system

Abstract

We provide the system. [Solution] A means for receiving service and campaign design proposals based on user input, A means of transmitting the aforementioned design proposal to a generative artificial intelligence and conducting an evaluation based on relevant laws and guidelines, A means of receiving evaluation results from a generative artificial intelligence and displaying them to the user, Means for generating multiple variations based on the aforementioned design proposal, A means for conducting A / B testing from multiple angles using the aforementioned variations, A means for aggregating the results of the aforementioned A / B test and selecting the optimal variation, A system that includes this.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a 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

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the design of services and campaigns, there is a need for a method to efficiently confirm problems based on laws and guidelines and conduct multi-faceted evaluations. However, until now, the confirmation work has mainly been done by humans, and it has been difficult to determine the optimal service design that reflects legal omissions and user preferences. As a result, deficiencies in the design process may cause claims and legal problems later. The present invention aims to solve such problems and provide a system that can efficiently design legally appropriate and user-optimal services and campaigns.

Means for Solving the Problems

[0005] The present invention provides the following means.

[0006] First, the system provides a means for receiving service and campaign design proposals based on user input. Next, it provides a means for transmitting the design proposals to a generative artificial intelligence and conducting evaluations based on relevant laws and guidelines. The system also includes means for receiving the evaluation results from the generative AI and displaying them to the user, and means for generating multiple variations based on the design proposals. Furthermore, it provides means for conducting multifaceted A / B testing using the variations, and includes means for aggregating the results of the A / B testing and selecting the optimal variation. This provides a system that makes it easy to design optimal services and campaigns through multifaceted testing while ensuring that there are no legal issues.

[0007] A "user" is a person who operates the system and inputs and confirms design proposals for services and campaigns.

[0008] A "terminal" is a device used by users to access the system, input design proposals, check evaluation results, and run tests.

[0009] A "server" is a computer system that handles the main processing and data management of a system, and also performs tasks such as coordinating with generative artificial intelligence, aggregating evaluation results, and generating variations.

[0010] "Generative artificial intelligence" refers to an AI system that receives design proposals for services or campaigns, evaluates them based on relevant laws and guidelines, and returns the results.

[0011] A "design proposal" refers to a plan or suggestion that outlines the content and structure of a service or campaign entered by a user.

[0012] "Evaluation" is the process by which a generative artificial intelligence determines whether a design proposal is problematic based on relevant laws and guidelines.

[0013] A "variation" refers to a design proposal that includes multiple different settings and conditions, generated based on the original design.

[0014] "AB testing" refers to a type of test that involves trying out multiple variations on different user groups and comparing their performance.

[0015] "Result aggregation" is the process of collecting performance data for each variation of an A / B test and statistically analyzing it to select the optimal variation.

[0016] "Optimal variation" refers to the settings and conditions of a service or campaign that are most effective and well-received by users, based on the results of A / B testing. [Brief explanation of the drawing]

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

Embodiments for Carrying Out the Invention

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

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

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

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

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

[0023] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0024] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0025] [First Embodiment]

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

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

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

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

[0030] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0031] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0032] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

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

[0034] As shown in Figure 2, in the data processing device 12, a specific processing 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" related to the technology of this 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 according to the specific processing program 56 executed on the RAM 30.

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

[0036] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

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

[0038] This invention provides a system in which users input service and campaign design proposals, and the system designs the optimal service through verification of laws and guidelines using generative artificial intelligence and multifaceted A / B testing. The operation of the system in each processing step is described in detail below.

[0039] User input

[0040] Users input service and campaign design proposals using their devices. The input design proposals are then sent from the device to the server.

[0041] Checking laws and guidelines

[0042] The server sends the design proposal received from the user to a generative artificial intelligence (AI) system, which then performs an evaluation based on relevant laws and guidelines. The AI ​​analyzes the design proposal and determines whether there are any legal issues. The evaluation results are returned to the server, which then sends them to the terminal. The terminal then displays the evaluation results to the user.

