store clerks in a brick-and-mortar store can be adopted. The following describes the specific roles of the server, terminal, and user.

[2104] server

[2105] The server performs the following operations using the following specific hardware and software:

[2106] 1. Generative AI and Natural Language Processing Technology

[2107] Software used: Python language, natural language processing library (e.g., SpaCy)

[2108] The server acquires the user's requirements and uses generation AI to extract and analyze keywords from the requirements.

[2109] For example, from the requirement "Please tell me your purchase history for the past three months," extract keywords such as "past three months" and "purchase history."

[2110] 2. Emotion Recognition Engine

[2111] Hardware used: Smart glasses camera

[2112] Software used: Emotion-FerPlus

[2113] The server analyzes the video data acquired through the glasses' camera and monitors the customer's emotional state in real time.

[2114] For example, determining whether a customer is stressed.

[2115] 3. Data mart selection and SQL query generation

[2116] Software used: SQLAlchemy

[2117] The server selects a data mart based on the analyzed requirements and automatically generates the necessary SQL queries.

[2118] For example, generate a query to extract data from a purchase history data mart within a specific period.

[2119] Terminal (smart glasses)

[2120] The smart glasses worn by the store clerk function as a terminal and support the following processes:

[2121] 1. Input customer requirements

[2122] Enter customer requirements using voice input or touch operation.

[2123] For example, a request such as "I want to know the purchase history for the past three months" is input.

[2124] 2. Displaying emotional states

[2125] The analysis results of the emotion recognition engine are displayed in real time.

[2126] For example, if a customer is feeling stressed, the glasses' display will say "Customer is feeling stressed."

[2127] 3. Displaying query results and suggesting how to respond

[2128] The data returned from the server and the response method based on the emotional state are displayed on the glasses' display.

[2129] For example, along with the purchase history as a query result, the message "The customer is stressed. Please respond calmly" may be displayed.

[2130] User (store clerk)

[2131] Store staff use the above system to efficiently handle customer service.

[2132] 1. Obtaining and inputting user requirements

[2133] Customer requirements are registered into the smart glasses using voice or touch input.

[2134] An example of a specific prompt might be: "The customer wants purchase history data for the past three months. Please generate a query."

[2135] 2. Checking and responding to emotional states

[2136] Check the customer's emotional state in real time and respond accordingly.

[2137] Specific example of how to respond: "The customer is stressed. Please respond calmly."

[2138] This system allows store clerks to efficiently obtain customer information without having SQL skills and to respond optimally according to the customer's emotional state, significantly improving customer satisfaction and store clerk work efficiency.

[2139] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[2140] Step 1:

[2141] The user uses the smart glasses to input customer requirements. At this time, the user can use voice input or touch operation. Specifically, the user can say, "I would like to know my purchase history for the past three months." The input data is acquired through the smart glasses' built-in microphone or touchpad. The input content is converted into text format and sent to the server.

[2142] Step 2:

[2143] The server analyzes the received user requirements using a generative AI model and natural language processing technology. The input data is in text format, and the generative AI model extracts keywords. In this process, keywords such as "past 3 months" and "purchase history" are identified.

[2144] Step 3:

[2145] The server selects an appropriate data mart based on the analysis results. The server refers to past consultation history and recommends a data mart that matches the analyzed keywords. For example, the "purchase history data mart" may be selected. This information is also returned to the user in text format.

[2146] Step 4:

[2147] The server automatically generates SQL queries using the selected data mart. The generative AI model creates appropriate queries based on keywords, for example, the following SQL queries are generated:

[2148] sql

[2149] SELECT customer_id, purchase_date, product_id, amount FROM purchase_history WHERE purchase_date >= DATE_SUB(CURDATE(), INTERVAL 3 MONTH);

[2150] The generated query is presented to the user in text form.

[2151] Step 5:

[2152] The user checks the generated SQL query through the smart glasses and makes corrections if necessary. The user visually checks the query displayed on the screen, accepts it as is if there are no problems, and makes corrections by voice or touch if necessary. The corrected query is then sent back to the server.

[2153] Step 6:

[2154] The server executes the verified or modified SQL query and retrieves the results from the data warehouse. The results are then retrieved from the database and converted to text format, which is then sent back to the user.

[2155] Step 7:

[2156] At the same time, the server monitors the customer's emotional state through the smart glasses' camera, and an emotion recognition engine analyzes the image data in real time to determine the customer's state, such as "feeling stressed."

[2157] Step 8:

[2158] The server generates the optimal response based on the customer's emotional state and displays it on the smart glasses. The generative AI model then uses the emotional data to devise a response method. For example, a message such as "The customer is feeling stressed. Please respond calmly" is displayed.

[2159] Step 9:

[2160] The user checks the execution results and response methods displayed on the smart glasses display and responds appropriately to the customer. Based on the results, the necessary information is conveyed to the customer, and appropriate communication is carried out based on emotional data. This optimizes customer response.

[2161] 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.

[2162] 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.

[2163] 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 robot 414.

[2164] 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.

[2165] 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.

[2166] 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.

[2167] 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).

[2168] 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.

[2169] 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."

[2170] 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.

[2171] 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).

[2172] 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.

[2173] 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.

[2174] 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.

[2175] 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.

[2176] 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.

[2177] 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.

[2178] 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.

[2179] 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.

[2180] 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.

[2181] 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.

[2182] The following is further disclosed regarding the above embodiment.

