System

Abstract

To provide a system.SOLUTION: A system comprises: means of receiving analysis indication in Japanese, performing corresponding data analysis, and visualizing its result; a feeling engine that recognizes a feeling of a customer on the basis of a tone of voice and an expression of the customer existing in a real store; means of performing a proposal for sales promotion and customer acquisition that are performed for the customer on the basis of customer reception method considering the recognized feeling of the customer and the recognized feeling of the customer; and means of proposing an optimum advertisement on the basis of the recognized feeling of the customer by inputting a specific prompt sentence into a generative AI model.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] There is a need for systems that can easily analyze and utilize data, even for people without SQL knowledge or visualization skills using BI tools, and that can constantly provide new insights based on patterns and trends in the data and the analysis instructions. [Means for solving the problem]

[0005] The system disclosed herein includes a means for accepting analysis instructions in Japanese, performing corresponding data analysis, and visualizing the results; an emotion engine for recognizing customer emotions based on the tone of voice and facial expression of the customer in a physical store; a customer service method that takes into account the recognized customer emotions; a means for making suggestions to the customer for sales promotion and customer acquisition based on the recognized customer emotions; and a means for suggesting optimal advertisements based on the recognized customer emotions by inputting specific prompt sentences into a generative AI model. [Brief explanation of the drawings]

[0006] [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. 2 is a sequence diagram showing a flow of processing in the data processing system according to the first embodiment of the first form example. [Figure 12]FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Embodiment 1. [Figure 13] FIG. 10 is a sequence diagram showing a processing flow of a data processing system in a second embodiment of the second form example. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 of Embodiment Example 2. [Figure 15] FIG. 10 is a sequence diagram showing the flow of processing in a data processing system according to a third embodiment of the third embodiment. [Figure 16] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 3 of Embodiment 3. [Figure 17] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the first embodiment of the first form example when an emotion engine is combined. [Figure 18] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Form Example 1 when an emotion engine is combined. [Figure 19] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the second embodiment of the second form example when an emotion engine is combined. [Figure 20] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 of Form Example 2 when an emotion engine is combined. [Figure 21] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the third embodiment of the third form example when an emotion engine is combined. [Figure 22] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 3 of Form Example 3 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0007] 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.

[0008] First, the terms used in the following description will be explained.

[0009] 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, the 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), an APU (Accelerated Processing Unit), or a TPU (TENSOR PROCESSING UNIT (registered trademark)).

[0010] 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.

[0011] 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.

[0012] 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.

[0013] 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."

[0014] [First embodiment]

[0015] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0016] 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.

[0017] 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).

[0018] 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.

[0019] 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.

[0020] 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.

[0021] 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.

[0022] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0023] 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.

[0024] 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.

[0025] 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.

[0026] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0027] "Example 1"

[0028] This embodiment of the present invention provides an interface that allows a user to input analysis instructions in Japanese. This interface can include a text input field, a voice input function, etc. For example, if a user inputs an instruction such as "Show me monthly sales trends," the system will analyze the corresponding data and visualize the results in the form of graphs or charts.

[0029] "Example 2"

[0030] Furthermore, the present invention has a function of making marketing proposals and audience targeting based on the analysis results. For example, based on the analysis result that sales of a specific product are declining, a promotion targeting that product is proposed. Also, based on the analysis result that a specific customer segment tends to purchase a specific product, a marketing activity targeting that customer segment is proposed. In the present disclosure, a customer may be interpreted as a user, and a user may be interpreted as a customer.

[0031] "Example 3"

[0032] This embodiment of the present invention also has the ability to provide new insights based on patterns and trends in data and instructed analyses. For example, it predicts what information the user will want to know next based on patterns in past analysis instructions and analysis results, and provides that information. This allows the user to obtain the information they need without having to specifically specify what they are looking for.

[0033] The processing flow of each embodiment will be described below.

[0034] "Example 1"

[0035] Step 1: The user inputs an analysis instruction in Japanese through the interface. For example, the user inputs an instruction such as "Show me the monthly sales trend."

[0036] Step 2: The system analyzes the relevant data. This analysis involves extracting the necessary data from the database and calculating monthly sales data.

[0037] Step 3: The system visualizes the analysis results, which includes displaying monthly sales data in the form of graphs and charts.

[0038] "Example 2"

[0039] Step 1: The system makes marketing proposals and audience targeting based on the analysis results. For example, if the analysis results show that sales of a particular product are declining, the system will propose a promotion targeting that product.

[0040] Step 2: The system performs audience targeting based on the analysis results. For example, if the analysis results show that a certain customer group tends to purchase a certain product, it will suggest marketing activities targeted at that customer group.

[0041] "Example 3"

[0042] Step 1: The system provides new insights based on patterns and trends in the data and the analysis instructions. For example, it predicts what the user will want to know next based on patterns in past analysis instructions and results.

[0043] Step 2: The system provides the user with the predicted information, allowing the user to get the information they need without having to specifically specify what they want.

[0044] Example 1

[0045] Next, a description will be given of Example 1 of Form 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."

[0046] Conventional data analysis systems have the problem that they are difficult to operate unless the user has specialized knowledge, and that it takes a lot of time and effort to acquire, process, and visualize data. Furthermore, marketing proposals, audience targeting, and new insights based on the analysis results are often not fully implemented.

[0047] 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.

[0048] In this invention, the server includes means for a user to input analysis instructions in natural language, means for analyzing the user instructions using a generative AI model, means for retrieving necessary data from a database based on the analysis results, means for formatting the retrieved data using a data processing tool, means for converting the formatted data into graphs or charts using a visualization tool, means for sending and displaying the generated graphs or charts on the user's terminal, means for making marketing proposals and audience targeting based on the analysis results, and means for providing new insights based on patterns and trends in the data and the instructed analysis. This enables users to easily perform data analysis even without specialized knowledge and to obtain marketing proposals and new insights based on the results.

[0049] A "user" is an entity that operates the system and inputs analysis instructions in natural language.

[0050] A "natural language" is a language that humans use on a daily basis and does not require a specific programming language or specialized terminology.

[0051] A "generative AI model" is a model that uses artificial intelligence technology to analyze natural language and understand user instructions.

[0052] A "database" is a system for systematically storing data and retrieving the data as needed.

[0053] A "data processing tool" is software that formats acquired data and converts it into a format suitable for analysis and visualization.

[0054] A "visualization tool" is software that converts formatted data into a visual format such as a graph or chart.

[0055] A "marketing proposal" is a proposal of strategies and measures for sales promotion and customer acquisition based on the results of analysis.

[0056] "Audience targeting" refers to conducting marketing activities aimed at specific customer groups or market segments.

[0057] "New insights" are new findings and understandings that are gained based on patterns and trends in data and analysis.

[0058] The present invention is a system in which a user inputs analysis instructions in natural language, and data is acquired, processed, and visualized based on the instructions. Specific embodiments of this system are described below.

[0059] Users input analysis instructions in natural language through the system's interface, which includes text input fields and voice input capabilities. For example, a user might type, "Show me monthly sales trends."

[0060] The server receives instructions from the user. The server analyzes the user's instructions using a generative AI model (e.g., GPT-4 (registered trademark)). As a result of the analysis, it identifies what kind of data analysis the user's instructions require.

[0061] Next, the server retrieves the necessary data from a database based on the analysis results. For example, it retrieves the relevant monthly sales data from a database (e.g., MySQL (registered trademark) or PostgreSQL) that stores sales data.

[0062] The acquired data is then formatted using a data processing tool (e.g., the Pandas library), aggregated monthly, and converted into a format suitable for analysis.

[0063] The formatted data is then converted into graphs and charts using a visualization tool (e.g., Matplotlib or Plotly). For example, a line graph showing monthly sales trends is generated.

[0064] The generated graphs and charts are sent from the server to the user's device, which then displays them on the user's device's interface, allowing the user to visually confirm the analysis results based on the instructions.

[0065] As a specific example, if a user inputs "Show me monthly sales trends," the server will process the data using the following procedure.

[0066] 1. The user enters "Show me monthly sales trends" in the text input field and clicks the submit button.

[0067] 2. The server uses a generative AI model (GPT-4) to analyze this instruction and determine that the request is for monthly sales data.

[0068] 3. The server retrieves sales data from the database (MySQL) and aggregates it monthly using the Pandas library.

[0069] 4. The server generates a line graph of monthly sales using Matplotlib.

[0070] 5. The server sends the generated graph to the user's device, which displays it on the interface.

[0071] An example of a prompt is as follows:

[0072] "Show me the monthly sales trends."

[0073] "Compare this month's sales with the same month last year's sales."

[0074] "Show me sales trends for a specific product"

[0075] This allows users to easily analyze data without specialized knowledge and to make marketing proposals and obtain new insights based on the results.

[0076] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0077] Step 1:

[0078] The user inputs analysis instructions in natural language.

[0079] Specific behavior: The user enters "Show me monthly sales trends" into the text input field in the system interface and clicks the submit button.

[0080] Input: User's natural language instructions for analysis.

[0081] Output: User instructions are sent to the server.

[0082] Step 2:

[0083] The server receives the user's instructions.

[0084] Specific operations: The server receives instructions sent by the user.

[0085] Input: User's natural language instructions for analysis.

[0086] Output: Received instructions are passed to the generative AI model.

[0087] Step 3:

[0088] The server uses the generative AI model to analyze the user's instructions.

[0089] Specific operation: The server inputs the user's instruction into a generative AI model (e.g., GPT-4) and determines that the instruction "Show me monthly sales trends" is requesting monthly sales data trends.

[0090] Input: User's natural language instructions for analysis.

[0091] Output: The analysis results identify the necessary data analysis.

[0092] Step 4:

[0093] The server retrieves the necessary data from the database based on the analysis results.

[0094] Specific operation: The server generates an SQL query and executes a query such as "SELECT FROM sales WHERE date BETWEEN (registered trademark) '2022-01-01' AND '2022-12-31'" against a database (e.g., MySQL).

[0095] Input: The data analysis results identified.

[0096] Output: Sales data retrieved from the database.

[0097] Step 5:

[0098] The data acquired by the server is formatted using a data processing tool.

[0099] Specific operation: The server creates a data frame using the Pandas library and executes code to aggregate sales by month, for example, by executing "df.groupby('month').sum()".

[0100] Input: Sales data retrieved from the database.

[0101] Output: Formatted monthly sales data.

[0102] Step 6:

[0103] The server converts the formatted data into graphs and charts using a visualization tool.

[0104] Specific operation: The server executes code that generates a line graph using Matplotlib, for example, by executing "plt.plot(df['month'], df['sales'])".

[0105] Input: Formatted monthly sales data.

[0106] Output: The generated line graph.

[0107] Step 7:

[0108] The server sends the generated graphs and charts to the user's device for display.

[0109] Specific operation: The server saves the generated graph as an image file and sends its URL to the user's device, which uses the URL to display the graph on the interface.

[0110] Input: The generated line chart.

[0111] Output: The graph displayed on the user's device.

[0112] (Application example 1)

[0113] Next, a description will be given of Application Example 1 of Embodiment 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."

[0114] Logistics centers are required to analyze data such as inventory status and shipping status in real time and visualize it quickly and accurately. However, with conventional systems, data acquisition and analysis takes time, making efficient business operations difficult. In addition, few systems allow for voice input, making them difficult to use. To solve these issues, a system is needed that has an interface that includes voice input and can analyze and visualize data in real time.

[0115] 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.

[0116] In this invention, the server includes means for accepting analysis instructions in Japanese, performing the corresponding data analysis, and visualizing the results; means for making marketing proposals and audience targeting based on the analysis results; means for providing new insights based on patterns and trends in the data and the instructed analysis; means for converting voice input into text; means for retrieving necessary data from a database and analyzing it in real time; and means for displaying the analysis results in graphs and charts. This enables voice input instructions at logistics centers, and enables real-time data analysis and visualization. This improves operational efficiency and speeds decision-making.

[0117] "Analysis instructions in Japanese" means that the user gives instructions for data analysis in Japanese.

[0118] "Data analysis" is the analysis of collected data and the extraction of meaningful information.

[0119] "Visualization" is the presentation of the results of data analysis in a visual format such as a graph or chart.

[0120] "Marketing proposals" refer to proposals regarding sales promotion and market strategies based on the results of the analysis.

[0121] "Audience targeting" refers to conducting marketing activities aimed at a specific customer demographic.

[0122] "New insights" refer to new findings and understanding gained as a result of data analysis.

[0123] "Voice input" means that the user inputs instructions by voice through a microphone.

[0124] "Convert to text" means converting voice-input instructions into text information.

[0125] A "database" is a system that stores data systematically and allows it to be searched and retrieved as needed.

[0126] "Real-time analysis" means analyzing data as it is generated.

[0127] "Graphs and charts" are diagrams that visually represent data.

[0128] A system for implementing this invention is configured as follows: The server has means for accepting analysis instructions in Japanese, executing the corresponding data analysis, and visualizing the results. The user can use a smartphone to give instructions by voice input or text input. The voice input is converted into text using the SpeechRecognition library. The converted text instructions are sent to the server, which retrieves the necessary data from a database.

[0129] The server analyzes the data retrieved from the database in real time and visualizes the analysis results in graphs and charts using the Matplotlib library. The analysis results are then displayed on the smartphone screen.

[0130] For example, if a user says "Tell me the shipping status this month," the server retrieves shipping data and displays the monthly shipping status in a bar graph. If a user says "Show me the inventory trend," the server retrieves inventory data and displays the monthly inventory trend in a line graph.

[0131] The system's hardware includes a smartphone (microphone, speaker, and display) and a server, while the software uses Python, the SpeechRecognition library, the Matplotlib library, the Pandas library, and SQLite.

[0132] An example of a prompt is as follows:

[0133] "Please tell me the shipping status for this month."

[0134] "Show me the inventory trends."

[0135] "Show me last month's stock availability."

[0136] This system allows for voice input of instructions at logistics centers and enables real-time data analysis and visualization, which improves operational efficiency and enables faster decision-making.

[0137] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0138] Step 1:

[0139] The user gives instructions by voice input or text input using a smartphone.

[0140] Input: User's voice or text instructions

[0141] Output: Audio data or text data

[0142] Specific action: The user speaks into the smartphone microphone and says, "Tell me about this month's shipping status," or types the same thing into the text field.

[0143] Step 2:

[0144] The device converts the voice input into text.

[0145] Input: Audio data

[0146] Output: Text data

[0147] Specific operation: The device uses the SpeechRecognition library to convert voice data into text data. For example, the speech "Tell me about this month's shipping status" is converted into text "Tell me about this month's shipping status."

[0148] Step 3:

[0149] The terminal transmits the text data to the server.

[0150] Input: Text data

[0151] Output: Request to server

[0152] Specific operation: The terminal sends the converted text data to the server as an HTTP request.

[0153] Step 4:

[0154] The server parses the text data and generates the appropriate SQL queries.

[0155] Input: Text data

[0156] Output: SQL query

[0157] Specific operation: The server parses the text data "Tell me about this month's shipping status" and generates an SQL query to retrieve shipping data. For example, the query generated is "SELECT FROM shipments WHERE date >= '2023-01-01'".

[0158] Step 5:

[0159] The server retrieves the required data from the database.

[0160] Input: SQL query

[0161] Output: Data retrieved from the database

[0162] Specific operation: The server executes the generated SQL query and retrieves the relevant shipping data from the database. For example, shipping data from January 2023 onwards is retrieved.

[0163] Step 6:

[0164] The server analyzes the acquired data in real time.

[0165] Input: Data retrieved from a database

[0166] Output: Analysis results

[0167] Specific operation: The server analyzes the acquired shipping data using the Pandas library and calculates the monthly shipping status.

[0168] Step 7:

[0169] The server visualizes the analysis results in graphs and charts.

[0170] Input: Analysis results

[0171] Output: Graphs and charts

[0172] Specific operation: The server uses the Matplotlib library to visualize the analysis results as bar graphs and line graphs. For example, monthly shipping status is displayed as a bar graph.

[0173] Step 8:

[0174] The server sends the visualized results to the terminal.

[0175] Input: Graphs and charts

[0176] Output: Response to the terminal

[0177] Specific operation: The server sends the generated graphs and charts to the terminal as image data.

[0178] Step 9:

[0179] The terminal displays the visualized results on the display.

[0180] Input: Image data

[0181] Output: Graphs and charts displayed on the screen

[0182] Specific operation: The device displays the received image data on the screen and provides the results to the user. For example, a bar graph of monthly shipment status is displayed on the smartphone screen.

[0183] Example 2

[0184] Next, a description will be given of Example 2 of Form 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."

[0185] With conventional data analysis systems, data collection, preprocessing, analysis, and the generation and notification of marketing proposals are rarely performed consistently, and each process is fragmented, making it difficult to carry out efficient marketing activities.In addition, the provision of new insights based on analysis results is insufficient, making it difficult to optimize marketing strategies.

[0186] 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.

[0187] In this invention, the server includes means for accepting analysis instructions in Japanese, executing the corresponding data analysis, and visualizing the results, means for making marketing proposals and audience targeting based on the analysis results, means for providing new insights based on patterns and trends in the data and the instructed analysis, means for collecting sales data and customer data from a database, means for preprocessing the collected data, means for analyzing the preprocessed data, means for generating marketing proposals based on the analysis results, and means for notifying the generated marketing proposals. This makes it possible to consistently perform a series of processes from data collection to generating and notifying marketing proposals, enabling efficient marketing activities and the provision of new insights.

[0188] "Analysis instructions in Japanese" refers to a user issuing a data analysis request to the system in Japanese.

[0189] "Data analysis" refers to the process of using collected data to find specific patterns or trends.

[0190] "Visualization" refers to displaying the results of data analysis in a visual format such as a graph or chart.

[0191] "Marketing proposal" refers to proposing a strategy or action plan for promoting a specific product or service based on the results of data analysis.

[0192] "Audience targeting" refers to conducting marketing activities aimed at a specific customer demographic.

[0193] "New insights" refers to new findings and discoveries gained as a result of data analysis.

[0194] A "database" refers to a system for systematically storing and managing data.

[0195] "Sales Data" means data containing information regarding sales of products.

[0196] "Customer data" refers to data that includes information about customer attributes and purchasing history.

[0197] "Preprocessing" refers to cleaning and shaping data before analyzing it.

[0198] "Analysis results" refers to the conclusions and insights obtained through the data analysis process.

[0199] "Notification" refers to informing the user of generated marketing proposals and analysis results.

[0200] The present invention relates to a system for consistently performing data analysis and marketing proposals. A specific embodiment of this system will be described below.

[0201] First, the server collects sales and customer data from a database, which stores product sales information and customer purchase history. The server uses SQL queries to retrieve the required data.

[0202] Next, the server preprocesses the collected data using the Python Pandas library. Specifically, it performs tasks such as filling in missing values, removing outliers, and normalizing the data. This improves the quality of the data and increases the accuracy of the analysis.

[0203] Once the preprocessing is complete, the data is then used for data analysis. The server uses the Scikit-learn library to analyze sales trends and cluster customer purchasing behavior. For example, it analyzes whether sales of a particular product have declined over the past few months. It also uses clustering techniques to identify whether a particular customer demographic tends to purchase a particular product.

[0204] Based on the analysis results, the server generates marketing proposals. For example, it may propose discount campaigns for products with declining sales. It may also propose marketing activities targeted at specific customer segments. This allows for the creation of efficient marketing strategies.

[0205] Finally, the server notifies the user of the generated marketing proposals via email or dashboard. By receiving these notifications, the user can take appropriate marketing actions.

[0206] As a concrete example, consider the following scenario.

[0207] 1. The server collects sales data from the database for the past year.

[0208] 2. Use the Pandas library to impute missing values ​​and normalize the data.

[0209] 3. Use the Scikit-learn library to perform sales trend analysis, for example, discovering that sales of a particular product A have dropped by 20% over the past three months.

[0210] 4. The server generates a promotion proposal to address the decline in sales of product A. For example, it makes a proposal such as "Implement a discount campaign for product A."

[0211] 5. You also analyze customer data and discover that specific customer group B tends to purchase product A.

[0212] 6. The server proposes marketing activities targeting customer group B. For example, it may propose "send a special offer for product A to customer group B."

[0213] Examples of prompts to input to a generative AI model include:

[0214] "Analyze sales data from the past year to identify products with declining sales. Also, generate promotional suggestions for those products."

[0215] Using this prompt, the generative AI model can make appropriate analysis and suggestions.

[0216] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0217] Step 1:

[0218] The server collects sales and customer data from the database. It receives database connection information and SQL queries as input and obtains sales and customer data as output. Specifically, the server connects to the database and executes SQL queries to retrieve the required data.

[0219] Step 2:

[0220] The server preprocesses the collected data. It receives sales and customer data as input and produces preprocessed data as output. Specifically, the server uses the Python Pandas library to impute missing values, remove outliers, and normalize the data.

[0221] Step 3:

[0222] The server analyzes the preprocessed data. It receives the preprocessed data as input and generates analytical results as output. Specifically, the server uses the Scikit-learn library to perform sales trend analysis and cluster customer purchasing behavior.

[0223] Step 4:

[0224] The server generates marketing proposals based on the analysis results. It receives the analysis results as input and generates marketing proposals as output. Specifically, the server proposes promotions for products with declining sales and proposes marketing activities targeting specific customer segments.

[0225] Step 5:

[0226] The server notifies the user of the generated marketing proposal. It receives the marketing proposal as input and sends a notification to the user as output. Specifically, the server notifies the user of the proposal content by email or dashboard.

[0227] (Application example 2)

[0228] Next, a description will be given of Application Example 2 of Form 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."

[0229] In conventional marketing systems, sales data analysis and customer segmentation are often done manually, resulting in inefficiencies. Furthermore, because promotion strategies for declining sales products and targeted advertising proposals are not made in real time, it is difficult to respond quickly. Furthermore, there is a lack of a way to quickly notify marketing personnel of the generated proposals, which means that marketing activities cannot be carried out at the appropriate time.

[0230] 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 accepting analysis instructions in Japanese, executing the corresponding data analysis, and visualizing the results, means for making marketing proposals and audience targeting based on the analysis results, means for providing new insights based on patterns and trends in the data and the instructed analysis, means for analyzing sales data in real time and proposing promotion strategies for products with declining sales, means for identifying customer segments using a clustering algorithm and proposing targeted advertising, and means for notifying the generated proposals. This enables efficient analysis of sales data and rapid proposals of promotion strategies and targeted advertising.

[0231] "Analysis instructions in Japanese" are data analysis instructions given by the user in Japanese.

[0232] "Data analysis" is the analysis of collected data using statistical methods and algorithms to extract useful information and patterns.

[0233] "Visualization" is the presentation of the results of data analysis in a visual format such as a graph or chart.

[0234] A "marketing proposal" is a proposal of strategies or measures aimed at promoting sales of specific products or services based on the results of analysis.

[0235] "Audience targeting" refers to conducting marketing activities aimed at a specific customer demographic.

[0236] "New insights" are new findings and understandings gained based on data analysis and pattern recognition.

[0237] "Sales data" is numerical information about the sales of a particular product or service.

[0238] "Real-time analysis" means analyzing data as it is generated.

[0239] A "promotion strategy" is a plan or measure to increase product sales.

[0240] A "clustering algorithm" is a computational method for dividing data into groups based on similarity.

[0241] A "customer demographic" is a group of customers who share certain common characteristics.

[0242] "Targeted advertising" refers to advertising activities aimed at specific customer segments.

[0243] "Notification" means informing the user of generated suggestions or information.

[0244] A system for implementing this invention is configured as follows: The server has means for accepting analysis instructions in Japanese, executing the corresponding data analysis, and visualizing the results. A user can input analysis instructions in Japanese using a smartphone. The server performs data analysis based on the accepted instructions and displays the results in a visual format such as a graph or chart.

[0245] The server then provides a means for making marketing proposals and audience targeting based on the analysis results. For example, it analyzes sales data in real time and proposes promotion strategies for products with declining sales. It also uses a clustering algorithm to identify customer segments and propose targeted advertising for those segments.

