System

Abstract

A system is provided.SOLUTION: A system comprising: means for collecting historical behavior data of a user; means for analyzing the historical behavior data and generating a profile of the user; means for generating a plurality of user copies based on the profile; means for conducting a virtual interview or questionnaire on the user copies regarding a new service or a new function; means for aggregating and analyzing results of the interview or questionnaire; and means for feeding back the results.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] When developing new services or features, obtaining appropriate feedback from target users requires conducting interviews and surveys with many actual users, which is time-consuming and costly. Furthermore, it is not always possible to obtain appropriate feedback from all users, and there are also issues with the reliability and reproducibility of survey results. A system that can resolve these issues and efficiently and accurately reflect user opinions is needed. [Means for solving the problem]

[0005] The present invention provides a means for collecting user behavior history data, analyzing it, and generating a user profile. It also provides a means for generating multiple user copies based on the profile and conducting virtual interviews or surveys on new services or features for these user copies. By building a system that aggregates and analyzes the results and ultimately provides feedback to the relevant teams, it is possible to obtain highly reliable feedback while reducing the time and cost associated with surveying actual users.

[0006] "User behavioral history data" refers to records of operations and activities performed by users on the Internet, and specifically includes web page browsing history, search keywords, click data, purchase history, etc.

[0007] A "profile" refers to a set of information that indicates a user's interests, values, and attributes, generated based on the user's behavioral history data.

[0008] "User copy" refers to a simulated virtual user based on a generated profile that replicates the characteristics of a real user.

[0009] "Interviews or surveys" refers to methods of asking questions about new services or features and collecting responses.

[0010] "Collection and analysis" refers to the process of organizing and statistically processing collected interview and questionnaire response data.

[0011] "Feedback" refers to the act of providing improvements and new insights to the relevant team based on the results of interviews and surveys.

[0012] "Database means" refers to a device or system for storing and managing behavior history data. [Brief explanation of the drawings]

[0013] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0014] 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.

[0015] First, the terms used in the following description will be explained.

[0016] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0017] 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.

[0018] 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.

[0019] 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.

[0020] 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."

[0021] [First embodiment]

[0022] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0023] 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.

[0024] 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).

[0025] 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.

[0026] 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.

[0027] 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.

[0028] 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.

[0029] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0030] 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.

[0031] 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.

[0032] 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.

[0033] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0034] The present invention relates to a system for generating a user copy based on user behavior history data, thereby improving the efficiency of market research for new services and functions. Specific embodiments for carrying out the present invention are as follows.

[0035] Collection of user behavior history data

[0036] When a user engages in online activities, the device records this behavioral history data, specifically, the web pages viewed, search keywords, links clicked, products purchased, etc. This behavioral history data is sent to a server in real time or periodically.

[0037] Saving and analyzing behavioral history data

[0038] The server then stores the received behavioral history data in a database. The server then analyzes this data and applies machine learning algorithms to extract patterns such as user interests and purchasing trends. For example, a user who frequently views products in a particular category can be inferred to be particularly interested in that category.

[0039] User profile generation

[0040] The server generates a user profile based on the analysis results, which includes the user's interests, values ​​(e.g., whether they prioritize cost or new technology), and demographic information (e.g., age, gender, region).

[0041] Generate a user copy

[0042] The server then creates multiple copies of the user based on the profile. These copies are virtual users that simulate different situations and values. For example, there could be multiple types, such as a "technology-loving user who prioritizes new features over price" and a "cost-conscious user who prioritizes price."

[0043] Conducting a virtual interview or survey

[0044] The server prepares interviews and questionnaires about new services and features and conducts them on the generated user copies. Because the user copies are virtual, the server automatically generates answers based on pre-set scenarios. As a result, a wide variety of feedback can be obtained without interacting with real users.

[0045] Calculation and analysis of results

[0046] The server aggregates and analyzes the results of virtual interviews and surveys. For example, it statistically processes user evaluations of new features to identify points of interest and areas that need improvement.

[0047] Feedback of results

[0048] Finally, the server compiles the analysis results into a report and provides feedback to the development and marketing teams, which helps them develop new services, improve new features, and formulate marketing strategies.

[0049] As described above, the present invention utilizes user behavior history data to obtain feedback efficiently and accurately, thereby significantly reducing time and costs. As a specific example, if a user shows interest in a particular gadget, a virtual interview can be conducted about new features related to the gadget, and a marketing strategy can be built based on the results.

[0050] The processing flow will be explained below.

[0051] Step 1:

[0052] A user logs in to an internet platform. The device recognizes the user's user ID, such as their Yahoo! JAPAN ID, and begins collecting behavioral history data associated with that ID.

[0053] Step 2:

[0054] The device records user behavior data, specifically, the web pages the user views, the keywords they search for, the links they click, the products they purchase, and other data collected in real time or periodically.

[0055] Step 3:

[0056] The behavioral history data collected by the device is sent to the server. Data transfer is performed by batch processing or streaming processing.

[0057] Step 4:

[0058] The server stores the received behavioral history data in a database, which then stores the data appropriately and creates an index for fast access.

[0059] Step 5:

[0060] The server preprocesses the stored data, cleaning out invalid and noisy data and extracting only the necessary data. The server then classifies and organizes the data by category.

[0061] Step 6:

[0062] The server analyzes the data and applies machine learning algorithms to extract user behavioral patterns, interests, purchasing tendencies, etc. For example, if a user frequently browses a particular product category, it can be assumed that they have a high interest in that category.

[0063] Step 7:

[0064] The server generates a user profile. Based on the analysis results from the previous step, the profile is compiled based on the user's interests, categories of interest, values, and attribute information (age, gender, region, etc.).

[0065] Step 8:

[0066] The server generates multiple user copies. Based on the generated profiles, it creates several virtual users with different scenarios and values. For example, it creates a user copy that is interested in technology and prioritizes functionality over price, and a user copy that is cost-sensitive.

[0067] Step 9:

[0068] The server conducts interviews and surveys about new services and features with the user copy, asking the user copy simulated questions according to a scenario set within the system and collecting their responses.

[0069] Step 10:

[0070] The server compiles and analyzes the results of interviews and surveys, statistically processing the collected data to analyze how users respond to new services and features.

[0071] Step 11:

[0072] The server compiles the results of the analysis and compiles them into a report. The server extracts insights and potential problems and organizes them in report format.

[0073] Step 12:

[0074] The server will then provide the generated reports to the development and marketing teams, who will use the feedback to improve new services and features, and to develop marketing strategies.

[0075] This series of specific processes makes it possible to effectively utilize user behavior history data and conduct market research efficiently and accurately.

[0076] Example 1

[0077] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0078] Modern market research and new product feedback is time-consuming and costly, and the amount of feedback available from users is limited, creating a need for a fast and efficient method of gathering diverse user opinions.

[0079] 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.

[0080] In this invention, the server includes means for collecting user behavior history data, means for transmitting the behavior history data to the server, means for saving the behavior history data in a database, means for analyzing the behavior history data and generating a user profile, means for generating a plurality of user copies based on the generated profiles, means for conducting virtual interviews or questionnaires regarding new services or new functions with the user copies, means for aggregating and analyzing the results of the interviews or questionnaires, and means for compiling the results into a report and providing feedback to the development team and the marketing team. This makes it possible to quickly and efficiently collect diverse user opinions and greatly improve the efficiency of market research and new product feedback.

[0081] "User behavioral history data" refers to information about a user's activities on the Internet, specifically web pages viewed, search keywords, links clicked, products purchased, etc.

[0082] The term "server" refers to a central computer system that stores, analyzes, and processes the behavioral history data, collects data from multiple users, and performs various processes based on that data.

[0083] A "database" is a storage system that organizes and stores the behavioral history data, and is used for subsequent analysis and profile generation.

[0084] A "user profile" is a set of data generated by a machine learning algorithm that includes a user's interests, concerns, purchasing trends, values, and attribute information.

[0085] A "user copy" is a virtual user simulated based on the profile of the user, and is a modeled entity with different situations and values.

[0086] A "virtual interview or survey" is a process in which user copy is asked questions about a new service or feature and hypothetical answers are generated.

[0087] "Means for aggregation and analysis" refers to a method or system for statistically processing the results of virtual interviews or questionnaires and extracting user evaluations and opinions.

[0088] "Feedback means" refers to the process of creating a report based on the results of the above-mentioned compilation and analysis, and transmitting this to the relevant team.

[0089] The present invention relates to a system for generating a user copy based on user behavior history data, thereby improving the efficiency of market research for new services and functions. Specific embodiments for carrying out the present invention will be described below.

[0090] Collection of user behavior history data

[0091] When a user engages in online activities, their device records this behavioral history data. Specifically, it collects data on the web pages they visit, search keywords, links they click, and products they purchase. This data is collected through web browsers, mobile applications, and cloud services. This behavioral history data is sent to a server in real time or periodically.

[0092] Examples:

[0093] A user searches for "smartwatch" on an e-commerce website and browses multiple product pages. At this point, the device collects information such as the search keyword "smartwatch," the viewed page URL, and click path.

[0094] Sending behavioral history data

[0095] The device sends the collected behavioral history data to the server using an encrypted protocol (e.g., HTTPS), which protects the user's privacy.

[0096] Examples:

[0097] When the user finishes browsing the site, data is automatically sent from the device to the server.

[0098] Saving and analyzing behavioral history data

[0099] The server stores the received behavioral history data in a database. Based on the stored data, the server applies a machine learning algorithm (e.g., a sparse classifier) ​​to extract the user's interests and purchasing trends.

[0100] Examples:

[0101] Using datasets stored on a server, machine learning algorithms create profiles of users who are interested in "smartwatches."

[0102] Generate a user profile

[0103] The server generates a user profile based on the analysis of the behavioral history data, which includes the user's interests, values ​​(e.g., whether they prioritize cost or new technology), and demographic information (e.g., age, gender, region).

[0104] Examples:

[0105] If a user frequently searches for information related to "smartwatches," their profile might include "interest category: gadgets," "values: emphasis on new technology," and "demographics: 30 years old, male, living in an urban area."

[0106] Generate a user copy

[0107] The server then creates multiple copies of the user based on the profile, simulating the user as a virtual entity with different situations and values.

[0108] Examples:

[0109] The server generates two copies of the user: one that prioritizes new features and one that prioritizes costs.

[0110] Conducting a virtual interview or survey

[0111] The server prepares interviews and questionnaires about new services and features and conducts them on the virtual user copy, who then automatically generates answers based on scenarios preset by the server.

[0112] Example prompt sentence:

[0113] We generated the following profile based on historical user behavior data:

[0114] Interests: Smartwatches

[0115] Demographic information: 30 years old, male, urban resident

[0116] Values: Emphasis on new features

[0117] Create a user copy with the following conditions:

[0118] Copy 1: New feature-focused virtual user

[0119] Copy 2: Price-sensitive virtual users

[0120] For each user copy, answer the following questions:

[0121] 1. What do you think about the new tracking features on smartwatches?

[0122] 2. What is the maximum price you are willing to pay for this new feature?

[0123] Calculation and analysis of results

[0124] The server aggregates and statistically analyzes the results of the virtual interviews and surveys, and uses statistical software to extract user interests and areas for improvement based on the analysis results.

[0125] Examples:

[0126] The server records the responses collected from user copies in a database and analyzes the data using statistical software such as R or Python, extracting, for example, satisfaction scores for new features and price sensitivity.

[0127] Feedback of results

[0128] The server compiles the analysis results into a report and provides feedback to the development and marketing teams, including specific suggestions for improvements and market strategies.

[0129] Examples:

[0130] The server generates a PDF report of the analysis results and sends it to the development and marketing teams via email. The report includes the user copy evaluation results, points of interest, and improvement suggestions.

[0131] The above is a specific embodiment for carrying out the present invention.

[0132] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0133] Step 1:

[0134] The device collects data about the user's online activities, including the web pages the user views, search keywords, links clicked, and products purchased. The device temporarily stores this data in local storage.

[0135] Specific behavior:

[0136] A user searches for "smartwatch" on an e-commerce website and browses multiple product pages. At this point, the device collects information such as the search keyword "smartwatch," the viewed page URL, and click path.

[0137] Step 2:

[0138] The device sends the collected behavioral history data to the server. The input includes all the behavioral history data collected in the previous step. The data is sent using an encrypted protocol (e.g., HTTPS).

[0139] Specific behavior:

[0140] When a user browses a website, the device automatically sends activity history data to the server. For example, the data is sent when the user finishes browsing the website.

[0141] Step 3:

[0142] The server stores the received behavioral history data in a database. The input includes behavioral history data sent from the terminal. The database organizes and stores this data.

[0143] Specific behavior:

[0144] The server uses a database system (e.g., SQL database) to efficiently store user behavior history data, which makes it possible to reference the data required for subsequent analysis processing.

[0145] Step 4:

[0146] The server analyzes the stored behavioral history data and generates a user profile. The input includes the behavioral history data stored in the database. Machine learning algorithms are applied to analyze the data and extract the user's interests and purchasing trends. The output is the generated user profile.

[0147] Specific behavior:

[0148] The server uses a machine learning model (e.g., a sparse classifier) ​​based on the behavioral history dataset to generate a profile of the user's interest categories (e.g., "smartwatch") and values ​​(e.g., "emphasis on new technology").

[0149] Step 5:

[0150] The server generates multiple user copies based on the generated profile. The input includes the user's profile. Each user copy is modeled as a virtual entity with different situations and values. The output is the generated user copy.

[0151] Specific behavior:

[0152] Based on the characteristics of the user profile, the server generates two types of user copies: a "new feature-oriented type" and a "cost-oriented type."

[0153] Step 6:

[0154] The server prepares a virtual interview or questionnaire about the new service or feature and administers it to the generated user copy. The input includes questions about the new service or feature and the user copy. The server automatically generates answers based on a pre-set scenario. The output is the generated answers.

[0155] Specific behavior:

[0156] The server poses questions to the user copy, such as "What do you think about the new tracking features on your smartwatch?" or "What is the maximum price you would be willing to pay for this new feature?", and collects hypothetical responses.

[0157] Step 7:

[0158] The server aggregates and statistically analyzes the results of the virtual interviews and surveys. The input includes user-copied responses. The server processes these data using statistical software to derive user ratings and opinions. The output is the analysis results.

[0159] Specific behavior:

[0160] The server stores the response data obtained from user copies in a database and performs statistical analysis using statistical software such as R or Python, for example, to extract the distribution of satisfaction scores for new features and price sensitivity.

[0161] Step 8:

[0162] The server compiles the analysis results as a report and feeds it back to the development and marketing teams. The input includes the results of the statistical analysis. The output is a report.

[0163] Specific behavior:

[0164] The server generates a PDF report based on the analysis results and sends it to the development and marketing teams by email, including the "User Copy Evaluation Results," "Points of Interest," and "Improvement Suggestions."

[0165] (Application example 1)

[0166] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0167] Conventional methods of market research and product recommendation are not only time-consuming and costly, but also have the drawback of being difficult to respond quickly to diverse user needs. Furthermore, when using user behavior data to recommend products, it is difficult to provide personalized and accurate recommendations, which hinders the expectation of improving user satisfaction.

[0168] 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.

[0169] In this invention, the server includes means for collecting user behavior history data, means for analyzing the behavior history data and generating a user profile, means for generating multiple user copies based on the profile, means for recommending products or services based on the user copies, means for providing the user with the recommendation results, means for aggregating and analyzing the recommendation results, and means for feeding back the results. This makes it possible to quickly respond to diverse user needs and recommend personalized and accurate products and services, thereby improving user satisfaction and making market research more efficient.

[0170] "User behavioral history data" refers to data about a user's activities on the Internet, such as web pages viewed, search keywords, links clicked, and products purchased.

[0171] "Means for collecting behavioral history data" refers to means that has the function of recording a user's Internet activity and transmitting that data to a server in real time or periodically.

[0172] "Means for analyzing behavioral history data and generating user profiles" refers to means for analyzing collected behavioral history data using machine learning algorithms, etc., and extracting patterns such as user interests, concerns, and purchasing tendencies to generate profiles.

[0173] The "means for generating multiple user copies based on a profile" is a means for creating multiple virtual users that simulate different situations and values ​​based on the generated profile.

[0174] The "means for recommending a product or service based on a user copy" is a means for recommending an optimal product or service based on a profile of a virtual user copy.

[0175] The "means of providing the user with the results of the recommendation" refers to a means of providing the recommended products or services to the user through a smartphone application or the like so that the user can view them.

[0176] The "means for aggregating and analyzing the results of recommendations" refers to the means for collecting data on users' reactions to recommended products and services, and statistically processing and analyzing the data.

[0177] "Means of feedback of results" refers to compiling the analysis results in the form of a report or similar and providing them to the development team or marketing team.

[0178] The present invention relates to a system that generates a user copy based on user behavior history data and improves the efficiency of market research for new services and functions. Specifically, the system is implemented as follows.

[0179] Collecting behavioral history data

[0180] When a user browses the internet or makes a purchase, the device records this behavioral history data, including the web pages viewed, search keywords, links clicked, and products purchased. This data is sent to a server in real time or periodically via a smartphone application.

[0181] Saving and analyzing behavioral history data

[0182] The server stores the transmitted behavioral history data in a database. The stored data is analyzed using machine learning algorithms. These algorithms include clustering techniques (e.g., KMeans) and recommendation systems (e.g., Recommendation Engine). This allows patterns such as user interests and purchasing trends to be extracted.

[0183] Generate a user profile

[0184] The server generates a user profile based on the analysis results, which includes the user's interests, values ​​(e.g., whether they prioritize cost or new technology), and demographic information (e.g., age, gender, region).

[0185] Generate a user copy

[0186] The server then generates multiple user copies based on the generated profile. These copies are virtual users that simulate different situations and values. For example, various user copies can be generated, such as a "technology-loving user who prioritizes new features over price" and a "cost-conscious user who prioritizes price."

[0187] Product and service recommendations

[0188] Based on the generated user copy, the server recommends the most suitable products and services. This recommendation is performed using a recommendation engine. The products recommended for each user copy are provided to the user. For example, they are displayed in the form of a pop-up notification or a list on a smartphone application.

[0189] Calculation and analysis of results

[0190] The data obtained as a result of the recommendations is collected and statistically processed by the server. The server analyzes this data to help correct and optimize the recommendation algorithm, and also to extract areas for improvement and user reactions.

[0191] Feedback of results

[0192] Finally, the server compiles the analysis results into a report and provides feedback to the development and marketing teams, which helps them develop new services, improve new features, and formulate marketing strategies.

[0193] Specific Examples

[0194] If a user frequently browses smartphone accessories (e.g., cases, chargers), this behavioral history data is analyzed to generate a profile of "interested in smartphone accessories." Next, a virtual copy of the user is generated based on this profile, and related new smartphone accessories are recommended using the recommendation engine. The recommendation results are notified to the user's smartphone, allowing the user to browse or purchase new accessories.

[0195] An example of a specific prompt is:

[0196] text

[0197] Recommend the next likely purchase for a user based on their smartphone accessory browsing history over the past month. Past purchase history is below: [Product List]. Use this data to suggest three new product recommendations.

[0198] By using this prompt, the AI ​​can make more accurate product recommendations.

[0199] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0200] Step 1:

[0201] Collection of user behavior history data

[0202] When a user navigates online, their device records behavioral history data. Specifically, the device collects the user's web page browsing history, search keywords, clicked links, purchased products, etc. This data is sent to a server in real time or periodically via a smartphone application.

[0203] Input: User internet activity data.

[0204] Output: Behavioral history data sent to the server.

[0205] Step 2:

[0206] Saving and analyzing behavioral history data

[0207] The server stores the transmitted behavioral history data in a database. The stored data is analyzed using machine learning algorithms such as clustering techniques (e.g., KMeans) and recommendation systems (e.g., Recommendation Engine). During this analysis, patterns such as the user's interests and purchasing trends are extracted.

[0208] Input: Behavioral history data.

[0209] Output: Analyzed data and generated user profile.

[0210] Step 3:

[0211] Generate a user copy

[0212] Based on the generated user profile, the server generates multiple virtual copies of the user that simulate different situations and values, creating a diverse range of fake users to improve the accuracy of market research and product recommendations.

[0213] Input: User profile data.

[0214] Output: Multiple user copies.

[0215] Step 4:

[0216] Product and service recommendations

[0217] Based on the generated user copy, the server uses a recommendation engine to recommend the most suitable products and services. The products recommended for each user copy are provided to the user in real time. Specifically, they are displayed in the form of a pop-up notification or list on the smartphone application.