[0043] Generating Variations

[0044] If the evaluation result is determined to be "no problems," the server generates multiple variations based on the design proposal. These variations refer to multiple proposals containing different settings and conditions, and the server sends these to the terminal.

[0045] Conducting A / B testing

[0046] The device sets up test scenarios based on the received variations and conducts A / B testing from multiple angles. Different variations are presented to multiple user groups, and test results are collected. For example, variations with 15% cashback, 20% cashback, and a free coupon are set up, and performance data (click-through rate, registration rate, purchase rate, etc.) is collected for each.

[0047] Summary of results and selection of the optimal variation

[0048] After the test is complete, the terminal sends the obtained performance data to the server. The server aggregates this data, analyzes it statistically, and selects the optimal variation.

[0049] Notification of optimal variation

[0050] Once the optimal variation is selected, the server sends the result to the terminal. The terminal then notifies the user of the optimal service design and releases the final service based on it.

[0051] Specific examples of operation

[0052] The following is a concrete example of how it works. The user designs a "new points reward campaign" and inputs it into the terminal. This design proposal is sent to the server, where it is legally checked by generative artificial intelligence. It is confirmed that it does not violate consumer protection laws or the Premiums and Representations Act, and the results are notified to the user. The server then generates variations such as 15% reward, 20% reward, and with a free coupon, and A / B testing is conducted on the terminal. The test results are sent to the server, which selects the "20% reward" option that is most popular with the user, and notifies the user of the result.

[0053] As described above, the system based on the present invention can efficiently perform verification in accordance with laws and guidelines, and design optimal services that reflect user preferences through A / B testing.

[0054] The following describes the processing flow.

[0055] Step 1:

[0056] The user inputs a design proposal for a service or campaign into the terminal. The terminal then sends the input design proposal to the server.

[0057] Step 2:

[0058] The server sends the received design proposal to the generative artificial intelligence. The generative AI analyzes the design proposal and conducts a legal evaluation based on relevant laws and guidelines.

[0059] Step 3:

[0060] The evaluation results from the generative artificial intelligence are returned to the server. The server receives the evaluation results and sends them to the terminal.

[0061] Step 4:

[0062] The device displays the evaluation results to the user. The user reviews the evaluation results and understands that there are no legal issues.

[0063] Step 5:

[0064] The server generates multiple variations based on the design proposal that was deemed "problem-free" in the evaluation results. These generated variations include different settings and conditions.

[0065] Step 6:

[0066] The server sends the generated variations to the device. The device then prepares to use the variations to conduct an A / B test.

[0067] Step 7:

[0068] The device presents variations to multiple user groups. For example, it might offer variations such as 15% cashback, 20% cashback, or a free coupon, and collects performance data (click-through rate, registration rate, purchase rate, etc.) for each variation.

[0069] Step 8:

[0070] Once the A / B test is complete, the device sends the collected performance data to the server. The server then aggregates and statistically analyzes the received data.

[0071] Step 9:

[0072] The server selects the most effective variation based on the aggregated data. Once the optimal variation is determined, the server sends the result to the terminal.

[0073] Step 10:

[0074] The device notifies the user of the optimal service design. Based on the notified optimal design, the user releases the final service or campaign.

[0075] The above describes the specific actions of each processing step in the program. This system makes it possible to comply with laws and guidelines and efficiently provide the best possible service to users.

[0076] (Example 1)

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

[0078] In recent years, when designing services and campaigns that appeal to consumers, users are required to experiment with multiple design options while avoiding legal risks. However, checking legal issues and rationally testing various variations to select the optimal design is a multifaceted and complex process that requires considerable effort and time. This has created a challenge in that it is difficult for users to quickly and efficiently find the optimal design.