[0246] The server also provides a means to provide new insights based on patterns and trends in the data and the analysis you specify, allowing you to leverage new insights gained from data analysis to develop more effective marketing strategies.

[0247] The server also has a means for notifying the generated proposals. For example, the generated promotion proposals and targeted advertising proposals can be notified to the marketing personnel in real time by email, which enables the marketing personnel to respond promptly.

[0248] Consider the following scenario as a specific example: A user uses a smartphone to input a prompt statement such as, "If sales of a specific product are declining, please generate a promotion proposal for that product. Also, if a specific customer segment has a tendency to purchase a specific product, please propose targeted advertisements for that customer segment." The server analyzes data based on this instruction, proposes a promotion strategy for the product with declining sales, and generates targeted advertisements for the specific customer segment. The generated proposals are notified to the marketing personnel via email.

[0249] The following hardware and software are used to realize this system. For hardware, a smartphone and a server are required. For software, Python, Pandas, Scikit-learn, and the SMTP library are used. Python is used to implement data analysis and machine learning algorithms, and Pandas is used to preprocess and manipulate data. Scikit-learn is used to implement the clustering algorithm, and the SMTP library is used to implement the email notification function.

[0250] As described above, the system for implementing this invention realizes a series of processes in which a user inputs analysis instructions in Japanese, the server performs data analysis based on those instructions, visualizes the results, makes marketing proposals and audience targeting, and notifies the user of the generated proposals.

[0251] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0252] Step 1:

[0253] The user inputs analysis instructions in Japanese using a smartphone. The input instructions include promotion suggestions for declining sales of a specific product and the generation of targeted advertisements for a specific customer demographic. The input data is then sent to the server.

[0254] Step 2:

[0255] The server interprets the received Japanese analysis instructions and performs the corresponding data analysis. Specifically, it reads the sales data and standardizes it. The input data is the sales data, and the output is the standardized data.

[0256] Step 3:

[0257] The server runs a clustering algorithm on the standardized data to identify customer demographics. Specifically, it uses the KMeans algorithm to divide the data into clusters. The input data is the standardized data, and the output is data with cluster labels.

[0258] Step 4:

[0259] The server identifies products with declining sales based on the cluster-labeled data. Specifically, it extracts products with sales below average. The input data is data with cluster labels, and the output is a list of products with declining sales.

[0260] Step 5:

[0261] The server generates a promotion strategy based on the list of declining sales items. Specifically, it creates promotion proposals for declining sales items. The input data is the list of declining sales items, and the output is the promotion proposals.

[0262] Step 6:

[0263] The server generates targeted advertising proposals based on the cluster-labeled data. Specifically, it proposes optimal advertising strategies for each cluster. The input data is the cluster-labeled data, and the output is targeted advertising proposals.

[0264] Step 7:

[0265] The server notifies the marketer of the generated promotion proposals and targeted advertising proposals. Specifically, it uses an SMTP library to send email notifications. The input data are the promotion proposals and targeted advertising proposals, and the output is a notification email.

[0266] The above steps realize a series of processes in which the server analyzes data based on the analysis instructions entered by the user in Japanese, generates a promotion strategy and targeted advertisements, and notifies the marketing personnel.

[0267] Example 3

[0268] Next, a description will be given of a third embodiment of the third embodiment. 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."

[0269] Conventional data analysis systems require users to issue specific analytical instructions, and a great deal of effort is required to gain new insights based on past data and analytical results. Furthermore, they lack the functionality to predict what information users will want to know next, making efficient data utilization difficult. This slows down the user's decision-making process, making it difficult for businesses to respond quickly.

[0270] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0271] In this invention, the server includes means for accepting analysis instructions in Japanese, performing the corresponding data analysis, and visualizing the results, means for making marketing proposals and audience targeting based on the analysis results, means for providing new insights based on patterns and trends in the data and the instructed analysis, means for collecting and preprocessing past data, means for performing pattern analysis using the preprocessed data, means for generating a predictive model using a generative AI model and predicting the information the user will want to know next, and means for providing the predicted information to the user. This allows the user to obtain the information they need without having to give specific instructions, enabling efficient data utilization and rapid decision-making.

[0272] "Analysis instructions in Japanese" are data analysis instructions given by the user in Japanese.

[0273] "Data analysis" is the process of using collected data to extract and understand information for a specific purpose.

[0274] "Visualization" is the presentation of the results of data analysis in a visual format such as a graph or chart.

[0275] A "marketing proposal" is a specific proposal for promoting products or services based on the results of data analysis.

[0276] "Audience targeting" refers to developing a marketing strategy that targets a specific customer demographic.

[0277] "New insights" are new findings and understandings gained from past data and analytical results.

[0278] "Historical data" is data previously collected and used in the current analysis.

[0279] "Preprocessing" refers to the process of formatting data before data analysis, filling in missing values, normalizing data, etc.

[0280] "Pattern analysis" is an analytical method for finding patterns and trends in data.

[0281] A "generative AI model" is a model that uses artificial intelligence technology to generate new information and predictions from data.

[0282] A "predictive model" is a model that predicts future events or outcomes based on past data.

[0283] "The information that the user wants to know next" is information that is highly likely to be needed next by the user.

[0284] "Predicted Information" means information about future events or outcomes generated by a predictive model.

[0285] MODE FOR CARRYING OUT THE INVENTION

[0286] The present invention relates to a data analysis system that allows a user to obtain necessary information without issuing specific instructions. A specific embodiment of this system will be described below.

[0287] 1. Generate the system program

[0288] The server generates programs for data analysis and prediction based on user requests. These programs use generative AI models to learn patterns from past data and the analysis instructions, and provide new insights.

[0289] 2. Explain the program's processing in natural language

[0290] The server processes the data as follows:

[0291] 1. Data Collection:

[0292] The server collects the analysis instructions and results that the user has performed in the past from a database. Specifically, it uses a relational database (e.g., MySQL or PostgreSQL) to acquire the user's past analysis history.

[0293] 2. Data Preprocessing:

[0294] The server preprocesses the collected data. Specifically, it fills in missing values ​​and normalizes the data. This process uses the Python Pandas library. For example, if there are missing values, they are filled in with the mean value, and the data is standardized to unify the scale.

[0295] 3. Pattern Analysis:

[0296] The server uses the preprocessed data to analyze the user's past analysis instructions and result patterns. This analysis uses machine learning libraries such as Scikit-learn. Specifically, it performs clustering and regression analysis to extract user behavior patterns.

[0297] 4. Generate predictive models:

[0298] The server uses a generative AI model (e.g., GPT-4) to predict what the user will want to know next. This model generates new insights based on past data and pattern analysis. For example, if the user frequently analyzes monthly sales data in the past, it can predict the next month's sales.

[0299] 5. Information provision:

[0300] The server provides the predicted information to the user. Specifically, it sends a notification to the user's device and displays the prediction results. For example, it displays the sales forecast for the next month in a graph format.

[0301] 3. Add examples to your description

[0302] For example, if a user frequently performs "monthly sales data analysis" in the past, the server will learn that pattern. Then, without the user having to issue a new analysis command, the server will automatically provide "next month's sales forecast."

[0303] Example prompts to input to the generative AI model

[0304] "Please forecast next month's sales based on past sales data."

[0305] In this way, the user can obtain the necessary information without having to specifically specify what he or she is looking for. The flow of the identification process in the third embodiment will be described with reference to FIG.

[0306] Step 1:

[0307] The server collects the analysis instructions and results of the user's past executions from a database. Specifically, it executes SQL queries on a relational database (e.g., MySQL or PostgreSQL) to retrieve the user's past analysis history. The input is filter conditions such as the user ID and period, and the output is a dataset of the corresponding analysis instructions and their results.

[0308] Step 2:

[0309] The server preprocesses the collected data. Specifically, it uses the Python Pandas library to impute missing values ​​and normalize the data. For example, if there are missing values, they are imputed with the mean value and standardized to unify the scale of the data. The input is the dataset obtained in step 1, and the output is the preprocessed, clean dataset.

[0310] Step 3:

[0311] The server performs pattern analysis using the preprocessed data. Specifically, it uses a machine learning library such as Scikit-learn to perform clustering and regression analysis. For example, it uses K-means clustering to classify the data into several groups and extract user behavior patterns. The input is the dataset preprocessed in step 2, and the output is data showing the analyzed patterns and trends.

[0312] Step 4:

[0313] The server uses a generative AI model (e.g., GPT-4) to generate a predictive model and predict the information the user will want to know next. Specifically, it uses past data and the results of pattern analysis as input to generate new insights, such as sales forecasts for the next month. The inputs are the results of the pattern analysis obtained in step 3 and past data, and the output is predicted information. For example, it generates a prompt sentence such as, "Please forecast sales for the next month based on past sales data."

[0314] Step 5:

[0315] The server provides the predicted information to the user. Specifically, it sends a notification to the user's device and displays the prediction results. For example, it displays the sales forecast for the next month in a graph format and sends a notification to the user. The input is the prediction information generated in step 4, and the output is the prediction results displayed on the user's device.

[0316] (Application example 3)

[0317] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[0318] Conventional data analysis systems require users to provide specific analysis instructions, and marketing proposals and audience targeting based on analysis results are limited. Furthermore, they lack the ability to predict which products users are likely to purchase next based on their past purchase and browsing history, making it difficult to improve the user shopping experience. This creates a problem in that users spend a lot of time and effort trying to obtain the information they need.

[0319] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[0320] In this invention, the server includes means for accepting analysis instructions in Japanese, performing the corresponding data analysis, and visualizing the results, means for making marketing proposals and audience targeting based on the analysis results, means for providing new insights based on patterns and trends in the data and the instructed analysis, means for analyzing the user's past purchase history and browsing history and predicting and suggesting products that the user is likely to purchase next, means for automatically notifying the user of new products and sale information that the user is likely to be interested in, and means for providing related information and reviews when the user asks a question about a specific product. This makes it possible to predict and suggest necessary information and products that the user is likely to purchase next, even without the user providing specific instructions, thereby improving the user's shopping experience.

[0321] An "analysis instruction" is a request for data analysis made by a user to the system.

[0322] "Data analysis" is the process of processing collected data and extracting meaningful information.

[0323] "Visualization" is the presentation of the results of data analysis in a visual format such as a graph or chart.

[0324] "Marketing proposals" are sales promotion proposals made to users based on the analysis results.

[0325] "Audience targeting" is a method for conducting effective marketing to a specific user group.

[0326] "New insights" are new findings and understandings that emerge from patterns in data and analysis.

[0327] "Purchase history" is a record of products a user has purchased in the past.

[0328] "Browse history" is a record of products that a user has viewed in the past.

[0329] "Product prediction" is the process of predicting the next product a user is likely to purchase based on their past behavioral data.

[0330] "New product notification" means automatically notifying users of information about new products that may be of interest to them.

[0331] "Sales information notification" means automatically notifying users of sales information that may be of interest to them.

[0332] "Related information" is information related to a particular product that is provided when a user asks a question about that product.

[0333] A "review" is a rating or opinion of a product provided by other users when a user asks a question about that product.

[0334] A system for implementing this invention includes a server, a user terminal, and a database. The server accepts analysis instructions in Japanese from a user, performs the corresponding data analysis, and visualizes the results. The server then makes marketing proposals and audience targeting based on the analysis results. The server also provides new insights based on patterns and trends in the data and the analysis instructions.

[0335] The user terminal is a device such as a smartphone or PC that the user uses to input analysis instructions and check the results. The user terminal receives notifications from the server and displays them to the user.

[0336] The database stores the user's past purchase history and browsing history, and the server analyzes this data. The server analyzes the user's past purchase history and browsing history to predict and suggest products that the user is likely to purchase next. The server also automatically notifies the user of new products and sales information that may interest the user. Furthermore, when a user asks about a specific product, the server provides related information and reviews.

[0337] In a specific example, the server is implemented using the Python programming language. The Pandas library is used for data processing, and the Scikit-learn library is used for the machine learning model. The server reads the user's purchase history data, separates it into features and targets, and splits the data into a training set and a test set. A random forest classifier is used to train the model and predict the next product that the user is likely to purchase based on new user data.

[0338] For example, if user A has purchased many smartphone-related products in the past, the server will suggest accessories or new smartphone models that they are likely to purchase next.Also, if the user inputs a prompt such as "Please predict the product that I am likely to purchase next," the server will predict and suggest related products.

[0339] Example prompt sentence:

[0340] "Based on User A's past purchase history, please predict the product that he is likely to purchase next."

[0341] In this way, the user's shopping experience is improved because the information they need and the products they are likely to purchase next are predicted and suggested without them having to give any specific instructions.

[0342] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[0343] Step 1:

[0344] The server accepts analysis instructions in Japanese from the user terminal.

[0345] Input: Japanese analysis instructions entered by the user into the terminal

[0346] Output: Analysis instructions received by the server

[0347] Specific operation: A user inputs specific analysis instructions using a smartphone or PC and sends them to the server, which receives the instructions and prepares them for analysis.

[0348] Step 2:

[0349] The server performs the corresponding data analysis based on the received analysis instructions.

[0350] Input: Analysis instructions received by the server, data stored in the database

[0351] Output: Results of data analysis

[0352] Specific operation: The server uses the Pandas library to extract the necessary data from the database and process it according to the analytical instructions, for example, aggregating sales data for a specific product and analyzing trends.

[0353] Step 3:

[0354] The server visualizes the results of the data analysis.

[0355] Input: Results of data analysis

[0356] Output: visualized data (graphs and charts)

[0357] Specific operation: The server uses visualization libraries such as Matplotlib and Seaborn to generate analysis results as graphs and charts and send them to the user's device.

[0358] Step 4:

[0359] The server makes marketing suggestions and audience targeting based on the analysis results.

[0360] Input: Results of data analysis

[0361] Output: Marketing proposals and targeting information

[0362] Specific operation: Based on the analysis results, the server generates and proposes effective marketing strategies for specific user groups. For example, it proposes related products to users who have purchased a specific product.

[0363] Step 5:

[0364] The server provides new insights based on patterns and trends in the data and directed analysis.

[0365] Input: Past analysis instructions and analysis results

[0366] Output: New insights

[0367] How it works: The server inputs historical data and analytical results into a machine learning model to generate new insights, such as predicting what products a customer is likely to buy next.

[0368] Step 6:

[0369] The server analyzes the user's past purchase history and browsing history, predicts and suggests products that the user is likely to purchase next.

[0370] Input: User purchase history and browsing history

[0371] Output: A list of products that are likely to be purchased next

[0372] What it does: The server uses the Scikit-learn library to train a random forest classifier to predict the next likely purchase based on the user's past data.

[0373] Step 7:

[0374] The server automatically notifies users of new products and sales information that may be of interest to them.

[0375] Input: User interest data, new product and sale information

[0376] Output: Notification of new products and sales information

[0377] Specific operation: The server analyzes the user's interest data and automatically notifies the user's device of related new products and sales information.

[0378] Step 8:

[0379] The server provides relevant information and reviews when users ask about specific products.

[0380] Input: User questions, product information and reviews stored in the database

[0381] Output: Related information and reviews

[0382] Specific operation: When a user asks a question about a specific product, the server searches the database for related information and reviews and provides them to the user's device.

[0383] 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.

[0384] "Example 1"

[0385] One embodiment of the present invention is a system incorporating an emotion engine. This system recognizes a user's emotions and performs data analysis and marketing proposals based on those emotions. Specifically, suppose a user instructs the system to "show me monthly sales trends." At this time, the emotion engine reads the user's emotions from their tone of voice, facial expressions, and choice of words. For example, if the system determines that the user is feeling impatient or anxious, it quickly outputs the analysis results. Alternatively, if the system determines that the user is feeling joy or relief, it makes marketing proposals that will amplify that joy.

[0386] "Example 2"

[0387] The emotion engine also provides new insights based on the user's emotions. For example, if the system determines that the user is feeling excited or excited, it can provide new insights that complement those emotions. Specifically, it can provide the insight that sales of a particular product are skyrocketing, further increasing the user's excitement and anticipation.

[0388] "Example 3"

[0389] Furthermore, the emotion engine performs audience targeting based on the user's emotions. For example, if the system determines that the user is feeling angry or frustrated, it will target the audience in a way that alleviates those emotions. Specifically, it will suggest products or services that may alleviate the frustration of a dissatisfied user.

[0390] The processing flow of each embodiment will be described below.

[0391] "Example 1"

[0392] Step 1: The user issues a command to the system. For example, the user might say, "Show me the monthly sales trend."

[0393] Step 2: The emotion engine reads the user's emotions from their tone of voice, facial expressions, and choice of words. Step 3: The system outputs the analysis results based on the emotions read by the emotion engine. For example, if the system determines that the user is feeling impatient or anxious, it will quickly output the analysis results.

[0394] Step 4: The system makes marketing suggestions based on the emotions detected by the emotion engine. For example, if the system determines that the user is feeling joy or relief, it will make marketing suggestions that will amplify that joy.

[0395] "Example 2"

[0396] Step 1: The emotion engine reads the user's emotions, for example, determining whether the user is feeling excited or excited.

[0397] Step 2: Based on the emotions detected by the emotion engine, the system provides new insights, such as an insight that sales of a particular product are skyrocketing, further increasing user excitement and anticipation.

[0398] "Example 3"

[0399] Step 1: The emotion engine reads the user's emotions, for example, determining that the user is angry or frustrated.

[0400] Step 2: Based on the emotions detected by the emotion engine, the system will target the audience, for example, targeting the audience to alleviate those emotions.

[0401] Step 3: The system suggests products or services that may resolve the complaint.

[0402] Example 1

[0403] Next, a description will be given of Example 1 of Form 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."

[0404] With conventional data analysis systems, even if users input instructions in Japanese, it is difficult to accurately analyze the instructions and perform appropriate data analysis. Furthermore, they are unable to provide responses that take the user's emotions into account, which makes it difficult to increase user satisfaction. Furthermore, marketing proposals and audience targeting based on data analysis results are not adequately implemented, making it difficult to provide new insights.

[0405] The identification process by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means. In this invention, the server includes means for accepting analysis instructions in Japanese, executing the corresponding data analysis, and visualizing the results; means for making marketing proposals and audience targeting based on the analysis results; means for providing new insights based on patterns and trends of the data and the instructed analysis; means for recognizing the user's emotions and adjusting the output method of the data analysis results based on the emotions; and means for analyzing the user's emotions using voice input and facial expression recognition. This makes it possible to accurately analyze instructions and perform appropriate data analysis, even when the user inputs instructions in Japanese. Furthermore, by providing a response that takes the user's emotions into consideration, user satisfaction can be increased. Furthermore, it is possible to make marketing proposals and audience targeting based on the data analysis results and provide new insights.

[0406] "Analysis instructions in Japanese" are requests for data analysis made to the system by the user in Japanese.

[0407] "Data analysis" is the process of analyzing collected data using statistical methods and algorithms to extract useful information and patterns.

[0408] "Visualization" refers to displaying the results of data analysis in a visual format such as a graph or chart.

[0409] "Marketing proposals" are proposals made to users regarding sales promotion and market strategies based on the results of data analysis.

[0410] "Audience targeting" refers to conducting marketing activities aimed at specific customer groups or market segments.

[0411] "New insights" refers to new findings and understanding gained from data analysis and user instruction patterns.

[0412] "Recognizing emotions" refers to analyzing and judging a user's emotional state from their tone of voice, facial expressions, choice of words, etc.

[0413] "Voice input" is a method in which a user inputs instructions to a system by voice through a microphone.

[0414] "Facial expression recognition" is a technology that analyzes a user's facial expressions and determines their emotional state.

[0415] This invention is a system with an interface that allows users to input analysis instructions in Japanese. Users can input instructions through a text input field or a voice input function. For example, a user can input an instruction such as "Show me monthly sales trends."

[0416] The server receives instructions from the user and analyzes them using a natural language processing engine. Specifically, a general API is used as the natural language processing engine. As a result of the analysis, the content of the instruction is converted into a specific action, such as "display monthly sales data trends."

[0417] Next, the server retrieves the required data from the database. A typical relational database management system (RDBMS) is used as the database. For example, sales data is retrieved using an SQL query. An example query is "SELECT FROM sales WHERE date >= '2023-01-01' AND date <= '2023-12-31'".

[0418] The server processes the acquired data using data analysis tools. Specifically, it uses programming languages ​​such as Python and R and aggregates the data by month using the Pandas library. It then generates graphs using the Matplotlib and Seaborn libraries. For example, it displays monthly sales as a bar graph.

[0419] The server then uses an emotion engine to recognize the user's emotions. For voice input, it uses a speech recognition API to convert the voice into text and analyzes emotions from the text and tone of the voice. For facial expression recognition, it uses a facial expression recognition API to read emotions from the user's facial expressions.

[0420] The server adjusts how it outputs the data analysis results based on the user's emotions recognized by the emotion engine. For example, if it determines that the user is feeling impatient or anxious, the server will quickly output the analysis results. On the other hand, if it determines that the user is feeling joy or relief, the server will make marketing suggestions that will amplify that joy.

[0421] Finally, the server displays the generated graphs and charts to the user, often via a dashboard in a web browser, for example, by using a web framework to render the generated graphs as HTML pages and serve them to the user.

[0422] As a concrete example, a user commands the system to "show me monthly sales trends." At this time, the emotion engine senses the user's impatience from their tone of voice and facial expression. The server quickly retrieves sales data from the database, aggregates the data by month using Python, and generates a graph using the Matplotlib library. The generated graph is then rendered as a web page using a web framework and displayed to the user.

[0423] An example of a prompt is as follows:

[0424] "Show me the monthly sales trends."

[0425] "Analyze this month's sales data."

[0426] "What are the sales trends for the past six months?"

[0427] In this way, based on instructions entered by the user in Japanese, the server can analyze the data and use its emotion engine to provide an appropriate response.

[0428] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0429] Step 1:

[0430] The user enters instructions into the interface, such as "Show me monthly sales figures" through a text input field or voice input function. The input instructions are sent to the system.

[0431] Step 2:

[0432] The server receives and analyzes the instructions. The server uses a natural language processing engine to analyze the user's instructions. Specifically, the natural language processing engine converts the instructions into a specific action, such as "display monthly trends in sales data." The input is the user's instructions, and the output is the analyzed action.

[0433] Step 3:

[0434] The server retrieves the required data from the database. The server sends a query to the database based on the parsed action. For example, sales data is retrieved using an SQL query. The input is the parsed action, and the output is the retrieved sales data. Specifically, the query "SELECT FROM sales WHERE date >= '2023-01-01' AND date <= '2023-12-31'" is executed.

[0435] Step 4:

[0436] The server analyzes and visualizes the data. The server processes the acquired data using data analysis tools such as Python and R. Specifically, the data is aggregated by month using the Pandas library, and graphs are generated using the Matplotlib and Seaborn libraries. The input is the acquired sales data, and the output is the generated graph. Specifically, the sales data is aggregated by month and displayed as a bar graph.

[0437] Step 5:

[0438] The server recognizes the user's emotions using an emotion engine. The server analyzes the user's emotions using voice input and facial expression recognition. In the case of voice input, a voice recognition API is used to convert the voice into text, and the emotion is analyzed from the text and tone of the voice. In the case of facial expression recognition, an facial expression recognition API is used to read the emotion from the user's facial expression. The input is the user's voice and facial expression, and the output is the analyzed emotion.

[0439] Step 6:

[0440] The server adjusts the response based on the emotion. The server adjusts how it outputs the data analysis results based on the user's emotion recognized by the emotion engine. For example, if it determines that the user is feeling impatient or anxious, the server will quickly output the analysis results. On the other hand, if it determines that the user is feeling joy or relief, the server will make marketing suggestions that will amplify that joy. The input is the analyzed emotion, and the output is the adjusted response.

[0441] Step 7:

[0442] The server displays the results to the user. The server displays the generated graphs and charts to the user. A dashboard using a web browser is often used for display. Specifically, the graph generated using a web framework is rendered as an HTML page and provided to the user. The input is the generated graph, and the output is the graph displayed to the user.