[0218] Input: User copy.

[0219] Output: A list of recommended products and services.

[0220] Step 5:

[0221] Recommendation results compilation and analysis

[0222] The data obtained as a result of the recommendations is collected by the server and processed and analyzed using statistical methods. This data includes user responses and purchase history, and is used to improve and optimize the recommendation algorithm.

[0223] Input: User response data to the recommended products.

[0224] Output: Statistically processed result data.

[0225] Step 6:

[0226] Feedback of results

[0227] Finally, the server compiles the analysis results into a report and provides feedback to the development and marketing teams, which can be used to develop new services, improve new features, and formulate marketing strategies.

[0228] Input: Statistically processed result data.

[0229] Output: Report and feedback information.

[0230] 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.

[0231] The present invention relates to a system that generates user copies based on user behavior history data, thereby improving the efficiency of market research for new services and functions. Furthermore, by combining this invention with an emotion engine that recognizes user emotions, it is possible to obtain more accurate feedback. Specific embodiments for implementing this invention are as follows.

[0232] Collection of user behavior history data

[0233] When a user logs in to an internet platform, the device recognizes the user's ID and begins collecting behavioral data, such as the web pages viewed, search keywords, links clicked, and products purchased, either in real time or periodically.

[0234] Saving and analyzing behavioral history data

[0235] The device sends the collected behavioral history data to a server, which stores the received data in a database and performs a cleaning process. The server then uses machine learning algorithms to extract behavioral patterns, interests, purchasing trends, and more.

[0236] Applying the Emotion Engine

[0237] The server is equipped with an emotion engine that analyzes user emotions from behavioral history data, user text input, voice data, etc. The emotion engine can identify emotional states such as happiness, surprise, anger, and sadness.

[0238] User profile generation

[0239] The server combines the analysis results with emotional data from the emotion engine to generate a user profile, which includes the user's interests, values, and demographic information (age, gender, region, etc.), as well as emotional data.

[0240] Generate a user copy

[0241] The server then generates multiple copies of the user based on the profile. These copies are virtual users that simulate different situations and values ​​of the user. Furthermore, it adjusts the reactions of each copy based on the user's emotional data.

[0242] Conducting a virtual interview or survey

[0243] The server prepares interviews and surveys about new services and features and administers them to the generated user copy, which generates hypothetical questions and answers that reflect the emotional data and automatically responds based on scenarios set within the system.

[0244] Calculation and analysis of results

[0245] The server compiles and analyzes the results of virtual interviews and surveys, statistically processing them, including sentiment data, to analyze how user copy reacts to new services and features.

[0246] Feedback of results

[0247] Finally, the server compiles the analysis results into a report and provides feedback to the development and marketing teams, including sentiment data, which can be used to improve new services and features, and to develop marketing strategies.

[0248] For example, if a user expresses interest in a particular gadget, sentiment analysis can determine whether that interest is positive or negative. Based on the results, virtual interviews can be conducted about new features, providing detailed feedback that includes the user's emotional reactions. This allows for the development of services and products that are more user-friendly.

[0249] The processing flow will be explained below.

[0250] Step 1:

[0251] A user logs into an internet platform. The device recognizes the user's ID and begins collecting behavioral data, including web page browsing history, search keywords, links clicked, and products purchased.

[0252] Step 2:

[0253] The device collects and transmits the behavioral history data to a server. Data transfer can be done in batch or real time. The device non-invasively monitors the user's behavior and continuously records the data.

[0254] Step 3:

[0255] The server stores the received behavioral history data in a database, where it indexes the data appropriately and manages it for quick access.

[0256] Step 4:

[0257] The server preprocesses the stored data, which includes cleaning the data and removing invalid and noisy data, extracting only the data necessary for analysis.

[0258] Step 5:

[0259] The server analyzes the pre-processed data and uses machine learning algorithms to extract user behavioral patterns, interests, purchasing tendencies, etc. For example, it can determine that a user who frequently browses products in a particular category has a strong interest in a particular product.

[0260] Step 6:

[0261] The emotion engine on the server recognizes emotions from user behavioral data, text input, and voice data. The emotion engine uses machine learning models to determine the user's emotional state (e.g., happiness, surprise, anger, sadness).

[0262] Step 7:

[0263] The server combines the analysis results with emotional data from the emotion engine to generate a user profile, which includes the user's interests, values, and demographic information (age, gender, region, etc.), as well as emotional data.

[0264] Step 8:

[0265] The server generates multiple copies of the user based on the generated profile. These copies are virtual users that simulate different situations and values ​​of the user, and adjust each copy's reactions based on emotional data.

[0266] Step 9:

[0267] The server prepares interviews and surveys about new services and features and conducts them on the generated user copy. The user copy generates virtual questions and answers that reflect the emotional data, and automatically responds based on scenarios set within the system.

[0268] Step 10:

[0269] The server compiles and analyzes the results of interviews and surveys. It statistically processes the data, including emotional data, to analyze how users react to new services and features. For example, the emotion engine identifies emotions such as "satisfied" and "dissatisfied" and calculates their percentages.

[0270] Step 11:

[0271] The server compiles the results of the analysis and compiles them into a report. The server extracts insights and potential problems and organizes them in a report format. This report also includes analysis results based on emotion data.

[0272] Step 12:

[0273] The server will then provide the generated report to the development and marketing teams. This feedback can be used to improve new services and features, and to develop marketing strategies. For example, if a user shows interest in a particular gadget and their interest is positive, they can conduct a virtual interview about the new feature and provide detailed feedback, including the user's emotional response.

[0274] Through these steps, market research can be conducted efficiently and accurately by utilizing user behavioral history data and emotion data, reducing time and costs. This system will enable the development of services and products that are in tune with user emotions.

[0275] Example 2

[0276] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0277] In today's consumer market, rapid and accurate market research is required when launching new services or features. However, traditional market research methods tend to be time-consuming and costly. Furthermore, more advanced analysis is required to accurately reflect users' emotions and behavioral patterns, but existing technologies have limitations in their accuracy. The present invention aims to solve these problems and provide a system that can conduct rapid and accurate market research based on users' detailed behavioral history and emotion data.

[0278] 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.

[0279] In this invention, the server includes means for collecting user behavioral history, transmitting it to the server, storing it, and cleaning it, means for extracting behavioral patterns, interests, and purchasing tendencies using a machine learning algorithm, and means for analyzing user emotions using an emotion engine and generating emotion data. This allows for the generation of detailed and accurate profiles based on the user behavioral history and emotion data, enabling market research using multiple virtual user copies.

[0280] "User behavioral history data" refers to data that includes information about the actions a user takes on an internet platform, such as the web pages they view, search keywords, links they click, and products they purchase.

[0281] A "server" is a central processing unit used to receive, store, and analyze behavioral history data.

[0282] A "machine learning algorithm" is a computational method for automatically learning patterns and relationships from data and predicting future behaviors and outcomes.

[0283] An "emotion engine" is a system that analyzes text input, voice data, and behavioral history data to identify a user's emotional state.

[0284] "Emotion data" is data that indicates the user's emotional state as analyzed by the emotion engine.

[0285] A "user profile" is information generated by integrating a user's behavioral patterns, interests, purchasing tendencies, attribute information (age, gender, region, etc.) and emotional data.

[0286] "User copies" are multiple virtual users generated based on a user profile and used to simulate different situations and values.

[0287] A "virtual interview or survey" is a simulated research method for new services or features conducted on user copies.

[0288] "Interview or Survey Results" are data and responses obtained through virtual interviews or surveys.

[0289] "Feedback" is information that is provided to the development and marketing teams after the results of the interviews or surveys have been analyzed and compiled into a report.

[0290] The present invention relates to a system for generating user copies based on user behavior history data and emotion data, thereby improving the efficiency of market research for new services and functions. The system of the present invention includes the following main components and associated means:

[0291] 1. Collection of user behavior history data

[0292] When a user logs into an internet platform, the device recognizes the user's ID and collects behavioral history data, including the web pages viewed, search keywords, links clicked, and products purchased. For example, if a user searches for "latest smartwatches," browses several product pages, adds products to their cart, but ultimately cancels the purchase, a series of behavioral history data will be collected.

[0293] 2. Transmission and storage of behavioral history data

[0294] The device sends the collected behavioral history data to a server, which stores the data in a database and uses Apache Hadoop to clean noise and duplicate data. This process generates a clean dataset.

[0295] 3. Data analysis

[0296] The server analyzes the collected data using machine learning algorithms, specifically TensorFlow and Scikit-learn, to extract user behavioral patterns, interests, and purchasing trends. For example, it uses a random forest algorithm to predict what purchasing behavior a particular behavioral pattern will lead to.

[0297] 4. Applying the Emotion Engine

[0298] The server is equipped with an emotion engine that analyzes the user's text input, voice data, and behavioral history data to identify the user's emotional state. The engine converts voice data into text using the Google Cloud Speech-to-Text API and analyzes emotions using a natural language processing library (e.g., NLTK).

[0299] 5. Generate a user profile

[0300] The server combines the analysis results with emotional data to create a user profile, which includes demographic information such as the user's interests, categories of interest, values, age, gender, and region, along with emotional data, resulting in a detailed and accurate user profile.

[0301] 6. Generate a user copy

[0302] The server generates multiple copies of the user based on the generated user profile. These virtual users are used to simulate different situations and values. The server adjusts the reactions of each copy of the user based on the emotional data. For example, it generates copies that show positive reactions and copies that show negative reactions.

[0303] 7. Conducting Virtual Interviews or Surveys

[0304] The server conducts virtual interviews and surveys about new services and features with the user copy, generating virtual questions and answers that reflect the user's emotional data and responding based on the set scenario.

[0305] 8. Calculation and analysis of results

[0306] The server aggregates and analyzes the results of interviews and surveys using statistical methods, such as using Hadoop's MapReduce to aggregate data and R programming language to perform statistical analysis and visualization of the data.

[0307] 9. Providing Feedback

[0308] Finally, the server compiles the analysis results into a report and provides feedback to the development and marketing teams. This feedback, including sentiment data, provides useful information for developing new services, improving new features, and formulating marketing strategies.

[0309] For example, if a user expresses interest in a particular gadget, sentiment analysis can determine whether that interest is positive or negative. Based on this result, the server can conduct virtual interviews about new features and provide detailed feedback, including the user's emotional reactions, to the development team, allowing them to develop services and products that are more user-oriented.

[0310] An example of an input prompt for a generative AI model is as follows:

[0311] "Show how a user feels about the latest smartwatch. The user's behavioral history includes viewing the product page, adding to cart once, and then canceling the purchase. What emotional state can the emotion engine determine from this behavioral history?"

[0312] The above is a specific embodiment for carrying out the present invention.

[0313] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0314] Step 1:

[0315] When a user logs in to an internet platform, the device recognizes the user's ID and begins collecting behavioral history data, such as the web pages the user viewed, search keywords, links clicked, and products purchased. The collected data is temporarily stored on the device.

[0316] Input: User login information and internet activity data

[0317] Output: Collected behavioral history data

[0318] Step 2:

[0319] The behavioral history data collected by the device is sent to a server. The server stores the received data in a database and cleans up noise and duplicate data. As a specific example, Apache Hadoop is used to process large amounts of data in a distributed manner and store it in an SQL database.

[0320] Input: Collected behavioral history data

[0321] Output: Clean and preserved behavioral history data

[0322] Step 3:

[0323] The server uses machine learning algorithms to analyze the stored behavioral history data. Specifically, TensorFlow and Scikit-learn are used to extract behavioral patterns, interests, and purchasing trends. For example, a random forest algorithm is used to predict user purchasing trends.

[0324] Input: Clean behavioral history data

[0325] Output: Analyzed behavioral patterns, interests, and purchasing tendencies

[0326] Step 4:

[0327] The server is equipped with an emotion engine that analyzes behavioral history data, user text input, and voice data to identify the user's emotional state. This engine includes a function to convert voice data into text using the Google Cloud Speech-to-Text API. A natural language processing library (e.g., NLTK) is used to classify the emotional state.

[0328] Input: Behavioral history data, user text input, voice data

[0329] Output: Identified emotion data

[0330] Step 5:

[0331] The server combines the analysis results with the emotional data to generate a user profile, which includes the user's interests, values, demographic information (age, gender, region, etc.), and emotional data.

[0332] Input: Analyzed behavioral patterns, interests, purchasing tendencies, and identified emotional data

[0333] Output: Unified user profile

[0334] Step 6:

[0335] The server generates multiple copies of the user based on the generated user profile. These virtual users are used to simulate different situations and values. The server adjusts the reactions of each copy of the user based on the emotional data.

[0336] Input: User Profile

[0337] Output: Multiple user copies

[0338] Step 7:

[0339] The server conducts virtual interviews and surveys about new services and features with the user copy, who then generates virtual questions and answers based on the emotional data and reacts based on the scenario.

[0340] Input: Interviews or surveys about new services or features

[0341] Output: User-copied hypothetical questions and answers

[0342] Step 8:

[0343] The server aggregates and analyzes the results of the virtual interviews and surveys using statistical methods. Specifically, it aggregates the data using Hadoop's MapReduce and performs statistical analysis and visualization of the data using the R programming language.

[0344] Input: Results of virtual interviews and surveys

[0345] Output: Aggregated and analyzed data

[0346] Step 9:

[0347] The server compiles the analysis results into a report and provides feedback to the development and marketing teams. For example, it can organize the results into a slide-format report and visually display emotional data and behavioral history data.

[0348] Input: Aggregated and analyzed data

[0349] Output: Feedback compiled into a report

[0350] (Application example 2)

[0351] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0352] Conventional ad delivery technologies could only deliver ads based on user behavioral history data, limiting their effectiveness. Furthermore, because they could not take user emotions into account, it was difficult to maximize the effectiveness of ads and user satisfaction. Furthermore, when conducting market research for new services and features, there was a lack of means to obtain feedback that took user emotions into account, making it difficult to improve the accuracy of development and marketing strategies.

[0353] 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 collecting user behavior history data and emotion data, means for analyzing the behavior history data and emotion data to generate a user profile, and means for generating multiple user copies based on the profile and generating and delivering personalized advertisements. This enables the delivery of personalized advertisements that take user emotions into consideration, maximizing the effectiveness of advertisements and improving user satisfaction. Furthermore, in market research for new services and functions, detailed feedback that reflects user emotions can be obtained, thereby improving the accuracy of development and marketing strategies.

[0354] "User behavioral history data" refers to data about the actions a user takes on an internet platform, specifically information such as the web pages viewed, search keywords, links clicked, and products purchased.

[0355] "Emotional data" refers to data on emotional states analyzed from user text input, voice data, facial expression analysis results, etc., and specifically refers to information obtained by identifying emotional states such as happiness, surprise, anger, and sadness.

[0356] "Profile" refers to detailed information that combines a user's interests, categories of interest, values, attribute information (age, gender, region, etc.) and emotional data, generated based on the user's behavioral history data and emotional data.

[0357] "User copies" refer to virtual users that simulate different situations and values ​​based on the generated profile. User copies can adjust their reactions based on emotional data.

[0358] "Personalized advertising" refers to advertising that is optimized for individual users and is generated based on historical behavioral and emotional data of a specific user or user group.

[0359] "Virtual interviews or surveys" refers to the process of asking generated user copies questions about new services or features and obtaining their responses in a simulated manner.

[0360] "Feedback Data" refers to data compiled and analyzed based on the results of hypothetical interviews or surveys and user responses to the delivery of advertisements.

[0361] The present invention relates to a system for generating and delivering personalized advertisements by collecting and analyzing user behavior history data and emotion data, and generating a user profile and user copy. This system can be realized using a terminal, a server, and a database.

[0362] Collection of user behavioral history data and emotional data

[0363] When a user logs into an internet platform, the device recognizes the user's ID and begins collecting behavioral history data. Specifically, data such as the web pages the user viewed, search keywords, links clicked, and products purchased is collected in real time or periodically. Emotional data is also collected through the user's text input, voice data, and facial expression analysis.

[0364] Storage and analysis of behavioral history data and emotional data

[0365] The device sends the collected behavioral history data and emotional data to a server. The server stores the received data in a database and performs a cleaning process. The server then uses machine learning algorithms to extract the user's behavioral patterns, interests, purchasing tendencies, etc., and analyzes the emotional data using an emotional engine.

[0366] Generating Profiles and User Copies

[0367] The server combines the analysis results with emotional data from the emotion engine to generate a user profile. This profile includes the user's interests, values, attribute information (age, gender, region, etc.), and emotional data. Based on the generated profile, the server generates multiple user copies. These user copies are virtual users that simulate different situations and values ​​of the user, and adjust the reactions of each user copy based on the emotional data.

[0368] Generating and delivering personalized ads

[0369] The server generates personalized ads using an advertising generation model (e.g., a generative AI model) based on the generated user profile. The server then delivers these ads to users' smartphones and other devices in real time. After delivering the ads, the server collects user responses and analyzes the feedback data.

[0370] Conducting virtual interviews or surveys and providing feedback on the results

[0371] The server prepares virtual interviews or questionnaires about new services or features and conducts them on the generated user copy. The user copy generates virtual questions and answers that reflect the emotional data and automatically reacts based on the scenarios set in the system. The server then compiles and analyzes the results of the virtual interviews or questionnaires and reports them as feedback.

[0372] Examples of usage

[0373] For example, if a user shows interest in a particular gadget, emotional data can be used to determine whether that interest is positive or negative. Based on this, personalized ads can be generated and delivered, resulting in more effective ad delivery. Virtual interviews about new features can also be conducted, providing detailed feedback, including emotional reactions.

[0374] Prompt Sentence Examples

[0375] Below are some examples of prompt sentences to input to the generative AI model.

[0376] "Create a program that generates personalized ads based on user behavioral history data and emotional data, and delivers them to smartphones. Behavioral data includes web pages viewed, search keywords, links clicked, and products purchased, while emotional data includes text input and voice data. Use a library for analyzing emotional data and generate a user profile using an appropriate machine learning model. Based on the generated profile, create a system that generates ads using an ad generation model and delivers them."

[0377] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0378] Step 1:

[0379] When a user logs in to an internet platform, the device recognizes the user ID and starts collecting behavioral history data. The input is the user ID, and the output is the collected behavioral history data (web pages viewed, search keywords, links clicked, products purchased, etc.).

[0380] Step 2:

[0381] The device collects emotion data using the user's text input, voice data, or camera. The input is the user's text input, voice data, or camera footage, and the output is analyzed emotion data (happiness, surprise, anger, sadness, etc.).

[0382] Step 3:

[0383] The terminal transmits the collected behavioral history data and emotion data to the server. The input is the behavioral history data and emotion data, and the output is the result of data transmission to the server.

[0384] Step 4:

[0385] The server stores the received behavioral history data and emotion data in a database and performs a cleaning process. The input is behavioral history data and emotion data, and the output is the cleaned data. Specific operations include data correction, such as deleting or supplementing characters.

[0386] Step 5:

[0387] The server uses machine learning algorithms to extract user behavioral patterns, interests, and purchasing tendencies from behavioral data, and then uses an emotion engine to analyze the emotional data. The input is cleaned behavioral data and emotional data, and the output is the analysis results. Specific operations include clustering, classification, and regression analysis.

[0388] Step 6:

[0389] The server combines the analysis results and emotional data to generate a user profile. The input is the analysis results and emotional data, and the output is a user profile. Specifically, the server integrates each item to create a profile.

[0390] Step 7:

[0391] The server generates multiple user copies based on the generated profile. The input is the user profile, and the output is the user copy. Specifically, it creates virtual users with different situations and values ​​based on the profile data.

[0392] Step 8:

[0393] The server uses the generated user copy to generate a personalized ad using an ad generation model (generative AI model). The input is the user copy, and the output is a personalized ad. Specific operations include prompts to generate ad text and images.

[0394] Step 9:

[0395] The server delivers the generated personalized advertisement to the user's smartphone. The input is the personalized advertisement, and the output is the advertisement delivery result.

[0396] Step 10:

[0397] After delivering the ad, the server collects user responses and analyzes the feedback data. The input is the user's response after the ad is delivered, and the output is the analysis results. Specific operations include analyzing click-through rates, viewing times, and changes in user emotions.

[0398] Step 11:

[0399] The server prepares a virtual interview or questionnaire about the new service or function and administers it to the generated user copy, where the input is the interview or questionnaire content and the user copy, and the output is the virtual answers.