[0079] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0080] In this invention, the server includes means for receiving design proposals for services and campaigns based on user input; means for transmitting the design proposals to a generative artificial intelligence and conducting evaluations based on relevant laws and guidelines; means for receiving evaluation results from the generative artificial intelligence and displaying them to the user; means for generating multiple variations based on the design proposals; means for conducting comparative tests from multiple perspectives using the variations; means for aggregating the results of the comparative tests and selecting the optimal variation; and means for notifying the user of the optimal variation. This makes it possible to efficiently design optimal services and campaigns while avoiding legal risks.

[0081] A "user" refers to an individual or legal entity that inputs design proposals for services or campaigns and utilizes the system.

[0082] "Terminal" refers to an input device used by a user, and includes electronic devices such as personal computers, smartphones, and tablets.

[0083] A "server" is a computer system that performs the central processing of a system and provides multiple functions in an integrated manner.

[0084] A "design proposal" refers to a plan that describes the specific content and conditions of a service or campaign that users will input.

[0085] "Generative artificial intelligence" refers to an artificial intelligence system that performs advanced analytical processing, such as legal evaluation and variation generation, from given input data.

[0086] "Laws and guidelines" refer to the laws and industry standards that apply to the service or campaign, including consumer protection laws and the Premiums and Representations Act.

[0087] "Evaluation results" refer to the results obtained when a generative artificial intelligence analyzes a design proposal and determines whether there are any legal issues or whether it conforms to guidelines.

[0088] "Variations" refer to multiple design options with different conditions and settings, generated based on the original design proposal.

[0089] A "comparative test" refers to a testing method in which multiple variations are presented to different user groups, and their response data is collected and compared.

[0090] "Performance data" refers to data showing user responses to each variation, including click-through rates, registration rates, and purchase rates.

[0091] The "optimal variation" refers to the variation that showed the highest performance data in comparative tests.

[0092] "Notifying" refers to the process of transmitting information from a server to a user via a terminal.

[0093] This invention relates to a system in which a user inputs a design proposal for a service or campaign, and the system performs optimal service design through legal evaluation and multifaceted comparative testing using generative artificial intelligence. Specific embodiments for carrying out this invention are described in detail below.

[0094] Users input service and campaign design proposals using their devices. For example, for a "new points reward campaign," they would describe the campaign details (campaign period, reward rate, etc.) in an input form on their device. This design proposal is then sent from the device to the server.

[0095] The server sends the received design proposal to a generative artificial intelligence (e.g., OpenAI® GPT-4®). The generative AI analyzes the design proposal, which is input as a prompt: "Please check whether this design proposal violates consumer protection laws or the Premiums and Representations Act." Based on the analysis, it determines whether there are any legal issues and returns the evaluation result to the server. The evaluation result is resent from the server to the terminal and displayed to the user on the terminal. For example, the evaluation result displayed might be: "This campaign does not violate consumer protection laws."

[0096] If the evaluation result is determined to be "no problems," the server generates multiple variations based on the design proposal. Generative artificial intelligence and algorithms are used to create variations with different settings and conditions. For example, it might generate three variations: "15% cashback," "20% cashback," and "with a free coupon." The generated variations are then sent from the server to the terminal.

[0097] The device sets up a comparative test scenario based on the received variations. This scenario presents different variations to multiple user groups and collects user response data (click-through rate, registration rate, purchase rate, etc.) for each variation. For example, user group A is presented with "15% cashback," user group B with "20% cashback," and user group C with "free coupon."

[0098] Once the results of each comparison test are collected, the device sends this performance data to the server. The server aggregates the data and performs statistical analysis to select the most effective variation. For example, by comparing the click-through rate and purchase rate of each variation, the "20% cashback" variation may be found to have the best performance. In this process, the R or Python pandas library may be used.

[0099] Once the optimal variation is selected, the server notifies the terminal of the result. The terminal then notifies the user of the optimal variation, and the user ultimately releases the service based on that variation. For example, the user might see a notification that says, "Test results indicate that a 20% cashback has been selected as the most effective variation."