[0443] (Application example 1)

[0444] Next, a description will be given of Application Example 1 of Embodiment 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."

[0445] Conventional data analysis systems analyze data and provide results without considering user emotions, which means they are unable to respond appropriately to user emotions. Furthermore, in brick-and-mortar stores, it is difficult to recognize customer emotions in real time and provide appropriate customer service and marketing suggestions based on those emotions. This has led to problems such as the inability to improve customer satisfaction or implement effective marketing.

[0446] 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.

[0447] In this invention, the server includes means for accepting analysis instructions in Japanese, executing the corresponding data analysis, and visualizing the results, means for making marketing proposals and audience targeting based on the analysis results, means for providing new insights based on patterns and trends of the data and the instructed analysis, means for recognizing user emotions and making data analysis and marketing proposals based on those emotions, and means for recognizing customer emotions in physical stores and making appropriate customer service and marketing proposals in real time. This enables appropriate data analysis and marketing proposals based on user emotions, and enables effective customer service and marketing that takes customer emotions into consideration even in physical stores.

[0448] "Analysis instructions in Japanese" are data analysis instructions given to the system by the user in Japanese.

[0449] "Data analysis" is the process of analyzing collected data and extracting meaningful information and patterns.

[0450] "Visualization" refers to displaying the results of data analysis in a visual format such as a graph or chart.

[0451] "Marketing proposals" are proposals made to users for sales promotion and customer acquisition based on the analysis results.

[0452] "Audience targeting" is a method for conducting effective marketing to a specific customer segment.

[0453] "New insights" are new findings and understandings gained from data and analytical patterns.

[0454] "Emotion recognition" is the process of reading a user's emotions from their tone of voice, facial expressions, and choice of words.

[0455] "Brick and mortar store" means a retail or service location that is located in a physical location.

[0456] "Real-time" refers to data and information being processed immediately and results being provided immediately.

[0457] "Customer service" refers to the service and response provided by store staff to customers.

[0458] The following system configuration is used as an embodiment of the present invention: The system is a data analysis and marketing proposal system equipped with an interface that allows users to input analysis instructions in Japanese and incorporates an emotion engine.

[0459] System Configuration

[0460] 1. Hardware:

[0461] Smartphone: A device that allows users to input voice commands.

[0462] Microphone: A device used to collect the user's voice.

[0463] 2. Software:

[0464] speech_recognition library: A library for converting user speech into text.

[0465] emotion_recognition library: A library for recognizing user emotions.

[0466] matplotlib library: A library for visualizing data analysis results as graphs and charts.

[0467] Processing flow

[0468] 1. Speech Recognition:

[0469] The user inputs analysis instructions in Japanese through the smartphone's microphone.

[0470] Use the speech_recognition library to convert spoken instructions to text.

[0471] 2. Emotion recognition:

[0472] Use the emotion_recognition library to recognize emotions from the user's tone of voice and facial expressions.

[0473] Tailor your data analysis and marketing recommendations based on perceived sentiment.

[0474] 3. Data analysis and visualization:

[0475] Perform data analysis based on user instructions.

[0476] Use the matplotlib library to visualize the analysis results as graphs and charts.

[0477] 4. Marketing Proposal:

[0478] Make appropriate marketing suggestions to users based on the perceived emotions.

[0479] In physical stores, it recognizes customer emotions in real time and provides appropriate customer service and marketing suggestions.

[0480] Specific examples

[0481] When a user inputs a voice command such as "Show me monthly sales trends," the system converts the voice into text, analyzes the sales data, and displays the monthly trends as a graph.

[0482] When a customer says, "Tell me about this product," in a physical store, the system recognizes the customer's emotions from their tone of voice and facial expressions, determining whether they are interested or anxious, and then suggests an appropriate response to the store clerk.

[0483] Prompt Sentence Examples

[0484] "Show me the monthly sales trends."

[0485] Please tell me more about this product

[0486] "Propose a new promotion"

[0487] In this way, the system can perform appropriate data analysis and marketing suggestions based on the user's emotions, enabling effective customer service and marketing that takes customer emotions into account even in physical stores.

[0488] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0489] Step 1:

[0490] The user inputs analysis instructions in Japanese through the smartphone's microphone.

[0491] Input: User's voice commands

[0492] Output: Audio data

[0493] Step 2:

[0494] The device uses the speech_recognition library to convert the audio data into text.

[0495] Input: Audio data

[0496] Data processing: Converts voice data into text data using a voice recognition algorithm

[0497] Output: Text data

[0498] Step 3:

[0499] The device uses the emotion_recognition library to recognize the user's emotions from text and voice data.

[0500] Input: Text data, audio data

[0501] Data Computing: Emotion recognition algorithms are used to analyze emotions from tone of voice and vocabulary

[0502] Output: Emotion data (e.g., anxiety, joy)

[0503] Step 4:

[0504] The server performs data analysis based on the text data.

[0505] Input: Text data

[0506] Data calculation: Extract relevant data from the database based on text data and perform analysis

[0507] Output: Analysis result data

[0508] Step 5:

[0509] The server uses the matplotlib library to visualize the analysis results data as graphs and charts.

[0510] Input: Analysis result data

[0511] Data processing: Converting analytical results data into a visually understandable format

[0512] Output: Visualized data (graphs, charts)

[0513] Step 6:

[0514] The server generates appropriate marketing offers for the user based on the emotion data.

[0515] Input: Emotion data, analysis result data

[0516] Data calculation: Combining emotional data and analytical results data to generate marketing proposals based on user emotions

[0517] Output: Marketing proposal data

[0518] Step 7:

[0519] The terminal displays the visualization data and the marketing proposal data to the user.

[0520] Input: Visualization data, marketing proposal data

[0521] What it does: Display graphs, charts, and marketing suggestions on your smartphone screen

[0522] Output: Providing visual information to the user

[0523] In this way, the system can perform appropriate data analysis and marketing proposals based on the user's emotions, enabling effective customer service and marketing that takes customer emotions into account even in physical stores.

[0524] Example 2

[0525] Next, a description will be given of Example 2 of Form 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."

[0526] In conventional data analysis systems, the acquisition and preprocessing of sales and customer data is often done manually, resulting in inefficiencies. Furthermore, it is difficult to provide marketing proposals and insights that take user emotions into account, resulting in the inability to provide proposals that meet user needs. Furthermore, marketing proposals and audience targeting based on analysis results are formulaic, requiring flexible responses that reflect user emotions.

[0527] 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.

[0528] In this invention, the server includes means for acquiring and preprocessing sales data and customer data, means for analyzing user emotions using natural language processing technology, and means for providing new insights according to the user's emotions based on the results of the emotion analysis. This automates and efficiently performs data acquisition and preprocessing, and enables flexible marketing proposals and insights that take user emotions into account.

[0529] "Data analysis" is the process of analyzing collected data using statistical methods and algorithms to extract useful information and patterns.

[0530] "Visualization" is a method of displaying the results of data analysis in a visual format such as graphs or charts to make them easier to understand.

[0531] A "marketing proposal" is a proposal of strategies or measures aimed at promoting sales of specific products or services based on the results of analysis.

[0532] "Audience targeting" is a method of analyzing customer data to narrow down targets in order to conduct marketing activities aimed at specific customer segments.

[0533] "Insights" are new findings and understandings gained as a result of data analysis and sentiment analysis.

[0534] "Sales data" is data that indicates the sales performance of a product during a specific period.

[0535] "Customer data" refers to data that includes information about customers, such as customer attributes and purchasing history.

[0536] "Preprocessing" is the process of filling in missing values, normalizing data, removing outliers, etc. before conducting data analysis.

[0537] "Natural language processing technology" is a technology that uses computers to understand, analyze, and generate human language.

[0538] "Sentiment analysis" is the process of analyzing text data and behavioral logs to determine a user's emotional state.

[0539] "New insights" are new findings and understandings gained from existing data and analytical results.

[0540] This invention is a system that automates data analysis and marketing proposals, and provides new insights based on user sentiment. This system is composed of a server, a terminal, and a user.

[0541] The server first retrieves sales and customer data from a database using a relational database management system (RDBMS) such as MySQL. The retrieved data is then preprocessed using Pandas, which involves filling in missing values ​​and normalizing the data.

[0542] Next, the server uses the Transformers library to analyze the user's emotions. The user's emotion data is obtained from the text entered by the user and from the user's behavior log. The result of the emotion analysis indicates the user's emotional state, such as whether they are excited or expectant.

[0543] The server then makes marketing proposals and audience targeting based on the analysis results. For example, if sales of a particular product are declining, it will propose a promotion for that product. Also, if a particular customer group tends to purchase a particular product, it will propose marketing activities targeted at that customer group.

[0544] Additionally, the emotion engine provides new insights based on the user's emotions. For example, if the system determines that the user is feeling excited or excited, it can provide new insights that complement those emotions. Specifically, it can provide the insight that sales of a particular product are skyrocketing, further increasing the user's excitement and anticipation.

[0545] As a concrete example, consider the following scenario.

[0546] Scenario 1: Proposing a promotion for a declining sales product

[0547] The server analyzes data that indicates that sales of a particular product are declining.

[0548] The server generates promotional offers targeted to the product.

[0549] The terminal displays to the user, "Sales of product A are declining. We suggest that you run a special discount campaign."

[0550] Scenario 2: Demographic Targeting

[0551] The server analyzes data that shows that a particular customer group tends to purchase a particular product.

[0552] The server generates marketing suggestions targeted to that demographic.

[0553] The device displays to the user, "Women in their 20s frequently purchase product B. We suggest social media advertising targeted at this customer demographic."

[0554] Prompt Sentence Examples

[0555] "Generate promotional suggestions if sales of a particular product are declining."

[0556] "Generate marketing suggestions for specific customer segments that tend to purchase specific products."

[0557] "Generate new insights that you can deliver to your users when they're excited."

[0558] This system automates and efficiently acquires and preprocesses data, while also enabling the provision of flexible marketing proposals and insights that take user sentiment into account.

[0559] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0560] Step 1:

[0561] The server retrieves sales and customer data from a database, specifically a relational database management system (RDBMS) such as MySQL, by executing SQL queries.

[0562] Input: Database connection information and SQL query

[0563] Output: A data frame of sales and customer data

[0564] Specific operation: The server executes an SQL query such as SELECT FROM sales_data WHERE date >= '2023-01-01'; and stores the results in a Pandas data frame.

[0565] Step 2:

[0566] The server uses Pandas to perform data preprocessing, such as imputing missing values, normalizing data, and removing outliers.

[0567] Input: A data frame of sales and customer data

[0568] Output: The preprocessed data frame

[0569] Specific operation: The server performs operations such as df.fillna(0, inplace=True) (filling missing values ​​with 0) and df = df[df['sales'] > 0] (removing outliers).

[0570] Step 3:

[0571] The server receives the user's input text and behavioral logs, including reviews and comments entered by the user, click logs, etc.

[0572] Input: User input text and action logs

[0573] Output: User emotion data

[0574] Specific operation: The server collects text data entered by the user and prepares it for sentiment analysis.

[0575] Step 4:

[0576] The server performs sentiment analysis using the Transformers library, specifically using the pre-trained BERT model to classify the sentiment of the text.

[0577] Input: User emotion data

[0578] Output: Sentiment analysis results

[0579] Specific behavior: The server performs sentiment analysis as follows: from transformers import pipeline; sentiment_analysis = pipeline('sentiment-analysis'); result = sentiment_analysis("This product is amazing!").

[0580] Step 5:

[0581] The server generates marketing proposals based on the sales data and the results of the sentiment analysis, specifically, proposing promotions for products with declining sales.

[0582] Input: Sales data and sentiment analysis results

[0583] Output: Marketing proposal

[0584] Specific operation: The server generates a proposal such as "Sales of product A are declining. We propose that you run a special discount campaign."

[0585] Step 6:

[0586] The server analyzes customer data to identify trends in which specific customer segments tend to purchase specific products. For example, if women in their 20s frequently purchase a particular product, it will suggest marketing activities targeted at that segment.

[0587] Input: Customer Data

[0588] Output: Audience targeting suggestions

[0589] Specific operation: The server generates a proposal such as, "Women in their 20s often purchase product B. We will propose social media advertising targeted at this customer demographic."

[0590] Step 7:

[0591] The server then provides new insights based on the user's emotions based on the results of the sentiment analysis. For example, if the user is excited, it will provide information about products that are experiencing a surge in sales.

[0592] Input: Sentiment analysis results

[0593] Output: New insights

[0594] What happens: The server generates an insight such as, "Sales for product C are skyrocketing. Now's the time to buy!"

[0595] The above is a concrete explanation of the program processing of this system.

[0596] (Application example 2)

[0597] Next, a description will be given of Application Example 2 of Form 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."

[0598] Conventional marketing systems provide uniform advertisements without considering the user's emotional state, making it difficult to draw the user's interest. Furthermore, there are problems with declining sales of specific products and a lack of promotional proposals based on trending products, making it difficult to carry out effective marketing activities. To solve these problems, a system is needed that analyzes the user's emotional state in real time and proposes optimal advertisements based on that analysis.

[0599] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0600] In this invention, the server includes means for accepting analysis instructions in Japanese, executing the corresponding data analysis, and visualizing the results, means for making marketing proposals and audience targeting based on the analysis results, means for providing new insights based on patterns and trends of the data and the instructed analysis, means for analyzing the user's emotional state in real time and proposing optimal advertisements based on the emotions, and means for proposing promotions based on declining sales of specific products or trending products. This enables effective advertisement proposals based on the user's emotional state, and also makes it possible to realize promotion proposals based on declining sales of specific products or trending products.

[0601] "Analysis instructions in Japanese" means giving instructions for data analysis using Japanese.

[0602] "Data analysis" is the analysis of collected data and the extraction of useful information and patterns.

[0603] "Visualization" is the presentation of the results of data analysis in a visual format such as a graph or chart.

[0604] "Marketing proposal" means proposing effective marketing strategies and measures based on the results of analysis.

[0605] "Audience targeting" refers to conducting marketing activities aimed at a specific customer demographic.

[0606] "New insights" refer to new knowledge and understanding gained from data and analytical results.

[0607] "User emotional state" refers to the emotions the user is feeling, such as excitement, anticipation, or neutrality.

[0608] "Real-time analysis" means analyzing data instantly and providing results immediately.

[0609] A "best fit ad" is an ad that best suits a user's emotional state and interests.

[0610] "Decreased sales" refers to a decrease in sales of a particular product or service.

[0611] "Trending products" are products that are currently popular and are being purchased by many consumers.

[0612] "Promotion" refers to advertising and campaigns aimed at promoting product sales.

[0613] A system for implementing the present invention is configured as follows.

[0614] System configuration

[0615] 1. Hardware

[0616] Smartphone: A device equipped with a camera and heart rate sensor.

[0617] Server: A high-performance computer for data processing and analysis.

[0618] 2. Software

[0619] EmotionEngine: Software for analyzing users' emotional states in real time.

[0620] MarketingAnalyzer: Software that analyzes product sales data and provides marketing insights.

[0621] Program processing

[0622] Analyzing the user's emotional state

[0623] The smartphone's camera and heart rate sensor are used to capture the user's emotional state in real time, and the EmotionEngine processes this data to determine the user's emotional state (excitement, anticipation, neutral, etc.).

[0624] Gain marketing insights

[0625] The server uses MarketingAnalyzer to analyze product sales data, thereby identifying products with declining sales and trending products.

[0626] Generate ad proposals

[0627] The server generates optimal ads based on the user's emotional state and marketing insights. For example, if a user is excited, it will suggest ads for trending products that will further increase that excitement. For products whose sales are declining, it will suggest ads with special offers.

[0628] Specific examples

[0629] If the user is excited: "Check out this trending product!"

[0630] If the user is neutral: "Check out our products on special offer!"

[0631] Prompt Sentence Examples

[0632] If the user's emotional state is "excited," generate an ad suggesting trending products.

[0633] If the user's emotional state is "neutral," generate an ad suggesting a product that's on special offer.

[0634] In this way, it is possible to make effective advertising suggestions based on the user's emotional state, and it is also possible to make promotion suggestions based on declining sales of specific products or trending products.

[0635] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0636] Step 1:

[0637] The user uses a smartphone to capture real-time data from the camera and heart rate sensor.

[0638] Input: User's face image data, heart rate data

[0639] Output: Emotional state data (excitement, anticipation, neutral, etc.)

[0640] How it works: The smartphone camera takes a picture of the user's face, and the heart rate sensor measures their heart rate. These data are sent to the Emotion Engine, which analyzes their emotional state.

[0641] Step 2:

[0642] The server uses the EmotionEngine to analyze the user's emotional state in real time.

[0643] Input: Face image data, heart rate data

[0644] Output: Emotional state (excited, anticipatory, neutral, etc.)

[0645] Specific operation: The server launches the EmotionEngine and analyzes the facial image data and heart rate data. As a result of the analysis, the user's emotional state is determined.

[0646] Step 3:

[0647] The server uses MarketingAnalyzer to analyze the product sales data.

[0648] Input: Sales data

[0649] Output: Marketing insights (low-selling products, trending products, etc.)

[0650] Specific operation: The server launches MarketingAnalyzer and analyzes the sales data. As a result of the analysis, products with declining sales and trending products are identified.

[0651] Step 4:

[0652] The server generates optimal advertisements based on the user's emotional state and marketing insights.

[0653] Inputs: Emotional state, marketing insights

[0654] Output: Ad proposal

[0655] What it does: The server combines emotional state and marketing insights to generate optimal ads for the user, for example, advertising trending products if the user is excited, or offering products on special offer if the user is neutral.

[0656] Step 5:

[0657] The advertisement generated by the server is sent to the smartphone and displayed to the user.

[0658] Enter: Ad Proposal

[0659] Output: Advertisement displayed on smartphone screen

[0660] How it works: The server sends the generated advertisement to the smartphone, which then displays it to the user. The user can then view the advertisement and, if interested, view further details.

[0661] Example 3

[0662] Next, a description will be given of a third embodiment of the third embodiment. 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."

[0663] Conventional data analysis systems have the drawback of being difficult to accurately meet user needs because users must provide specific analysis instructions and audience targeting is not based on user emotions. Furthermore, they lack the functionality to predict and provide the information users will want to know next, making it difficult to improve user satisfaction.

[0664] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0665] In this invention, the server includes means for accepting analysis instructions in Japanese, performing the corresponding data analysis, and visualizing the results, means for making marketing proposals and audience targeting based on the analysis results, means for providing new insights based on patterns and trends of the data and the instructed analysis, means for analyzing user emotions and performing audience targeting based on those emotions, and means for predicting the information the user will want to know next using a generative AI model and providing that information. This allows the user to obtain the information they need without giving specific instructions, and by making appropriate proposals based on the user's emotions, it is possible to improve user satisfaction.

[0666] "Analysis instructions in Japanese" are data analysis instructions given by the user in Japanese.

[0667] "Data analysis" is the process of analyzing collected data and extracting useful information and patterns.

[0668] "Visualization" refers to displaying the results of data analysis in a visual format such as a graph or chart.

[0669] "Marketing proposal" refers to making specific proposals for promoting products and services based on the results of data analysis.

[0670] "Audience targeting" is a method for delivering effective messages and advertisements to specific user groups.

[0671] "New insights" refers to new knowledge and understanding gained from patterns and trends in data and analysis.

[0672] "Sentiment analysis" is the process of analyzing the emotions of a user's input or feedback to determine a specific emotional state.

[0673] A "generative AI model" refers to an algorithm or system that uses artificial intelligence to generate new information or suggestions.

[0674] "Next Information" is information that predicts what the user will need next and provides it to them.

[0675] This invention is a system that allows users to provide analytical instructions in Japanese, executes data analysis based on those instructions, and visualizes the results. It also provides marketing proposals and audience targeting based on the analysis results, providing new insights based on patterns and trends in the data and the analysis instructions. It also has the ability to analyze user emotions and target audiences based on those emotions. Using a generative AI model, it can predict what information the user will want to know next and provide that information.

[0676] Hardware and software used

[0677] Hardware:

[0678] Server: A high-performance server for running database management systems and analytical engines.

[0679] Terminal: A computer or smartphone where the user inputs analysis instructions

[0680] software:

[0681] Database management systems: MySQL, PostgreSQL

[0682] Data analysis libraries: Python pandas, scikit-learn

[0683] Sentiment analysis engine: IBM Watson(R) Tone Analyzer

[0684] Generative AI models: Generative AI models such as GPT-4

[0685] Program processing explanation

[0686] Data collection:

[0687] The server collects the user's past analysis instructions and analysis results from the database. For example, it connects to a MySQL database and executes the query "SELECT FROM analysis_history WHERE user_id = '12345'".

[0688] Data Analysis:

[0689] The server converts the collected data into a pandas data frame and uses a scikit-learn clustering algorithm to extract patterns, such as K-means clustering, to classify patterns in the user's analysis instructions.

[0690] Generate insights:

[0691] The server inputs a prompt to the generative AI model (GPT-4) saying, "Please predict what information the user wants to know next." The generative AI model predicts the next need based on past data and outputs a request to know "market reaction to new products."

[0692] Emotion analysis:

[0693] The server inputs the user's recent feedback into IBM Watson Tone Analyzer and analyzes the sentiment, for example, obtaining a result such as "The feedback on the recent analysis results is negative."

[0694] Proposal generation:

[0695] The server inputs a prompt to the generative AI model saying, "Please generate a proposal to resolve the user's dissatisfaction." The generative AI model then proposes a "promotion strategy for the new product" and outputs its details.

[0696] Specific examples

[0697] For example, if a user has frequently requested "sales analysis" or "customer satisfaction surveys" in the past, the server predicts that the user will next want to know about "market reaction to a new product." Also, if the user is dissatisfied with the results of a recent analysis, the server will suggest a "new product promotion strategy" to resolve that dissatisfaction.

[0698] Example prompt sentence:

[0699] "Predict what users want to know next based on past sales analysis and customer satisfaction survey results. Also, if users are dissatisfied with recent analysis results, generate suggestions to resolve their dissatisfaction."

[0700] In this way, the server can predict the user's needs and provide appropriate information and suggestions, thereby improving the user's satisfaction. The flow of the identification process in the third embodiment will be described with reference to FIG.

[0701] Step 1:

[0702] Data collection

[0703] The server collects the user's past analysis instructions and analysis results from the database. As input, the user ID is given. The server connects to the MySQL database and executes the query "SELECT FROM analysis_history WHERE user_id = 'user ID'". As output, it obtains a dataset containing the user's past analysis instructions and results.

[0704] Step 2:

[0705] Data analysis

[0706] The server analyzes the collected data. As input, it receives the dataset obtained in step 1. The server uses Python's pandas library to format the data and extracts patterns using a clustering algorithm from scikit-learn (e.g., K-means clustering). As output, it obtains analysis results that contain patterns and trends according to the user's analysis instructions.

[0707] Step 3:

[0708] Generate insights

[0709] The server uses a generative AI model (e.g., GPT-4) to predict the information the user will want to know next. The analysis results obtained in step 2 are given as input. The server inputs a prompt to the generative AI model saying, "Please predict the information the user will want to know next." The output is the information the user will want to know next (e.g., "Market reaction to a new product").

[0710] Step 4:

[0711] Emotion analysis

[0712] The server analyzes the user's emotions. As input, the user's recent feedback is given. The server analyzes the user's emotions using an emotion analysis engine (e.g., IBM Watson Tone Analyzer). As output, it obtains the user's emotional state (e.g., "negative").

[0713] Step 5:

[0714] Proposal Generation

[0715] The server generates a proposal for the user based on the results of the sentiment analysis. As input, it receives the information the user wants to know next, obtained in step 3, and the user's emotional state, obtained in step 4. The server inputs a prompt to the generative AI model saying, "Please generate a proposal to resolve the user's dissatisfaction." As output, it obtains a specific proposal to resolve the user's dissatisfaction (e.g., "promotion strategy for a new product").