[0400] Step 12:

[0401] The server aggregates and analyzes the results of virtual interviews and surveys, as well as advertising feedback data, and provides feedback to the development team and marketing team. The input is the results of virtual interviews and surveys, as well as advertising feedback data, and the output is a feedback report. Specific operations include statistical processing and graph generation.

[0402] 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.

[0403] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0404] 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.

[0405] [Second embodiment]

[0406] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0407] 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.

[0408] 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).

[0409] 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.

[0410] 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.

[0411] 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).

[0412] 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.

[0413] 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.

[0414] 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.

[0415] 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.

[0416] 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.

[0417] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0418] The present invention relates to a system for generating a user copy based on user behavior history data, thereby improving the efficiency of market research for new services and functions. Specific embodiments for carrying out the present invention are as follows.

[0419] Collection of user behavior history data

[0420] When a user engages in online activities, the device records this behavioral history data, specifically, the web pages viewed, search keywords, links clicked, products purchased, etc. This behavioral history data is sent to a server in real time or periodically.

[0421] Saving and analyzing behavioral history data

[0422] The server then stores the received behavioral history data in a database. The server then analyzes this data and applies machine learning algorithms to extract patterns such as user interests and purchasing trends. For example, a user who frequently views products in a particular category can be inferred to be particularly interested in that category.

[0423] User profile generation

[0424] The server generates a user profile based on the analysis results, which includes the user's interests, values ​​(e.g., whether they prioritize cost or new technology), and demographic information (e.g., age, gender, region).

[0425] Generate a user copy

[0426] The server then creates multiple copies of the user based on the profile. These copies are virtual users that simulate different situations and values. For example, there could be multiple types, such as a "technology-loving user who prioritizes new features over price" and a "cost-conscious user who prioritizes price."

[0427] Conducting a virtual interview or survey

[0428] The server prepares interviews and questionnaires about new services and features and conducts them on the generated user copies. Because the user copies are virtual, the server automatically generates answers based on pre-set scenarios. As a result, a wide variety of feedback can be obtained without interacting with real users.

[0429] Calculation and analysis of results

[0430] The server aggregates and analyzes the results of virtual interviews and surveys. For example, it statistically processes user evaluations of new features to identify points of interest and areas that need improvement.

[0431] Feedback of results

[0432] Finally, the server compiles the analysis results into a report and provides feedback to the development and marketing teams, which helps them develop new services, improve new features, and formulate marketing strategies.

[0433] As described above, the present invention utilizes user behavior history data to obtain feedback efficiently and accurately, thereby significantly reducing time and costs. As a specific example, if a user shows interest in a particular gadget, a virtual interview can be conducted about new features related to the gadget, and a marketing strategy can be built based on the results.

[0434] The processing flow will be explained below.

[0435] Step 1:

[0436] A user logs in to an internet platform. The device recognizes the user's user ID, such as their Yahoo! JAPAN ID, and begins collecting behavioral history data associated with that ID.

[0437] Step 2:

[0438] The device records user behavior data, specifically, the web pages the user views, the keywords they search for, the links they click, the products they purchase, and other data collected in real time or periodically.

[0439] Step 3:

[0440] The behavioral history data collected by the device is sent to the server. Data transfer is performed by batch processing or streaming processing.

[0441] Step 4:

[0442] The server stores the received behavioral history data in a database, which then stores the data appropriately and creates an index for fast access.

[0443] Step 5:

[0444] The server preprocesses the stored data, cleaning out invalid and noisy data and extracting only the necessary data. The server then classifies and organizes the data by category.

[0445] Step 6:

[0446] The server analyzes the data and applies machine learning algorithms to extract user behavioral patterns, interests, purchasing tendencies, etc. For example, if a user frequently browses a particular product category, it can be assumed that they have a high interest in that category.

[0447] Step 7:

[0448] The server generates a user profile. Based on the analysis results from the previous step, the profile is compiled based on the user's interests, categories of interest, values, and attribute information (age, gender, region, etc.).

[0449] Step 8:

[0450] The server generates multiple user copies. Based on the generated profiles, it creates several virtual users with different scenarios and values. For example, it creates a user copy that is interested in technology and prioritizes functionality over price, and a user copy that is cost-sensitive.

[0451] Step 9:

[0452] The server conducts interviews and surveys about new services and features with the user copy, asking the user copy simulated questions according to a scenario set within the system and collecting their responses.

[0453] Step 10:

[0454] The server compiles and analyzes the results of interviews and surveys, statistically processing the collected data to analyze how users respond to new services and features.

[0455] Step 11:

[0456] The server compiles the results of the analysis and compiles them into a report. The server extracts insights and potential problems and organizes them in report format.

[0457] Step 12:

[0458] The server will then provide the generated reports to the development and marketing teams, who will use the feedback to improve new services and features, and to develop marketing strategies.

[0459] This series of specific processes makes it possible to effectively utilize user behavior history data and conduct market research efficiently and accurately.

[0460] Example 1

[0461] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0462] Modern market research and new product feedback is time-consuming and costly, and the amount of feedback available from users is limited, creating a need for a fast and efficient method of gathering diverse user opinions.

[0463] 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.

[0464] In this invention, the server includes means for collecting user behavior history data, means for transmitting the behavior history data to the server, means for saving the behavior history data in a database, means for analyzing the behavior history data and generating a user profile, means for generating a plurality of user copies based on the generated profiles, means for conducting virtual interviews or questionnaires regarding new services or new functions with the user copies, means for aggregating and analyzing the results of the interviews or questionnaires, and means for compiling the results into a report and providing feedback to the development team and the marketing team. This makes it possible to quickly and efficiently collect diverse user opinions and greatly improve the efficiency of market research and new product feedback.

[0465] "User behavioral history data" refers to information about a user's activities on the Internet, specifically web pages viewed, search keywords, links clicked, products purchased, etc.

[0466] The term "server" refers to a central computer system that stores, analyzes, and processes the behavioral history data, collects data from multiple users, and performs various processes based on that data.

[0467] A "database" is a storage system that organizes and stores the behavioral history data, and is used for subsequent analysis and profile generation.

[0468] A "user profile" is a set of data generated by a machine learning algorithm that includes a user's interests, concerns, purchasing trends, values, and attribute information.

[0469] A "user copy" is a virtual user simulated based on the profile of the user, and is a modeled entity with different situations and values.

[0470] A "virtual interview or survey" is a process in which user copy is asked questions about a new service or feature and hypothetical answers are generated.

[0471] "Means for aggregation and analysis" refers to a method or system for statistically processing the results of virtual interviews or questionnaires and extracting user evaluations and opinions.

[0472] "Feedback means" refers to the process of creating a report based on the results of the above-mentioned compilation and analysis, and transmitting this to the relevant team.

[0473] The present invention relates to a system for generating a user copy based on user behavior history data, thereby improving the efficiency of market research for new services and functions. Specific embodiments for carrying out the present invention will be described below.

[0474] Collection of user behavior history data

[0475] When a user engages in online activities, their device records this behavioral history data. Specifically, it collects data on the web pages they visit, search keywords, links they click, and products they purchase. This data is collected through web browsers, mobile applications, and cloud services. This behavioral history data is sent to a server in real time or periodically.

[0476] Examples:

[0477] A user searches for "smartwatch" on an e-commerce website and browses multiple product pages. At this point, the device collects information such as the search keyword "smartwatch," the viewed page URL, and click path.

[0478] Sending behavioral history data

[0479] The device sends the collected behavioral history data to the server using an encrypted protocol (e.g., HTTPS), which protects the user's privacy.

[0480] Examples:

[0481] When the user finishes browsing the site, data is automatically sent from the device to the server.

[0482] Saving and analyzing behavioral history data

[0483] The server stores the received behavioral history data in a database. Based on the stored data, the server applies a machine learning algorithm (e.g., a sparse classifier) ​​to extract the user's interests and purchasing trends.

[0484] Examples:

[0485] Using datasets stored on a server, machine learning algorithms create profiles of users who are interested in "smartwatches."

[0486] Generate a user profile

[0487] The server generates a user profile based on the analysis of the behavioral history data, which includes the user's interests, values ​​(e.g., whether they prioritize cost or new technology), and demographic information (e.g., age, gender, region).

[0488] Examples:

[0489] If a user frequently searches for information related to "smartwatches," their profile might include "interest category: gadgets," "values: emphasis on new technology," and "demographics: 30 years old, male, living in an urban area."

[0490] Generate a user copy

[0491] The server then creates multiple copies of the user based on the profile, simulating the user as a virtual entity with different situations and values.

[0492] Examples:

[0493] The server generates two copies of the user: one that prioritizes new features and one that prioritizes costs.

[0494] Conducting a virtual interview or survey

[0495] The server prepares interviews and questionnaires about new services and features and conducts them on the virtual user copy, who then automatically generates answers based on scenarios preset by the server.

[0496] Example prompt sentence:

[0497] We generated the following profile based on historical user behavior data:

[0498] Interests: Smartwatches

[0499] Demographic information: 30 years old, male, urban resident

[0500] Values: Emphasis on new features

[0501] Create a user copy with the following conditions:

[0502] Copy 1: New feature-focused virtual user

[0503] Copy 2: Price-sensitive virtual users

[0504] For each user copy, answer the following questions:

[0505] 1. What do you think about the new tracking features on smartwatches?

[0506] 2. What is the maximum price you are willing to pay for this new feature?

[0507] Calculation and analysis of results

[0508] The server aggregates and statistically analyzes the results of the virtual interviews and surveys, and uses statistical software to extract user interests and areas for improvement based on the analysis results.

[0509] Examples:

[0510] The server records the responses collected from user copies in a database and analyzes the data using statistical software such as R or Python, extracting, for example, satisfaction scores for new features and price sensitivity.

[0511] Feedback of results

[0512] The server compiles the analysis results into a report and provides feedback to the development and marketing teams, including specific suggestions for improvements and market strategies.

[0513] Examples:

[0514] The server generates a PDF report of the analysis results and sends it to the development and marketing teams via email. The report includes the user copy evaluation results, points of interest, and improvement suggestions.

[0515] The above is a specific embodiment for carrying out the present invention.

[0516] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0517] Step 1:

[0518] The device collects data about the user's online activities, including the web pages the user views, search keywords, links clicked, and products purchased. The device temporarily stores this data in local storage.

[0519] Specific behavior:

[0520] A user searches for "smartwatch" on an e-commerce website and browses multiple product pages. At this point, the device collects information such as the search keyword "smartwatch," the viewed page URL, and click path.

[0521] Step 2:

[0522] The device sends the collected behavioral history data to the server. The input includes all the behavioral history data collected in the previous step. The data is sent using an encrypted protocol (e.g., HTTPS).

[0523] Specific behavior:

[0524] When a user browses a website, the device automatically sends activity history data to the server. For example, the data is sent when the user finishes browsing the website.

[0525] Step 3:

[0526] The server stores the received behavioral history data in a database. The input includes behavioral history data sent from the terminal. The database organizes and stores this data.

[0527] Specific behavior:

[0528] The server uses a database system (e.g., SQL database) to efficiently store user behavior history data, which makes it possible to reference the data required for subsequent analysis processing.

[0529] Step 4:

[0530] The server analyzes the stored behavioral history data and generates a user profile. The input includes the behavioral history data stored in the database. Machine learning algorithms are applied to analyze the data and extract the user's interests and purchasing trends. The output is the generated user profile.

[0531] Specific behavior:

[0532] The server uses a machine learning model (e.g., a sparse classifier) ​​based on the behavioral history dataset to generate a profile of the user's interest categories (e.g., "smartwatch") and values ​​(e.g., "emphasis on new technology").

[0533] Step 5:

[0534] The server generates multiple user copies based on the generated profile. The input includes the user's profile. Each user copy is modeled as a virtual entity with different situations and values. The output is the generated user copy.

[0535] Specific behavior:

[0536] Based on the characteristics of the user profile, the server generates two types of user copies: a "new feature-oriented type" and a "cost-oriented type."

[0537] Step 6:

[0538] The server prepares a virtual interview or questionnaire about the new service or feature and administers it to the generated user copy. The input includes questions about the new service or feature and the user copy. The server automatically generates answers based on a pre-set scenario. The output is the generated answers.

[0539] Specific behavior:

[0540] The server poses questions to the user copy, such as "What do you think about the new tracking features on your smartwatch?" or "What is the maximum price you would be willing to pay for this new feature?", and collects hypothetical responses.

[0541] Step 7:

[0542] The server aggregates and statistically analyzes the results of the virtual interviews and surveys. The input includes user-copied responses. The server processes these data using statistical software to derive user ratings and opinions. The output is the analysis results.

[0543] Specific behavior:

[0544] The server stores the response data obtained from user copies in a database and performs statistical analysis using statistical software such as R or Python, for example, to extract the distribution of satisfaction scores for new features and price sensitivity.

[0545] Step 8:

[0546] The server compiles the analysis results as a report and feeds it back to the development and marketing teams. The input includes the results of the statistical analysis. The output is a report.

[0547] Specific behavior:

[0548] The server generates a PDF report based on the analysis results and sends it to the development and marketing teams by email, including the "User Copy Evaluation Results," "Points of Interest," and "Improvement Suggestions."

[0549] (Application example 1)

[0550] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0551] Conventional methods of market research and product recommendation are not only time-consuming and costly, but also have the drawback of being difficult to respond quickly to diverse user needs. Furthermore, when using user behavior data to recommend products, it is difficult to provide personalized and accurate recommendations, which hinders the expectation of improving user satisfaction.

[0552] 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.

[0553] In this invention, the server includes means for collecting user behavior history data, means for analyzing the behavior history data and generating a user profile, means for generating multiple user copies based on the profile, means for recommending products or services based on the user copies, means for providing the user with the recommendation results, means for aggregating and analyzing the recommendation results, and means for feeding back the results. This makes it possible to quickly respond to diverse user needs and recommend personalized and accurate products and services, thereby improving user satisfaction and making market research more efficient.

[0554] "User behavioral history data" refers to data about a user's activities on the Internet, such as web pages viewed, search keywords, links clicked, and products purchased.

[0555] "Means for collecting behavioral history data" refers to means that has the function of recording a user's Internet activity and transmitting that data to a server in real time or periodically.

[0556] "Means for analyzing behavioral history data and generating user profiles" refers to means for analyzing collected behavioral history data using machine learning algorithms, etc., and extracting patterns such as user interests, concerns, and purchasing tendencies to generate profiles.

[0557] The "means for generating multiple user copies based on a profile" is a means for creating multiple virtual users that simulate different situations and values ​​based on the generated profile.

[0558] The "means for recommending a product or service based on a user copy" is a means for recommending an optimal product or service based on a profile of a virtual user copy.

[0559] The "means of providing the user with the results of the recommendation" refers to a means of providing the recommended products or services to the user through a smartphone application or the like so that the user can view them.

[0560] The "means for aggregating and analyzing the results of recommendations" refers to the means for collecting data on users' reactions to recommended products and services, and statistically processing and analyzing the data.

[0561] "Means of feedback of results" refers to compiling the analysis results in the form of a report or similar and providing them to the development team or marketing team.

[0562] The present invention relates to a system that generates a user copy based on user behavior history data and improves the efficiency of market research for new services and functions. Specifically, the system is implemented as follows.

[0563] Collecting behavioral history data

[0564] When a user browses the internet or makes a purchase, the device records this behavioral history data, including the web pages viewed, search keywords, links clicked, and products purchased. This data is sent to a server in real time or periodically via a smartphone application.

[0565] Saving and analyzing behavioral history data

[0566] The server stores the transmitted behavioral history data in a database. The stored data is analyzed using machine learning algorithms. These algorithms include clustering techniques (e.g., KMeans) and recommendation systems (e.g., Recommendation Engine). This allows patterns such as user interests and purchasing trends to be extracted.

[0567] Generate a user profile

[0568] The server generates a user profile based on the analysis results, which includes the user's interests, values ​​(e.g., whether they prioritize cost or new technology), and demographic information (e.g., age, gender, region).

[0569] Generate a user copy

[0570] The server then generates multiple user copies based on the generated profile. These copies are virtual users that simulate different situations and values. For example, various user copies can be generated, such as a "technology-loving user who prioritizes new features over price" and a "cost-conscious user who prioritizes price."

[0571] Product and service recommendations

[0572] Based on the generated user copy, the server recommends the most suitable products and services. This recommendation is performed using a recommendation engine. The products recommended for each user copy are provided to the user. For example, they are displayed in the form of a pop-up notification or a list on a smartphone application.

[0573] Calculation and analysis of results

[0574] The data obtained as a result of the recommendations is collected and statistically processed by the server. The server analyzes this data to help correct and optimize the recommendation algorithm, and also to extract areas for improvement and user reactions.

[0575] Feedback of results

[0576] Finally, the server compiles the analysis results into a report and provides feedback to the development and marketing teams, which helps them develop new services, improve new features, and formulate marketing strategies.

[0577] Specific Examples

[0578] If a user frequently browses smartphone accessories (e.g., cases, chargers), this behavioral history data is analyzed to generate a profile of "interested in smartphone accessories." Next, a virtual copy of the user is generated based on this profile, and related new smartphone accessories are recommended using the recommendation engine. The recommendation results are notified to the user's smartphone, allowing the user to browse or purchase new accessories.

[0579] An example of a specific prompt is:

[0580] text

[0581] Recommend the next likely purchase for a user based on their smartphone accessory browsing history over the past month. Past purchase history is below: [Product List]. Use this data to suggest three new product recommendations.

[0582] By using this prompt, the AI ​​can make more accurate product recommendations.

[0583] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0584] Step 1:

[0585] Collection of user behavior history data

[0586] When a user navigates online, their device records behavioral history data. Specifically, the device collects the user's web page browsing history, search keywords, clicked links, purchased products, etc. This data is sent to a server in real time or periodically via a smartphone application.

[0587] Input: User internet activity data.

[0588] Output: Behavioral history data sent to the server.

[0589] Step 2:

[0590] Saving and analyzing behavioral history data

[0591] The server stores the transmitted behavioral history data in a database. The stored data is analyzed using machine learning algorithms such as clustering techniques (e.g., KMeans) and recommendation systems (e.g., Recommendation Engine). During this analysis, patterns such as the user's interests and purchasing trends are extracted.

[0592] Input: Behavioral history data.

[0593] Output: Analyzed data and generated user profile.

[0594] Step 3:

[0595] Generate a user copy

[0596] Based on the generated user profile, the server generates multiple virtual copies of the user that simulate different situations and values, creating a diverse range of fake users to improve the accuracy of market research and product recommendations.

[0597] Input: User profile data.

[0598] Output: Multiple user copies.

[0599] Step 4:

[0600] Product and service recommendations

[0601] Based on the generated user copy, the server uses a recommendation engine to recommend the most suitable products and services. The products recommended for each user copy are provided to the user in real time. Specifically, they are displayed in the form of a pop-up notification or list on the smartphone application.

[0602] Input: User copy.

[0603] Output: A list of recommended products and services.

[0604] Step 5:

[0605] Recommendation results compilation and analysis

[0606] The data obtained as a result of the recommendations is collected by the server and processed and analyzed using statistical methods. This data includes user responses and purchase history, and is used to improve and optimize the recommendation algorithm.

[0607] Input: User response data to the recommended products.

[0608] Output: Statistically processed result data.

[0609] Step 6:

[0610] Feedback of results

[0611] Finally, the server compiles the analysis results into a report and provides feedback to the development and marketing teams, which can be used to develop new services, improve new features, and formulate marketing strategies.

[0612] Input: Statistically processed result data.

[0613] Output: Report and feedback information.

[0614] 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.

[0615] The present invention relates to a system that generates user copies based on user behavior history data, thereby improving the efficiency of market research for new services and functions. Furthermore, by combining this invention with an emotion engine that recognizes user emotions, it is possible to obtain more accurate feedback. Specific embodiments for implementing this invention are as follows.

[0616] Collection of user behavior history data

[0617] When a user logs in to an internet platform, the device recognizes the user's ID and begins collecting behavioral data, such as the web pages viewed, search keywords, links clicked, and products purchased, either in real time or periodically.

[0618] Saving and analyzing behavioral history data

[0619] The device sends the collected behavioral history data to a server, which stores the received data in a database and performs a cleaning process. The server then uses machine learning algorithms to extract behavioral patterns, interests, purchasing trends, and more.

[0620] Applying the Emotion Engine

[0621] The server is equipped with an emotion engine that analyzes user emotions from behavioral history data, user text input, voice data, etc. The emotion engine can identify emotional states such as happiness, surprise, anger, and sadness.