[2183] (Claim 1)

[2184] A means for analyzing input requirements using a generative AI;

[2185] A means for selecting an appropriate data mart based on the analyzed requirements;

[2186] A means for automatically generating SQL queries based on the selected data mart;

[2187] means for presenting the generated SQL query to a user and accepting user confirmation and modification;

[2188] means for executing the confirmed or modified SQL query and presenting the execution results to the user;

[2189] A system including:

[2190] (Claim 2)

[2191] 2. The system according to claim 1, wherein the analyzing means analyzes the user's requirements using natural language processing technology.

[2192] (Claim 3)

[2193] 2. The system according to claim 1, wherein the means for selecting a data mart refers to past consultation history to recommend an appropriate data mart.

[2194] "Example 1"

[2195] (Claim 1)

[2196] A means for analyzing input requirements using a generative AI;

[2197] a means for selecting an appropriate data structure based on the analyzed requirements;

[2198] means for automatically generating a data extraction query based on the selected data structure;

[2199] means for presenting the generated data extraction query to a user and accepting confirmation and modification by the user;

[2200] means for executing the confirmed or modified data extraction query and presenting the execution results to the user;

[2201] A system including:

[2202] (Claim 2)

[2203] 2. The system according to claim 1, wherein the analyzing means analyzes the user's requirements using natural language processing technology.

[2204] (Claim 3)

[2205] 2. The system according to claim 1, wherein the means for selecting a data structure refers to past correspondence history to recommend an appropriate data structure.

[2206] "Application Example 1"

[2207] (Claim 1)

[2208] A means for analyzing input requirements using a generative AI;

[2209] A means for selecting an appropriate data mart based on the analyzed requirements;

[2210] A means for automatically generating SQL queries based on the selected data mart;

[2211] means for presenting the generated SQL query to a user and accepting user confirmation and modification;

[2212] means for executing the confirmed or modified SQL query and presenting the execution results to the user;

[2213] Specializing in data analysis for logistics centers, it provides a means to automatically obtain past shipping records and inventory data,

[2214] A system including:

[2215] (Claim 2)

[2216] 2. The system according to claim 1, wherein the analyzing means analyzes the user's requirements using natural language processing technology.

[2217] (Claim 3)

[2218] 2. The system according to claim 1, wherein the means for selecting a data mart refers to past consultation history to recommend an appropriate data mart.

[2219] "Example 2: Combining Emotion Engines"

[2220] (Claim 1)

[2221] A means for a user to input requirements into an input form on the terminal;

[2222] means for the emotion engine to monitor the user's emotional state during input of the user requirements;

[2223] means for transmitting the input requirement data and user emotion data to a server;

[2224] A means of sending requirement data to a generation AI and extracting and structuring requirement keywords using natural language processing technology;

[2225] a means for analyzing the user's emotional state based on data from the emotion engine and determining whether the user is experiencing stress;

[2226] A means for selecting the optimal data mart based on the requirements and emotional state analyzed by referring to the past consultation history;

[2227] A means for receiving recommendation data mart information from the generative AI and presenting interface adjustments and additional support content according to the user's emotional state;

[2228] A means to automatically generate the required SQL queries based on the selected data mart;

[2229] means for presenting the generated SQL query to a user and accepting user confirmation and modification;

[2230] means for executing the confirmed or modified SQL query and presenting the execution results to the user;

[2231] A system including:

[2232] (Claim 2)

[2233] 10. The system of claim 1, wherein natural language processing techniques are used to analyze the user's requirements.

[2234] (Claim 3)

[2235] The system according to claim 1, which recommends an appropriate data mart by referring to past consultation history.

[2236] "Application example 2 when combining emotion engines"

[2237] Extracting new parts

[2238] The novel aspects of the application are:

[2239] 1. An emotion recognition engine that monitors the emotional state of the customer.

[2240] 2. The process of generating a response based on emotional state.

[2241] Based on this, we will create a claim.

[2242] Rewritten claims

[2243] (Claim 1)

[2244] A means for analyzing input requirements using a generative AI;

[2245] A means for selecting an appropriate data mart based on the analyzed requirements;

[2246] A means for automatically generating SQL queries based on the selected data mart;

[2247] means for presenting the generated SQL query to a user and accepting user confirmation and modification;

[2248] means for executing the confirmed or modified SQL query and presenting the execution results to the user;

[2249] an emotion recognition engine that monitors the user's emotional state;

[2250] means for generating and presenting a response based on the emotional state;

[2251] A system including:

[2252] (Claim 2)

[2253] 2. The system according to claim 1, wherein the analyzing means analyzes the user's requirements using natural language processing technology.

[2254] (Claim 3)

[2255] 2. The system according to claim 1, wherein the means for selecting a data mart refers to past consultation history to recommend an appropriate data mart. [Explanation of symbols]

[2256] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means for analyzing input requirements using a generative AI; A means for selecting an appropriate data mart based on the analyzed requirements; A means for automatically generating SQL queries based on the selected data mart; means for presenting the generated SQL query to a user and accepting user confirmation and modification; means for executing the confirmed or modified SQL query and presenting the execution results to the user; A system including:

2. 2. The system according to claim 1, wherein the analyzing means analyzes the user's requirements using natural language processing techniques.

3. 2. The system according to claim 1, wherein the means for selecting a data mart refers to a past consultation history to recommend an appropriate data mart.

Images