[0716] In this way, the server can anticipate the user's needs and provide appropriate information and suggestions, thereby improving user satisfaction.

[0717] (Application example 3)

[0718] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[0719] Conventional data analysis systems require users to clearly specify what they specifically want, placing a heavy burden on users. Furthermore, they lack personalized suggestions based on user emotions, which hinders the user experience. Furthermore, they lack appropriate product and service suggestions to alleviate user dissatisfaction and anger.

[0720] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[0721] In this invention, the server includes means for accepting analysis instructions in Japanese, performing the corresponding data analysis, and visualizing the results, means for making marketing proposals and audience targeting based on the analysis results, means for providing new insights based on patterns and trends in the data and the instructed analysis, means for analyzing the user's past behavioral history and emotional data and predicting and suggesting products that the user is likely to be interested in next, and means for suggesting products and services that will alleviate dissatisfaction based on the user's emotions. This allows the user to obtain the information and products they need without giving specific instructions, improving the user experience and enabling appropriate suggestions to alleviate the user's dissatisfaction or anger.

[0722] "Analysis instructions in Japanese" refers to requests or commands for data analysis given by the user in Japanese.

[0723] "Data analysis" is the process of analyzing collected data using statistical methods and algorithms to extract useful information and patterns.

[0724] "Visualization" is a method of displaying the results of data analysis in a visual format such as graphs or charts to make them easier for users to understand.

[0725] A "marketing proposal" is the presentation of specific strategies and action plans for promoting products and services based on the results of data analysis.

[0726] "Audience targeting" is a method for delivering optimal advertisements and content to specific user groups.

[0727] "New insights" are new findings and understandings gained from data and analytical results, providing useful information to users.

[0728] "Behavioral history" is a record of a user's past actions, such as purchases and browsing.

[0729] "Emotional data" is data that indicates the emotional state of a user, and is information extracted from reviews, comments, etc.

[0730] "Predicting and suggesting products" means predicting the next product a user will be interested in based on the user's past behavioral history and emotional data, and recommending that product to the user.

[0731] "Easing frustration" means suggesting appropriate products or services to reduce the frustration or anger the user feels.

[0732] The system for implementing this invention consists of a server and a user terminal. The server runs programs for data analysis, sentiment analysis, and product recommendation. The user terminal accepts input from the user and displays the results from the server.

[0733] The server first collects the user's past behavioral history and emotional data, including purchase history, browsing history, reviews, comments, etc. The collected data is then stored in a database.

[0734] The server then performs data analysis using software such as Python, Pandas, and Scikit-learn. The data is standardized and clustering algorithms (e.g., KMeans) are used to analyze user behavior patterns.

[0735] Sentiment analysis is performed to extract emotional data from user reviews and comments. Natural language processing techniques are used for sentiment analysis. Specifically, a custom module for sentiment analysis (sentiment_analysis) is used.

[0736] Based on the results of data analysis and sentiment analysis, the server predicts and suggests products that the user will be interested in next. This recommendation engine uses the recommendation_engine module. Furthermore, if the user is feeling dissatisfied or angry, the server suggests appropriate products or services to alleviate their feelings.

[0737] The user terminal receives the recommendation results sent from the server and displays them to the user, allowing the user to obtain the information and products they need without having to give any specific instructions.

[0738] For example, if a user recently purchased a smartphone case and complained in their review that the case was fragile, the system would recommend a durable smartphone case or a case with a warranty.

[0739] Example prompts to input to the generative AI model:

[0740] Based on the user's past purchase history and recent reviews, recommend the next product they might be interested in. The user recently purchased a smartphone case and complained that the case was "easy to break."

[0741] In this way, personalized product recommendations based on the user's emotions and behavioral patterns can be realized.

[0742] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[0743] Step 1:

[0744] The server collects users' past behavioral history and emotional data. Specifically, it retrieves data such as users' purchase history, browsing history, reviews, and comments from a database. The input is the user's behavioral history data and emotional data, and the output is a collection of these data.

[0745] Step 2:

[0746] The server preprocesses the collected data. Specifically, it standardizes the data, imputes missing values, and removes outliers. The input is the data collected in step 1, and the output is the preprocessed data.

[0747] Step 3:

[0748] The server performs clustering using the preprocessed data. Specifically, it uses the KMeans clustering algorithm to analyze user behavior patterns. The input is the preprocessed data, and the output is the clustering results.

[0749] Step 4:

[0750] The server extracts emotional data from user reviews and comments. Specifically, it uses natural language processing technology to analyze the text data of reviews and comments and calculate an emotional score. The input is the text data of user reviews and comments, and the output is an emotional score.

[0751] Step 5:

[0752] The server predicts and suggests products that the user will be interested in next based on the clustering results and emotion scores. Specifically, it uses a recommendation engine to recommend products based on the user's behavioral patterns and emotions. The input is the clustering results and emotion scores, and the output is a list of recommended products.

[0753] Step 6:

[0754] If a user feels dissatisfied or angry, the server suggests appropriate products or services to alleviate those emotions. Specifically, if the emotion score exceeds a certain threshold, the server recommends products or services that are expected to have an effect of alleviating the emotion. The input is the emotion score, and the output is a list of products that can alleviate the emotion.

[0755] Step 7:

[0756] The terminal receives the recommendation results sent from the server and displays them to the user. Specifically, it displays a list of recommended products on the user's screen so that the user can view or purchase those products. The input is the recommendation results sent from the server, and the output is the product list displayed to the user.

[0757] In this way, personalized product recommendations based on the user's emotions and behavioral patterns can be realized.

[0758] 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.

[0759] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ) 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.

[0760] Another example of generative AI is Gemini (registered trademark) (Internet search engine). <url: https: gemini.google.com ?hl="ja">) are mentioned.

[0761] 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.

[0762] [Second embodiment]

[0763] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0764] 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.

[0765] 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).

[0766] 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.

[0767] 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.

[0768] 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).

[0769] 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.

[0770] 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.

[0771] 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.

[0772] 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.

[0773] 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.

[0774] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0775] "Example 1"

[0776] This embodiment of the present invention provides an interface that allows a user to input analysis instructions in Japanese. This interface can include a text input field, a voice input function, etc. For example, if a user inputs an instruction such as "Show me monthly sales trends," the system will analyze the corresponding data and visualize the results in the form of graphs or charts.

[0777] "Example 2"

[0778] Furthermore, this embodiment of the present invention has the function of making marketing proposals and audience targeting based on the analysis results. For example, based on the analysis result that sales of a specific product are declining, a promotion targeting that product is proposed. Also, based on the analysis result that a specific customer segment tends to purchase a specific product, a marketing activity targeting that customer segment is proposed.

[0779] "Example 3"

[0780] This embodiment of the present invention also has the ability to provide new insights based on patterns and trends in data and instructed analyses. For example, it predicts what information the user will want to know next based on patterns in past analysis instructions and analysis results, and provides that information. This allows the user to obtain the information they need without having to specifically specify what they are looking for.

[0781] The processing flow of each embodiment will be described below.

[0782] "Example 1"

[0783] Step 1: The user inputs an analysis instruction in Japanese through the interface. For example, the user inputs an instruction such as "Show me the monthly sales trend."

[0784] Step 2: The system analyzes the relevant data. This analysis involves extracting the necessary data from the database and calculating monthly sales data.

[0785] Step 3: The system visualizes the analysis results, which includes displaying monthly sales data in the form of graphs and charts.

[0786] "Example 2"

[0787] Step 1: The system makes marketing proposals and audience targeting based on the analysis results. For example, if the analysis results show that sales of a particular product are declining, the system will propose a promotion targeting that product.

[0788] Step 2: The system performs audience targeting based on the analysis results. For example, if the analysis results show that a certain customer group tends to purchase a certain product, it will suggest marketing activities targeted at that customer group.

[0789] "Example 3"

[0790] Step 1: The system provides new insights based on patterns and trends in the data and the analysis instructions. For example, it predicts what the user will want to know next based on patterns in past analysis instructions and results.

[0791] Step 2: The system provides the user with the predicted information, allowing the user to get the information they need without having to specifically specify what they want.

[0792] Example 1

[0793] Next, a description will be given of Example 1 of Form 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."

[0794] Conventional data analysis systems have the problem that they are difficult to operate unless the user has specialized knowledge, and that it takes a lot of time and effort to acquire, process, and visualize data. Furthermore, marketing proposals, audience targeting, and new insights based on the analysis results are often not fully implemented.

[0795] 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.

[0796] In this invention, the server includes means for a user to input analysis instructions in natural language, means for analyzing the user instructions using a generative AI model, means for retrieving necessary data from a database based on the analysis results, means for formatting the retrieved data using a data processing tool, means for converting the formatted data into graphs or charts using a visualization tool, means for sending and displaying the generated graphs or charts on the user's terminal, means for making marketing proposals and audience targeting based on the analysis results, and means for providing new insights based on patterns and trends in the data and the instructed analysis. This enables users to easily perform data analysis even without specialized knowledge and to obtain marketing proposals and new insights based on the results.

[0797] A "user" is an entity that operates the system and inputs analysis instructions in natural language.

[0798] A "natural language" is a language that humans use on a daily basis and does not require a specific programming language or specialized terminology.

[0799] A "generative AI model" is a model that uses artificial intelligence technology to analyze natural language and understand user instructions.

[0800] A "database" is a system for systematically storing data and retrieving the data as needed.

[0801] A "data processing tool" is software that formats acquired data and converts it into a format suitable for analysis and visualization.

[0802] A "visualization tool" is software that converts formatted data into a visual format such as a graph or chart.

[0803] A "marketing proposal" is a proposal of strategies and measures for sales promotion and customer acquisition based on the results of analysis.

[0804] "Audience targeting" refers to conducting marketing activities aimed at specific customer groups or market segments.

[0805] "New insights" are new findings and understandings that are gained based on patterns and trends in data and analysis.

[0806] The present invention is a system in which a user inputs analysis instructions in natural language, and data is acquired, processed, and visualized based on the instructions. A specific embodiment of this system will be described below.

[0807] Users input analysis instructions in natural language through the system's interface, which includes text input fields and voice input capabilities. For example, a user might type, "Show me monthly sales trends."

[0808] The server receives instructions from the user, analyzes the user's instructions using a generative AI model (e.g., GPT-4), and determines what kind of data analysis the user's instructions require.

[0809] Next, the server retrieves the necessary data from a database based on the analysis results. For example, it retrieves the relevant monthly sales data from a database (e.g., MySQL or PostgreSQL) that stores sales data.

[0810] The acquired data is then formatted using a data processing tool (e.g., the Pandas library), aggregated monthly, and converted into a format suitable for analysis.

[0811] The formatted data is then converted into graphs and charts using visualization tools (e.g., Matplotlib or Plotly). For example, a line graph showing monthly sales trends is generated.

[0812] The generated graphs and charts are sent from the server to the user's device, which then displays them on the user's device's interface, allowing the user to visually confirm the analysis results based on the instructions.

[0813] As a specific example, if a user inputs "Show me monthly sales trends," the server will process the data using the following procedure.

[0814] 1. The user enters "Show me monthly sales trends" in the text input field and clicks the submit button.

[0815] 2. The server uses a generative AI model (GPT-4) to analyze this instruction and determine that the request is for monthly sales data.

[0816] 3. The server retrieves sales data from the database (MySQL) and aggregates it monthly using the Pandas library.

[0817] 4. The server generates a line graph of monthly sales using Matplotlib.

[0818] 5. The server sends the generated graph to the user's device, which displays it on the interface.

[0819] An example of a prompt is as follows:

[0820] "Show me the monthly sales trends."

[0821] "Compare this month's sales with the same month last year's sales."

[0822] "Show me sales trends for a specific product"

[0823] This allows users to easily analyze data without specialized knowledge and to make marketing proposals and obtain new insights based on the results.

[0824] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0825] Step 1:

[0826] The user inputs analysis instructions in natural language.

[0827] Specific behavior: The user enters "Show me monthly sales trends" into the text input field in the system interface and clicks the submit button.

[0828] Input: User's natural language instructions for analysis.

[0829] Output: User instructions are sent to the server.

[0830] Step 2:

[0831] The server receives the user's instructions.

[0832] Specific operations: The server receives instructions sent by the user.

[0833] Input: User's natural language instructions for analysis.

[0834] Output: Received instructions are passed to the generative AI model.

[0835] Step 3:

[0836] The server uses the generative AI model to analyze the user's instructions.

[0837] Specific operation: The server inputs the user's instruction into a generative AI model (e.g., GPT-4) and determines that the instruction "Show me monthly sales trends" is requesting monthly sales data trends.

[0838] Input: User's natural language instructions for analysis.

[0839] Output: The analysis results identify the necessary data analysis.

[0840] Step 4:

[0841] The server retrieves the necessary data from the database based on the analysis results.

[0842] Specific operation: The server generates an SQL query and executes it against the database (e.g., MySQL) using a query such as "SELECT FROM sales WHERE date BETWEEN '2022-01-01' AND '2022-12-31'".

[0843] Input: The data analysis results identified.

[0844] Output: Sales data retrieved from the database.

[0845] Step 5:

[0846] The data acquired by the server is formatted using a data processing tool.

[0847] Specific operation: The server creates a data frame using the Pandas library and executes code to aggregate sales by month, for example, by executing "df.groupby('month').sum()".

[0848] Input: Sales data retrieved from the database.

[0849] Output: Formatted monthly sales data.

[0850] Step 6:

[0851] The server converts the formatted data into graphs and charts using a visualization tool.

[0852] Specific operation: The server executes code that generates a line graph using Matplotlib, for example, by executing "plt.plot(df['month'], df['sales'])".

[0853] Input: Formatted monthly sales data.

[0854] Output: The generated line graph.

[0855] Step 7:

[0856] The server sends the generated graphs and charts to the user's device for display.

[0857] Specific operation: The server saves the generated graph as an image file and sends its URL to the user's device, which uses the URL to display the graph on the interface.

[0858] Input: The generated line chart.

[0859] Output: The graph displayed on the user's device.

[0860] (Application example 1)

[0861] Next, a description will be given of Application Example 1 of Form 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."

[0862] Logistics centers are required to analyze data such as inventory status and shipping status in real time and visualize it quickly and accurately. However, with conventional systems, data acquisition and analysis takes time, making efficient business operations difficult. In addition, few systems allow for voice input, making them difficult to use. To solve these issues, a system is needed that has an interface that includes voice input and can analyze and visualize data in real time.

[0863] 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.

[0864] In this invention, the server includes means for accepting analysis instructions in Japanese, performing the corresponding data analysis, and visualizing the results; means for making marketing proposals and audience targeting based on the analysis results; means for providing new insights based on patterns and trends in the data and the instructed analysis; means for converting voice input into text; means for retrieving necessary data from a database and analyzing it in real time; and means for displaying the analysis results in graphs and charts. This enables voice input instructions at logistics centers, and enables real-time data analysis and visualization. This improves operational efficiency and speeds decision-making.

[0865] "Analysis instructions in Japanese" means that the user gives instructions for data analysis in Japanese.

[0866] "Data analysis" is the analysis of collected data and the extraction of meaningful information.

[0867] "Visualization" is the presentation of the results of data analysis in a visual format such as a graph or chart.

[0868] "Marketing proposals" refer to proposals regarding sales promotion and market strategies based on the results of the analysis.

[0869] "Audience targeting" refers to conducting marketing activities aimed at a specific customer demographic.

[0870] "New insights" refer to new findings and understanding gained as a result of data analysis.

[0871] "Voice input" means that the user inputs instructions by voice through a microphone.

[0872] "Convert to text" means converting voice-input instructions into text information.

[0873] A "database" is a system that stores data systematically and allows it to be searched and retrieved as needed.

[0874] "Real-time analysis" means analyzing data as it is generated.

[0875] "Graphs and charts" are diagrams that visually represent data.

[0876] A system for implementing this invention is configured as follows: The server has means for accepting analysis instructions in Japanese, executing the corresponding data analysis, and visualizing the results. The user can use a smartphone to give instructions by voice input or text input. The voice input is converted into text using the SpeechRecognition library. The converted text instructions are sent to the server, which retrieves the necessary data from a database.

[0877] The server analyzes the data retrieved from the database in real time and visualizes the analysis results in graphs and charts using the Matplotlib library. The analysis results are then displayed on the smartphone screen.

[0878] For example, if a user says "Tell me the shipping status this month," the server retrieves shipping data and displays the monthly shipping status in a bar graph. Also, if a user says "Show me the inventory trend," the server retrieves inventory data and displays the monthly inventory trend in a line graph.

[0879] The system's hardware includes a smartphone (microphone, speaker, and display) and a server, while the software uses Python, the SpeechRecognition library, the Matplotlib library, the Pandas library, and SQLite.

[0880] An example of a prompt is as follows:

[0881] "Please tell me the shipping status for this month."

[0882] "Show me the inventory trends."

[0883] "Show me last month's stock availability."

[0884] This system allows for voice input of instructions at logistics centers and enables real-time data analysis and visualization, which improves operational efficiency and enables faster decision-making.

[0885] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0886] Step 1:

[0887] The user gives instructions by voice input or text input using a smartphone.

[0888] Input: User's voice or text instructions

[0889] Output: Audio data or text data

[0890] Specific action: The user speaks into the smartphone microphone and says, "Tell me about this month's shipping status," or types the same thing into the text field.

[0891] Step 2:

[0892] The device converts the voice input into text.

[0893] Input: Audio data

[0894] Output: Text data

[0895] Specific operation: The device uses the SpeechRecognition library to convert voice data into text data. For example, the speech "Tell me about this month's shipping status" is converted into text "Tell me about this month's shipping status."

[0896] Step 3:

[0897] The terminal transmits the text data to the server.

[0898] Input: Text data

[0899] Output: Request to server

[0900] Specific operation: The terminal sends the converted text data to the server as an HTTP request.

[0901] Step 4:

[0902] The server parses the text data and generates the appropriate SQL queries.

[0903] Input: Text data

[0904] Output: SQL query

[0905] Specific operation: The server parses the text data "Tell me about this month's shipping status" and generates an SQL query to retrieve shipping data. For example, the query generated is "SELECT FROM shipments WHERE date >= '2023-01-01'".

[0906] Step 5:

[0907] The server retrieves the required data from the database.

[0908] Input: SQL query

[0909] Output: Data retrieved from the database

[0910] Specific operation: The server executes the generated SQL query and retrieves the relevant shipping data from the database. For example, shipping data from January 2023 onwards is retrieved.

[0911] Step 6:

[0912] The server analyzes the acquired data in real time.

[0913] Input: Data retrieved from a database

[0914] Output: Analysis results

[0915] Specific operation: The server analyzes the acquired shipping data using the Pandas library and calculates the monthly shipping status.

[0916] Step 7:

[0917] The server visualizes the analysis results in graphs and charts.

[0918] Input: Analysis results

[0919] Output: Graphs and charts

[0920] Specific operation: The server uses the Matplotlib library to visualize the analysis results as bar graphs and line graphs. For example, monthly shipping status is displayed as a bar graph.

[0921] Step 8:

[0922] The server sends the visualized results to the terminal.

[0923] Input: Graphs and charts

[0924] Output: Response to the terminal

[0925] Specific operation: The server sends the generated graphs and charts to the terminal as image data.

[0926] Step 9:

[0927] The terminal displays the visualized results on the display.

[0928] Input: Image data

[0929] Output: Graphs and charts displayed on the screen

[0930] Specific operation: The device displays the received image data on the display and provides the results to the user. For example, a bar graph of monthly shipping status is displayed on the smartphone screen.

[0931] Example 2

[0932] Next, a description will be given of Example 2 of Form 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."

[0933] With conventional data analysis systems, data collection, preprocessing, analysis, and the generation and notification of marketing proposals are rarely performed consistently, and each process is fragmented, making it difficult to carry out efficient marketing activities.In addition, the provision of new insights based on analysis results is insufficient, making it difficult to optimize marketing strategies.

[0934] 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.

[0935] In this invention, the server includes means for accepting analysis instructions in Japanese, executing the corresponding data analysis, and visualizing the results, means for making marketing proposals and audience targeting based on the analysis results, means for providing new insights based on patterns and trends in the data and the instructed analysis, means for collecting sales data and customer data from a database, means for preprocessing the collected data, means for analyzing the preprocessed data, means for generating marketing proposals based on the analysis results, and means for notifying the generated marketing proposals. This makes it possible to consistently perform a series of processes from data collection to generating and notifying marketing proposals, enabling efficient marketing activities and the provision of new insights.

[0936] "Analysis instructions in Japanese" refers to a user issuing a data analysis request to the system in Japanese.

[0937] "Data analysis" refers to the process of using collected data to find specific patterns or trends.

[0938] "Visualization" refers to displaying the results of data analysis in a visual format such as a graph or chart.

[0939] "Marketing proposal" refers to proposing a strategy or action plan for promoting a specific product or service based on the results of data analysis.

[0940] "Audience targeting" refers to conducting marketing activities aimed at a specific customer demographic.

[0941] "New insights" refers to new findings and discoveries gained as a result of data analysis.

[0942] A "database" refers to a system for systematically storing and managing data.

[0943] "Sales Data" means data containing information regarding sales of products.

[0944] "Customer data" refers to data that includes information about customer attributes and purchasing history.

[0945] "Preprocessing" refers to cleaning and shaping data before analyzing it.

[0946] "Analysis results" refers to the conclusions and insights obtained through the data analysis process.

[0947] "Notification" refers to informing the user of generated marketing proposals and analysis results.

[0948] The present invention relates to a system for consistently performing data analysis and marketing proposals. A specific embodiment of this system will be described below.

[0949] First, the server collects sales and customer data from a database, which stores product sales information and customer purchase history. The server uses SQL queries to retrieve the required data.

[0950] Next, the server preprocesses the collected data using the Python Pandas library. Specifically, it performs tasks such as filling in missing values, removing outliers, and normalizing the data. This improves the quality of the data and increases the accuracy of the analysis.

[0951] Once the preprocessing is complete, the data is then used for data analysis. The server uses the Scikit-learn library to analyze sales trends and cluster customer purchasing behavior. For example, it analyzes whether sales of a particular product have declined over the past few months. It also uses clustering techniques to identify whether a particular customer demographic tends to purchase a particular product.

[0952] Based on the analysis results, the server generates marketing proposals. For example, it may propose discount campaigns for products with declining sales. It may also propose marketing activities targeted at specific customer segments. This allows for the creation of efficient marketing strategies.

[0953] Finally, the server notifies the user of the generated marketing proposals via email or dashboard. By receiving these notifications, the user can take appropriate marketing actions.

[0954] As a concrete example, consider the following scenario.

[0955] 1. The server collects sales data from the database for the past year.

[0956] 2. Use the Pandas library to impute missing values ​​and normalize the data.

[0957] 3. Use the Scikit-learn library to perform sales trend analysis, for example, discovering that sales of a particular product A have dropped by 20% over the past three months.

[0958] 4. The server generates a promotion proposal to address the decline in sales of product A. For example, it makes a proposal such as "Implement a discount campaign for product A."

[0959] 5. You also analyze customer data and discover that specific customer group B tends to purchase product A.

[0960] 6. The server proposes marketing activities targeting customer group B. For example, it may propose "send a special offer for product A to customer group B."

[0961] Examples of prompts to input to a generative AI model include:

[0962] "Analyze sales data from the past year to identify products with declining sales. Also, generate promotional suggestions for those products."

[0963] Using this prompt, the generative AI model can make appropriate analysis and suggestions.

[0964] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0965] Step 1:

[0966] The server collects sales and customer data from the database. It receives database connection information and SQL queries as input and obtains sales and customer data as output. Specifically, the server connects to the database and executes SQL queries to retrieve the required data.

[0967] Step 2:

[0968] The server preprocesses the collected data. It receives sales and customer data as input and produces preprocessed data as output. Specifically, the server uses the Python Pandas library to impute missing values, remove outliers, and normalize the data.

[0969] Step 3:

[0970] The server analyzes the preprocessed data. It receives the preprocessed data as input and generates analytical results as output. Specifically, the server uses the Scikit-learn library to perform sales trend analysis and cluster customer purchasing behavior.