[0622] User profile generation

[0623] The server combines the analysis results with emotional data from the emotion engine to generate a user profile, which includes the user's interests, values, and demographic information (age, gender, region, etc.), as well as emotional data.

[0624] Generate a user copy

[0625] The server then generates multiple copies of the user based on the profile. These copies are virtual users that simulate different situations and values ​​of the user. Furthermore, it adjusts the reactions of each copy based on the user's emotional data.

[0626] Conducting a virtual interview or survey

[0627] The server prepares interviews and surveys about new services and features and administers them to the generated user copy, which generates hypothetical questions and answers that reflect the emotional data and automatically responds based on scenarios set within the system.

[0628] Calculation and analysis of results

[0629] The server compiles and analyzes the results of virtual interviews and surveys, statistically processing them, including sentiment data, to analyze how user copy reacts to new services and features.

[0630] Feedback of results

[0631] Finally, the server compiles the analysis results into a report and provides feedback to the development and marketing teams, including sentiment data, which can be used to improve new services and features, and to develop marketing strategies.

[0632] For example, if a user expresses interest in a particular gadget, sentiment analysis can determine whether that interest is positive or negative. Based on the results, virtual interviews can be conducted about new features, providing detailed feedback that includes the user's emotional reactions. This allows for the development of services and products that are more user-friendly.

[0633] The processing flow will be explained below.

[0634] Step 1:

[0635] A user logs into an internet platform. The device recognizes the user's ID and begins collecting behavioral data, including web page browsing history, search keywords, links clicked, and products purchased.

[0636] Step 2:

[0637] The device collects and transmits the behavioral history data to a server. Data transfer can be done in batch or real time. The device non-invasively monitors the user's behavior and continuously records the data.

[0638] Step 3:

[0639] The server stores the received behavioral history data in a database, where it indexes the data appropriately and manages it for quick access.

[0640] Step 4:

[0641] The server preprocesses the stored data, which includes cleaning the data and removing invalid and noisy data, extracting only the data necessary for analysis.

[0642] Step 5:

[0643] The server analyzes the pre-processed data and uses machine learning algorithms to extract user behavioral patterns, interests, purchasing tendencies, etc. For example, it can determine that a user who frequently browses products in a particular category has a strong interest in a particular product.

[0644] Step 6:

[0645] The emotion engine on the server recognizes emotions from user behavioral data, text input, and voice data. The emotion engine uses machine learning models to determine the user's emotional state (e.g., happiness, surprise, anger, sadness).

[0646] Step 7:

[0647] The server combines the analysis results with emotional data from the emotion engine to generate a user profile, which includes the user's interests, values, and demographic information (age, gender, region, etc.), as well as emotional data.

[0648] Step 8:

[0649] The server generates multiple copies of the user based on the generated profile. These copies are virtual users that simulate different situations and values ​​of the user, and adjust each copy's reactions based on emotional data.

[0650] Step 9:

[0651] The server prepares interviews and surveys about new services and features and conducts them on the generated user copy. The user copy generates virtual questions and answers that reflect the emotional data, and automatically responds based on scenarios set within the system.

[0652] Step 10:

[0653] The server compiles and analyzes the results of interviews and surveys. It statistically processes the data, including emotional data, to analyze how users react to new services and features. For example, the emotion engine identifies emotions such as "satisfied" and "dissatisfied" and calculates their percentages.

[0654] Step 11:

[0655] The server compiles the results of the analysis and compiles them into a report. The server extracts insights and potential problems and organizes them in a report format. This report also includes analysis results based on emotion data.

[0656] Step 12:

[0657] The server will then provide the generated report to the development and marketing teams. This feedback can be used to improve new services and features, and to develop marketing strategies. For example, if a user shows interest in a particular gadget and their interest is positive, they can conduct a virtual interview about the new feature and provide detailed feedback, including the user's emotional response.

[0658] Through these steps, market research can be conducted efficiently and accurately by utilizing user behavioral history data and emotion data, reducing time and costs. This system will enable the development of services and products that are in tune with user emotions.

[0659] Example 2

[0660] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0661] In today's consumer market, rapid and accurate market research is required when launching new services or features. However, traditional market research methods tend to be time-consuming and costly. Furthermore, more advanced analysis is required to accurately reflect users' emotions and behavioral patterns, but existing technologies have limitations in their accuracy. The present invention aims to solve these problems and provide a system that can conduct rapid and accurate market research based on users' detailed behavioral history and emotion data.

[0662] 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.

[0663] In this invention, the server includes means for collecting user behavioral history, transmitting it to the server, storing it, and cleaning it, means for extracting behavioral patterns, interests, and purchasing tendencies using a machine learning algorithm, and means for analyzing user emotions using an emotion engine and generating emotion data. This allows for the generation of detailed and accurate profiles based on the user behavioral history and emotion data, enabling market research using multiple virtual user copies.

[0664] "User behavioral history data" refers to data that includes information about the actions a user takes on an internet platform, such as the web pages they view, search keywords, links they click, and products they purchase.

[0665] A "server" is a central processing unit used to receive, store, and analyze behavioral history data.

[0666] A "machine learning algorithm" is a computational method for automatically learning patterns and relationships from data and predicting future behaviors and outcomes.

[0667] An "emotion engine" is a system that analyzes text input, voice data, and behavioral history data to identify a user's emotional state.

[0668] "Emotion data" is data that indicates the user's emotional state as analyzed by the emotion engine.

[0669] A "user profile" is information generated by integrating a user's behavioral patterns, interests, purchasing tendencies, attribute information (age, gender, region, etc.) and emotional data.

[0670] "User copies" are multiple virtual users generated based on a user profile and used to simulate different situations and values.

[0671] A "virtual interview or survey" is a simulated research method for new services or features conducted on user copies.

[0672] "Interview or Survey Results" are data and responses obtained through virtual interviews or surveys.

[0673] "Feedback" is information that is provided to the development and marketing teams after the results of the interviews or surveys have been analyzed and compiled into a report.

[0674] The present invention relates to a system for generating user copies based on user behavior history data and emotion data, thereby improving the efficiency of market research for new services and functions. The system of the present invention includes the following main components and associated means:

[0675] 1. Collection of user behavior history data

[0676] When a user logs into an internet platform, the device recognizes the user's ID and collects behavioral history data, including the web pages viewed, search keywords, links clicked, and products purchased. For example, if a user searches for "latest smartwatches," browses several product pages, adds products to their cart, but ultimately cancels the purchase, a series of behavioral history data will be collected.

[0677] 2. Transmission and storage of behavioral history data

[0678] The device sends the collected behavioral history data to a server, which stores the data in a database and uses Apache Hadoop to clean noise and duplicate data. This process generates a clean dataset.

[0679] 3. Data analysis

[0680] The server analyzes the collected data using machine learning algorithms, specifically TensorFlow and Scikit-learn, to extract user behavioral patterns, interests, and purchasing trends. For example, it uses a random forest algorithm to predict what purchasing behavior a particular behavioral pattern will lead to.

[0681] 4. Applying the Emotion Engine

[0682] The server is equipped with an emotion engine that analyzes the user's text input, voice data, and behavioral history data to identify the user's emotional state. The engine converts voice data into text using the Google Cloud Speech-to-Text API and analyzes emotions using a natural language processing library (e.g., NLTK).

[0683] 5. Generate a user profile

[0684] The server combines the analysis results with emotional data to create a user profile, which includes demographic information such as the user's interests, categories of interest, values, age, gender, and region, along with emotional data, resulting in a detailed and accurate user profile.

[0685] 6. Generate a user copy

[0686] The server generates multiple copies of the user based on the generated user profile. These virtual users are used to simulate different situations and values. The server adjusts the reactions of each copy of the user based on the emotional data. For example, it generates copies that show positive reactions and copies that show negative reactions.

[0687] 7. Conducting Virtual Interviews or Surveys

[0688] The server conducts virtual interviews and surveys about new services and features with the user copy, generating virtual questions and answers that reflect the user's emotional data and responding based on the set scenario.

[0689] 8. Calculation and analysis of results

[0690] The server aggregates and analyzes the results of interviews and surveys using statistical methods, such as using Hadoop's MapReduce to aggregate data and R programming language to perform statistical analysis and visualization of the data.

[0691] 9. Providing Feedback

[0692] Finally, the server compiles the analysis results into a report and provides feedback to the development and marketing teams. This feedback, including sentiment data, provides useful information for developing new services, improving new features, and formulating marketing strategies.

[0693] For example, if a user expresses interest in a particular gadget, sentiment analysis can determine whether that interest is positive or negative. Based on this result, the server can conduct virtual interviews about new features and provide detailed feedback, including the user's emotional reactions, to the development team, allowing them to develop services and products that are more user-oriented.

[0694] An example of an input prompt for a generative AI model is as follows:

[0695] "Show how a user feels about the latest smartwatch. The user's behavioral history includes viewing the product page, adding to cart once, and then canceling the purchase. What emotional state can the emotion engine determine from this behavioral history?"

[0696] The above is a specific embodiment for carrying out the present invention.

[0697] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0698] Step 1:

[0699] When a user logs in to an internet platform, the device recognizes the user's ID and begins collecting behavioral history data, such as the web pages the user viewed, search keywords, links clicked, and products purchased. The collected data is temporarily stored on the device.

[0700] Input: User login information and internet activity data

[0701] Output: Collected behavioral history data

[0702] Step 2:

[0703] The behavioral history data collected by the device is sent to a server. The server stores the received data in a database and cleans up noise and duplicate data. As a specific example, Apache Hadoop is used to process large amounts of data in a distributed manner and store it in an SQL database.

[0704] Input: Collected behavioral history data

[0705] Output: Clean and preserved behavioral history data

[0706] Step 3:

[0707] The server uses machine learning algorithms to analyze the stored behavioral history data. Specifically, TensorFlow and Scikit-learn are used to extract behavioral patterns, interests, and purchasing trends. For example, a random forest algorithm is used to predict user purchasing trends.

[0708] Input: Clean behavioral history data

[0709] Output: Analyzed behavioral patterns, interests, and purchasing tendencies

[0710] Step 4:

[0711] The server is equipped with an emotion engine that analyzes behavioral history data, user text input, and voice data to identify the user's emotional state. This engine includes a function to convert voice data into text using the Google Cloud Speech-to-Text API. A natural language processing library (e.g., NLTK) is used to classify the emotional state.

[0712] Input: Behavioral history data, user text input, voice data

[0713] Output: Identified emotion data

[0714] Step 5:

[0715] The server combines the analysis results with the emotional data to generate a user profile, which includes the user's interests, values, demographic information (age, gender, region, etc.), and emotional data.

[0716] Input: Analyzed behavioral patterns, interests, purchasing tendencies, and identified emotional data

[0717] Output: Unified user profile

[0718] Step 6:

[0719] The server generates multiple copies of the user based on the generated user profile. These virtual users are used to simulate different situations and values. The server adjusts the reactions of each copy of the user based on the emotional data.

[0720] Input: User Profile

[0721] Output: Multiple user copies

[0722] Step 7:

[0723] The server conducts virtual interviews and surveys about new services and features with the user copy, who then generates virtual questions and answers based on the emotional data and reacts based on the scenario.

[0724] Input: Interviews or surveys about new services or features

[0725] Output: User-copied hypothetical questions and answers

[0726] Step 8:

[0727] The server aggregates and analyzes the results of the virtual interviews and surveys using statistical methods. Specifically, it aggregates the data using Hadoop's MapReduce and performs statistical analysis and visualization of the data using the R programming language.

[0728] Input: Results of virtual interviews and surveys

[0729] Output: Aggregated and analyzed data

[0730] Step 9:

[0731] The server compiles the analysis results into a report and provides feedback to the development and marketing teams. For example, it can organize the results into a slide-format report and visually display emotional data and behavioral history data.

[0732] Input: Aggregated and analyzed data

[0733] Output: Feedback compiled into a report

[0734] (Application example 2)

[0735] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0736] Conventional ad delivery technologies could only deliver ads based on user behavioral history data, limiting their effectiveness. Furthermore, because they could not take user emotions into account, it was difficult to maximize the effectiveness of ads and user satisfaction. Furthermore, when conducting market research for new services and features, there was a lack of means to obtain feedback that took user emotions into account, making it difficult to improve the accuracy of development and marketing strategies.

[0737] 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 collecting user behavior history data and emotion data, means for analyzing the behavior history data and emotion data to generate a user profile, and means for generating multiple user copies based on the profile and generating and delivering personalized advertisements. This enables the delivery of personalized advertisements that take user emotions into consideration, maximizing the effectiveness of advertisements and improving user satisfaction. Furthermore, in market research for new services and functions, detailed feedback that reflects user emotions can be obtained, thereby improving the accuracy of development and marketing strategies.

[0738] "User behavioral history data" refers to data about the actions a user takes on an internet platform, specifically information such as the web pages viewed, search keywords, links clicked, and products purchased.

[0739] "Emotional data" refers to data on emotional states analyzed from user text input, voice data, facial expression analysis results, etc., and specifically refers to information obtained by identifying emotional states such as happiness, surprise, anger, and sadness.

[0740] "Profile" refers to detailed information that combines a user's interests, categories of interest, values, attribute information (age, gender, region, etc.) and emotional data, generated based on the user's behavioral history data and emotional data.

[0741] "User copies" refer to virtual users that simulate different situations and values ​​based on the generated profile. User copies can adjust their reactions based on emotional data.

[0742] "Personalized advertising" refers to advertising that is optimized for individual users and is generated based on historical behavioral and emotional data of a specific user or user group.

[0743] "Virtual interviews or surveys" refers to the process of asking generated user copies questions about new services or features and obtaining their responses in a simulated manner.

[0744] "Feedback Data" refers to data compiled and analyzed based on the results of hypothetical interviews or surveys and user responses to the delivery of advertisements.

[0745] The present invention relates to a system for generating and delivering personalized advertisements by collecting and analyzing user behavior history data and emotion data, and generating a user profile and user copy. This system can be realized using a terminal, a server, and a database.

[0746] Collection of user behavioral history data and emotional data

[0747] When a user logs into an internet platform, the device recognizes the user's ID and begins collecting behavioral history data. Specifically, data such as the web pages the user viewed, search keywords, links clicked, and products purchased is collected in real time or periodically. Emotional data is also collected through the user's text input, voice data, and facial expression analysis.

[0748] Storage and analysis of behavioral history data and emotional data

[0749] The device sends the collected behavioral history data and emotional data to a server. The server stores the received data in a database and performs a cleaning process. The server then uses machine learning algorithms to extract the user's behavioral patterns, interests, purchasing tendencies, etc., and analyzes the emotional data using an emotional engine.

[0750] Generating Profiles and User Copies

[0751] The server combines the analysis results with emotional data from the emotion engine to generate a user profile. This profile includes the user's interests, values, attribute information (age, gender, region, etc.), and emotional data. Based on the generated profile, the server generates multiple user copies. These user copies are virtual users that simulate different situations and values ​​of the user, and adjust the reactions of each user copy based on the emotional data.

[0752] Generating and delivering personalized ads

[0753] The server generates personalized ads using an advertising generation model (e.g., a generative AI model) based on the generated user profile. The server then delivers these ads to users' smartphones and other devices in real time. After delivering the ads, the server collects user responses and analyzes the feedback data.

[0754] Conducting virtual interviews or surveys and providing feedback on the results

[0755] The server prepares virtual interviews or questionnaires about new services or features and conducts them on the generated user copy. The user copy generates virtual questions and answers that reflect the emotional data and automatically reacts based on the scenarios set in the system. The server then compiles and analyzes the results of the virtual interviews or questionnaires and reports them as feedback.

[0756] Examples of usage

[0757] For example, if a user shows interest in a particular gadget, emotional data can be used to determine whether that interest is positive or negative. Based on this, personalized ads can be generated and delivered, resulting in more effective ad delivery. Virtual interviews about new features can also be conducted, providing detailed feedback, including emotional reactions.

[0758] Prompt Sentence Examples

[0759] Below are some examples of prompt sentences to input to the generative AI model.

[0760] "Create a program that generates personalized ads based on user behavioral history data and emotional data, and delivers them to smartphones. Behavioral data includes web pages viewed, search keywords, links clicked, and products purchased, while emotional data includes text input and voice data. Use a library for analyzing emotional data and generate a user profile using an appropriate machine learning model. Based on the generated profile, create a system that generates ads using an ad generation model and delivers them."

[0761] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0762] Step 1:

[0763] When a user logs in to an internet platform, the device recognizes the user ID and starts collecting behavioral history data. The input is the user ID, and the output is the collected behavioral history data (web pages viewed, search keywords, links clicked, products purchased, etc.).

[0764] Step 2:

[0765] The device collects emotion data using the user's text input, voice data, or camera. The input is the user's text input, voice data, or camera footage, and the output is analyzed emotion data (happiness, surprise, anger, sadness, etc.).

[0766] Step 3:

[0767] The terminal transmits the collected behavioral history data and emotion data to the server. The input is the behavioral history data and emotion data, and the output is the result of data transmission to the server.

[0768] Step 4:

[0769] The server stores the received behavioral history data and emotion data in a database and performs a cleaning process. The input is behavioral history data and emotion data, and the output is the cleaned data. Specific operations include data correction, such as deleting or supplementing characters.

[0770] Step 5:

[0771] The server uses machine learning algorithms to extract user behavioral patterns, interests, and purchasing tendencies from behavioral data, and then uses an emotion engine to analyze the emotional data. The input is cleaned behavioral data and emotional data, and the output is the analysis results. Specific operations include clustering, classification, and regression analysis.

[0772] Step 6:

[0773] The server combines the analysis results and emotional data to generate a user profile. The input is the analysis results and emotional data, and the output is a user profile. Specifically, the server integrates each item to create a profile.

[0774] Step 7:

[0775] The server generates multiple user copies based on the generated profile. The input is the user profile, and the output is the user copy. Specifically, it creates virtual users with different situations and values ​​based on the profile data.

[0776] Step 8:

[0777] The server uses the generated user copy to generate a personalized ad using an ad generation model (generative AI model). The input is the user copy, and the output is a personalized ad. Specific operations include prompts to generate ad text and images.

[0778] Step 9:

[0779] The server delivers the generated personalized advertisement to the user's smartphone. The input is the personalized advertisement, and the output is the advertisement delivery result.

[0780] Step 10:

[0781] After delivering the ad, the server collects user responses and analyzes the feedback data. The input is the user's response after the ad is delivered, and the output is the analysis results. Specific operations include analyzing click-through rates, viewing times, and changes in user emotions.

[0782] Step 11:

[0783] The server prepares a virtual interview or questionnaire about the new service or function and administers it to the generated user copy, where the input is the interview or questionnaire content and the user copy, and the output is the virtual answers.

[0784] Step 12:

[0785] The server aggregates and analyzes the results of virtual interviews and surveys, as well as advertising feedback data, and provides feedback to the development team and marketing team. The input is the results of virtual interviews and surveys, as well as advertising feedback data, and the output is a feedback report. Specific operations include statistical processing and graph generation.

[0786] 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.

[0787] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0788] 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.

[0789] [Third embodiment]

[0790] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0791] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0792] 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).

[0793] 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.

[0794] 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.

[0795] 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).

[0796] 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.

[0797] 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.

[0798] 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.

[0799] 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.

[0800] 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.

[0801] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0802] The present invention relates to a system for generating a user copy based on user behavior history data, thereby improving the efficiency of market research for new services and functions. Specific embodiments for carrying out the present invention are as follows.

[0803] Collection of user behavior history data

[0804] When a user engages in online activities, the device records this behavioral history data, specifically, the web pages viewed, search keywords, links clicked, products purchased, etc. This behavioral history data is sent to a server in real time or periodically.

[0805] Saving and analyzing behavioral history data

[0806] The server then stores the received behavioral history data in a database. The server then analyzes this data and applies machine learning algorithms to extract patterns such as user interests and purchasing trends. For example, a user who frequently views products in a particular category can be inferred to be particularly interested in that category.

[0807] User profile generation

[0808] The server generates a user profile based on the analysis results, which includes the user's interests, values ​​(e.g., whether they prioritize cost or new technology), and demographic information (e.g., age, gender, region).

[0809] Generate a user copy

[0810] The server then creates multiple copies of the user based on the profile. These copies are virtual users that simulate different situations and values. For example, there could be multiple types, such as a "technology-loving user who prioritizes new features over price" and a "cost-conscious user who prioritizes price."

[0811] Conducting a virtual interview or survey

[0812] The server prepares interviews and questionnaires about new services and features and conducts them on the generated user copies. Because the user copies are virtual, the server automatically generates answers based on pre-set scenarios. As a result, a wide variety of feedback can be obtained without interacting with real users.

[0813] Calculation and analysis of results

[0814] The server aggregates and analyzes the results of virtual interviews and surveys. For example, it statistically processes user evaluations of new features to identify points of interest and areas that need improvement.

[0815] Feedback of results

[0816] Finally, the server compiles the analysis results into a report and provides feedback to the development and marketing teams, which helps them develop new services, improve new features, and formulate marketing strategies.

[0817] As described above, the present invention utilizes user behavior history data to obtain feedback efficiently and accurately, thereby significantly reducing time and costs. As a specific example, if a user shows interest in a particular gadget, a virtual interview can be conducted about new features related to the gadget, and a marketing strategy can be built based on the results.

[0818] The processing flow will be explained below.

[0819] Step 1:

[0820] A user logs in to an internet platform. The device recognizes the user's user ID, such as their Yahoo! JAPAN ID, and begins collecting behavioral history data associated with that ID.

[0821] Step 2:

[0822] The device records user behavior data, specifically, the web pages the user views, the keywords they search for, the links they click, the products they purchase, and other data collected in real time or periodically.

[0823] Step 3:

[0824] The behavioral history data collected by the device is sent to the server. Data transfer is performed by batch processing or streaming processing.

[0825] Step 4:

[0826] The server stores the received behavioral history data in a database, which then stores the data appropriately and creates an index for fast access.

[0827] Step 5:

[0828] The server preprocesses the stored data, cleaning out invalid and noisy data and extracting only the necessary data. The server then classifies and organizes the data by category.

[0829] Step 6:

[0830] The server analyzes the data and applies machine learning algorithms to extract user behavioral patterns, interests, purchasing tendencies, etc. For example, if a user frequently browses a particular product category, it can be assumed that they have a high interest in that category.

[0831] Step 7:

[0832] The server generates a user profile. Based on the analysis results from the previous step, the profile is compiled based on the user's interests, categories of interest, values, and attribute information (age, gender, region, etc.).

[0833] Step 8:

[0834] The server generates multiple user copies. Based on the generated profiles, it creates several virtual users with different scenarios and values. For example, it creates a user copy that is interested in technology and prioritizes functionality over price, and a user copy that is cost-sensitive.

[0835] Step 9:

[0836] The server conducts interviews and surveys about new services and features with the user copy, asking the user copy simulated questions according to a scenario set within the system and collecting their responses.

[0837] Step 10:

[0838] The server compiles and analyzes the results of interviews and surveys, statistically processing the collected data to analyze how users respond to new services and features.

[0839] Step 11:

[0840] The server compiles the results of the analysis and compiles them into a report. The server extracts insights and potential problems and organizes them in report format.

[0841] Step 12:

[0842] The server will then provide the generated reports to the development and marketing teams, who will use the feedback to improve new services and features, and to develop marketing strategies.

[0843] This series of specific processes makes it possible to effectively utilize user behavior history data and conduct market research efficiently and accurately.

[0844] Example 1

[0845] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0846] Modern market research and new product feedback is time-consuming and costly, and the amount of feedback available from users is limited, creating a need for a fast and efficient method of gathering diverse user opinions.

[0847] 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.

[0848] In this invention, the server includes means for collecting user behavior history data, means for transmitting the behavior history data to the server, means for saving the behavior history data in a database, means for analyzing the behavior history data and generating a user profile, means for generating a plurality of user copies based on the generated profiles, means for conducting virtual interviews or questionnaires regarding new services or new functions with the user copies, means for aggregating and analyzing the results of the interviews or questionnaires, and means for compiling the results into a report and providing feedback to the development team and the marketing team. This makes it possible to quickly and efficiently collect diverse user opinions and greatly improve the efficiency of market research and new product feedback.

[0849] "User behavioral history data" refers to information about a user's activities on the Internet, specifically web pages viewed, search keywords, links clicked, products purchased, etc.

[0850] The term "server" refers to a central computer system that stores, analyzes, and processes the behavioral history data, collects data from multiple users, and performs various processes based on that data.

[0851] A "database" is a storage system that organizes and stores the behavioral history data, and is used for subsequent analysis and profile generation.

[0852] A "user profile" is a set of data generated by a machine learning algorithm that includes a user's interests, concerns, purchasing trends, values, and attribute information.

[0853] A "user copy" is a virtual user simulated based on the profile of the user, and is a modeled entity with different situations and values.

[0854] A "virtual interview or survey" is a process in which user copy is asked questions about a new service or feature and hypothetical answers are generated.

[0855] "Means for aggregation and analysis" refers to a method or system for statistically processing the results of virtual interviews or questionnaires and extracting user evaluations and opinions.

[0856] "Feedback means" refers to the process of creating a report based on the results of the above-mentioned compilation and analysis, and transmitting this to the relevant team.

[0857] The present invention relates to a system for generating a user copy based on user behavior history data, thereby improving the efficiency of market research for new services and functions. Specific embodiments for carrying out the present invention will be described below.

[0858] Collection of user behavior history data

[0859] When a user engages in online activities, their device records this behavioral history data. Specifically, it collects data on the web pages they visit, search keywords, links they click, and products they purchase. This data is collected through web browsers, mobile applications, and cloud services. This behavioral history data is sent to a server in real time or periodically.

[0860] Examples:

[0861] A user searches for "smartwatch" on an e-commerce website and browses multiple product pages. At this point, the device collects information such as the search keyword "smartwatch," the viewed page URL, and click path.

[0862] Sending behavioral history data

[0863] The device sends the collected behavioral history data to the server using an encrypted protocol (e.g., HTTPS), which protects the user's privacy.

[0864] Examples:

[0865] When the user finishes browsing the site, data is automatically sent from the device to the server.

[0866] Saving and analyzing behavioral history data

[0867] The server stores the received behavioral history data in a database. Based on the stored data, the server applies a machine learning algorithm (e.g., a sparse classifier) ​​to extract the user's interests and purchasing trends.

[0868] Examples:

[0869] Using datasets stored on a server, machine learning algorithms create profiles of users who are interested in "smartwatches."

[0870] Generate a user profile

[0871] The server generates a user profile based on the analysis of the behavioral history data, which includes the user's interests, values ​​(e.g., whether they prioritize cost or new technology), and demographic information (e.g., age, gender, region).

[0872] Examples:

[0873] If a user frequently searches for information related to "smartwatches," their profile might include "interest category: gadgets," "values: emphasis on new technology," and "demographics: 30 years old, male, living in an urban area."

[0874] Generate a user copy

[0875] The server then creates multiple copies of the user based on the profile, simulating the user as a virtual entity with different situations and values.

[0876] Examples:

[0877] The server generates two copies of the user: one that prioritizes new features and one that prioritizes costs.

[0878] Conducting a virtual interview or survey

[0879] The server prepares interviews and questionnaires about new services and features and conducts them on the virtual user copy, who then automatically generates answers based on scenarios preset by the server.

[0880] Example prompt sentence:

[0881] We generated the following profile based on historical user behavior data:

[0882] Interests: Smartwatches

[0883] Demographic information: 30 years old, male, urban resident

[0884] Values: Emphasis on new features

[0885] Create a user copy with the following conditions:

[0886] Copy 1: New feature-focused virtual user

[0887] Copy 2: Price-sensitive virtual users

[0888] For each user copy, answer the following questions:

[0889] 1. What do you think about the new tracking features on smartwatches?

[0890] 2. What is the maximum price you are willing to pay for this new feature?

[0891] Calculation and analysis of results

[0892] The server aggregates and statistically analyzes the results of the virtual interviews and surveys, and uses statistical software to extract user interests and areas for improvement based on the analysis results.

[0893] Examples:

[0894] The server records the responses collected from user copies in a database and analyzes the data using statistical software such as R or Python, extracting, for example, satisfaction scores for new features and price sensitivity.

[0895] Feedback of results

[0896] The server compiles the analysis results into a report and provides feedback to the development and marketing teams, including specific suggestions for improvements and market strategies.

[0897] Examples:

[0898] The server generates a PDF report of the analysis results and sends it to the development and marketing teams via email. The report includes the user copy evaluation results, points of interest, and improvement suggestions.

[0899] The above is a specific embodiment for carrying out the present invention.

[0900] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0901] Step 1:

[0902] The device collects data about the user's online activities, including the web pages the user views, search keywords, links clicked, and products purchased. The device temporarily stores this data in local storage.

[0903] Specific behavior:

[0904] A user searches for "smartwatch" on an e-commerce website and browses multiple product pages. At this point, the device collects information such as the search keyword "smartwatch," the viewed page URL, and click path.

[0905] Step 2:

[0906] The device sends the collected behavioral history data to the server. The input includes all the behavioral history data collected in the previous step. The data is sent using an encrypted protocol (e.g., HTTPS).

[0907] Specific behavior:

[0908] When a user browses a website, the device automatically sends activity history data to the server. For example, the data is sent when the user finishes browsing the website.

[0909] Step 3:

[0910] The server stores the received behavioral history data in a database. The input includes behavioral history data sent from the terminal. The database organizes and stores this data.

[0911] Specific behavior:

[0912] The server uses a database system (e.g., SQL database) to efficiently store user behavior history data, which makes it possible to reference the data required for subsequent analysis processing.

[0913] Step 4:

[0914] The server analyzes the stored behavioral history data and generates a user profile. The input includes the behavioral history data stored in the database. Machine learning algorithms are applied to analyze the data and extract the user's interests and purchasing trends. The output is the generated user profile.

[0915] Specific behavior:

[0916] The server uses a machine learning model (e.g., a sparse classifier) ​​based on the behavioral history dataset to generate a profile of the user's interest categories (e.g., "smartwatch") and values ​​(e.g., "emphasis on new technology").

[0917] Step 5:

[0918] The server generates multiple user copies based on the generated profile. The input includes the user's profile. Each user copy is modeled as a virtual entity with different situations and values. The output is the generated user copy.

[0919] Specific behavior:

[0920] Based on the characteristics of the user profile, the server generates two types of user copies: a "new feature-oriented type" and a "cost-oriented type."

[0921] Step 6:

[0922] The server prepares a virtual interview or questionnaire about the new service or feature and administers it to the generated user copy. The input includes questions about the new service or feature and the user copy. The server automatically generates answers based on a pre-set scenario. The output is the generated answers.

[0923] Specific behavior:

[0924] The server poses questions to the user copy, such as "What do you think about the new tracking features on your smartwatch?" or "What is the maximum price you would be willing to pay for this new feature?", and collects hypothetical responses.

[0925] Step 7:

[0926] The server aggregates and statistically analyzes the results of the virtual interviews and surveys. The input includes user-copied responses. The server processes these data using statistical software to derive user ratings and opinions. The output is the analysis results.

[0927] Specific behavior:

[0928] The server stores the response data obtained from user copies in a database and performs statistical analysis using statistical software such as R or Python, for example, to extract the distribution of satisfaction scores for new features and price sensitivity.

[0929] Step 8:

[0930] The server compiles the analysis results as a report and feeds it back to the development and marketing teams. The input includes the results of the statistical analysis. The output is a report.

[0931] Specific behavior:

[0932] The server generates a PDF report based on the analysis results and sends it to the development and marketing teams by email, including the "User Copy Evaluation Results," "Points of Interest," and "Improvement Suggestions."

[0933] (Application example 1)

[0934] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0935] Conventional methods of market research and product recommendation are not only time-consuming and costly, but also have the drawback of being difficult to respond quickly to diverse user needs. Furthermore, when using user behavior data to recommend products, it is difficult to provide personalized and accurate recommendations, which hinders the expectation of improving user satisfaction.

[0936] 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.

[0937] In this invention, the server includes means for collecting user behavior history data, means for analyzing the behavior history data and generating a user profile, means for generating multiple user copies based on the profile, means for recommending products or services based on the user copies, means for providing the user with the recommendation results, means for aggregating and analyzing the recommendation results, and means for feeding back the results. This makes it possible to quickly respond to diverse user needs and recommend personalized and accurate products and services, thereby improving user satisfaction and making market research more efficient.

[0938] "User behavioral history data" refers to data about a user's activities on the Internet, such as web pages viewed, search keywords, links clicked, and products purchased.

[0939] "Means for collecting behavioral history data" refers to means that has the function of recording a user's Internet activity and transmitting that data to a server in real time or periodically.

[0940] "Means for analyzing behavioral history data and generating user profiles" refers to means for analyzing collected behavioral history data using machine learning algorithms, etc., and extracting patterns such as user interests, concerns, and purchasing tendencies to generate profiles.

[0941] The "means for generating multiple user copies based on a profile" is a means for creating multiple virtual users that simulate different situations and values ​​based on the generated profile.

[0942] The "means for recommending a product or service based on a user copy" is a means for recommending an optimal product or service based on a profile of a virtual user copy.

[0943] The "means of providing the user with the results of the recommendation" refers to a means of providing the recommended products or services to the user through a smartphone application or the like so that the user can view them.

[0944] The "means for aggregating and analyzing the results of recommendations" refers to the means for collecting data on users' reactions to recommended products and services, and statistically processing and analyzing the data.

[0945] "Means of feedback of results" refers to compiling the analysis results in the form of a report or similar and providing them to the development team or marketing team.

[0946] The present invention relates to a system that generates a user copy based on user behavior history data and improves the efficiency of market research for new services and functions. Specifically, the system is implemented as follows.

[0947] Collecting behavioral history data

[0948] When a user browses the internet or makes a purchase, the device records this behavioral history data, including the web pages viewed, search keywords, links clicked, and products purchased. This data is sent to a server in real time or periodically via a smartphone application.

[0949] Saving and analyzing behavioral history data

[0950] The server stores the transmitted behavioral history data in a database. The stored data is analyzed using machine learning algorithms. These algorithms include clustering techniques (e.g., KMeans) and recommendation systems (e.g., Recommendation Engine). This allows patterns such as user interests and purchasing trends to be extracted.

[0951] Generate a user profile

[0952] The server generates a user profile based on the analysis results, which includes the user's interests, values ​​(e.g., whether they prioritize cost or new technology), and demographic information (e.g., age, gender, region).

[0953] Generate a user copy

[0954] The server then generates multiple user copies based on the generated profile. These copies are virtual users that simulate different situations and values. For example, various user copies can be generated, such as a "technology-loving user who prioritizes new features over price" and a "cost-conscious user who prioritizes price."

[0955] Product and service recommendations

[0956] Based on the generated user copy, the server recommends the most suitable products and services. This recommendation is performed using a recommendation engine. The products recommended for each user copy are provided to the user. For example, they are displayed in the form of a pop-up notification or a list on a smartphone application.

[0957] Calculation and analysis of results

[0958] The data obtained as a result of the recommendations is collected and statistically processed by the server. The server analyzes this data to help correct and optimize the recommendation algorithm, and also to extract areas for improvement and user reactions.

[0959] Feedback of results

[0960] Finally, the server compiles the analysis results into a report and provides feedback to the development and marketing teams, which helps them develop new services, improve new features, and formulate marketing strategies.

[0961] Specific Examples

[0962] If a user frequently browses smartphone accessories (e.g., cases, chargers), this behavioral history data is analyzed to generate a profile of "interested in smartphone accessories." Next, a virtual copy of the user is generated based on this profile, and related new smartphone accessories are recommended using the recommendation engine. The recommendation results are notified to the user's smartphone, allowing the user to browse or purchase new accessories.

[0963] An example of a specific prompt is:

[0964] text

[0965] Recommend the next likely purchase for a user based on their smartphone accessory browsing history over the past month. Past purchase history is below: [Product List]. Use this data to suggest three new product recommendations.

[0966] By using this prompt, the AI ​​can make more accurate product recommendations.

[0967] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0968] Step 1:

[0969] Collection of user behavior history data

[0970] When a user navigates online, their device records behavioral history data. Specifically, the device collects the user's web page browsing history, search keywords, clicked links, purchased products, etc. This data is sent to a server in real time or periodically via a smartphone application.

[0971] Input: User internet activity data.

[0972] Output: Behavioral history data sent to the server.

[0973] Step 2:

[0974] Saving and analyzing behavioral history data

[0975] The server stores the transmitted behavioral history data in a database. The stored data is analyzed using machine learning algorithms such as clustering techniques (e.g., KMeans) and recommendation systems (e.g., Recommendation Engine). During this analysis, patterns such as the user's interests and purchasing trends are extracted.

[0976] Input: Behavioral history data.

[0977] Output: Analyzed data and generated user profile.

[0978] Step 3:

[0979] Generate a user copy

[0980] Based on the generated user profile, the server generates multiple virtual copies of the user that simulate different situations and values, creating a diverse range of fake users to improve the accuracy of market research and product recommendations.

[0981] Input: User profile data.

[0982] Output: Multiple user copies.

[0983] Step 4:

[0984] Product and service recommendations

[0985] Based on the generated user copy, the server uses a recommendation engine to recommend the most suitable products and services. The products recommended for each user copy are provided to the user in real time. Specifically, they are displayed in the form of a pop-up notification or list on the smartphone application.

[0986] Input: User copy.

[0987] Output: A list of recommended products and services.

[0988] Step 5:

[0989] Recommendation results compilation and analysis

[0990] The data obtained as a result of the recommendations is collected by the server and processed and analyzed using statistical methods. This data includes user responses and purchase history, and is used to improve and optimize the recommendation algorithm.

[0991] Input: User response data to the recommended products.

[0992] Output: Statistically processed result data.

[0993] Step 6:

[0994] Feedback of results

[0995] Finally, the server compiles the analysis results into a report and provides feedback to the development and marketing teams, which can be used to develop new services, improve new features, and formulate marketing strategies.

[0996] Input: Statistically processed result data.

[0997] Output: Report and feedback information.

[0998] 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.

[0999] The present invention relates to a system that generates user copies based on user behavior history data, thereby improving the efficiency of market research for new services and functions. Furthermore, by combining this invention with an emotion engine that recognizes user emotions, it is possible to obtain more accurate feedback. Specific embodiments for implementing this invention are as follows.

[1000] Collection of user behavior history data

[1001] When a user logs in to an internet platform, the device recognizes the user's ID and begins collecting behavioral data, such as the web pages viewed, search keywords, links clicked, and products purchased, either in real time or periodically.

[1002] Saving and analyzing behavioral history data

[1003] The device sends the collected behavioral history data to a server, which stores the received data in a database and performs a cleaning process. The server then uses machine learning algorithms to extract behavioral patterns, interests, purchasing trends, and more.

[1004] Applying the Emotion Engine

[1005] The server is equipped with an emotion engine that analyzes user emotions from behavioral history data, user text input, voice data, etc. The emotion engine can identify emotional states such as happiness, surprise, anger, and sadness.

[1006] User profile generation

[1007] The server combines the analysis results with emotional data from the emotion engine to generate a user profile, which includes the user's interests, values, and demographic information (age, gender, region, etc.), as well as emotional data.

[1008] Generate a user copy

[1009] The server then generates multiple copies of the user based on the profile. These copies are virtual users that simulate different situations and values ​​of the user. Furthermore, it adjusts the reactions of each copy based on the user's emotional data.

[1010] Conducting a virtual interview or survey

[1011] The server prepares interviews and surveys about new services and features and administers them to the generated user copy, which generates hypothetical questions and answers that reflect the emotional data and automatically responds based on scenarios set within the system.

[1012] Calculation and analysis of results

[1013] The server compiles and analyzes the results of virtual interviews and surveys, statistically processing them, including sentiment data, to analyze how user copy reacts to new services and features.

[1014] Feedback of results

[1015] Finally, the server compiles the analysis results into a report and provides feedback to the development and marketing teams, including sentiment data, which can be used to improve new services and features, and to develop marketing strategies.

[1016] For example, if a user expresses interest in a particular gadget, sentiment analysis can determine whether that interest is positive or negative. Based on the results, virtual interviews can be conducted about new features, providing detailed feedback that includes the user's emotional reactions. This allows for the development of services and products that are more user-friendly.

[1017] The processing flow will be explained below.

[1018] Step 1:

[1019] A user logs into an internet platform. The device recognizes the user's ID and begins collecting behavioral data, including web page browsing history, search keywords, links clicked, and products purchased.

[1020] Step 2:

[1021] The device collects and transmits the behavioral history data to a server. Data transfer can be done in batch or real time. The device non-invasively monitors the user's behavior and continuously records the data.