[0971] Step 4:

[0972] The server generates marketing proposals based on the analysis results. It receives the analysis results as input and generates marketing proposals as output. Specifically, the server proposes promotions for products with declining sales and proposes marketing activities targeting specific customer segments.

[0973] Step 5:

[0974] The server notifies the user of the generated marketing proposal. It receives the marketing proposal as input and sends a notification to the user as output. Specifically, the server notifies the user of the proposal content by email or dashboard.

[0975] (Application example 2)

[0976] Next, a description will be given of Application Example 2 of Form 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."

[0977] In conventional marketing systems, sales data analysis and customer segmentation are often done manually, resulting in inefficiencies. Furthermore, because promotion strategies for declining sales products and targeted advertising proposals are not made in real time, it is difficult to respond quickly. Furthermore, there is a lack of a way to quickly notify marketing personnel of the generated proposals, which means that marketing activities cannot be carried out at the appropriate time.

[0978] 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 accepting analysis instructions in Japanese, executing the corresponding data analysis, and visualizing the results, means for making marketing proposals and audience targeting based on the analysis results, means for providing new insights based on patterns and trends in the data and the instructed analysis, means for analyzing sales data in real time and proposing promotion strategies for products with declining sales, means for identifying customer segments using a clustering algorithm and proposing targeted advertising, and means for notifying the generated proposals. This enables efficient analysis of sales data and rapid proposals of promotion strategies and targeted advertising.

[0979] "Analysis instructions in Japanese" are data analysis instructions given by the user in Japanese.

[0980] "Data analysis" is the analysis of collected data using statistical methods and algorithms to extract useful information and patterns.

[0981] "Visualization" is the presentation of the results of data analysis in a visual format such as a graph or chart.

[0982] A "marketing proposal" is a proposal of strategies or measures aimed at promoting sales of specific products or services based on the results of analysis.

[0983] "Audience targeting" refers to conducting marketing activities aimed at a specific customer demographic.

[0984] "New insights" are new findings and understandings gained based on data analysis and pattern recognition.

[0985] "Sales data" is numerical information about the sales of a particular product or service.

[0986] "Real-time analysis" means analyzing data as it is generated.

[0987] A "promotion strategy" is a plan or measure to increase product sales.

[0988] A "clustering algorithm" is a computational method for dividing data into groups based on similarity.

[0989] A "customer demographic" is a group of customers who share certain common characteristics.

[0990] "Targeted advertising" refers to advertising activities aimed at specific customer segments.

[0991] "Notification" means informing the user of generated suggestions or information.

[0992] A system for implementing this invention is configured as follows: The server has means for accepting analysis instructions in Japanese, executing the corresponding data analysis, and visualizing the results. A user can input analysis instructions in Japanese using a smartphone. The server performs data analysis based on the accepted instructions and displays the results in a visual format such as a graph or chart.

[0993] The server then provides a means for making marketing proposals and audience targeting based on the analysis results. For example, it analyzes sales data in real time and proposes promotion strategies for products with declining sales. It also uses a clustering algorithm to identify customer segments and propose targeted advertising for those segments.

[0994] The server also provides a means to provide new insights based on patterns and trends in the data and the analysis you specify, allowing you to leverage new insights gained from data analysis to develop more effective marketing strategies.

[0995] The server also has a means for notifying the generated proposals. For example, the generated promotion proposals and targeted advertising proposals can be notified to the marketing personnel in real time by email, which enables the marketing personnel to respond promptly.

[0996] Consider the following scenario as a specific example: A user uses a smartphone to input a prompt statement such as, "If sales of a specific product are declining, please generate a promotion proposal for that product. Also, if a specific customer segment has a tendency to purchase a specific product, please propose targeted advertisements for that customer segment." The server analyzes data based on this instruction, proposes a promotion strategy for the product with declining sales, and generates targeted advertisements for the specific customer segment. The generated proposals are notified to the marketing personnel via email.

[0997] The following hardware and software are used to realize this system. For hardware, a smartphone and a server are required. For software, Python, Pandas, Scikit-learn, and the SMTP library are used. Python is used to implement data analysis and machine learning algorithms, and Pandas is used to preprocess and manipulate data. Scikit-learn is used to implement the clustering algorithm, and the SMTP library is used to implement the email notification function.

[0998] As described above, the system for implementing this invention realizes a series of processes in which a user inputs analysis instructions in Japanese, the server performs data analysis based on those instructions, visualizes the results, makes marketing proposals and audience targeting, and notifies the user of the generated proposals.

[0999] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1000] Step 1:

[1001] The user inputs analysis instructions in Japanese using a smartphone. The input instructions include promotion suggestions for declining sales of a specific product and the generation of targeted advertisements for a specific customer demographic. The input data is then sent to the server.

[1002] Step 2:

[1003] The server interprets the received Japanese analysis instructions and performs the corresponding data analysis. Specifically, it reads the sales data and standardizes it. The input data is the sales data, and the output is the standardized data.

[1004] Step 3:

[1005] The server runs a clustering algorithm on the standardized data to identify customer demographics. Specifically, it uses the KMeans algorithm to divide the data into clusters. The input data is the standardized data, and the output is data with cluster labels.

[1006] Step 4:

[1007] The server identifies products with declining sales based on the cluster-labeled data. Specifically, it extracts products with sales below average. The input data is data with cluster labels, and the output is a list of products with declining sales.

[1008] Step 5:

[1009] The server generates a promotion strategy based on the list of declining sales items. Specifically, it creates promotion proposals for declining sales items. The input data is the list of declining sales items, and the output is the promotion proposals.

[1010] Step 6:

[1011] The server generates targeted advertising proposals based on the cluster-labeled data. Specifically, it proposes optimal advertising strategies for each cluster. The input data is the cluster-labeled data, and the output is targeted advertising proposals.

[1012] Step 7:

[1013] The server notifies the marketer of the generated promotion proposals and targeted advertising proposals. Specifically, it uses an SMTP library to send email notifications. The input data are the promotion proposals and targeted advertising proposals, and the output is a notification email.

[1014] The above steps realize a series of processes in which the server analyzes data based on the analysis instructions entered by the user in Japanese, generates a promotion strategy and targeted advertisements, and notifies the marketing personnel.

[1015] Example 3

[1016] Next, a description will be given of Example 3 of Form Example 3. 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."

[1017] Conventional data analysis systems require users to issue specific analytical instructions, and a great deal of effort is required to gain new insights based on past data and analytical results. Furthermore, they lack the functionality to predict what information users will want to know next, making efficient data utilization difficult. This slows down the user's decision-making process, making it difficult for businesses to respond quickly.

[1018] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1019] In this invention, the server includes means for accepting analysis instructions in Japanese, performing the corresponding data analysis, and visualizing the results, means for making marketing proposals and audience targeting based on the analysis results, means for providing new insights based on patterns and trends in the data and the instructed analysis, means for collecting and preprocessing past data, means for performing pattern analysis using the preprocessed data, means for generating a predictive model using a generative AI model and predicting the information the user will want to know next, and means for providing the predicted information to the user. This allows the user to obtain the information they need without having to give specific instructions, enabling efficient data utilization and rapid decision-making.

[1020] "Analysis instructions in Japanese" are data analysis instructions given by the user in Japanese.

[1021] "Data analysis" is the process of using collected data to extract and understand information for a specific purpose.

[1022] "Visualization" is the presentation of the results of data analysis in a visual format such as a graph or chart.

[1023] A "marketing proposal" is a specific proposal for promoting products or services based on the results of data analysis.

[1024] "Audience targeting" refers to developing a marketing strategy that targets a specific customer demographic.

[1025] "New insights" are new findings and understandings gained from past data and analytical results.

[1026] "Historical data" is data previously collected and used in the current analysis.

[1027] "Preprocessing" refers to the process of formatting data before data analysis, filling in missing values, normalizing data, etc.

[1028] "Pattern analysis" is an analytical method for finding patterns and trends in data.

[1029] A "generative AI model" is a model that uses artificial intelligence technology to generate new information and predictions from data.

[1030] A "predictive model" is a model that predicts future events or outcomes based on past data.

[1031] "The information that the user wants to know next" is information that is highly likely to be needed next by the user.

[1032] "Predicted Information" means information about future events or outcomes generated by a predictive model.

[1033] MODE FOR CARRYING OUT THE INVENTION

[1034] The present invention relates to a data analysis system that allows a user to obtain necessary information without issuing specific instructions. A specific embodiment of this system will be described below.

[1035] 1. Generate the system program

[1036] The server generates programs for data analysis and prediction based on user requests. These programs use generative AI models to learn patterns from past data and the analysis instructions, and provide new insights.

[1037] 2. Explain the program's processing in natural language

[1038] The server processes the data as follows:

[1039] 1. Data Collection:

[1040] The server collects the analysis instructions and results that the user has performed in the past from a database. Specifically, it uses a relational database (e.g., MySQL or PostgreSQL) to acquire the user's past analysis history.

[1041] 2. Data Preprocessing:

[1042] The server preprocesses the collected data. Specifically, it fills in missing values ​​and normalizes the data. This process uses the Python Pandas library. For example, if there are missing values, they are filled in with the mean value, and the data is standardized to unify the scale.

[1043] 3. Pattern Analysis:

[1044] The server uses the preprocessed data to analyze the user's past analysis instructions and result patterns. This analysis uses machine learning libraries such as Scikit-learn. Specifically, it performs clustering and regression analysis to extract user behavior patterns.

[1045] 4. Generate predictive models:

[1046] The server uses a generative AI model (e.g., GPT-4) to predict what the user will want to know next. This model generates new insights based on past data and pattern analysis. For example, if the user frequently analyzes monthly sales data in the past, it can predict the next month's sales.

[1047] 5. Information provision:

[1048] The server provides the predicted information to the user. Specifically, it sends a notification to the user's device and displays the prediction results. For example, it displays the sales forecast for the next month in a graph format.

[1049] 3. Add examples to your description

[1050] For example, if a user frequently performs "monthly sales data analysis" in the past, the server will learn that pattern. Then, without the user having to issue a new analysis command, the server will automatically provide "next month's sales forecast."

[1051] Example prompts to input to the generative AI model

[1052] "Please forecast next month's sales based on past sales data."

[1053] In this way, the user can obtain the necessary information without having to specifically specify what he or she is looking for. The flow of the identification process in the third embodiment will be described with reference to FIG.

[1054] Step 1:

[1055] The server collects the analysis instructions and results of the user's past executions from a database. Specifically, it executes SQL queries on a relational database (e.g., MySQL or PostgreSQL) to retrieve the user's past analysis history. The input is filter conditions such as the user ID and period, and the output is a dataset of the corresponding analysis instructions and their results.

[1056] Step 2:

[1057] The server preprocesses the collected data. Specifically, it uses the Python Pandas library to impute missing values ​​and normalize the data. For example, if there are missing values, they are imputed with the mean value and standardized to unify the scale of the data. The input is the dataset obtained in step 1, and the output is the preprocessed, clean dataset.

[1058] Step 3:

[1059] The server performs pattern analysis using the preprocessed data. Specifically, it uses a machine learning library such as Scikit-learn to perform clustering and regression analysis. For example, it uses K-means clustering to classify the data into several groups and extract user behavior patterns. The input is the dataset preprocessed in step 2, and the output is data showing the analyzed patterns and trends.

[1060] Step 4:

[1061] The server uses a generative AI model (e.g., GPT-4) to generate a predictive model and predict the information the user will want to know next. Specifically, it uses past data and the results of pattern analysis as input to generate new insights, such as sales forecasts for the next month. The inputs are the results of the pattern analysis obtained in step 3 and past data, and the output is predicted information. For example, it generates a prompt sentence such as, "Please forecast sales for the next month based on past sales data."

[1062] Step 5:

[1063] The server provides the predicted information to the user. Specifically, it sends a notification to the user's device and displays the prediction results. For example, it displays the sales forecast for the next month in a graph format and sends a notification to the user. The input is the prediction information generated in step 4, and the output is the prediction results displayed on the user's device.

[1064] (Application example 3)

[1065] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[1066] Conventional data analysis systems require users to provide specific analysis instructions, and marketing proposals and audience targeting based on analysis results are limited. Furthermore, they lack the ability to predict which products users are likely to purchase next based on their past purchase and browsing history, making it difficult to improve the user shopping experience. This creates a problem in that users spend a lot of time and effort trying to obtain the information they need.

[1067] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[1068] In this invention, the server includes means for accepting analysis instructions in Japanese, performing the corresponding data analysis, and visualizing the results, means for making marketing proposals and audience targeting based on the analysis results, means for providing new insights based on patterns and trends in the data and the instructed analysis, means for analyzing the user's past purchase history and browsing history and predicting and suggesting products that the user is likely to purchase next, means for automatically notifying the user of new products and sale information that the user is likely to be interested in, and means for providing related information and reviews when the user asks a question about a specific product. This makes it possible to predict and suggest necessary information and products that the user is likely to purchase next, even without the user providing specific instructions, thereby improving the user's shopping experience.

[1069] An "analysis instruction" is a request for data analysis made by a user to the system.

[1070] "Data analysis" is the process of processing collected data and extracting meaningful information.

[1071] "Visualization" is the presentation of the results of data analysis in a visual format such as a graph or chart.

[1072] "Marketing proposals" are sales promotion proposals made to users based on the analysis results.

[1073] "Audience targeting" is a method for conducting effective marketing to a specific user group.

[1074] "New insights" are new findings and understandings that emerge from patterns in data and analysis.

[1075] "Purchase history" is a record of products a user has purchased in the past.

[1076] "Browse history" is a record of products that a user has viewed in the past.

[1077] "Product prediction" is the process of predicting the next product a user is likely to purchase based on their past behavioral data.

[1078] "New product notification" means automatically notifying users of information about new products that may be of interest to them.

[1079] "Sales information notification" means automatically notifying users of sales information that may be of interest to them.

[1080] "Related information" is information related to a particular product that is provided when a user asks a question about that product.

[1081] A "review" is a rating or opinion of a product provided by other users when a user asks a question about that product.

[1082] A system for implementing this invention includes a server, a user terminal, and a database. The server accepts analysis instructions in Japanese from a user, performs the corresponding data analysis, and visualizes the results. The server then makes marketing proposals and audience targeting based on the analysis results. The server also provides new insights based on patterns and trends in the data and the analysis instructions.

[1083] The user terminal is a device such as a smartphone or PC that the user uses to input analysis instructions and check the results. The user terminal receives notifications from the server and displays them to the user.

[1084] The database stores the user's past purchase history and browsing history, and the server analyzes this data. The server analyzes the user's past purchase history and browsing history to predict and suggest products that the user is likely to purchase next. The server also automatically notifies the user of new products and sales information that may interest the user. Furthermore, when a user asks about a specific product, the server provides related information and reviews.

[1085] In a specific example, the server is implemented using the Python programming language. The Pandas library is used for data processing, and the Scikit-learn library is used for the machine learning model. The server reads the user's purchase history data, separates it into features and targets, and splits the data into a training set and a test set. A random forest classifier is used to train the model and predict the next product that the user is likely to purchase based on new user data.

[1086] For example, if user A has purchased many smartphone-related products in the past, the server will suggest accessories or new smartphone models that they are likely to purchase next.Also, if the user inputs a prompt such as "Please predict the product that I am likely to purchase next," the server will predict and suggest related products.

[1087] Example prompt sentence:

[1088] "Based on User A's past purchase history, please predict the product that he is likely to purchase next."

[1089] In this way, the user's shopping experience is improved because the information they need and the products they are likely to purchase next are predicted and suggested without them having to give any specific instructions.

[1090] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[1091] Step 1:

[1092] The server accepts analysis instructions in Japanese from the user terminal.

[1093] Input: Japanese analysis instructions entered by the user into the terminal

[1094] Output: Analysis instructions received by the server

[1095] Specific operation: A user inputs specific analysis instructions using a smartphone or PC and sends them to the server, which receives the instructions and prepares them for analysis.

[1096] Step 2:

[1097] The server performs the corresponding data analysis based on the received analysis instructions.

[1098] Input: Analysis instructions received by the server, data stored in the database

[1099] Output: Results of data analysis

[1100] Specific operation: The server uses the Pandas library to extract the necessary data from the database and process it according to the analytical instructions, for example, aggregating sales data for a specific product and analyzing trends.

[1101] Step 3:

[1102] The server visualizes the results of the data analysis.

[1103] Input: Results of data analysis

[1104] Output: visualized data (graphs and charts)

[1105] Specific operation: The server uses visualization libraries such as Matplotlib and Seaborn to generate analysis results as graphs and charts and send them to the user's device.

[1106] Step 4:

[1107] The server makes marketing suggestions and audience targeting based on the analysis results.

[1108] Input: Results of data analysis

[1109] Output: Marketing proposals and targeting information

[1110] Specific operation: Based on the analysis results, the server generates and proposes effective marketing strategies for specific user groups. For example, it proposes related products to users who have purchased a specific product.

[1111] Step 5:

[1112] The server provides new insights based on patterns and trends in the data and directed analysis.

[1113] Input: Past analysis instructions and analysis results

[1114] Output: New insights

[1115] How it works: The server inputs historical data and analytical results into a machine learning model to generate new insights, such as predicting what products a customer is likely to buy next.

[1116] Step 6:

[1117] The server analyzes the user's past purchase history and browsing history, predicts and suggests products that the user is likely to purchase next.

[1118] Input: User purchase history and browsing history

[1119] Output: A list of products that are likely to be purchased next

[1120] What it does: The server uses the Scikit-learn library to train a random forest classifier to predict the next likely purchase based on the user's past data.

[1121] Step 7:

[1122] The server automatically notifies users of new products and sales information that may be of interest to them.

[1123] Input: User interest data, new product and sale information

[1124] Output: Notification of new products and sales information

[1125] Specific operation: The server analyzes the user's interest data and automatically notifies the user's device of related new products and sales information.

[1126] Step 8:

[1127] The server provides relevant information and reviews when users ask about specific products.

[1128] Input: User questions, product information and reviews stored in the database

[1129] Output: Related information and reviews

[1130] Specific operation: When a user asks a question about a specific product, the server searches the database for related information and reviews and provides them to the user's device.

[1131] 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.

[1132] "Example 1"

[1133] One embodiment of the present invention is a system incorporating an emotion engine. This system recognizes a user's emotions and performs data analysis and marketing proposals based on those emotions. Specifically, suppose a user instructs the system to "show me monthly sales trends." At this time, the emotion engine reads the user's emotions from their tone of voice, facial expressions, and choice of words. For example, if the system determines that the user is feeling impatient or anxious, it quickly outputs the analysis results. Alternatively, if the system determines that the user is feeling joy or relief, it makes marketing proposals that will amplify that joy.

[1134] "Example 2"

[1135] The emotion engine also provides new insights based on the user's emotions. For example, if the system determines that the user is feeling excited or excited, it can provide new insights that complement those emotions. Specifically, it can provide the insight that sales of a particular product are skyrocketing, further increasing the user's excitement and anticipation.

[1136] "Example 3"

[1137] Furthermore, the emotion engine performs audience targeting based on the user's emotions. For example, if the system determines that the user is feeling angry or frustrated, it will target the audience in a way that alleviates those emotions. Specifically, it will suggest products or services that may alleviate the frustration of a dissatisfied user.

[1138] The processing flow of each embodiment will be described below.

[1139] "Example 1"

[1140] Step 1: The user issues a command to the system. For example, the user might say, "Show me the monthly sales trend."

[1141] Step 2: The emotion engine reads the user's emotions from their tone of voice, facial expressions, and choice of words. Step 3: The system outputs the analysis results based on the emotions read by the emotion engine. For example, if the system determines that the user is feeling impatient or anxious, it will quickly output the analysis results.

[1142] Step 4: The system makes marketing suggestions based on the emotions detected by the emotion engine. For example, if the system determines that the user is feeling joy or relief, it will make marketing suggestions that will amplify that joy.

[1143] "Example 2"

[1144] Step 1: The emotion engine reads the user's emotions, for example, determining whether the user is feeling excited or excited.

[1145] Step 2: Based on the emotions detected by the emotion engine, the system provides new insights, such as an insight that sales of a particular product are skyrocketing, further increasing user excitement and anticipation.

[1146] "Example 3"

[1147] Step 1: The emotion engine reads the user's emotions. For example, the user is angry or frustrated.

[1148] It is determined that the person is feeling this way.

[1149] Step 2: Based on the emotions detected by the emotion engine, the system performs audience targeting, for example, targeting audiences that will alleviate those emotions.

[1150] Step 3: The system suggests products or services that may resolve the complaint.

[1151] Example 1

[1152] Next, a description will be given of Example 1 of Form 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."

[1153] With conventional data analysis systems, even if users input instructions in Japanese, it is difficult to accurately analyze the instructions and perform appropriate data analysis. Furthermore, they are unable to provide responses that take the user's emotions into account, which makes it difficult to increase user satisfaction. Furthermore, marketing proposals and audience targeting based on data analysis results are not adequately implemented, making it difficult to provide new insights.

[1154] The identification process by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means. In this invention, the server includes means for accepting analysis instructions in Japanese, executing the corresponding data analysis, and visualizing the results; means for making marketing proposals and audience targeting based on the analysis results; means for providing new insights based on patterns and trends of the data and the instructed analysis; means for recognizing the user's emotions and adjusting the output method of the data analysis results based on the emotions; and means for analyzing the user's emotions using voice input and facial expression recognition. This makes it possible to accurately analyze instructions and perform appropriate data analysis, even when the user inputs instructions in Japanese. Furthermore, by providing a response that takes the user's emotions into consideration, user satisfaction can be increased. Furthermore, it is possible to make marketing proposals and audience targeting based on the data analysis results and provide new insights.

[1155] "Analysis instructions in Japanese" are requests for data analysis made to the system by the user in Japanese.

[1156] "Data analysis" is the process of analyzing collected data using statistical methods and algorithms to extract useful information and patterns.

[1157] "Visualization" refers to displaying the results of data analysis in a visual format such as a graph or chart.

[1158] "Marketing proposals" are proposals made to users regarding sales promotion and market strategies based on the results of data analysis.

[1159] "Audience targeting" refers to conducting marketing activities aimed at specific customer groups or market segments.

[1160] "New insights" refers to new findings and understanding gained from data analysis and user instruction patterns.

[1161] "Recognizing emotions" refers to analyzing and judging a user's emotional state from their tone of voice, facial expressions, choice of words, etc.

[1162] "Voice input" is a method in which a user inputs instructions to a system by voice through a microphone.

[1163] "Facial expression recognition" is a technology that analyzes a user's facial expressions and determines their emotional state.

[1164] This invention is a system with an interface that allows users to input analysis instructions in Japanese. Users can input instructions through a text input field or a voice input function. For example, a user can input an instruction such as "Show me monthly sales trends."

[1165] The server receives instructions from the user and analyzes them using a natural language processing engine. Specifically, a general API is used as the natural language processing engine. As a result of the analysis, the content of the instruction is converted into a specific action, such as "display monthly sales data trends."

[1166] Next, the server retrieves the required data from the database. A typical relational database management system (RDBMS) is used as the database. For example, sales data is retrieved using an SQL query. An example query is "SELECT FROM sales WHERE date >= '2023-01-01' AND date <= '2023-12-31'".

[1167] The server processes the acquired data using data analysis tools. Specifically, it uses programming languages ​​such as Python and R and aggregates the data by month using the Pandas library. It then generates graphs using the Matplotlib and Seaborn libraries. For example, it displays monthly sales as a bar graph.

[1168] The server then uses an emotion engine to recognize the user's emotions. For voice input, it uses a speech recognition API to convert the voice into text and analyzes emotions from the text and tone of the voice. For facial expression recognition, it uses a facial expression recognition API to read emotions from the user's facial expressions.

[1169] The server adjusts how it outputs the data analysis results based on the user's emotions recognized by the emotion engine. For example, if it determines that the user is feeling impatient or anxious, the server will quickly output the analysis results. On the other hand, if it determines that the user is feeling joy or relief, the server will make marketing suggestions that will amplify that joy.