[1022] Step 3:

[1023] The server stores the received behavioral history data in a database, where it indexes the data appropriately and manages it for quick access.

[1024] Step 4:

[1025] The server preprocesses the stored data, which includes cleaning the data and removing invalid and noisy data, extracting only the data necessary for analysis.

[1026] Step 5:

[1027] The server analyzes the pre-processed data and uses machine learning algorithms to extract user behavioral patterns, interests, purchasing tendencies, etc. For example, it can determine that a user who frequently browses products in a particular category has a strong interest in a particular product.

[1028] Step 6:

[1029] The emotion engine on the server recognizes emotions from user behavioral data, text input, and voice data. The emotion engine uses machine learning models to determine the user's emotional state (e.g., happiness, surprise, anger, sadness).

[1030] Step 7:

[1031] The server combines the analysis results with emotional data from the emotion engine to generate a user profile, which includes the user's interests, values, and demographic information (age, gender, region, etc.), as well as emotional data.

[1032] Step 8:

[1033] The server generates multiple copies of the user based on the generated profile. These copies are virtual users that simulate different situations and values ​​of the user, and adjust each copy's reactions based on emotional data.

[1034] Step 9:

[1035] The server prepares interviews and surveys about new services and features and conducts them on the generated user copy. The user copy generates virtual questions and answers that reflect the emotional data, and automatically responds based on scenarios set within the system.

[1036] Step 10:

[1037] The server compiles and analyzes the results of interviews and surveys. It statistically processes the data, including emotional data, to analyze how users react to new services and features. For example, the emotion engine identifies emotions such as "satisfied" and "dissatisfied" and calculates their percentages.

[1038] Step 11:

[1039] The server compiles the results of the analysis and compiles them into a report. The server extracts insights and potential problems and organizes them in a report format. This report also includes analysis results based on emotion data.

[1040] Step 12:

[1041] The server will then provide the generated report to the development and marketing teams. This feedback can be used to improve new services and features, and to develop marketing strategies. For example, if a user shows interest in a particular gadget and their interest is positive, they can conduct a virtual interview about the new feature and provide detailed feedback, including the user's emotional response.

[1042] Through these steps, market research can be conducted efficiently and accurately by utilizing user behavioral history data and emotion data, reducing time and costs. This system will enable the development of services and products that are in tune with user emotions.

[1043] Example 2

[1044] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1045] In today's consumer market, rapid and accurate market research is required when launching new services or features. However, traditional market research methods tend to be time-consuming and costly. Furthermore, more advanced analysis is required to accurately reflect users' emotions and behavioral patterns, but existing technologies have limitations in their accuracy. The present invention aims to solve these problems and provide a system that can conduct rapid and accurate market research based on users' detailed behavioral history and emotion data.

[1046] 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.

[1047] In this invention, the server includes means for collecting user behavioral history, transmitting it to the server, storing it, and cleaning it, means for extracting behavioral patterns, interests, and purchasing tendencies using a machine learning algorithm, and means for analyzing user emotions using an emotion engine and generating emotion data. This allows for the generation of detailed and accurate profiles based on the user behavioral history and emotion data, enabling market research using multiple virtual user copies.

[1048] "User behavioral history data" refers to data that includes information about the actions a user takes on an internet platform, such as the web pages they view, search keywords, links they click, and products they purchase.

[1049] A "server" is a central processing unit used to receive, store, and analyze behavioral history data.

[1050] A "machine learning algorithm" is a computational method for automatically learning patterns and relationships from data and predicting future behaviors and outcomes.

[1051] An "emotion engine" is a system that analyzes text input, voice data, and behavioral history data to identify a user's emotional state.

[1052] "Emotion data" is data that indicates the user's emotional state as analyzed by the emotion engine.

[1053] A "user profile" is information generated by integrating a user's behavioral patterns, interests, purchasing tendencies, attribute information (age, gender, region, etc.) and emotional data.

[1054] "User copies" are multiple virtual users generated based on a user profile and used to simulate different situations and values.

[1055] A "virtual interview or survey" is a simulated research method for new services or features conducted on user copies.

[1056] "Interview or Survey Results" are data and responses obtained through virtual interviews or surveys.

[1057] "Feedback" is information that is provided to the development and marketing teams after the results of the interviews or surveys have been analyzed and compiled into a report.

[1058] The present invention relates to a system for generating user copies based on user behavior history data and emotion data, thereby improving the efficiency of market research for new services and functions. The system of the present invention includes the following main components and associated means:

[1059] 1. Collection of user behavior history data

[1060] When a user logs into an internet platform, the device recognizes the user's ID and collects behavioral history data, including the web pages viewed, search keywords, links clicked, and products purchased. For example, if a user searches for "latest smartwatches," browses several product pages, adds products to their cart, but ultimately cancels the purchase, a series of behavioral history data will be collected.

[1061] 2. Transmission and storage of behavioral history data

[1062] The device sends the collected behavioral history data to a server, which stores the data in a database and uses Apache Hadoop to clean noise and duplicate data. This process generates a clean dataset.

[1063] 3. Data analysis

[1064] The server analyzes the collected data using machine learning algorithms, specifically TensorFlow and Scikit-learn, to extract user behavioral patterns, interests, and purchasing trends. For example, it uses a random forest algorithm to predict what purchasing behavior a particular behavioral pattern will lead to.

[1065] 4. Applying the Emotion Engine

[1066] The server is equipped with an emotion engine that analyzes the user's text input, voice data, and behavioral history data to identify the user's emotional state. The engine converts voice data into text using the Google Cloud Speech-to-Text API and analyzes emotions using a natural language processing library (e.g., NLTK).

[1067] 5. Generate a user profile

[1068] The server combines the analysis results with emotional data to create a user profile, which includes demographic information such as the user's interests, categories of interest, values, age, gender, and region, along with emotional data, resulting in a detailed and accurate user profile.

[1069] 6. Generate a user copy

[1070] The server generates multiple copies of the user based on the generated user profile. These virtual users are used to simulate different situations and values. The server adjusts the reactions of each copy of the user based on the emotional data. For example, it generates copies that show positive reactions and copies that show negative reactions.

[1071] 7. Conducting Virtual Interviews or Surveys

[1072] The server conducts virtual interviews and surveys about new services and features with the user copy, generating virtual questions and answers that reflect the user's emotional data and responding based on the set scenario.

[1073] 8. Calculation and analysis of results

[1074] The server aggregates and analyzes the results of interviews and surveys using statistical methods, such as using Hadoop's MapReduce to aggregate data and R programming language to perform statistical analysis and visualization of the data.

[1075] 9. Providing Feedback

[1076] Finally, the server compiles the analysis results into a report and provides feedback to the development and marketing teams. This feedback, including sentiment data, provides useful information for developing new services, improving new features, and formulating marketing strategies.

[1077] For example, if a user expresses interest in a particular gadget, sentiment analysis can determine whether that interest is positive or negative. Based on this result, the server can conduct virtual interviews about new features and provide detailed feedback, including the user's emotional reactions, to the development team, allowing them to develop services and products that are more user-oriented.

[1078] An example of an input prompt for a generative AI model is as follows:

[1079] "Show how a user feels about the latest smartwatch. The user's behavioral history includes viewing the product page, adding to cart once, and then canceling the purchase. What emotional state can the emotion engine determine from this behavioral history?"

[1080] The above is a specific embodiment for carrying out the present invention.

[1081] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1082] Step 1:

[1083] When a user logs in to an internet platform, the device recognizes the user's ID and begins collecting behavioral history data, such as the web pages the user viewed, search keywords, links clicked, and products purchased. The collected data is temporarily stored on the device.

[1084] Input: User login information and internet activity data

[1085] Output: Collected behavioral history data

[1086] Step 2:

[1087] The behavioral history data collected by the device is sent to a server. The server stores the received data in a database and cleans up noise and duplicate data. As a specific example, Apache Hadoop is used to process large amounts of data in a distributed manner and store it in an SQL database.

[1088] Input: Collected behavioral history data

[1089] Output: Clean and preserved behavioral history data

[1090] Step 3:

[1091] The server uses machine learning algorithms to analyze the stored behavioral history data. Specifically, TensorFlow and Scikit-learn are used to extract behavioral patterns, interests, and purchasing trends. For example, a random forest algorithm is used to predict user purchasing trends.

[1092] Input: Clean behavioral history data

[1093] Output: Analyzed behavioral patterns, interests, and purchasing tendencies

[1094] Step 4:

[1095] The server is equipped with an emotion engine that analyzes behavioral history data, user text input, and voice data to identify the user's emotional state. This engine includes a function to convert voice data into text using the Google Cloud Speech-to-Text API. A natural language processing library (e.g., NLTK) is used to classify the emotional state.

[1096] Input: Behavioral history data, user text input, voice data

[1097] Output: Identified emotion data

[1098] Step 5:

[1099] The server combines the analysis results with the emotional data to generate a user profile, which includes the user's interests, values, demographic information (age, gender, region, etc.), and emotional data.

[1100] Input: Analyzed behavioral patterns, interests, purchasing tendencies, and identified emotional data

[1101] Output: Unified user profile

[1102] Step 6:

[1103] The server generates multiple copies of the user based on the generated user profile. These virtual users are used to simulate different situations and values. The server adjusts the reactions of each copy of the user based on the emotional data.

[1104] Input: User Profile

[1105] Output: Multiple user copies

[1106] Step 7:

[1107] The server conducts virtual interviews and surveys about new services and features with the user copy, who then generates virtual questions and answers based on the emotional data and reacts based on the scenario.

[1108] Input: Interviews or surveys about new services or features

[1109] Output: User-copied hypothetical questions and answers

[1110] Step 8:

[1111] The server aggregates and analyzes the results of the virtual interviews and surveys using statistical methods. Specifically, it aggregates the data using Hadoop's MapReduce and performs statistical analysis and visualization of the data using the R programming language.

[1112] Input: Results of virtual interviews and surveys

[1113] Output: Aggregated and analyzed data

[1114] Step 9:

[1115] The server compiles the analysis results into a report and provides feedback to the development and marketing teams. For example, it can organize the results into a slide-format report and visually display emotional data and behavioral history data.

[1116] Input: Aggregated and analyzed data

[1117] Output: Feedback compiled into a report

[1118] (Application example 2)

[1119] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1120] Conventional ad delivery technologies could only deliver ads based on user behavioral history data, limiting their effectiveness. Furthermore, because they could not take user emotions into account, it was difficult to maximize the effectiveness of ads and user satisfaction. Furthermore, when conducting market research for new services and features, there was a lack of means to obtain feedback that took user emotions into account, making it difficult to improve the accuracy of development and marketing strategies.

[1121] 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 collecting user behavior history data and emotion data, means for analyzing the behavior history data and emotion data to generate a user profile, and means for generating multiple user copies based on the profile and generating and delivering personalized advertisements. This enables the delivery of personalized advertisements that take user emotions into consideration, maximizing the effectiveness of advertisements and improving user satisfaction. Furthermore, in market research for new services and functions, detailed feedback that reflects user emotions can be obtained, thereby improving the accuracy of development and marketing strategies.

[1122] "User behavioral history data" refers to data about the actions a user takes on an internet platform, specifically information such as the web pages viewed, search keywords, links clicked, and products purchased.

[1123] "Emotional data" refers to data on emotional states analyzed from user text input, voice data, facial expression analysis results, etc., and specifically refers to information obtained by identifying emotional states such as happiness, surprise, anger, and sadness.

[1124] "Profile" refers to detailed information that combines a user's interests, categories of interest, values, attribute information (age, gender, region, etc.) and emotional data, generated based on the user's behavioral history data and emotional data.

[1125] "User copies" refer to virtual users that simulate different situations and values ​​based on the generated profile. User copies can adjust their reactions based on emotional data.

[1126] "Personalized advertising" refers to advertising that is optimized for individual users and is generated based on historical behavioral and emotional data of a specific user or user group.

[1127] "Virtual interviews or surveys" refers to the process of asking generated user copies questions about new services or features and obtaining their responses in a simulated manner.

[1128] "Feedback Data" refers to data compiled and analyzed based on the results of hypothetical interviews or surveys and user responses to the delivery of advertisements.

[1129] The present invention relates to a system for generating and delivering personalized advertisements by collecting and analyzing user behavior history data and emotion data, and generating a user profile and user copy. This system can be realized using a terminal, a server, and a database.

[1130] Collection of user behavioral history data and emotional data

[1131] When a user logs into an internet platform, the device recognizes the user's ID and begins collecting behavioral history data. Specifically, data such as the web pages the user viewed, search keywords, links clicked, and products purchased is collected in real time or periodically. Emotional data is also collected through the user's text input, voice data, and facial expression analysis.

[1132] Storage and analysis of behavioral history data and emotional data

[1133] The device sends the collected behavioral history data and emotional data to a server. The server stores the received data in a database and performs a cleaning process. The server then uses machine learning algorithms to extract the user's behavioral patterns, interests, purchasing tendencies, etc., and analyzes the emotional data using an emotional engine.

[1134] Generating Profiles and User Copies

[1135] The server combines the analysis results with emotional data from the emotion engine to generate a user profile. This profile includes the user's interests, values, attribute information (age, gender, region, etc.), and emotional data. Based on the generated profile, the server generates multiple user copies. These user copies are virtual users that simulate different situations and values ​​of the user, and adjust the reactions of each user copy based on the emotional data.

[1136] Generating and delivering personalized ads

[1137] The server generates personalized ads using an advertising generation model (e.g., a generative AI model) based on the generated user profile. The server then delivers these ads to users' smartphones and other devices in real time. After delivering the ads, the server collects user responses and analyzes the feedback data.

[1138] Conducting virtual interviews or surveys and providing feedback on the results

[1139] The server prepares virtual interviews or questionnaires about new services or features and conducts them on the generated user copy. The user copy generates virtual questions and answers that reflect the emotional data and automatically reacts based on the scenarios set in the system. The server then compiles and analyzes the results of the virtual interviews or questionnaires and reports them as feedback.

[1140] Examples of usage

[1141] For example, if a user shows interest in a particular gadget, emotional data can be used to determine whether that interest is positive or negative. Based on this, personalized ads can be generated and delivered, resulting in more effective ad delivery. Virtual interviews about new features can also be conducted, providing detailed feedback, including emotional reactions.

[1142] Prompt Sentence Examples

[1143] Below are some examples of prompt sentences to input to the generative AI model.

[1144] "Create a program that generates personalized ads based on user behavioral history data and emotional data, and delivers them to smartphones. Behavioral data includes web pages viewed, search keywords, links clicked, and products purchased, while emotional data includes text input and voice data. Use a library for analyzing emotional data and generate a user profile using an appropriate machine learning model. Based on the generated profile, create a system that generates ads using an ad generation model and delivers them."

[1145] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1146] Step 1:

[1147] When a user logs in to an internet platform, the device recognizes the user ID and starts collecting behavioral history data. The input is the user ID, and the output is the collected behavioral history data (web pages viewed, search keywords, links clicked, products purchased, etc.).

[1148] Step 2:

[1149] The device collects emotion data using the user's text input, voice data, or camera. The input is the user's text input, voice data, or camera footage, and the output is analyzed emotion data (happiness, surprise, anger, sadness, etc.).

[1150] Step 3:

[1151] The terminal transmits the collected behavioral history data and emotion data to the server. The input is the behavioral history data and emotion data, and the output is the result of data transmission to the server.

[1152] Step 4:

[1153] The server stores the received behavioral history data and emotion data in a database and performs a cleaning process. The input is behavioral history data and emotion data, and the output is the cleaned data. Specific operations include data correction, such as deleting or supplementing characters.

[1154] Step 5:

[1155] The server uses machine learning algorithms to extract user behavioral patterns, interests, and purchasing tendencies from behavioral data, and then uses an emotion engine to analyze the emotional data. The input is cleaned behavioral data and emotional data, and the output is the analysis results. Specific operations include clustering, classification, and regression analysis.

[1156] Step 6:

[1157] The server combines the analysis results and emotional data to generate a user profile. The input is the analysis results and emotional data, and the output is a user profile. Specifically, the server integrates each item to create a profile.

[1158] Step 7:

[1159] The server generates multiple user copies based on the generated profile. The input is the user profile, and the output is the user copy. Specifically, it creates virtual users with different situations and values ​​based on the profile data.

[1160] Step 8:

[1161] The server uses the generated user copy to generate a personalized ad using an ad generation model (generative AI model). The input is the user copy, and the output is a personalized ad. Specific operations include prompts to generate ad text and images.

[1162] Step 9:

[1163] The server delivers the generated personalized advertisement to the user's smartphone. The input is the personalized advertisement, and the output is the advertisement delivery result.

[1164] Step 10:

[1165] After delivering the ad, the server collects user responses and analyzes the feedback data. The input is the user's response after the ad is delivered, and the output is the analysis results. Specific operations include analyzing click-through rates, viewing times, and changes in user emotions.

[1166] Step 11:

[1167] The server prepares a virtual interview or questionnaire about the new service or function and administers it to the generated user copy, where the input is the interview or questionnaire content and the user copy, and the output is the virtual answers.

[1168] Step 12:

[1169] The server aggregates and analyzes the results of virtual interviews and surveys, as well as advertising feedback data, and provides feedback to the development team and marketing team. The input is the results of virtual interviews and surveys, as well as advertising feedback data, and the output is a feedback report. Specific operations include statistical processing and graph generation.

[1170] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1171] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1172] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1173] [Fourth embodiment]

[1174] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1175] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1176] 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).

[1177] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1178] 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.

[1179] 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).

[1180] 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.

[1181] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1182] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1183] 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.

[1184] 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.

[1185] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1186] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1187] The present invention relates to a system for generating a user copy based on user behavior history data, thereby improving the efficiency of market research for new services and functions. Specific embodiments for carrying out the present invention are as follows.

[1188] Collection of user behavior history data

[1189] When a user engages in online activities, the device records this behavioral history data, specifically, the web pages viewed, search keywords, links clicked, products purchased, etc. This behavioral history data is sent to a server in real time or periodically.

[1190] Saving and analyzing behavioral history data

[1191] The server then stores the received behavioral history data in a database. The server then analyzes this data and applies machine learning algorithms to extract patterns such as user interests and purchasing trends. For example, a user who frequently views products in a particular category can be inferred to be particularly interested in that category.

[1192] User profile generation

[1193] The server generates a user profile based on the analysis results, which includes the user's interests, values ​​(e.g., whether they prioritize cost or new technology), and demographic information (e.g., age, gender, region).

[1194] Generate a user copy

[1195] The server then creates multiple copies of the user based on the profile. These copies are virtual users that simulate different situations and values. For example, there could be multiple types, such as a "technology-loving user who prioritizes new features over price" and a "cost-conscious user who prioritizes price."

[1196] Conducting a virtual interview or survey

[1197] The server prepares interviews and questionnaires about new services and features and conducts them on the generated user copies. Because the user copies are virtual, the server automatically generates answers based on pre-set scenarios. As a result, a wide variety of feedback can be obtained without interacting with real users.

[1198] Calculation and analysis of results

[1199] The server aggregates and analyzes the results of virtual interviews and surveys. For example, it statistically processes user evaluations of new features to identify points of interest and areas that need improvement.

[1200] Feedback of results

[1201] Finally, the server compiles the analysis results into a report and provides feedback to the development and marketing teams, which helps them develop new services, improve new features, and formulate marketing strategies.

[1202] As described above, the present invention utilizes user behavior history data to obtain feedback efficiently and accurately, thereby significantly reducing time and costs. As a specific example, if a user shows interest in a particular gadget, a virtual interview can be conducted about new features related to the gadget, and a marketing strategy can be built based on the results.

[1203] The processing flow will be explained below.

[1204] Step 1:

[1205] A user logs in to an internet platform. The device recognizes the user's user ID, such as their Yahoo! JAPAN ID, and begins collecting behavioral history data associated with that ID.

[1206] Step 2:

[1207] The device records user behavior data, specifically, the web pages the user views, the keywords they search for, the links they click, the products they purchase, and other data collected in real time or periodically.

[1208] Step 3:

[1209] The behavioral history data collected by the device is sent to the server. Data transfer is performed by batch processing or streaming processing.

[1210] Step 4:

[1211] The server stores the received behavioral history data in a database, which then stores the data appropriately and creates an index for fast access.

[1212] Step 5:

[1213] The server preprocesses the stored data, cleaning out invalid and noisy data and extracting only the necessary data. The server then classifies and organizes the data by category.

[1214] Step 6:

[1215] The server analyzes the data and applies machine learning algorithms to extract user behavioral patterns, interests, purchasing tendencies, etc. For example, if a user frequently browses a particular product category, it can be assumed that they have a high interest in that category.

[1216] Step 7:

[1217] The server generates a user profile. Based on the analysis results from the previous step, the profile is compiled based on the user's interests, categories of interest, values, and attribute information (age, gender, region, etc.).

[1218] Step 8:

[1219] The server generates multiple user copies. Based on the generated profiles, it creates several virtual users with different scenarios and values. For example, it creates a user copy that is interested in technology and prioritizes functionality over price, and a user copy that is cost-sensitive.

[1220] Step 9:

[1221] The server conducts interviews and surveys about new services and features with the user copy, asking the user copy simulated questions according to a scenario set within the system and collecting their responses.

[1222] Step 10:

[1223] The server compiles and analyzes the results of interviews and surveys, statistically processing the collected data to analyze how users respond to new services and features.

[1224] Step 11:

[1225] The server compiles the results of the analysis and compiles them into a report. The server extracts insights and potential problems and organizes them in report format.

[1226] Step 12:

[1227] The server will then provide the generated reports to the development and marketing teams, who will use the feedback to improve new services and features, and to develop marketing strategies.

[1228] This series of specific processes makes it possible to effectively utilize user behavior history data and conduct market research efficiently and accurately.

[1229] Example 1

[1230] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1231] Modern market research and new product feedback is time-consuming and costly, and the amount of feedback available from users is limited, creating a need for a fast and efficient method of gathering diverse user opinions.

[1232] 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.

[1233] In this invention, the server includes means for collecting user behavior history data, means for transmitting the behavior history data to the server, means for saving the behavior history data in a database, means for analyzing the behavior history data and generating a user profile, means for generating a plurality of user copies based on the generated profiles, means for conducting virtual interviews or questionnaires regarding new services or new functions with the user copies, means for aggregating and analyzing the results of the interviews or questionnaires, and means for compiling the results into a report and providing feedback to the development team and the marketing team. This makes it possible to quickly and efficiently collect diverse user opinions and greatly improve the efficiency of market research and new product feedback.

[1234] "User behavioral history data" refers to information about a user's activities on the Internet, specifically web pages viewed, search keywords, links clicked, products purchased, etc.

[1235] The term "server" refers to a central computer system that stores, analyzes, and processes the behavioral history data, collects data from multiple users, and performs various processes based on that data.

[1236] A "database" is a storage system that organizes and stores the behavioral history data, and is used for subsequent analysis and profile generation.

[1237] A "user profile" is a set of data generated by a machine learning algorithm that includes a user's interests, concerns, purchasing trends, values, and attribute information.

[1238] A "user copy" is a virtual user simulated based on the profile of the user, and is a modeled entity with different situations and values.

[1239] A "virtual interview or survey" is a process in which user copy is asked questions about a new service or feature and hypothetical answers are generated.

[1240] "Means for aggregation and analysis" refers to a method or system for statistically processing the results of virtual interviews or questionnaires and extracting user evaluations and opinions.

[1241] "Feedback means" refers to the process of creating a report based on the results of the above-mentioned compilation and analysis, and transmitting this to the relevant team.

[1242] The present invention relates to a system for generating a user copy based on user behavior history data, thereby improving the efficiency of market research for new services and functions. Specific embodiments for carrying out the present invention will be described below.

[1243] Collection of user behavior history data

[1244] When a user engages in online activities, their device records this behavioral history data. Specifically, it collects data on the web pages they visit, search keywords, links they click, and products they purchase. This data is collected through web browsers, mobile applications, and cloud services. This behavioral history data is sent to a server in real time or periodically.

[1245] Examples:

[1246] A user searches for "smartwatch" on an e-commerce website and browses multiple product pages. At this point, the device collects information such as the search keyword "smartwatch," the viewed page URL, and click path.

[1247] Sending behavioral history data

[1248] The device sends the collected behavioral history data to the server using an encrypted protocol (e.g., HTTPS), which protects the user's privacy.

[1249] Examples:

[1250] When the user finishes browsing the site, data is automatically sent from the device to the server.

[1251] Saving and analyzing behavioral history data

[1252] The server stores the received behavioral history data in a database. Based on the stored data, the server applies a machine learning algorithm (e.g., a sparse classifier) ​​to extract the user's interests and purchasing trends.

[1253] Examples:

[1254] Using datasets stored on a server, machine learning algorithms create profiles of users who are interested in "smartwatches."

[1255] Generate a user profile

[1256] The server generates a user profile based on the analysis of the behavioral history data, which includes the user's interests, values ​​(e.g., whether they prioritize cost or new technology), and demographic information (e.g., age, gender, region).

[1257] Examples:

[1258] If a user frequently searches for information related to "smartwatches," their profile might include "interest category: gadgets," "values: emphasis on new technology," and "demographics: 30 years old, male, living in an urban area."

[1259] Generate a user copy

[1260] The server then creates multiple copies of the user based on the profile, simulating the user as a virtual entity with different situations and values.

[1261] Examples:

[1262] The server generates two copies of the user: one that prioritizes new features and one that prioritizes costs.

[1263] Conducting a virtual interview or survey

[1264] The server prepares interviews and questionnaires about new services and features and conducts them on the virtual user copy, who then automatically generates answers based on scenarios preset by the server.

[1265] Example prompt sentence:

[1266] We generated the following profile based on historical user behavior data:

[1267] Interests: Smartwatches

[1268] Demographic information: 30 years old, male, urban resident

[1269] Values: Emphasis on new features

[1270] Create a user copy with the following conditions:

[1271] Copy 1: New feature-focused virtual user

[1272] Copy 2: Price-sensitive virtual users

[1273] For each user copy, answer the following questions:

[1274] 1. What do you think about the new tracking features on smartwatches?

[1275] 2. What is the maximum price you are willing to pay for this new feature?

[1276] Calculation and analysis of results

[1277] The server aggregates and statistically analyzes the results of the virtual interviews and surveys, and uses statistical software to extract user interests and areas for improvement based on the analysis results.

[1278] Examples:

[1279] The server records the responses collected from user copies in a database and analyzes the data using statistical software such as R or Python, extracting, for example, satisfaction scores for new features and price sensitivity.

[1280] Feedback of results

[1281] The server compiles the analysis results into a report and provides feedback to the development and marketing teams, including specific suggestions for improvements and market strategies.

[1282] Examples:

[1283] The server generates a PDF report of the analysis results and sends it to the development and marketing teams via email. The report includes the user copy evaluation results, points of interest, and improvement suggestions.

[1284] The above is a specific embodiment for carrying out the present invention.

[1285] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1286] Step 1:

[1287] The device collects data about the user's online activities, including the web pages the user views, search keywords, links clicked, and products purchased. The device temporarily stores this data in local storage.

[1288] Specific behavior:

[1289] A user searches for "smartwatch" on an e-commerce website and browses multiple product pages. At this point, the device collects information such as the search keyword "smartwatch," the viewed page URL, and click path.

[1290] Step 2:

[1291] The device sends the collected behavioral history data to the server. The input includes all the behavioral history data collected in the previous step. The data is sent using an encrypted protocol (e.g., HTTPS).

[1292] Specific behavior:

[1293] When a user browses a website, the device automatically sends activity history data to the server. For example, the data is sent when the user finishes browsing the website.

[1294] Step 3:

[1295] The server stores the received behavioral history data in a database. The input includes behavioral history data sent from the terminal. The database organizes and stores this data.

[1296] Specific behavior:

[1297] The server uses a database system (e.g., SQL database) to efficiently store user behavior history data, which makes it possible to reference the data required for subsequent analysis processing.

[1298] Step 4:

[1299] The server analyzes the stored behavioral history data and generates a user profile. The input includes the behavioral history data stored in the database. Machine learning algorithms are applied to analyze the data and extract the user's interests and purchasing trends. The output is the generated user profile.

[1300] Specific behavior:

[1301] The server uses a machine learning model (e.g., a sparse classifier) ​​based on the behavioral history dataset to generate a profile of the user's interest categories (e.g., "smartwatch") and values ​​(e.g., "emphasis on new technology").

[1302] Step 5:

[1303] The server generates multiple user copies based on the generated profile. The input includes the user's profile. Each user copy is modeled as a virtual entity with different situations and values. The output is the generated user copy.

[1304] Specific behavior:

[1305] Based on the characteristics of the user profile, the server generates two types of user copies: a "new feature-oriented type" and a "cost-oriented type."

[1306] Step 6:

[1307] The server prepares a virtual interview or questionnaire about the new service or feature and administers it to the generated user copy. The input includes questions about the new service or feature and the user copy. The server automatically generates answers based on a pre-set scenario. The output is the generated answers.

[1308] Specific behavior:

[1309] The server poses questions to the user copy, such as "What do you think about the new tracking features on your smartwatch?" or "What is the maximum price you would be willing to pay for this new feature?", and collects hypothetical responses.

[1310] Step 7:

[1311] The server aggregates and statistically analyzes the results of the virtual interviews and surveys. The input includes user-copied responses. The server processes these data using statistical software to derive user ratings and opinions. The output is the analysis results.

[1312] Specific behavior:

[1313] The server stores the response data obtained from user copies in a database and performs statistical analysis using statistical software such as R or Python, for example, to extract the distribution of satisfaction scores for new features and price sensitivity.

[1314] Step 8:

[1315] The server compiles the analysis results as a report and feeds it back to the development and marketing teams. The input includes the results of the statistical analysis. The output is a report.

[1316] Specific behavior:

[1317] The server generates a PDF report based on the analysis results and sends it to the development and marketing teams by email, including the "User Copy Evaluation Results," "Points of Interest," and "Improvement Suggestions."

[1318] (Application example 1)

[1319] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1320] Conventional methods of market research and product recommendation are not only time-consuming and costly, but also have the drawback of being difficult to respond quickly to diverse user needs. Furthermore, when using user behavior data to recommend products, it is difficult to provide personalized and accurate recommendations, which hinders the expectation of improving user satisfaction.

[1321] 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.

[1322] In this invention, the server includes means for collecting user behavior history data, means for analyzing the behavior history data and generating a user profile, means for generating multiple user copies based on the profile, means for recommending products or services based on the user copies, means for providing the user with the recommendation results, means for aggregating and analyzing the recommendation results, and means for feeding back the results. This makes it possible to quickly respond to diverse user needs and recommend personalized and accurate products and services, thereby improving user satisfaction and making market research more efficient.

[1323] "User behavioral history data" refers to data about a user's activities on the Internet, such as web pages viewed, search keywords, links clicked, and products purchased.

[1324] "Means for collecting behavioral history data" refers to means that has the function of recording a user's Internet activity and transmitting that data to a server in real time or periodically.

[1325] "Means for analyzing behavioral history data and generating user profiles" refers to means for analyzing collected behavioral history data using machine learning algorithms, etc., and extracting patterns such as user interests, concerns, and purchasing tendencies to generate profiles.

[1326] The "means for generating multiple user copies based on a profile" is a means for creating multiple virtual users that simulate different situations and values ​​based on the generated profile.

[1327] The "means for recommending a product or service based on a user copy" is a means for recommending an optimal product or service based on a profile of a virtual user copy.

[1328] The "means of providing the user with the results of the recommendation" refers to a means of providing the recommended products or services to the user through a smartphone application or the like so that the user can view them.

[1329] The "means for aggregating and analyzing the results of recommendations" refers to the means for collecting data on users' reactions to recommended products and services, and statistically processing and analyzing the data.

[1330] "Means of feedback of results" refers to compiling the analysis results in the form of a report or similar and providing them to the development team or marketing team.

[1331] The present invention relates to a system that generates a user copy based on user behavior history data and improves the efficiency of market research for new services and functions. Specifically, the system is implemented as follows.

[1332] Collecting behavioral history data

[1333] When a user browses the internet or makes a purchase, the device records this behavioral history data, including the web pages viewed, search keywords, links clicked, and products purchased. This data is sent to a server in real time or periodically via a smartphone application.

[1334] Saving and analyzing behavioral history data

[1335] The server stores the transmitted behavioral history data in a database. The stored data is analyzed using machine learning algorithms. These algorithms include clustering techniques (e.g., KMeans) and recommendation systems (e.g., Recommendation Engine). This allows patterns such as user interests and purchasing trends to be extracted.

[1336] Generate a user profile

[1337] The server generates a user profile based on the analysis results, which includes the user's interests, values ​​(e.g., whether they prioritize cost or new technology), and demographic information (e.g., age, gender, region).

[1338] Generate a user copy

[1339] The server then generates multiple user copies based on the generated profile. These copies are virtual users that simulate different situations and values. For example, various user copies can be generated, such as a "technology-loving user who prioritizes new features over price" and a "cost-conscious user who prioritizes price."

[1340] Product and service recommendations

[1341] Based on the generated user copy, the server recommends the most suitable products and services. This recommendation is performed using a recommendation engine. The products recommended for each user copy are provided to the user. For example, they are displayed in the form of a pop-up notification or a list on a smartphone application.

[1342] Calculation and analysis of results

[1343] The data obtained as a result of the recommendations is collected and statistically processed by the server. The server analyzes this data to help correct and optimize the recommendation algorithm, and also to extract areas for improvement and user reactions.

[1344] Feedback of results

[1345] Finally, the server compiles the analysis results into a report and provides feedback to the development and marketing teams, which helps them develop new services, improve new features, and formulate marketing strategies.

[1346] Specific Examples

[1347] If a user frequently browses smartphone accessories (e.g., cases, chargers), this behavioral history data is analyzed to generate a profile of "interested in smartphone accessories." Next, a virtual copy of the user is generated based on this profile, and related new smartphone accessories are recommended using the recommendation engine. The recommendation results are notified to the user's smartphone, allowing the user to browse or purchase new accessories.

[1348] An example of a specific prompt is:

[1349] text

[1350] Recommend the next likely purchase for a user based on their smartphone accessory browsing history over the past month. Past purchase history is below: [Product List]. Use this data to suggest three new product recommendations.

[1351] By using this prompt, the AI ​​can make more accurate product recommendations.

[1352] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1353] Step 1:

[1354] Collection of user behavior history data

[1355] When a user navigates online, their device records behavioral history data. Specifically, the device collects the user's web page browsing history, search keywords, clicked links, purchased products, etc. This data is sent to a server in real time or periodically via a smartphone application.

[1356] Input: User internet activity data.

[1357] Output: Behavioral history data sent to the server.

[1358] Step 2:

[1359] Saving and analyzing behavioral history data

[1360] The server stores the transmitted behavioral history data in a database. The stored data is analyzed using machine learning algorithms such as clustering techniques (e.g., KMeans) and recommendation systems (e.g., Recommendation Engine). During this analysis, patterns such as the user's interests and purchasing trends are extracted.

[1361] Input: Behavioral history data.

[1362] Output: Analyzed data and generated user profile.

[1363] Step 3:

[1364] Generate a user copy

[1365] Based on the generated user profile, the server generates multiple virtual copies of the user that simulate different situations and values, creating a diverse range of fake users to improve the accuracy of market research and product recommendations.

[1366] Input: User profile data.

[1367] Output: Multiple user copies.

[1368] Step 4:

[1369] Product and service recommendations

[1370] Based on the generated user copy, the server uses a recommendation engine to recommend the most suitable products and services. The products recommended for each user copy are provided to the user in real time. Specifically, they are displayed in the form of a pop-up notification or list on the smartphone application.

[1371] Input: User copy.

[1372] Output: A list of recommended products and services.

[1373] Step 5:

[1374] Recommendation results compilation and analysis

[1375] The data obtained as a result of the recommendations is collected by the server and processed and analyzed using statistical methods. This data includes user responses and purchase history, and is used to improve and optimize the recommendation algorithm.

[1376] Input: User response data to the recommended products.

[1377] Output: Statistically processed result data.

[1378] Step 6:

[1379] Feedback of results

[1380] Finally, the server compiles the analysis results into a report and provides feedback to the development and marketing teams, which can be used to develop new services, improve new features, and formulate marketing strategies.

[1381] Input: Statistically processed result data.

[1382] Output: Report and feedback information.

[1383] 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.

[1384] The present invention relates to a system that generates user copies based on user behavior history data, thereby improving the efficiency of market research for new services and functions. Furthermore, by combining this invention with an emotion engine that recognizes user emotions, it is possible to obtain more accurate feedback. Specific embodiments for implementing this invention are as follows.

[1385] Collection of user behavior history data

[1386] When a user logs in to an internet platform, the device recognizes the user's ID and begins collecting behavioral data, such as the web pages viewed, search keywords, links clicked, and products purchased, either in real time or periodically.

[1387] Saving and analyzing behavioral history data

[1388] The device sends the collected behavioral history data to a server, which stores the received data in a database and performs a cleaning process. The server then uses machine learning algorithms to extract behavioral patterns, interests, purchasing trends, and more.

[1389] Applying the Emotion Engine

[1390] The server is equipped with an emotion engine that analyzes user emotions from behavioral history data, user text input, voice data, etc. The emotion engine can identify emotional states such as happiness, surprise, anger, and sadness.

[1391] User profile generation

[1392] The server combines the analysis results with emotional data from the emotion engine to generate a user profile, which includes the user's interests, values, and demographic information (age, gender, region, etc.), as well as emotional data.

[1393] Generate a user copy

[1394] The server then generates multiple copies of the user based on the profile. These copies are virtual users that simulate different situations and values ​​of the user. Furthermore, it adjusts the reactions of each copy based on the user's emotional data.

[1395] Conducting a virtual interview or survey

[1396] The server prepares interviews and surveys about new services and features and administers them to the generated user copy, which generates hypothetical questions and answers that reflect the emotional data and automatically responds based on scenarios set within the system.

[1397] Calculation and analysis of results

[1398] The server compiles and analyzes the results of virtual interviews and surveys, statistically processing them, including sentiment data, to analyze how user copy reacts to new services and features.

[1399] Feedback of results

[1400] Finally, the server compiles the analysis results into a report and provides feedback to the development and marketing teams, including sentiment data, which can be used to improve new services and features, and to develop marketing strategies.

[1401] For example, if a user expresses interest in a particular gadget, sentiment analysis can determine whether that interest is positive or negative. Based on the results, virtual interviews can be conducted about new features, providing detailed feedback that includes the user's emotional reactions. This allows for the development of services and products that are more user-friendly.

[1402] The processing flow will be explained below.

[1403] Step 1:

[1404] A user logs into an internet platform. The device recognizes the user's ID and begins collecting behavioral data, including web page browsing history, search keywords, links clicked, and products purchased.

[1405] Step 2:

[1406] The device collects and transmits the behavioral history data to a server. Data transfer can be done in batch or real time. The device non-invasively monitors the user's behavior and continuously records the data.

[1407] Step 3:

[1408] The server stores the received behavioral history data in a database, where it indexes the data appropriately and manages it for quick access.

[1409] Step 4:

[1410] The server preprocesses the stored data, which includes cleaning the data and removing invalid and noisy data, extracting only the data necessary for analysis.

[1411] Step 5:

[1412] The server analyzes the pre-processed data and uses machine learning algorithms to extract user behavioral patterns, interests, purchasing tendencies, etc. For example, it can determine that a user who frequently browses products in a particular category has a strong interest in a particular product.

[1413] Step 6:

[1414] The emotion engine on the server recognizes emotions from user behavioral data, text input, and voice data. The emotion engine uses machine learning models to determine the user's emotional state (e.g., happiness, surprise, anger, sadness).

[1415] Step 7:

[1416] The server combines the analysis results with emotional data from the emotion engine to generate a user profile, which includes the user's interests, values, and demographic information (age, gender, region, etc.), as well as emotional data.

[1417] Step 8:

[1418] The server generates multiple copies of the user based on the generated profile. These copies are virtual users that simulate different situations and values ​​of the user, and adjust each copy's reactions based on emotional data.

[1419] Step 9:

[1420] The server prepares interviews and surveys about new services and features and conducts them on the generated user copy. The user copy generates virtual questions and answers that reflect the emotional data, and automatically responds based on scenarios set within the system.

[1421] Step 10:

[1422] The server compiles and analyzes the results of interviews and surveys. It statistically processes the data, including emotional data, to analyze how users react to new services and features. For example, the emotion engine identifies emotions such as "satisfied" and "dissatisfied" and calculates their percentages.

[1423] Step 11:

[1424] The server compiles the results of the analysis and compiles them into a report. The server extracts insights and potential problems and organizes them in a report format. This report also includes analysis results based on emotion data.