[1170] Finally, the server displays the generated graphs and charts to the user, often via a dashboard in a web browser, for example, by using a web framework to render the generated graphs as HTML pages and serve them to the user.

[1171] As a concrete example, a user commands the system to "show me monthly sales trends." At this time, the emotion engine senses the user's impatience from their tone of voice and facial expression. The server quickly retrieves sales data from the database, aggregates the data by month using Python, and generates a graph using the Matplotlib library. The generated graph is then rendered as a web page using a web framework and displayed to the user.

[1172] An example of a prompt is as follows:

[1173] "Show me the monthly sales trends."

[1174] "Analyze this month's sales data."

[1175] "What are the sales trends for the past six months?"

[1176] In this way, based on instructions entered by the user in Japanese, the server can analyze the data and use its emotion engine to provide an appropriate response.

[1177] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1178] Step 1:

[1179] The user enters instructions into the interface, such as "Show me monthly sales figures" through a text input field or voice input function. The input instructions are sent to the system.

[1180] Step 2:

[1181] The server receives and analyzes the instructions. The server uses a natural language processing engine to analyze the user's instructions. Specifically, the natural language processing engine converts the instructions into a specific action, such as "display monthly trends in sales data." The input is the user's instructions, and the output is the analyzed action.

[1182] Step 3:

[1183] The server retrieves the required data from the database. The server sends a query to the database based on the parsed action. For example, sales data is retrieved using an SQL query. The input is the parsed action, and the output is the retrieved sales data. Specifically, the query "SELECT FROM sales WHERE date >= '2023-01-01' AND date <= '2023-12-31'" is executed.

[1184] Step 4:

[1185] The server analyzes and visualizes the data. The server processes the acquired data using data analysis tools such as Python and R. Specifically, the data is aggregated by month using the Pandas library, and graphs are generated using the Matplotlib and Seaborn libraries. The input is the acquired sales data, and the output is the generated graph. Specifically, the sales data is aggregated by month and displayed as a bar graph.

[1186] Step 5:

[1187] The server recognizes the user's emotions using an emotion engine. The server analyzes the user's emotions using voice input and facial expression recognition. In the case of voice input, a voice recognition API is used to convert the voice into text, and the emotion is analyzed from the text and tone of the voice. In the case of facial expression recognition, an facial expression recognition API is used to read the emotion from the user's facial expression. The input is the user's voice and facial expression, and the output is the analyzed emotion.

[1188] Step 6:

[1189] The server adjusts the response based on the emotion. The server adjusts how it outputs the data analysis results based on the user's emotion recognized by the emotion engine. For example, if it determines that the user is feeling impatient or anxious, the server will quickly output the analysis results. On the other hand, if it determines that the user is feeling joy or relief, the server will make marketing suggestions that will amplify that joy. The input is the analyzed emotion, and the output is the adjusted response.

[1190] Step 7:

[1191] The server displays the results to the user. The server displays the generated graphs and charts to the user. A dashboard using a web browser is often used for display. Specifically, the graph generated using a web framework is rendered as an HTML page and provided to the user. The input is the generated graph, and the output is the graph displayed to the user.

[1192] (Application example 1)

[1193] Next, a description will be given of Application Example 1 of Form 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."

[1194] Conventional data analysis systems analyze data and provide results without considering user emotions, which means they are unable to respond appropriately to user emotions. Furthermore, in brick-and-mortar stores, it is difficult to recognize customer emotions in real time and provide appropriate customer service and marketing suggestions based on those emotions. This has led to problems such as the inability to improve customer satisfaction or implement effective marketing.

[1195] 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.

[1196] In this invention, the server includes means for accepting analysis instructions in Japanese, executing the corresponding data analysis, and visualizing the results, means for making marketing proposals and audience targeting based on the analysis results, means for providing new insights based on patterns and trends of the data and the instructed analysis, means for recognizing user emotions and making data analysis and marketing proposals based on those emotions, and means for recognizing customer emotions in physical stores and making appropriate customer service and marketing proposals in real time. This enables appropriate data analysis and marketing proposals based on user emotions, and enables effective customer service and marketing that takes customer emotions into consideration even in physical stores.

[1197] "Analysis instructions in Japanese" are data analysis instructions given to the system by the user in Japanese.

[1198] "Data analysis" is the process of analyzing collected data and extracting meaningful information and patterns.

[1199] "Visualization" refers to displaying the results of data analysis in a visual format such as a graph or chart.

[1200] "Marketing proposals" are proposals made to users for sales promotion and customer acquisition based on the analysis results.

[1201] "Audience targeting" is a method for conducting effective marketing to a specific customer segment.

[1202] "New insights" are new findings and understandings gained from data and analytical patterns.

[1203] "Emotion recognition" is the process of reading a user's emotions from their tone of voice, facial expressions, and choice of words.

[1204] "Brick and mortar store" means a retail or service location that is located in a physical location.

[1205] "Real-time" refers to data and information being processed immediately and results being provided immediately.

[1206] "Customer service" refers to the service and response provided by store staff to customers.

[1207] The following system configuration is used as an embodiment of the present invention: The system is a data analysis and marketing proposal system equipped with an interface that allows users to input analysis instructions in Japanese and incorporates an emotion engine.

[1208] System Configuration

[1209] 1. Hardware:

[1210] Smartphone: A device that allows users to input voice commands.

[1211] Microphone: A device used to collect the user's voice.

[1212] 2. Software:

[1213] speech_recognition library: A library for converting user speech into text.

[1214] emotion_recognition library: A library for recognizing user emotions.

[1215] matplotlib library: A library for visualizing data analysis results as graphs and charts.

[1216] Processing flow

[1217] 1. Speech Recognition:

[1218] The user inputs analysis instructions in Japanese through the smartphone's microphone.

[1219] Use the speech_recognition library to convert spoken instructions to text.

[1220] 2. Emotion recognition:

[1221] Use the emotion_recognition library to recognize emotions from the user's tone of voice and facial expressions.

[1222] Tailor your data analysis and marketing recommendations based on perceived sentiment.

[1223] 3. Data analysis and visualization:

[1224] Perform data analysis based on user instructions.

[1225] Use the matplotlib library to visualize the analysis results as graphs and charts.

[1226] 4. Marketing Proposal:

[1227] Make appropriate marketing suggestions to users based on the perceived emotions.

[1228] In physical stores, it recognizes customer emotions in real time and provides appropriate customer service and marketing suggestions.

[1229] Specific examples

[1230] When a user inputs a voice command such as "Show me monthly sales trends," the system converts the voice into text, analyzes the sales data, and displays the monthly trends as a graph.

[1231] When a customer says, "Tell me about this product," in a physical store, the system recognizes the customer's emotions from their tone of voice and facial expressions, determining whether they are interested or anxious, and then suggests an appropriate response to the store clerk.

[1232] Prompt Sentence Examples

[1233] "Show me the monthly sales trends."

[1234] Please tell me more about this product

[1235] "Propose a new promotion"

[1236] In this way, the system can perform appropriate data analysis and marketing suggestions based on the user's emotions, enabling effective customer service and marketing that takes customer emotions into account even in physical stores.

[1237] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1238] Step 1:

[1239] The user inputs analysis instructions in Japanese through the smartphone's microphone.

[1240] Input: User's voice commands

[1241] Output: Audio data

[1242] Step 2:

[1243] The device uses the speech_recognition library to convert the audio data into text.

[1244] Input: Audio data

[1245] Data processing: Converts voice data into text data using a voice recognition algorithm

[1246] Output: Text data

[1247] Step 3:

[1248] The device uses the emotion_recognition library to recognize the user's emotions from text and voice data.

[1249] Input: Text data, audio data

[1250] Data Computing: Emotion recognition algorithms are used to analyze emotions from tone of voice and vocabulary

[1251] Output: Emotion data (e.g., anxiety, joy)

[1252] Step 4:

[1253] The server performs data analysis based on the text data.

[1254] Input: Text data

[1255] Data calculation: Extract relevant data from the database based on text data and perform analysis

[1256] Output: Analysis result data

[1257] Step 5:

[1258] The server uses the matplotlib library to visualize the analysis results data as graphs and charts.

[1259] Input: Analysis result data

[1260] Data processing: Converting analytical results data into a visually understandable format

[1261] Output: Visualized data (graphs, charts)

[1262] Step 6:

[1263] The server generates appropriate marketing offers for the user based on the emotion data.

[1264] Input: Emotion data, analysis result data

[1265] Data calculation: Combining emotional data and analytical results data to generate marketing proposals based on user emotions

[1266] Output: Marketing proposal data

[1267] Step 7:

[1268] The terminal displays the visualization data and the marketing proposal data to the user.

[1269] Input: Visualization data, marketing proposal data

[1270] What it does: Display graphs, charts, and marketing suggestions on your smartphone screen

[1271] Output: Providing visual information to the user

[1272] In this way, the system can perform appropriate data analysis and marketing proposals based on the user's emotions, enabling effective customer service and marketing that takes customer emotions into account even in physical stores.

[1273] Example 2

[1274] Next, a description will be given of Example 2 of Form 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."

[1275] In conventional data analysis systems, the acquisition and preprocessing of sales and customer data is often done manually, resulting in inefficiencies. Furthermore, it is difficult to provide marketing proposals and insights that take user emotions into account, resulting in the inability to provide proposals that meet user needs. Furthermore, marketing proposals and audience targeting based on analysis results are formulaic, requiring flexible responses that reflect user emotions.

[1276] 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.

[1277] In this invention, the server includes means for acquiring and preprocessing sales data and customer data, means for analyzing user emotions using natural language processing technology, and means for providing new insights according to the user's emotions based on the results of the emotion analysis. This automates and efficiently performs data acquisition and preprocessing, and enables flexible marketing proposals and insights that take user emotions into account.

[1278] "Data analysis" is the process of analyzing collected data using statistical methods and algorithms to extract useful information and patterns.

[1279] "Visualization" is a method of displaying the results of data analysis in a visual format such as graphs or charts to make them easier to understand.

[1280] A "marketing proposal" is a proposal of strategies or measures aimed at promoting sales of specific products or services based on the results of analysis.

[1281] "Audience targeting" is a method of analyzing customer data to narrow down targets in order to conduct marketing activities aimed at specific customer segments.

[1282] "Insights" are new findings and understandings gained as a result of data analysis and sentiment analysis.

[1283] "Sales data" is data that indicates the sales performance of a product during a specific period.

[1284] "Customer data" refers to data that includes information about customers, such as customer attributes and purchasing history.

[1285] "Preprocessing" is the process of filling in missing values, normalizing data, removing outliers, etc. before conducting data analysis.

[1286] "Natural language processing technology" is a technology that uses computers to understand, analyze, and generate human language.

[1287] "Sentiment analysis" is the process of analyzing text data and behavioral logs to determine a user's emotional state.

[1288] "New insights" are new findings and understandings gained from existing data and analytical results.

[1289] This invention is a system that automates data analysis and marketing proposals, and provides new insights based on user sentiment. This system is composed of a server, a terminal, and a user.

[1290] The server first retrieves sales and customer data from a database using a relational database management system (RDBMS) such as MySQL. The retrieved data is then preprocessed using Pandas, which involves filling in missing values ​​and normalizing the data.

[1291] Next, the server uses the Transformers library to analyze the user's emotions. The user's emotion data is obtained from the text entered by the user and from the user's behavior log. The result of the emotion analysis indicates the user's emotional state, such as whether they are excited or expectant.

[1292] The server then makes marketing proposals and audience targeting based on the analysis results. For example, if sales of a particular product are declining, it will propose a promotion for that product. Also, if a particular customer group tends to purchase a particular product, it will propose marketing activities targeted at that customer group.

[1293] Additionally, the emotion engine provides new insights based on the user's emotions. For example, if the system determines that the user is feeling excited or excited, it can provide new insights that complement those emotions. Specifically, it can provide the insight that sales of a particular product are skyrocketing, further increasing the user's excitement and anticipation.

[1294] As a concrete example, consider the following scenario.

[1295] Scenario 1: Proposing a promotion for a declining sales product

[1296] The server analyzes data that indicates that sales of a particular product are declining.

[1297] The server generates promotional offers targeted to the product.

[1298] The terminal displays to the user, "Sales of product A are declining. We suggest that you run a special discount campaign."

[1299] Scenario 2: Demographic Targeting

[1300] The server analyzes data that shows that a particular customer group tends to purchase a particular product.

[1301] The server generates marketing suggestions targeted to that demographic.

[1302] The device displays to the user, "Women in their 20s frequently purchase product B. We suggest social media advertising targeted at this customer demographic."

[1303] Prompt Sentence Examples

[1304] "Generate promotional suggestions if sales of a particular product are declining."

[1305] "Generate marketing suggestions for specific customer segments that tend to purchase specific products."

[1306] "Generate new insights that you can deliver to your users when they're excited."

[1307] This system automates and efficiently acquires and preprocesses data, while also enabling the provision of flexible marketing proposals and insights that take user sentiment into account.

[1308] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1309] Step 1:

[1310] The server retrieves sales and customer data from a database, specifically a relational database management system (RDBMS) such as MySQL, by executing SQL queries.

[1311] Input: Database connection information and SQL query

[1312] Output: A data frame of sales and customer data

[1313] Specific operation: The server executes an SQL query such as SELECT FROM sales_data WHERE date >= '2023-01-01'; and stores the results in a Pandas data frame.

[1314] Step 2:

[1315] The server uses Pandas to perform data preprocessing, such as imputing missing values, normalizing data, and removing outliers.

[1316] Input: A data frame of sales and customer data

[1317] Output: The preprocessed data frame

[1318] Specific operation: The server performs operations such as df.fillna(0, inplace=True) (filling missing values ​​with 0) and df = df[df['sales'] > 0] (removing outliers).

[1319] Step 3:

[1320] The server receives the user's input text and behavioral logs, including reviews and comments entered by the user, click logs, etc.

[1321] Input: User input text and action logs

[1322] Output: User emotion data

[1323] Specific operation: The server collects text data entered by the user and prepares it for sentiment analysis.

[1324] Step 4:

[1325] The server performs sentiment analysis using the Transformers library, specifically using the pre-trained BERT model to classify the sentiment of the text.

[1326] Input: User emotion data

[1327] Output: Sentiment analysis results

[1328] Specific behavior: The server performs sentiment analysis as follows: from transformers import pipeline; sentiment_analysis = pipeline('sentiment-analysis'); result = sentiment_analysis("This product is amazing!").

[1329] Step 5:

[1330] The server generates marketing proposals based on the sales data and the results of the sentiment analysis, specifically, proposing promotions for products with declining sales.

[1331] Input: Sales data and sentiment analysis results

[1332] Output: Marketing proposal

[1333] Specific operation: The server generates a proposal such as "Sales of product A are declining. We propose that you run a special discount campaign."

[1334] Step 6:

[1335] The server analyzes customer data to identify trends in which specific customer segments tend to purchase specific products. For example, if women in their 20s frequently purchase a particular product, it will suggest marketing activities targeted at that segment.

[1336] Input: Customer Data

[1337] Output: Audience targeting suggestions

[1338] Specific operation: The server generates a proposal such as, "Women in their 20s often purchase product B. We will propose social media advertising targeted at this customer demographic."

[1339] Step 7:

[1340] The server then provides new insights based on the user's emotions based on the results of the sentiment analysis. For example, if the user is excited, it will provide information about products that are experiencing a surge in sales.

[1341] Input: Sentiment analysis results

[1342] Output: New insights

[1343] What happens: The server generates an insight such as, "Sales for product C are skyrocketing. Now's the time to buy!"

[1344] The above is a concrete explanation of the program processing of this system.

[1345] (Application example 2)

[1346] Next, a description will be given of Application Example 2 of Form 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."

[1347] Conventional marketing systems provide uniform advertisements without considering the user's emotional state, making it difficult to draw the user's interest. Furthermore, there are problems with declining sales of specific products and a lack of promotional proposals based on trending products, making it difficult to carry out effective marketing activities. To solve these problems, a system is needed that analyzes the user's emotional state in real time and proposes optimal advertisements based on that analysis.

[1348] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1349] In this invention, the server includes means for accepting analysis instructions in Japanese, executing the corresponding data analysis, and visualizing the results, means for making marketing proposals and audience targeting based on the analysis results, means for providing new insights based on patterns and trends of the data and the instructed analysis, means for analyzing the user's emotional state in real time and proposing optimal advertisements based on the emotions, and means for proposing promotions based on declining sales of specific products or trending products. This enables effective advertisement proposals based on the user's emotional state, and also makes it possible to realize promotion proposals based on declining sales of specific products or trending products.

[1350] "Analysis instructions in Japanese" means giving instructions for data analysis in Japanese.

[1351] "Data analysis" is the process of analyzing collected data and extracting useful information and patterns.

[1352] "Visualization" is the presentation of the results of data analysis in a visual format such as a graph or chart.

[1353] "Marketing proposal" means proposing effective marketing strategies and measures based on the results of analysis.

[1354] "Audience targeting" refers to conducting marketing activities aimed at a specific customer demographic.

[1355] "New insights" refer to new knowledge and understanding gained from data and analytical results.

[1356] "User emotional state" refers to the emotions the user is feeling, such as excitement, anticipation, or neutrality.

[1357] "Real-time analysis" means analyzing data instantly and providing results immediately.

[1358] A "best fit ad" is an ad that best suits a user's emotional state and interests.

[1359] "Decreased sales" refers to a decrease in sales of a particular product or service.

[1360] "Trending products" are products that are currently popular and are being purchased by many consumers.

[1361] "Promotion" refers to advertising and campaigns aimed at promoting product sales.

[1362] A system for implementing the present invention is configured as follows.

[1363] System configuration

[1364] 1. Hardware

[1365] Smartphone: A device equipped with a camera and heart rate sensor.

[1366] Server: A high-performance computer for data processing and analysis.

[1367] 2. Software

[1368] EmotionEngine: Software for analyzing users' emotional states in real time.

[1369] MarketingAnalyzer: Software that analyzes product sales data and provides marketing insights.

[1370] Program processing

[1371] Analyzing the user's emotional state

[1372] The smartphone's camera and heart rate sensor are used to capture the user's emotional state in real time, and the EmotionEngine processes this data to determine the user's emotional state (excitement, anticipation, neutral, etc.).

[1373] Gain marketing insights

[1374] The server uses MarketingAnalyzer to analyze product sales data, thereby identifying products with declining sales and trending products.

[1375] Generate ad proposals

[1376] The server generates optimal ads based on the user's emotional state and marketing insights. For example, if a user is excited, it will suggest ads for trending products that will further increase that excitement. For products whose sales are declining, it will suggest ads with special offers.

[1377] Specific examples

[1378] If the user is excited: "Check out this trending product!"

[1379] If the user is neutral: "Check out our products on special offer!"

[1380] Prompt Sentence Examples

[1381] If the user's emotional state is "excited," generate an ad suggesting trending products.

[1382] If the user's emotional state is "neutral," generate an ad suggesting a product that's on special offer.

[1383] In this way, it is possible to make effective advertising suggestions based on the user's emotional state, and it is also possible to make promotion suggestions based on declining sales of specific products or trending products.

[1384] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1385] Step 1:

[1386] The user uses a smartphone to capture real-time data from the camera and heart rate sensor.

[1387] Input: User's face image data, heart rate data

[1388] Output: Emotional state data (excitement, anticipation, neutral, etc.)

[1389] How it works: The smartphone camera takes a picture of the user's face, and the heart rate sensor measures their heart rate. These data are sent to the Emotion Engine, which analyzes their emotional state.

[1390] Step 2:

[1391] The server uses the EmotionEngine to analyze the user's emotional state in real time.

[1392] Input: Face image data, heart rate data

[1393] Output: Emotional state (excited, anticipatory, neutral, etc.)

[1394] Specific operation: The server launches the EmotionEngine and analyzes the facial image data and heart rate data. As a result of the analysis, the user's emotional state is determined.

[1395] Step 3:

[1396] The server uses MarketingAnalyzer to analyze the product sales data.

[1397] Input: Sales data

[1398] Output: Marketing insights (low-selling products, trending products, etc.)

[1399] Specific operation: The server launches MarketingAnalyzer and analyzes the sales data. As a result of the analysis, products with declining sales and trending products are identified.

[1400] Step 4:

[1401] The server generates optimal advertisements based on the user's emotional state and marketing insights.

[1402] Inputs: Emotional state, marketing insights

[1403] Output: Ad proposal

[1404] What it does: The server combines emotional state and marketing insights to generate optimal ads for the user, for example, advertising trending products if the user is excited, or offering products on special offer if the user is neutral.

[1405] Step 5:

[1406] The advertisement generated by the server is sent to the smartphone and displayed to the user.

[1407] Enter: Ad Proposal

[1408] Output: Advertisement displayed on smartphone screen

[1409] How it works: The server sends the generated advertisement to the smartphone, which then displays it to the user. The user can then view the advertisement and, if interested, view further details.

[1410] Example 3

[1411] Next, a description will be given of Example 3 of Form Example 3. 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."

[1412] Conventional data analysis systems have the drawback of being difficult to accurately meet user needs because users must provide specific analysis instructions and audience targeting is not based on user emotions. Furthermore, they lack the functionality to predict and provide the information users will want to know next, making it difficult to improve user satisfaction.

[1413] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1414] In this invention, the server includes means for accepting analysis instructions in Japanese, performing the corresponding data analysis, and visualizing the results, means for making marketing proposals and audience targeting based on the analysis results, means for providing new insights based on patterns and trends of the data and the instructed analysis, means for analyzing user emotions and performing audience targeting based on those emotions, and means for predicting the information the user will want to know next using a generative AI model and providing that information. This allows the user to obtain the information they need without giving specific instructions, and by making appropriate proposals based on the user's emotions, it is possible to improve user satisfaction.

[1415] "Analysis instructions in Japanese" are data analysis instructions given by the user in Japanese.

[1416] "Data analysis" is the process of analyzing collected data and extracting useful information and patterns.

[1417] "Visualization" refers to displaying the results of data analysis in a visual format such as a graph or chart.

[1418] "Marketing proposal" refers to making specific proposals for promoting products and services based on the results of data analysis.

[1419] "Audience targeting" is a method for delivering effective messages and advertisements to specific user groups.

[1420] "New insights" refers to new knowledge and understanding gained from patterns and trends in data and analysis.

[1421] "Sentiment analysis" is the process of analyzing the emotions of a user's input or feedback to determine a specific emotional state.

[1422] A "generative AI model" refers to an algorithm or system that uses artificial intelligence to generate new information or suggestions.

[1423] "Next Information" is information that predicts what the user will need next and provides it to them.

[1424] This invention is a system that allows users to provide analytical instructions in Japanese, executes data analysis based on those instructions, and visualizes the results. It also provides marketing proposals and audience targeting based on the analysis results, providing new insights based on patterns and trends in the data and the analysis instructions. It also has the ability to analyze user emotions and target audiences based on those emotions. Using a generative AI model, it can predict what information the user will want to know next and provide that information.

[1425] Hardware and software used

[1426] Hardware:

[1427] Server: A high-performance server for running database management systems and analytical engines.

[1428] Terminal: A computer or smartphone where the user inputs analysis instructions

[1429] software:

[1430] Database management systems: MySQL, PostgreSQL

[1431] Data analysis libraries: Python pandas, scikit-learn

[1432] Sentiment analysis engine: IBM Watson Tone Analyzer

[1433] Generative AI models: Generative AI models such as GPT-4

[1434] Program processing explanation

[1435] Data collection:

[1436] The server collects the user's past analysis instructions and analysis results from the database. For example, it connects to a MySQL database and executes the query "SELECT FROM analysis_history WHERE user_id = '12345'".

[1437] Data Analysis:

[1438] The server converts the collected data into a pandas data frame and uses a scikit-learn clustering algorithm to extract patterns, such as K-means clustering, to classify patterns in the user's analysis instructions.