[1425] Step 12:

[1426] The server will then provide the generated report to the development and marketing teams. This feedback can be used to improve new services and features, and to develop marketing strategies. For example, if a user shows interest in a particular gadget and their interest is positive, they can conduct a virtual interview about the new feature and provide detailed feedback, including the user's emotional response.

[1427] Through these steps, market research can be conducted efficiently and accurately by utilizing user behavioral history data and emotion data, reducing time and costs. This system will enable the development of services and products that are in tune with user emotions.

[1428] Example 2

[1429] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1430] In today's consumer market, rapid and accurate market research is required when launching new services or features. However, traditional market research methods tend to be time-consuming and costly. Furthermore, more advanced analysis is required to accurately reflect users' emotions and behavioral patterns, but existing technologies have limitations in their accuracy. The present invention aims to solve these problems and provide a system that can conduct rapid and accurate market research based on users' detailed behavioral history and emotion data.

[1431] 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.

[1432] In this invention, the server includes means for collecting user behavioral history, transmitting it to the server, storing it, and cleaning it, means for extracting behavioral patterns, interests, and purchasing tendencies using a machine learning algorithm, and means for analyzing user emotions using an emotion engine and generating emotion data. This allows for the generation of detailed and accurate profiles based on the user behavioral history and emotion data, enabling market research using multiple virtual user copies.

[1433] "User behavioral history data" refers to data that includes information about the actions a user takes on an internet platform, such as the web pages they view, search keywords, links they click, and products they purchase.

[1434] A "server" is a central processing unit used to receive, store, and analyze behavioral history data.

[1435] A "machine learning algorithm" is a computational method for automatically learning patterns and relationships from data and predicting future behaviors and outcomes.

[1436] An "emotion engine" is a system that analyzes text input, voice data, and behavioral history data to identify a user's emotional state.

[1437] "Emotion data" is data that indicates the user's emotional state as analyzed by the emotion engine.

[1438] A "user profile" is information generated by integrating a user's behavioral patterns, interests, purchasing tendencies, attribute information (age, gender, region, etc.) and emotional data.

[1439] "User copies" are multiple virtual users generated based on a user profile and used to simulate different situations and values.

[1440] A "virtual interview or survey" is a simulated research method for new services or features conducted on user copies.

[1441] "Interview or Survey Results" are data and responses obtained through virtual interviews or surveys.

[1442] "Feedback" is information that is provided to the development and marketing teams after the results of the interviews or surveys have been analyzed and compiled into a report.

[1443] The present invention relates to a system for generating user copies based on user behavior history data and emotion data, thereby improving the efficiency of market research for new services and functions. The system of the present invention includes the following main components and associated means:

[1444] 1. Collection of user behavior history data

[1445] When a user logs into an internet platform, the device recognizes the user's ID and collects behavioral history data, including the web pages viewed, search keywords, links clicked, and products purchased. For example, if a user searches for "latest smartwatches," browses several product pages, adds products to their cart, but ultimately cancels the purchase, a series of behavioral history data will be collected.

[1446] 2. Transmission and storage of behavioral history data

[1447] The device sends the collected behavioral history data to a server, which stores the data in a database and uses Apache Hadoop to clean noise and duplicate data. This process generates a clean dataset.

[1448] 3. Data analysis

[1449] The server analyzes the collected data using machine learning algorithms, specifically TensorFlow and Scikit-learn, to extract user behavioral patterns, interests, and purchasing trends. For example, it uses a random forest algorithm to predict what purchasing behavior a particular behavioral pattern will lead to.

[1450] 4. Applying the Emotion Engine

[1451] The server is equipped with an emotion engine that analyzes the user's text input, voice data, and behavioral history data to identify the user's emotional state. The engine converts voice data into text using the Google Cloud Speech-to-Text API and analyzes emotions using a natural language processing library (e.g., NLTK).

[1452] 5. Generate a user profile

[1453] The server combines the analysis results with emotional data to create a user profile, which includes demographic information such as the user's interests, categories of interest, values, age, gender, and region, along with emotional data, resulting in a detailed and accurate user profile.

[1454] 6. Generate a user copy

[1455] The server generates multiple copies of the user based on the generated user profile. These virtual users are used to simulate different situations and values. The server adjusts the reactions of each copy of the user based on the emotional data. For example, it generates copies that show positive reactions and copies that show negative reactions.

[1456] 7. Conducting Virtual Interviews or Surveys

[1457] The server conducts virtual interviews and surveys about new services and features with the user copy, generating virtual questions and answers that reflect the user's emotional data and responding based on the set scenario.

[1458] 8. Calculation and analysis of results

[1459] The server aggregates and analyzes the results of interviews and surveys using statistical methods, such as using Hadoop's MapReduce to aggregate data and R programming language to perform statistical analysis and visualization of the data.

[1460] 9. Providing Feedback

[1461] Finally, the server compiles the analysis results into a report and provides feedback to the development and marketing teams. This feedback, including sentiment data, provides useful information for developing new services, improving new features, and formulating marketing strategies.

[1462] For example, if a user expresses interest in a particular gadget, sentiment analysis can determine whether that interest is positive or negative. Based on this result, the server can conduct virtual interviews about new features and provide detailed feedback, including the user's emotional reactions, to the development team, allowing them to develop services and products that are more user-oriented.

[1463] An example of an input prompt for a generative AI model is as follows:

[1464] "Show how a user feels about the latest smartwatch. The user's behavioral history includes viewing the product page, adding to cart once, and then canceling the purchase. What emotional state can the emotion engine determine from this behavioral history?"

[1465] The above is a specific embodiment for carrying out the present invention.

[1466] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1467] Step 1:

[1468] When a user logs in to an internet platform, the device recognizes the user's ID and begins collecting behavioral history data, such as the web pages the user viewed, search keywords, links clicked, and products purchased. The collected data is temporarily stored on the device.

[1469] Input: User login information and internet activity data

[1470] Output: Collected behavioral history data

[1471] Step 2:

[1472] The behavioral history data collected by the device is sent to a server. The server stores the received data in a database and cleans up noise and duplicate data. As a specific example, Apache Hadoop is used to process large amounts of data in a distributed manner and store it in an SQL database.

[1473] Input: Collected behavioral history data

[1474] Output: Clean and preserved behavioral history data

[1475] Step 3:

[1476] The server uses machine learning algorithms to analyze the stored behavioral history data. Specifically, TensorFlow and Scikit-learn are used to extract behavioral patterns, interests, and purchasing trends. For example, a random forest algorithm is used to predict user purchasing trends.

[1477] Input: Clean behavioral history data

[1478] Output: Analyzed behavioral patterns, interests, and purchasing tendencies

[1479] Step 4:

[1480] The server is equipped with an emotion engine that analyzes behavioral history data, user text input, and voice data to identify the user's emotional state. This engine includes a function to convert voice data into text using the Google Cloud Speech-to-Text API. A natural language processing library (e.g., NLTK) is used to classify the emotional state.

[1481] Input: Behavioral history data, user text input, voice data

[1482] Output: Identified emotion data

[1483] Step 5:

[1484] The server combines the analysis results with the emotional data to generate a user profile, which includes the user's interests, values, demographic information (age, gender, region, etc.), and emotional data.

[1485] Input: Analyzed behavioral patterns, interests, purchasing tendencies, and identified emotional data

[1486] Output: Unified user profile

[1487] Step 6:

[1488] The server generates multiple copies of the user based on the generated user profile. These virtual users are used to simulate different situations and values. The server adjusts the reactions of each copy of the user based on the emotional data.

[1489] Input: User Profile

[1490] Output: Multiple user copies

[1491] Step 7:

[1492] The server conducts virtual interviews and surveys about new services and features with the user copy, who then generates virtual questions and answers based on the emotional data and reacts based on the scenario.

[1493] Input: Interviews or surveys about new services or features

[1494] Output: User-copied hypothetical questions and answers

[1495] Step 8:

[1496] The server aggregates and analyzes the results of the virtual interviews and surveys using statistical methods. Specifically, it aggregates the data using Hadoop's MapReduce and performs statistical analysis and visualization of the data using the R programming language.

[1497] Input: Results of virtual interviews and surveys

[1498] Output: Aggregated and analyzed data

[1499] Step 9:

[1500] The server compiles the analysis results into a report and provides feedback to the development and marketing teams. For example, it can organize the results into a slide-format report and visually display emotional data and behavioral history data.

[1501] Input: Aggregated and analyzed data

[1502] Output: Feedback compiled into a report

[1503] (Application example 2)

[1504] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1505] Conventional ad delivery technologies could only deliver ads based on user behavioral history data, limiting their effectiveness. Furthermore, because they could not take user emotions into account, it was difficult to maximize the effectiveness of ads and user satisfaction. Furthermore, when conducting market research for new services and features, there was a lack of means to obtain feedback that took user emotions into account, making it difficult to improve the accuracy of development and marketing strategies.

[1506] 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 collecting user behavior history data and emotion data, means for analyzing the behavior history data and emotion data to generate a user profile, and means for generating multiple user copies based on the profile and generating and delivering personalized advertisements. This enables the delivery of personalized advertisements that take user emotions into consideration, maximizing the effectiveness of advertisements and improving user satisfaction. Furthermore, in market research for new services and functions, detailed feedback that reflects user emotions can be obtained, thereby improving the accuracy of development and marketing strategies.

[1507] "User behavioral history data" refers to data about the actions a user takes on an internet platform, specifically information such as the web pages viewed, search keywords, links clicked, and products purchased.

[1508] "Emotional data" refers to data on emotional states analyzed from user text input, voice data, facial expression analysis results, etc., and specifically refers to information obtained by identifying emotional states such as happiness, surprise, anger, and sadness.

[1509] "Profile" refers to detailed information that combines a user's interests, categories of interest, values, attribute information (age, gender, region, etc.) and emotional data, generated based on the user's behavioral history data and emotional data.

[1510] "User copies" refer to virtual users that simulate different situations and values ​​based on the generated profile. User copies can adjust their reactions based on emotional data.

[1511] "Personalized advertising" refers to advertising that is optimized for individual users and is generated based on historical behavioral and emotional data of a specific user or user group.

[1512] "Virtual interviews or surveys" refers to the process of asking generated user copies questions about new services or features and obtaining their responses in a simulated manner.

[1513] "Feedback Data" refers to data compiled and analyzed based on the results of hypothetical interviews or surveys and user responses to the delivery of advertisements.

[1514] The present invention relates to a system for generating and delivering personalized advertisements by collecting and analyzing user behavior history data and emotion data, and generating a user profile and user copy. This system can be realized using a terminal, a server, and a database.

[1515] Collection of user behavioral history data and emotional data

[1516] When a user logs into an internet platform, the device recognizes the user's ID and begins collecting behavioral history data. Specifically, data such as the web pages the user viewed, search keywords, links clicked, and products purchased is collected in real time or periodically. Emotional data is also collected through the user's text input, voice data, and facial expression analysis.

[1517] Storage and analysis of behavioral history data and emotional data

[1518] The device sends the collected behavioral history data and emotional data to a server. The server stores the received data in a database and performs a cleaning process. The server then uses machine learning algorithms to extract the user's behavioral patterns, interests, purchasing tendencies, etc., and analyzes the emotional data using an emotional engine.

[1519] Generating Profiles and User Copies

[1520] The server combines the analysis results with emotional data from the emotion engine to generate a user profile. This profile includes the user's interests, values, attribute information (age, gender, region, etc.), and emotional data. Based on the generated profile, the server generates multiple user copies. These user copies are virtual users that simulate different situations and values ​​of the user, and adjust the reactions of each user copy based on the emotional data.

[1521] Generating and delivering personalized ads

[1522] The server generates personalized ads using an advertising generation model (e.g., a generative AI model) based on the generated user profile. The server then delivers these ads to users' smartphones and other devices in real time. After delivering the ads, the server collects user responses and analyzes the feedback data.

[1523] Conducting virtual interviews or surveys and providing feedback on the results

[1524] The server prepares virtual interviews or questionnaires about new services or features and conducts them on the generated user copy. The user copy generates virtual questions and answers that reflect the emotional data and automatically reacts based on the scenarios set in the system. The server then compiles and analyzes the results of the virtual interviews or questionnaires and reports them as feedback.

[1525] Examples of usage

[1526] For example, if a user shows interest in a particular gadget, emotional data can be used to determine whether that interest is positive or negative. Based on this, personalized ads can be generated and delivered, resulting in more effective ad delivery. Virtual interviews about new features can also be conducted, providing detailed feedback, including emotional reactions.

[1527] Prompt Sentence Examples

[1528] Below are some examples of prompt sentences to input to the generative AI model.

[1529] "Create a program that generates personalized ads based on user behavioral history data and emotional data, and delivers them to smartphones. Behavioral data includes web pages viewed, search keywords, links clicked, and products purchased, while emotional data includes text input and voice data. Use a library for analyzing emotional data and generate a user profile using an appropriate machine learning model. Based on the generated profile, create a system that generates ads using an ad generation model and delivers them."

[1530] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1531] Step 1:

[1532] When a user logs in to an internet platform, the device recognizes the user ID and starts collecting behavioral history data. The input is the user ID, and the output is the collected behavioral history data (web pages viewed, search keywords, links clicked, products purchased, etc.).

[1533] Step 2:

[1534] The device collects emotion data using the user's text input, voice data, or camera. The input is the user's text input, voice data, or camera footage, and the output is analyzed emotion data (happiness, surprise, anger, sadness, etc.).

[1535] Step 3:

[1536] The terminal transmits the collected behavioral history data and emotion data to the server. The input is the behavioral history data and emotion data, and the output is the result of data transmission to the server.

[1537] Step 4:

[1538] The server stores the received behavioral history data and emotion data in a database and performs a cleaning process. The input is behavioral history data and emotion data, and the output is the cleaned data. Specific operations include data correction, such as deleting or supplementing characters.

[1539] Step 5:

[1540] The server uses machine learning algorithms to extract user behavioral patterns, interests, and purchasing tendencies from behavioral data, and then uses an emotion engine to analyze the emotional data. The input is cleaned behavioral data and emotional data, and the output is the analysis results. Specific operations include clustering, classification, and regression analysis.

[1541] Step 6:

[1542] The server combines the analysis results and emotional data to generate a user profile. The input is the analysis results and emotional data, and the output is a user profile. Specifically, the server integrates each item to create a profile.

[1543] Step 7:

[1544] The server generates multiple user copies based on the generated profile. The input is the user profile, and the output is the user copy. Specifically, it creates virtual users with different situations and values ​​based on the profile data.

[1545] Step 8:

[1546] The server uses the generated user copy to generate a personalized ad using an ad generation model (generative AI model). The input is the user copy, and the output is a personalized ad. Specific operations include prompts to generate ad text and images.

[1547] Step 9:

[1548] The server delivers the generated personalized advertisement to the user's smartphone. The input is the personalized advertisement, and the output is the advertisement delivery result.

[1549] Step 10:

[1550] After delivering the ad, the server collects user responses and analyzes the feedback data. The input is the user's response after the ad is delivered, and the output is the analysis results. Specific operations include analyzing click-through rates, viewing times, and changes in user emotions.

[1551] Step 11:

[1552] The server prepares a virtual interview or questionnaire about the new service or function and administers it to the generated user copy, where the input is the interview or questionnaire content and the user copy, and the output is the virtual answers.

[1553] Step 12:

[1554] The server aggregates and analyzes the results of virtual interviews and surveys, as well as advertising feedback data, and provides feedback to the development team and marketing team. The input is the results of virtual interviews and surveys, as well as advertising feedback data, and the output is a feedback report. Specific operations include statistical processing and graph generation.

[1555] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1556] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1557] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1558] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1559] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1560] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1561] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1562] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1563] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1564] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1565] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1566] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1567] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1568] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1569] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1570] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1571] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1572] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1573] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1574] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1575] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1576] The following is further disclosed regarding the above embodiment.

[1577] (Claim 1)

[1578] A means for collecting user behavior history data;

[1579] means for analyzing the behavioral history data and generating a profile of the user;

[1580] means for generating a plurality of user copies based on said profile;

[1581] means for conducting a virtual interview or questionnaire regarding new services or new features with said user copy;

[1582] means for compiling and analyzing the results of said interviews or questionnaires;

[1583] A means for feeding back the results;

[1584] A system including:

[1585] (Claim 2)

[1586] The system of claim 1, further comprising means for generating the profile based on values ​​and attribute information of the user.

[1587] (Claim 3)

[1588] 2. The system according to claim 1, further comprising database means for storing said behavior history data.

[1589] "Example 1"

[1590] (Claim 1)

[1591] A means for collecting user behavior history data;

[1592] means for transmitting the behavior history data to a server;

[1593] means for storing the behavior history data in a database;

[1594] means for analyzing the behavioral history data and generating a profile of the user;

[1595] means for generating a plurality of user copies based on the generated profile;

[1596] means for conducting a virtual interview or questionnaire regarding new services or new features with said user copy;

[1597] means for compiling and analyzing the results of said interviews or questionnaires;

[1598] A means for compiling the results into a report and providing feedback to the development team and marketing team;

[1599] A system including:

[1600] (Claim 2)

[1601] The system of claim 1, further comprising means for generating the profile based on values ​​and attribute information of the user.

[1602] (Claim 3)

[1603] 10. The system of claim 1, further comprising means for statistically analyzing the results of said interviews or questionnaires.

[1604] "Application Example 1"

[1605] New Claims

[1606] (Claim 1)

[1607] A means for collecting user behavior history data;

[1608] means for analyzing the behavioral history data and generating a profile of the user;

[1609] means for generating a plurality of user copies based on said profile;

[1610] means for recommending goods or services based on said user copy;

[1611] means for providing the results of the recommendations to a user;

[1612] means for aggregating and analyzing the results of said recommendations;

[1613] A means for feeding back the results;

[1614] A system including:

[1615] (Claim 2)

[1616] The system of claim 1, further comprising means for generating the profile based on values ​​and attribute information of the user.

[1617] (Claim 3)

[1618] 2. The system according to claim 1, further comprising database means for storing said behavior history data.

[1619] "Example 2: Combining Emotion Engines"

[1620] (Claim 1)

[1621] A means for collecting user behavior history data;

[1622] a means for transmitting the behavior history data to a server, and storing and cleaning the data;

[1623] A means for analyzing the behavioral history data and extracting behavioral patterns, interests, and purchasing tendencies using a machine learning algorithm;

[1624] a means for analyzing a user's emotions using an emotion engine to generate emotion data;

[1625] means for combining the analysis results with emotion data to generate a profile of the user;

[1626] means for generating a plurality of user copies based on the profile and adjusting the responses of each user copy based on emotion data;

[1627] means for conducting a virtual interview or questionnaire regarding new services or new features with said user copy;

[1628] means for compiling and analyzing the results of said interviews or questionnaires;

[1629] A means of compiling the results into a report and providing feedback to the development team and marketing team;

[1630] A system including:

[1631] (Claim 2)

[1632] The system of claim 1, further comprising means for generating the profile based on values ​​and attribute information of the user.

[1633] (Claim 3)

[1634] 2. The system according to claim 1, further comprising database means for storing said behavior history data and emotion data.

[1635] "Application example 2 when combining emotion engines"

[1636] (Claim 1)

[1637] A means for collecting user behavior history data and emotion data;

[1638] means for analyzing the behavioral history data and emotion data to generate a profile of the user;

[1639] means for generating a plurality of user copies based on the profile and generating and delivering personalized advertisements;

[1640] means for conducting a virtual interview or questionnaire regarding new services or new features with said user copy;

[1641] means for compiling and analyzing the results of the interviews or questionnaires and advertising feedback data;

[1642] A means for feeding back the results;

[1643] A system including:

[1644] (Claim 2)

[1645] The system of claim 1 , further comprising means for generating the profile based on values, attribute information, and emotional data of the user.

[1646] (Claim 3)

[1647] 2. The system according to claim 1, further comprising database means for storing said behavior history data and emotion data. [Explanation of symbols]

[1648] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means for collecting user behavior history data; means for analyzing the behavioral history data and generating a profile of the user; means for generating a plurality of user copies based on said profile; means for conducting a virtual interview or questionnaire regarding new services or new features with said user copy; means for compiling and analyzing the results of said interviews or questionnaires; A means for feeding back the results; A system including:

2. The system according to claim 1 , further comprising means for generating the profile based on values ​​and attribute information of the user.

3. The system according to claim 1 , further comprising database means for storing said behavior history data.

Images