[1439] Generate insights:

[1440] The server inputs a prompt to the generative AI model (GPT-4) saying, "Please predict what information the user wants to know next." The generative AI model predicts the next need based on past data and outputs a request to know "market reaction to new products."

[1441] Emotion analysis:

[1442] The server inputs the user's recent feedback into IBM Watson Tone Analyzer and analyzes the sentiment, for example, obtaining a result such as "The feedback on the recent analysis results is negative."

[1443] Proposal generation:

[1444] The server inputs a prompt to the generative AI model saying, "Please generate a proposal to resolve the user's dissatisfaction." The generative AI model then proposes a "promotion strategy for the new product" and outputs its details.

[1445] Specific examples

[1446] For example, if a user has frequently requested "sales analysis" or "customer satisfaction surveys" in the past, the server predicts that the user will next want to know about "market reaction to a new product." Also, if the user is dissatisfied with the results of a recent analysis, the server will suggest a "new product promotion strategy" to resolve that dissatisfaction.

[1447] Example prompt sentence:

[1448] "Predict what users want to know next based on past sales analysis and customer satisfaction survey results. Also, if users are dissatisfied with recent analysis results, generate suggestions to resolve their dissatisfaction."

[1449] In this way, the server can predict the user's needs and provide appropriate information and suggestions, thereby improving the user's satisfaction. The flow of the identification process in the third embodiment will be described with reference to FIG.

[1450] Step 1:

[1451] Data collection

[1452] The server collects the user's past analysis instructions and analysis results from the database. As input, the user ID is given. The server connects to the MySQL database and executes the query "SELECT FROM analysis_history WHERE user_id = 'user ID'". As output, it obtains a dataset containing the user's past analysis instructions and results.

[1453] Step 2:

[1454] Data analysis

[1455] The server analyzes the collected data. As input, it receives the dataset obtained in step 1. The server uses Python's pandas library to format the data and extracts patterns using a clustering algorithm from scikit-learn (e.g., K-means clustering). As output, it obtains analysis results that contain patterns and trends according to the user's analysis instructions.

[1456] Step 3:

[1457] Generate insights

[1458] The server uses a generative AI model (e.g., GPT-4) to predict the information the user will want to know next. The analysis results obtained in step 2 are given as input. The server inputs a prompt to the generative AI model saying, "Please predict the information the user will want to know next." The output is the information the user will want to know next (e.g., "Market reaction to a new product").

[1459] Step 4:

[1460] Emotion analysis

[1461] The server analyzes the user's emotions. As input, the user's recent feedback is given. The server analyzes the user's emotions using an emotion analysis engine (e.g., IBM Watson Tone Analyzer). As output, it obtains the user's emotional state (e.g., "negative").

[1462] Step 5:

[1463] Proposal Generation

[1464] The server generates a proposal for the user based on the results of the sentiment analysis. As input, it receives the information the user wants to know next, obtained in step 3, and the user's emotional state, obtained in step 4. The server inputs a prompt to the generative AI model saying, "Please generate a proposal to resolve the user's dissatisfaction." As output, it obtains a specific proposal to resolve the user's dissatisfaction (e.g., "promotion strategy for a new product").

[1465] In this way, the server can anticipate the user's needs and provide appropriate information and suggestions, thereby improving user satisfaction.

[1466] (Application example 3)

[1467] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[1468] Conventional data analysis systems require users to clearly specify what they specifically want, placing a heavy burden on users. Furthermore, they lack personalized suggestions based on user emotions, which hinders the user experience. Furthermore, they lack appropriate product and service suggestions to alleviate user dissatisfaction and anger.

[1469] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[1470] In this invention, the server includes means for accepting analysis instructions in Japanese, performing the corresponding data analysis, and visualizing the results, means for making marketing proposals and audience targeting based on the analysis results, means for providing new insights based on patterns and trends in the data and the instructed analysis, means for analyzing the user's past behavioral history and emotional data and predicting and suggesting products that the user is likely to be interested in next, and means for suggesting products and services that will alleviate dissatisfaction based on the user's emotions. This allows the user to obtain the information and products they need without giving specific instructions, improving the user experience and enabling appropriate suggestions to alleviate the user's dissatisfaction or anger.

[1471] "Analysis instructions in Japanese" refers to requests or commands for data analysis given by the user in Japanese.

[1472] "Data analysis" is the process of analyzing collected data using statistical methods and algorithms to extract useful information and patterns.

[1473] "Visualization" is a method of displaying the results of data analysis in a visual format such as graphs or charts to make them easier for users to understand.

[1474] A "marketing proposal" is the presentation of specific strategies and action plans for promoting products and services based on the results of data analysis.

[1475] "Audience targeting" is a method for delivering optimal advertisements and content to specific user groups.

[1476] "New insights" are new findings and understandings gained from data and analytical results, providing useful information to users.

[1477] "Behavioral history" is a record of a user's past actions, such as purchases and browsing.

[1478] "Emotional data" is data that indicates the emotional state of a user, and is information extracted from reviews, comments, etc.

[1479] "Predicting and suggesting products" means predicting the next product a user will be interested in based on the user's past behavioral history and emotional data, and recommending that product to the user.

[1480] "Easing frustration" means suggesting appropriate products or services to reduce the frustration or anger the user feels.

[1481] The system for implementing this invention consists of a server and a user terminal. The server runs programs for data analysis, sentiment analysis, and product recommendation. The user terminal accepts input from the user and displays the results from the server.

[1482] The server first collects the user's past behavioral history and emotional data, including purchase history, browsing history, reviews, comments, etc. The collected data is then stored in a database.

[1483] The server then performs data analysis using software such as Python, Pandas, and Scikit-learn. The data is standardized and clustering algorithms (e.g., KMeans) are used to analyze user behavior patterns.

[1484] Sentiment analysis is performed to extract emotional data from user reviews and comments. Natural language processing techniques are used for sentiment analysis. Specifically, a custom module for sentiment analysis (sentiment_analysis) is used.

[1485] Based on the results of data analysis and sentiment analysis, the server predicts and suggests products that the user will be interested in next. This recommendation engine uses the recommendation_engine module. Furthermore, if the user is feeling dissatisfied or angry, the server suggests appropriate products or services to alleviate their feelings.

[1486] The user terminal receives the recommendation results sent from the server and displays them to the user, allowing the user to obtain the information and products they need without having to give any specific instructions.

[1487] For example, if a user recently purchased a smartphone case and complained in their review that the case was fragile, the system would recommend a durable smartphone case or a case with a warranty.

[1488] Example prompts to input to the generative AI model:

[1489] Based on the user's past purchase history and recent reviews, recommend the next product they might be interested in. The user recently purchased a smartphone case and complained that the case was "easy to break."

[1490] In this way, personalized product recommendations based on the user's emotions and behavioral patterns can be realized.

[1491] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[1492] Step 1:

[1493] The server collects users' past behavioral history and emotional data. Specifically, it retrieves data such as users' purchase history, browsing history, reviews, and comments from a database. The input is the user's behavioral history data and emotional data, and the output is a collection of these data.

[1494] Step 2:

[1495] The server preprocesses the collected data. Specifically, it standardizes the data, imputes missing values, and removes outliers. The input is the data collected in step 1, and the output is the preprocessed data.

[1496] Step 3:

[1497] The server performs clustering using the preprocessed data. Specifically, it uses the KMeans clustering algorithm to analyze user behavior patterns. The input is the preprocessed data, and the output is the clustering results.

[1498] Step 4:

[1499] The server extracts emotional data from user reviews and comments. Specifically, it uses natural language processing technology to analyze the text data of reviews and comments and calculate an emotional score. The input is the text data of user reviews and comments, and the output is an emotional score.

[1500] Step 5:

[1501] The server predicts and suggests products that the user will be interested in next based on the clustering results and emotion scores. Specifically, it uses a recommendation engine to recommend products based on the user's behavioral patterns and emotions. The input is the clustering results and emotion scores, and the output is a list of recommended products.

[1502] Step 6:

[1503] If a user feels dissatisfied or angry, the server suggests appropriate products or services to alleviate those emotions. Specifically, if the emotion score exceeds a certain threshold, the server recommends products or services that are expected to have an effect of alleviating the emotion. The input is the emotion score, and the output is a list of products that can alleviate the emotion.

[1504] Step 7:

[1505] The terminal receives the recommendation results sent from the server and displays them to the user. Specifically, it displays a list of recommended products on the user's screen so that the user can view or purchase those products. The input is the recommendation results sent from the server, and the output is the product list displayed to the user.

[1506] In this way, personalized product recommendations based on the user's emotions and behavioral patterns can be realized.

[1507] 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.

[1508] 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> ) 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.

[1509] Another example of generative AI is Gemini (internet search engine). <url: https: gemini.google.com ?hl="ja">) are mentioned.

[1510] 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.

[1511] [Third embodiment]

[1512] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1513] 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.

[1514] 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).

[1515] 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.

[1516] 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.

[1517] 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).

[1518] 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.

[1519] 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.

[1520] 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.

[1521] 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.

[1522] 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.

[1523] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[1524] "Example 1"

[1525] This embodiment of the present invention provides an interface that allows a user to input analysis instructions in Japanese. This interface can include a text input field, a voice input function, etc. For example, if a user inputs an instruction such as "Show me monthly sales trends," the system will analyze the corresponding data and visualize the results in the form of graphs or charts.

[1526] "Example 2"

[1527] Furthermore, this embodiment of the present invention has the function of making marketing proposals and audience targeting based on the analysis results. For example, based on the analysis result that sales of a specific product are declining, a promotion targeting that product is proposed. Also, based on the analysis result that a specific customer segment tends to purchase a specific product, a marketing activity targeting that customer segment is proposed.

[1528] "Example 3"

[1529] This embodiment of the present invention also has the ability to provide new insights based on patterns and trends in data and instructed analyses. For example, it predicts what information the user will want to know next based on patterns in past analysis instructions and analysis results, and provides that information. This allows the user to obtain the information they need without having to specifically specify what they are looking for.

[1530] The processing flow of each embodiment will be described below.

[1531] "Example 1"

[1532] Step 1: The user inputs an analysis instruction in Japanese through the interface. For example, the user inputs an instruction such as "Show me the monthly sales trend."

[1533] Step 2: The system analyzes the relevant data. This analysis involves extracting the necessary data from the database and calculating monthly sales data.

[1534] Step 3: The system visualizes the analysis results, which includes displaying monthly sales data in the form of graphs and charts.

[1535] "Example 2"

[1536] Step 1: The system makes marketing proposals and audience targeting based on the analysis results. For example, if the analysis results show that sales of a particular product are declining, the system will propose a promotion targeting that product.

[1537] Step 2: The system performs audience targeting based on the analysis results. For example, if the analysis results show that a certain customer group tends to purchase a certain product, it will suggest marketing activities targeted at that customer group.

[1538] "Example 3"

[1539] Step 1: The system provides new insights based on patterns and trends in the data and the analysis instructions. For example, it predicts what the user will want to know next based on patterns in past analysis instructions and results.

[1540] Step 2: The system provides the user with the predicted information, allowing the user to get the information they need without having to specifically specify what they want.

[1541] Example 1

[1542] Next, a description will be given of Example 1 of Form 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."

[1543] Conventional data analysis systems have the problem that they are difficult to operate unless the user has specialized knowledge, and that it takes a lot of time and effort to acquire, process, and visualize data. Furthermore, marketing proposals, audience targeting, and new insights based on the analysis results are often not fully implemented.

[1544] 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.

[1545] In this invention, the server includes means for a user to input analysis instructions in natural language, means for analyzing the user instructions using a generative AI model, means for retrieving necessary data from a database based on the analysis results, means for formatting the retrieved data using a data processing tool, means for converting the formatted data into graphs or charts using a visualization tool, means for sending and displaying the generated graphs or charts on the user's terminal, means for making marketing proposals and audience targeting based on the analysis results, and means for providing new insights based on patterns and trends in the data and the instructed analysis. This enables users to easily perform data analysis even without specialized knowledge and to obtain marketing proposals and new insights based on the results.

[1546] A "user" is an entity that operates the system and inputs analysis instructions in natural language.

[1547] A "natural language" is a language that humans use on a daily basis and does not require a specific programming language or specialized terminology.

[1548] A "generative AI model" is a model that uses artificial intelligence technology to analyze natural language and understand user instructions.

[1549] A "database" is a system for systematically storing data and retrieving the data as needed.

[1550] A "data processing tool" is software that formats acquired data and converts it into a format suitable for analysis and visualization.

[1551] A "visualization tool" is software that converts formatted data into a visual format such as a graph or chart.

[1552] A "marketing proposal" is a proposal of strategies and measures for sales promotion and customer acquisition based on the results of analysis.

[1553] "Audience targeting" refers to conducting marketing activities aimed at specific customer groups or market segments.

[1554] "New insights" are new findings and understandings that are gained based on patterns and trends in data and analysis.

[1555] The present invention is a system in which a user inputs analysis instructions in natural language, and data is acquired, processed, and visualized based on the instructions. A specific embodiment of this system will be described below.

[1556] Users input analysis instructions in natural language through the system's interface, which includes text input fields and voice input capabilities. For example, a user might type, "Show me monthly sales trends."

[1557] The server receives instructions from the user, analyzes the user's instructions using a generative AI model (e.g., GPT-4), and determines what kind of data analysis the user's instructions require.

[1558] Next, the server retrieves the necessary data from a database based on the analysis results. For example, it retrieves the relevant monthly sales data from a database (e.g., MySQL or PostgreSQL) that stores sales data.

[1559] The acquired data is then formatted using a data processing tool (e.g., the Pandas library), aggregated monthly, and converted into a format suitable for analysis.

[1560] The formatted data is then converted into graphs and charts using visualization tools (e.g., Matplotlib or Plotly). For example, a line graph showing monthly sales trends is generated.

[1561] The generated graphs and charts are sent from the server to the user's device, which then displays them on the user's device's interface, allowing the user to visually confirm the analysis results based on the instructions.

[1562] As a specific example, if a user inputs "Show me monthly sales trends," the server will process the data using the following procedure.

[1563] 1. The user enters "Show me monthly sales trends" in the text input field and clicks the submit button.

[1564] 2. The server uses a generative AI model (GPT-4) to analyze this instruction and determine that the request is for monthly sales data.

[1565] 3. The server retrieves sales data from the database (MySQL) and aggregates it monthly using the Pandas library.

[1566] 4. The server generates a line graph of monthly sales using Matplotlib.

[1567] 5. The server sends the generated graph to the user's device, which displays it on the interface.

[1568] An example of a prompt is as follows:

[1569] "Show me the monthly sales trends."

[1570] "Compare this month's sales with the same month last year's sales."

[1571] "Show me sales trends for a specific product"

[1572] This allows users to easily analyze data without specialized knowledge and to make marketing proposals and obtain new insights based on the results.

[1573] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1574] Step 1:

[1575] The user inputs analysis instructions in natural language.

[1576] Specific behavior: The user enters "Show me monthly sales trends" into the text input field in the system interface and clicks the submit button.

[1577] Input: User's natural language instructions for analysis.

[1578] Output: User instructions are sent to the server.

[1579] Step 2:

[1580] The server receives the user's instructions.

[1581] Specific operations: The server receives instructions sent by the user.

[1582] Input: User's natural language instructions for analysis.

[1583] Output: Received instructions are passed to the generative AI model.

[1584] Step 3:

[1585] The server uses the generative AI model to analyze the user's instructions.

[1586] Specific operation: The server inputs the user's instruction into a generative AI model (e.g., GPT-4) and determines that the instruction "Show me monthly sales trends" is requesting monthly sales data trends.

[1587] Input: User's natural language instructions for analysis.

[1588] Output: The analysis results identify the necessary data analysis.

[1589] Step 4:

[1590] The server retrieves the necessary data from the database based on the analysis results.

[1591] Specific operation: The server generates an SQL query and executes it against the database (e.g., MySQL) using a query such as "SELECT FROM sales WHERE date BETWEEN '2022-01-01' AND '2022-12-31'".

[1592] Input: The data analysis results identified.

[1593] Output: Sales data retrieved from the database.

[1594] Step 5:

[1595] The data acquired by the server is formatted using a data processing tool.

[1596] Specific operation: The server creates a data frame using the Pandas library and executes code to aggregate sales by month, for example, by executing "df.groupby('month').sum()".

[1597] Input: Sales data retrieved from the database.

[1598] Output: Formatted monthly sales data.

[1599] Step 6:

[1600] The server converts the formatted data into graphs and charts using a visualization tool.

[1601] Specific operation: The server executes code that generates a line graph using Matplotlib, for example, by executing "plt.plot(df['month'], df['sales'])".

[1602] Input: Formatted monthly sales data.

[1603] Output: The generated line graph.

[1604] Step 7:

[1605] The server sends the generated graphs and charts to the user's device for display.

[1606] Specific operation: The server saves the generated graph as an image file and sends its URL to the user's device, which uses the URL to display the graph on the interface.

[1607] Input: The generated line chart.

[1608] Output: The graph displayed on the user's device.

[1609] (Application example 1)

[1610] Next, a description will be given of Application Example 1 of Form 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."

[1611] Logistics centers are required to analyze data such as inventory status and shipping status in real time and visualize it quickly and accurately. However, with conventional systems, data acquisition and analysis takes time, making efficient business operations difficult. In addition, few systems allow for voice input, making them difficult to use. To solve these issues, a system is needed that has an interface that includes voice input and can analyze and visualize data in real time.

[1612] 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.

[1613] In this invention, the server includes means for accepting analysis instructions in Japanese, performing the corresponding data analysis, and visualizing the results; means for making marketing proposals and audience targeting based on the analysis results; means for providing new insights based on patterns and trends in the data and the instructed analysis; means for converting voice input into text; means for retrieving necessary data from a database and analyzing it in real time; and means for displaying the analysis results in graphs and charts. This enables voice input instructions at logistics centers, and enables real-time data analysis and visualization. This improves operational efficiency and speeds decision-making.

[1614] "Analysis instructions in Japanese" means that the user gives instructions for data analysis in Japanese.

[1615] "Data analysis" is the analysis of collected data and the extraction of meaningful information.

[1616] "Visualization" is the presentation of the results of data analysis in a visual format such as a graph or chart.

[1617] "Marketing proposals" refer to proposals regarding sales promotion and market strategies based on the results of the analysis.

[1618] "Audience targeting" refers to conducting marketing activities aimed at a specific customer demographic.

[1619] "New insights" refer to new findings and understanding gained as a result of data analysis.

[1620] "Voice input" means that the user inputs instructions by voice through a microphone.

[1621] "Convert to text" means converting voice-input instructions into text information.

[1622] A "database" is a system that stores data systematically and allows it to be searched and retrieved as needed.

[1623] "Real-time analysis" means analyzing data as it is generated.

[1624] "Graphs and charts" are diagrams that visually represent data.

[1625] A system for implementing this invention is configured as follows: The server has means for accepting analysis instructions in Japanese, executing the corresponding data analysis, and visualizing the results. The user can use a smartphone to give instructions by voice input or text input. The voice input is converted into text using the SpeechRecognition library. The converted text instructions are sent to the server, which retrieves the necessary data from a database.

[1626] The server analyzes the data retrieved from the database in real time and visualizes the analysis results in graphs and charts using the Matplotlib library. The analysis results are then displayed on the smartphone screen.

[1627] For example, if a user says "Tell me the shipping status this month," the server retrieves shipping data and displays the monthly shipping status in a bar graph. If a user says "Show me the inventory trend," the server retrieves inventory data and displays the monthly inventory trend in a line graph.

[1628] The system's hardware includes a smartphone (microphone, speaker, and display) and a server, while the software uses Python, the SpeechRecognition library, the Matplotlib library, the Pandas library, and SQLite.

[1629] An example of a prompt is as follows:

[1630] "Please tell me the shipping status for this month."

[1631] "Show me the inventory trends."

[1632] "Show me last month's stock availability."

[1633] This system allows for voice input of instructions at logistics centers and enables real-time data analysis and visualization, which improves operational efficiency and enables faster decision-making.

[1634] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1635] Step 1:

[1636] The user gives instructions by voice input or text input using a smartphone.

[1637] Input: User's voice or text instructions

[1638] Output: Audio data or text data

[1639] Specific action: The user speaks into the smartphone microphone and says, "Tell me about this month's shipping status," or types the same thing into the text field.

[1640] Step 2:

[1641] The device converts the voice input into text.

[1642] Input: Audio data

[1643] Output: Text data

[1644] Specific operation: The device uses the SpeechRecognition library to convert voice data into text data. For example, the speech "Tell me about this month's shipping status" is converted into text "Tell me about this month's shipping status."

[1645] Step 3:

[1646] The terminal transmits the text data to the server.

[1647] Input: Text data

[1648] Output: Request to server

[1649] Specific operation: The terminal sends the converted text data to the server as an HTTP request.

[1650] Step 4:

[1651] The server parses the text data and generates the appropriate SQL queries.

[1652] Input: Text data

[1653] Output: SQL query

[1654] Specific operation: The server parses the text data "Tell me about this month's shipping status" and generates an SQL query to retrieve shipping data. For example, the query generated is "SELECT FROM shipments WHERE date >= '2023-01-01'".

[1655] Step 5:

[1656] The server retrieves the required data from the database.

[1657] Input: SQL query

[1658] Output: Data retrieved from the database

[1659] Specific operation: The server executes the generated SQL query and retrieves the relevant shipping data from the database. For example, shipping data from January 2023 onwards is retrieved.

[1660] Step 6:

[1661] The server analyzes the acquired data in real time.

[1662] Input: Data retrieved from a database

[1663] Output: Analysis results

[1664] Specific operation: The server analyzes the acquired shipping data using the Pandas library and calculates the monthly shipping status.

[1665] Step 7:

[1666] The server visualizes the analysis results in graphs and charts.

[1667] Input: Analysis results

[1668] Output: Graphs and charts

[1669] Specific operation: The server uses the Matplotlib library to visualize the analysis results as bar graphs and line graphs. For example, monthly shipping status is displayed as a bar graph.

[1670] Step 8:

[1671] The server sends the visualized results to the terminal.

[1672] Input: Graphs and charts

[1673] Output: Response to the terminal

[1674] Specific operation: The server sends the generated graphs and charts to the terminal as image data.

[1675] Step 9:

[1676] The terminal displays the visualized results on the display.

[1677] Input: Image data

[1678] Output: Graphs and charts displayed on the screen

[1679] Specific operation: The device displays the received image data on the screen and provides the results to the user. For example, a bar graph of monthly shipment status is displayed on the smartphone screen.

[1680] Example 2

[1681] Next, a description will be given of Example 2 of Form 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."

[1682] With conventional data analysis systems, data collection, preprocessing, analysis, and the generation and notification of marketing proposals are rarely performed consistently, and each process is fragmented, making it difficult to carry out efficient marketing activities.In addition, the provision of new insights based on analysis results is insufficient, making it difficult to optimize marketing strategies.

[1683] 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.

[1684] In this invention, the server includes means for accepting analysis instructions in Japanese, executing the corresponding data analysis, and visualizing the results, means for making marketing proposals and audience targeting based on the analysis results, means for providing new insights based on patterns and trends in the data and the instructed analysis, means for collecting sales data and customer data from a database, means for preprocessing the collected data, means for analyzing the preprocessed data, means for generating marketing proposals based on the analysis results, and means for notifying the generated marketing proposals. This makes it possible to consistently perform a series of processes from data collection to generating and notifying marketing proposals, enabling efficient marketing activities and the provision of new insights.

[1685] "Analysis instructions in Japanese" refers to a user issuing a data analysis request to the system in Japanese.

[1686] "Data analysis" refers to the process of using collected data to find specific patterns or trends.

[1687] "Visualization" refers to displaying the results of data analysis in a visual format such as a graph or chart.

[1688] "Marketing proposal" refers to proposing a strategy or action plan for promoting a specific product or service based on the results of data analysis.

[1689] "Audience targeting" refers to conducting marketing activities aimed at a specific customer demographic.

[1690] "New insights" refers to new findings and discoveries gained as a result of data analysis.

[1691] A "database" refers to a system for systematically storing and managing data.

[1692] "Sales Data" means data containing information regarding sales of products.

[1693] "Customer data" refers to data that includes information about customer attributes and purchasing history.

[1694] "Preprocessing" refers to cleaning and shaping data before analyzing it.

[1695] "Analysis results" refers to the conclusions and insights obtained through the data analysis process.

[1696] "Notification" refers to informing the user of generated marketing proposals and analysis results.

[1697] The present invention relates to a system for consistently performing data analysis and marketing proposals. A specific embodiment of this system will be described below.

[1698] First, the server collects sales and customer data from a database, which stores product sales information and customer purchase history. The server uses SQL queries to retrieve the required data.

[1699] Next, the server preprocesses the collected data using the Python Pandas library. Specifically, it performs tasks such as filling in missing values, removing outliers, and normalizing the data. This improves the quality of the data and increases the accuracy of the analysis.

[1700] Once the preprocessing is complete, the data is then used for data analysis. The server uses the Scikit-learn library to analyze sales trends and cluster customer purchasing behavior. For example, it analyzes whether sales of a particular product have declined over the past few months. It also uses clustering techniques to identify whether a particular customer demographic tends to purchase a particular product.

[1701] Based on the analysis results, the server generates marketing proposals. For example, it may propose discount campaigns for products with declining sales. It may also propose marketing activities targeted at specific customer segments. This allows for the creation of efficient marketing strategies.

[1702] Finally, the server notifies the user of the generated marketing proposals via email or dashboard. By receiving these notifications, the user can take appropriate marketing actions.

[1703] As a concrete example, consider the following scenario.

[1704] 1. The server collects sales data from the database for the past year.

[1705] 2. Use the Pandas library to impute missing values ​​and normalize the data.

[1706] 3. Use the Scikit-learn library to perform sales trend analysis, for example, discovering that sales of a particular product A have dropped by 20% over the past three months.

[1707] 4. The server generates a promotion proposal to address the decline in sales of product A. For example, it makes a proposal such as "Implement a discount campaign for product A."

[1708] 5. You also analyze customer data and discover that specific customer group B tends to purchase product A.

[1709] 6. The server proposes marketing activities targeting customer group B. For example, it may propose "send a special offer for product A to customer group B."

[1710] Examples of prompts to input to a generative AI model include:

[1711] "Analyze sales data from the past year to identify products with declining sales. Also, generate promotional suggestions for those products."

[1712] Using this prompt, the generative AI model can make appropriate analysis and suggestions.

[1713] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1714] Step 1:

[1715] The server collects sales and customer data from the database. It receives database connection information and SQL queries as input and obtains sales and customer data as output. Specifically, the server connects to the database and executes SQL queries to retrieve the required data.

[1716] Step 2:

[1717] The server preprocesses the collected data. It receives sales and customer data as input and produces preprocessed data as output. Specifically, the server uses the Python Pandas library to impute missing values, remove outliers, and normalize the data.

[1718] Step 3:

[1719] The server analyzes the preprocessed data. It receives the preprocessed data as input and generates analytical results as output. Specifically, the server uses the Scikit-learn library to perform sales trend analysis and cluster customer purchasing behavior.

[1720] Step 4:

[1721] The server generates marketing proposals based on the analysis results. It receives the analysis results as input and generates marketing proposals as output. Specifically, the server proposes promotions for products with declining sales and proposes marketing activities targeting specific customer segments.

[1722] Step 5:

[1723] The server notifies the user of the generated marketing proposal. It receives the marketing proposal as input and sends a notification to the user as output. Specifically, the server notifies the user of the proposal content by email or dashboard.

[1724] (Application example 2)

[1725] Next, a description will be given of Application Example 2 of Form 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."

[1726] In conventional marketing systems, sales data analysis and customer segmentation are often done manually, resulting in inefficiencies. Furthermore, because promotion strategies for declining sales products and targeted advertising proposals are not made in real time, it is difficult to respond quickly. Furthermore, there is a lack of a way to quickly notify marketing personnel of the generated proposals, which means that marketing activities cannot be carried out at the appropriate time.

[1727] 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 accepting analysis instructions in Japanese, executing the corresponding data analysis, and visualizing the results, means for making marketing proposals and audience targeting based on the analysis results, means for providing new insights based on patterns and trends in the data and the instructed analysis, means for analyzing sales data in real time and proposing promotion strategies for products with declining sales, means for identifying customer segments using a clustering algorithm and proposing targeted advertising, and means for notifying the generated proposals. This enables efficient analysis of sales data and rapid proposals of promotion strategies and targeted advertising.

[1728] "Analysis instructions in Japanese" are data analysis instructions given by the user in Japanese.

[1729] "Data analysis" is the analysis of collected data using statistical methods and algorithms to extract useful information and patterns.

[1730] "Visualization" is the presentation of the results of data analysis in a visual format such as a graph or chart.

[1731] A "marketing proposal" is a proposal of strategies or measures aimed at promoting sales of specific products or services based on the results of analysis.

[1732] "Audience targeting" refers to conducting marketing activities aimed at a specific customer demographic.

[1733] "New insights" are new findings and understandings gained based on data analysis and pattern recognition.

[1734] "Sales data" is numerical information about the sales of a particular product or service.

[1735] "Real-time analysis" means analyzing data as it is generated.

[1736] A "promotion strategy" is a plan or measure to increase product sales.

[1737] A "clustering algorithm" is a computational method for dividing data into groups based on similarity.

[1738] A "customer demographic" is a group of customers who share certain common characteristics.

[1739] "Targeted advertising" refers to advertising activities aimed at specific customer segments.

[1740] "Notification" means informing the user of generated suggestions or information.

[1741] A system for implementing this invention is configured as follows: The server has means for accepting analysis instructions in Japanese, executing the corresponding data analysis, and visualizing the results. A user can input analysis instructions in Japanese using a smartphone. The server performs data analysis based on the accepted instructions and displays the results in a visual format such as a graph or chart.

[1742] The server then provides a means for making marketing proposals and audience targeting based on the analysis results. For example, it analyzes sales data in real time and proposes promotion strategies for products with declining sales. It also uses a clustering algorithm to identify customer segments and propose targeted advertising for those segments.

[1743] The server also provides a means to provide new insights based on patterns and trends in the data and the analysis you specify, allowing you to leverage new insights gained from data analysis to develop more effective marketing strategies.

[1744] The server also has a means for notifying the generated proposals. For example, the generated promotion proposals and targeted advertising proposals can be notified to the marketing personnel in real time by email, which enables the marketing personnel to respond promptly.

[1745] Consider the following scenario as a specific example: A user uses a smartphone to input a prompt statement such as, "If sales of a specific product are declining, please generate a promotion proposal for that product. Also, if a specific customer segment has a tendency to purchase a specific product, please propose targeted advertisements for that customer segment." The server analyzes data based on this instruction, proposes a promotion strategy for the product with declining sales, and generates targeted advertisements for the specific customer segment. The generated proposals are notified to the marketing personnel via email.

[1746] The following hardware and software are used to realize this system. For hardware, a smartphone and a server are required. For software, Python, Pandas, Scikit-learn, and the SMTP library are used. Python is used to implement data analysis and machine learning algorithms, and Pandas is used to preprocess and manipulate data. Scikit-learn is used to implement the clustering algorithm, and the SMTP library is used to implement the email notification function.

[1747] As described above, the system for implementing this invention realizes a series of processes in which a user inputs analysis instructions in Japanese, the server performs data analysis based on those instructions, visualizes the results, makes marketing proposals and audience targeting, and notifies the user of the generated proposals.

[1748] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1749] Step 1:

[1750] The user inputs analysis instructions in Japanese using a smartphone. The input instructions include promotion suggestions for declining sales of a specific product and the generation of targeted advertisements for a specific customer demographic. The input data is then sent to the server.

[1751] Step 2:

[1752] The server interprets the received Japanese analysis instructions and performs the corresponding data analysis. Specifically, it reads the sales data and standardizes it. The input data is the sales data, and the output is the standardized data.

[1753] Step 3:

[1754] The server runs a clustering algorithm on the standardized data to identify customer demographics. Specifically, it uses the KMeans algorithm to divide the data into clusters. The input data is the standardized data, and the output is data with cluster labels.

[1755] Step 4:

[1756] The server identifies products with declining sales based on the cluster-labeled data. Specifically, it extracts products with sales below average. The input data is data with cluster labels, and the output is a list of products with declining sales.

[1757] Step 5:

[1758] The server generates a promotion strategy based on the list of declining sales items. Specifically, it creates promotion proposals for declining sales items. The input data is the list of declining sales items, and the output is the promotion proposals.

[1759] Step 6:

[1760] The server generates targeted advertising proposals based on the cluster-labeled data. Specifically, it proposes optimal advertising strategies for each cluster. The input data is the cluster-labeled data, and the output is targeted advertising proposals.

[1761] Step 7:

[1762] The server notifies the marketer of the generated promotion proposals and targeted advertising proposals. Specifically, it uses an SMTP library to send email notifications. The input data are the promotion proposals and targeted advertising proposals, and the output is a notification email.

[1763] The above steps realize a series of processes in which the server analyzes data based on the analysis instructions entered by the user in Japanese, generates a promotion strategy and targeted advertisements, and notifies the marketing personnel.

[1764] Example 3

[1765] Next, a third embodiment of the third embodiment will be described. 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."

[1766] Conventional data analysis systems require users to issue specific analytical instructions, and a great deal of effort is required to gain new insights based on past data and analytical results. Furthermore, they lack the functionality to predict what information users will want to know next, making efficient data utilization difficult. This slows down the user's decision-making process, making it difficult for businesses to respond quickly.

[1767] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1768] In this invention, the server includes means for accepting analysis instructions in Japanese, performing the corresponding data analysis, and visualizing the results, means for making marketing proposals and audience targeting based on the analysis results, means for providing new insights based on patterns and trends in the data and the instructed analysis, means for collecting and preprocessing past data, means for performing pattern analysis using the preprocessed data, means for generating a predictive model using a generative AI model and predicting the information the user will want to know next, and means for providing the predicted information to the user. This allows the user to obtain the information they need without having to give specific instructions, enabling efficient data utilization and rapid decision-making.

[1769] "Analysis instructions in Japanese" are data analysis instructions given by the user in Japanese.

[1770] "Data analysis" is the process of using collected data to extract and understand information for a specific purpose.

[1771] "Visualization" is the presentation of the results of data analysis in a visual format such as a graph or chart.

[1772] A "marketing proposal" is a specific proposal for promoting products or services based on the results of data analysis.

[1773] "Audience targeting" refers to developing a marketing strategy that targets a specific customer demographic.

[1774] "New insights" are new findings and understandings gained from past data and analytical results.

[1775] "Historical data" is data previously collected and used in the current analysis.

[1776] "Preprocessing" refers to the process of formatting data before data analysis, filling in missing values, normalizing data, etc.

[1777] "Pattern analysis" is an analytical method for finding patterns and trends in data.

[1778] A "generative AI model" is a model that uses artificial intelligence technology to generate new information and predictions from data.

[1779] A "predictive model" is a model that predicts future events or outcomes based on past data.

[1780] "The information that the user wants to know next" is information that is highly likely to be needed next by the user.

[1781] "Predicted Information" is information about future events or outcomes generated by a predictive model.

[1782] MODE FOR CARRYING OUT THE INVENTION

[1783] The present invention relates to a data analysis system that allows a user to obtain necessary information without issuing specific instructions. A specific embodiment of this system will be described below.

[1784] 1. Generate the system program

[1785] The server generates programs for data analysis and prediction based on user requests. These programs use generative AI models to learn patterns from past data and the analysis instructions, and provide new insights.

[1786] 2. Explain the program's processing in natural language

[1787] The server processes the data as follows:

[1788] 1. Data Collection:

[1789] The server collects the analysis instructions and results that the user has performed in the past from a database. Specifically, it uses a relational database (e.g., MySQL or PostgreSQL) to acquire the user's past analysis history.

[1790] 2. Data Preprocessing:

[1791] The server preprocesses the collected data. Specifically, it fills in missing values ​​and normalizes the data. This process uses the Python Pandas library. For example, if there are missing values, they are filled in with the mean value, and the data is standardized to unify the scale.

[1792] 3. Pattern Analysis:

[1793] The server uses the preprocessed data to analyze the user's past analysis instructions and result patterns. This analysis uses machine learning libraries such as Scikit-learn. Specifically, it performs clustering and regression analysis to extract user behavior patterns.

[1794] 4. Generate predictive models:

[1795] The server uses a generative AI model (e.g., GPT-4) to predict what the user will want to know next. This model generates new insights based on past data and pattern analysis. For example, if the user frequently analyzes monthly sales data in the past, it can predict the next month's sales.

[1796] 5. Information provision:

[1797] The server provides the predicted information to the user. Specifically, it sends a notification to the user's device and displays the prediction results. For example, it displays the sales forecast for the next month in a graph format.

[1798] 3. Add examples to your description

[1799] For example, if a user frequently performs "monthly sales data analysis" in the past, the server will learn that pattern. Then, without the user having to issue a new analysis command, the server will automatically provide "next month's sales forecast."

[1800] Example prompts to be input to the generative AI model

[1801] "Please forecast next month's sales based on past sales data."

[1802] In this way, the user can obtain the necessary information without having to specifically specify what he or she is looking for. The flow of the identification process in the third embodiment will be described with reference to FIG.

[1803] Step 1:

[1804] The server collects the analysis instructions and results of the user's past executions from a database. Specifically, it executes SQL queries on a relational database (e.g., MySQL or PostgreSQL) to retrieve the user's past analysis history. The input is filter conditions such as the user ID and period, and the output is a dataset of the corresponding analysis instructions and their results.

[1805] Step 2:

[1806] The server preprocesses the collected data. Specifically, it uses the Python Pandas library to impute missing values ​​and normalize the data. For example, if there are missing values, they are imputed with the mean value and standardized to unify the scale of the data. The input is the dataset obtained in step 1, and the output is the preprocessed, clean dataset.

[1807] Step 3:

[1808] The server performs pattern analysis using the preprocessed data. Specifically, it uses a machine learning library such as Scikit-learn to perform clustering and regression analysis. For example, it uses K-means clustering to classify the data into several groups and extract user behavior patterns. The input is the dataset preprocessed in step 2, and the output is data showing the analyzed patterns and trends.

[1809] Step 4:

[1810] The server uses a generative AI model (e.g., GPT-4) to generate a predictive model and predict the information the user will want to know next. Specifically, it uses past data and the results of pattern analysis as input to generate new insights, such as sales forecasts for the next month. The inputs are the results of the pattern analysis obtained in step 3 and past data, and the output is predicted information. For example, it generates a prompt sentence such as, "Please forecast sales for the next month based on past sales data."

[1811] Step 5:

[1812] The server provides the predicted information to the user. Specifically, it sends a notification to the user's device and displays the prediction results. For example, it displays the sales forecast for the next month in a graph format and sends a notification to the user. The input is the prediction information generated in step 4, and the output is the prediction results displayed on the user's device.

[1813] (Application example 3)

[1814] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[1815] Conventional data analysis systems require users to provide specific analysis instructions, and marketing proposals and audience targeting based on analysis results are limited. Furthermore, they lack the ability to predict which products users are likely to purchase next based on their past purchase and browsing history, making it difficult to improve the user shopping experience. This creates a problem in that users spend a lot of time and effort trying to obtain the information they need.

[1816] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[1817] In this invention, the server includes means for accepting analysis instructions in Japanese, performing the corresponding data analysis, and visualizing the results, means for making marketing proposals and audience targeting based on the analysis results, means for providing new insights based on patterns and trends in the data and the instructed analysis, means for analyzing the user's past purchase history and browsing history and predicting and suggesting products that the user is likely to purchase next, means for automatically notifying the user of new products and sale information that the user is likely to be interested in, and means for providing related information and reviews when the user asks a question about a specific product. This makes it possible to predict and suggest necessary information and products that the user is likely to purchase next, even without the user providing specific instructions, thereby improving the user's shopping experience.

[1818] An "analysis instruction" is a request for data analysis made by a user to the system.

[1819] "Data analysis" is the process of processing collected data and extracting meaningful information.

[1820] "Visualization" is the presentation of the results of data analysis in a visual format such as a graph or chart.

[1821] "Marketing proposals" are sales promotion proposals made to users based on the analysis results.

[1822] "Audience targeting" is a method for conducting effective marketing to a specific user group.

[1823] "New insights" are new findings and understandings that emerge from patterns in data and analysis.

[1824] "Purchase history" is a record of products a user has purchased in the past.

[1825] "Browse history" is a record of products that a user has viewed in the past.

[1826] "Product prediction" is the process of predicting the next product a user is likely to purchase based on their past behavioral data.

[1827] "New product notification" means automatically notifying users of information about new products that may be of interest to them.

[1828] "Sales information notification" means automatically notifying users of sales information that may be of interest to them.

[1829] "Related information" is information related to a particular product that is provided when a user asks a question about that product.

[1830] A "review" is a rating or opinion of a product provided by other users when a user asks a question about that product.

[1831] A system for implementing this invention includes a server, a user terminal, and a database. The server accepts analysis instructions in Japanese from a user, performs the corresponding data analysis, and visualizes the results. The server then makes marketing proposals and audience targeting based on the analysis results. The server also provides new insights based on patterns and trends in the data and the analysis instructions.

[1832] The user terminal is a device such as a smartphone or PC that the user uses to input analysis instructions and check the results. The user terminal receives notifications from the server and displays them to the user.

[1833] The database stores the user's past purchase history and browsing history, and the server analyzes this data. The server analyzes the user's past purchase history and browsing history to predict and suggest products that the user is likely to purchase next. The server also automatically notifies the user of new products and sales information that may interest the user. Furthermore, when a user asks about a specific product, the server provides related information and reviews.

[1834] In a specific example, the server is implemented using the Python programming language. The Pandas library is used for data processing, and the Scikit-learn library is used for the machine learning model. The server reads the user's purchase history data, separates it into features and targets, and splits the data into a training set and a test set. A random forest classifier is used to train the model and predict the next product that the user is likely to purchase based on new user data.

[1835] For example, if user A has purchased many smartphone-related products in the past, the server will suggest accessories or new smartphone models that they are likely to purchase next.Also, if the user inputs a prompt such as "Please predict the product that I am likely to purchase next," the server will predict and suggest related products.

[1836] Example prompt sentence:

[1837] "Based on User A's past purchase history, please predict the product that he is likely to purchase next."

[1838] In this way, the user's shopping experience is improved because the information they need and the products they are likely to purchase next are predicted and suggested without them having to give any specific instructions.

[1839] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[1840] Step 1:

[1841] The server accepts analysis instructions in Japanese from the user terminal.

[1842] Input: Japanese analysis instructions entered by the user into the terminal

[1843] Output: Analysis instructions received by the server

[1844] Specific operation: A user inputs specific analysis instructions using a smartphone or PC and sends them to the server, which receives the instructions and prepares them for analysis.

[1845] Step 2:

[1846] The server performs the corresponding data analysis based on the received analysis instructions.

[1847] Input: Analysis instructions received by the server, data stored in the database

[1848] Output: Results of data analysis

[1849] Specific operation: The server uses the Pandas library to extract the necessary data from the database and process it according to the analytical instructions, for example, aggregating sales data for a specific product and analyzing trends.

[1850] Step 3:

[1851] The server visualizes the results of the data analysis.

[1852] Input: Results of data analysis

[1853] Output: visualized data (graphs and charts)

[1854] Specific operation: The server uses visualization libraries such as Matplotlib and Seaborn to generate analysis results as graphs and charts and send them to the user's device.

[1855] Step 4:

[1856] The server makes marketing suggestions and audience targeting based on the analysis results.

[1857] Input: Results of data analysis

[1858] Output: Marketing proposals and targeting information

[1859] Specific operation: Based on the analysis results, the server generates and proposes effective marketing strategies for specific user groups. For example, it proposes related products to users who have purchased a specific product.

[1860] Step 5:

[1861] The server provides new insights based on patterns and trends in the data and directed analysis.

[1862] Input: Past analysis instructions and analysis results

[1863] Output: New insights

[1864] How it works: The server inputs historical data and analytical results into a machine learning model to generate new insights, such as predicting what products a customer is likely to buy next.

[1865] Step 6:

[1866] The server analyzes the user's past purchase history and browsing history, predicts and suggests products that the user is likely to purchase next.

[1867] Input: User purchase history and browsing history

[1868] Output: A list of products that are likely to be purchased next

[1869] What it does: The server uses the Scikit-learn library to train a random forest classifier to predict the next likely purchase based on the user's past data.

[1870] Step 7:

[1871] The server automatically notifies users of new products and sales information that may be of interest to them.

[1872] Input: User interest data, new product and sale information

[1873] Output: Notifications of new products and sales information

[1874] Specific operation: The server analyzes the user's interest data and automatically notifies the user's device of related new products and sales information.

[1875] Step 8:

[1876] The server provides relevant information and reviews when users ask about specific products.

[1877] Input: User questions, product information and reviews stored in the database

[1878] Output: Related information and reviews

[1879] Specific operation: When a user asks a question about a specific product, the server searches the database for related information and reviews and provides them to the user's device.

[1880] 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.

[1881] "Example 1"

[1882] One embodiment of the present invention is a system incorporating an emotion engine. This system recognizes a user's emotions and performs data analysis and marketing proposals based on those emotions. Specifically, suppose a user instructs the system to "show me monthly sales trends." At this time, the emotion engine reads the user's emotions from their tone of voice, facial expressions, and choice of words. For example, if the system determines that the user is feeling impatient or anxious, it quickly outputs the analysis results. Alternatively, if the system determines that the user is feeling joy or relief, it makes marketing proposals that will amplify that joy.

[1883] "Example 2"

[1884] The emotion engine also provides new insights based on the user's emotions. For example, if the system determines that the user is feeling excited or excited, it can provide new insights that complement those emotions. Specifically, it can provide the insight that sales of a particular product are skyrocketing, further increasing the user's excitement and anticipation.

[1885] "Example 3"

[1886] Furthermore, the emotion engine performs audience targeting based on the user's emotions. For example, if the system determines that the user is feeling angry or frustrated, it will target the audience in a way that alleviates those emotions. Specifically, it will suggest products or services that may alleviate the frustration of a dissatisfied user.

[1887] The processing flow of each embodiment will be described below.

[1888] "Example 1"

[1889] Step 1: The user issues a command to the system. For example, the user might say, "Show me the monthly sales trend."

[1890] Step 2: The emotion engine reads the user's emotions from their tone of voice, facial expressions, and choice of words. Step 3: The system outputs the analysis results based on the emotions read by the emotion engine. For example, if the system determines that the user is feeling impatient or anxious, it will quickly output the analysis results.

[1891] Step 4: The system makes marketing suggestions based on the emotions detected by the emotion engine. For example, if the system determines that the user is feeling joy or relief, it will make marketing suggestions that will amplify that joy.

[1892] "Example 2"

[1893] Step 1: The emotion engine reads the user's emotions, for example, determining whether the user is feeling excited or excit...

Claims

1. A means to accept analysis instructions in Japanese, execute the corresponding data analysis, and visualize the results; An emotion engine that recognizes emotions of customers in a physical store based on their tone of voice and facial expressions; A customer service method that takes into consideration the recognized emotions of the customer, and a means for making a proposal for sales promotion and customer acquisition to the customer based on the recognized emotions of the customer; A means for suggesting optimal advertisements based on the recognized customer sentiment by inputting specific prompt sentences into a generative AI model; A system including:

2. The system of claim 1, wherein a specific prompt sentence is input to the generative AI model to suggest advertisements and campaigns aimed at promoting a list of products with declining sales.

3. A means for predicting and suggesting products that the customer may be interested in based on the customer's past behavior history and the customer's emotions recognized by the emotion engine; A means for suggesting products and services that reduce the customer's dissatisfaction or anger based on the customer's emotions recognized by the emotion engine; The system of claim 1 , comprising:

Images