[0100] In this way, the system based on the present invention can generate design proposals based on user input and efficiently design optimal services and campaigns through multifaceted comparative testing while avoiding legal risks.

[0101] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0102] Step 1:

[0103] Users input service and campaign design proposals using their devices. For example, they might describe a "new points reward campaign," detailing specific conditions such as the campaign period and reward rate. This design proposal is sent from the device to the server as input data. Once the server receives the design proposal as input, processing begins on the server side.

[0104] Step 2:

[0105] The server sends the received design proposal to a generative artificial intelligence (e.g., OpenAI GPT-4). Specifically, the design proposal is used as input data with the prompt message "Please check whether this design proposal violates consumer protection laws or the Premiums and Representations Act." The generative AI model analyzes this input data and determines whether there are any legal issues. The evaluation result is sent back to the server as output, and this result serves as information to confirm whether there are any legal problems.

[0106] Step 3:

[0107] The server receives the evaluation results from the generative artificial intelligence and sends them to the user's terminal for them to review. The terminal displays the legal evaluation results to the user, for example, a message such as "This campaign does not violate consumer protection laws." Information to be notified to the user is generated based on the evaluation results as input.

[0108] Step 4:

[0109] If the evaluation result is determined to be "no problems," the server generates multiple variations based on the design proposal. Generative artificial intelligence and algorithms are used to create multiple variations with different settings and conditions. For example, variations such as "15% cashback," "20% cashback," and "with a free coupon" may be included. The generated variations are sent from the server to the terminal as output data, preparing it for the next test.

[0110] Step 5:

[0111] The device sets up a comparative test scenario based on the received variations. Specifically, it creates test cases that present different variations to multiple user groups. User group A is presented with "15% cashback," user group B with "20% cashback," and user group C with "free coupon." A multifaceted comparative test is performed based on the variations as input. The output is the collection of response data from each group.

[0112] Step 6:

[0113] Once the results of the comparative tests are collected, the device sends this performance data (e.g., click-through rate, registration rate, purchase rate) to the server. The server aggregates this data and performs statistical analysis. Based on the performance data as input, it selects the most effective variation. For example, by comparing the click-through rate and purchase rate of each variation, the results might show that "20% cashback" performs the best. As output, the optimal variation is selected.

[0114] Step 7:

[0115] Once the optimal variation is selected, the server notifies the terminal of the result. The terminal then notifies the user of the optimal variation, and the user ultimately releases the service based on that variation. For example, a notification might appear stating, "Based on the test results, a 20% cashback has been selected as the most effective variation." This provides the user with information to take their next action based on the test results as input.

[0116] (Application Example 1)

[0117] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0118] Traditional advertising campaign design often involved manual processes for checking legal issues, generating multiple variations, testing, and selecting the optimal variation, resulting in inefficiencies and time-consuming tasks. Furthermore, there was a lack of systems providing user-friendly interfaces using smart devices. This made it difficult to quickly and effectively maximize marketing effectiveness.

[0119] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0120] In this invention, the server includes means for receiving service and campaign design proposals based on user input; means for transmitting the design proposals to a generative artificial intelligence and conducting evaluations based on relevant regulations and guidelines; means for receiving evaluation results from the generative artificial intelligence and displaying them to the user; means for generating multiple options based on the design proposals; means for conducting multifaceted A / B testing using the options; means for aggregating the results of the A / B testing and selecting the optimal option; means for notifying the user of the optimal option; and means including an application installed on a smart device. This enables efficient advertising campaign design, legal review, variation generation, testing, and optimization, and maximizes marketing effectiveness with a user-friendly interface.

[0121] A "user" refers to a person who uses the system to design services and campaigns.

[0122] "Input" refers to the act of a user providing design proposals and related information to the system.

[0123] A "design proposal" refers to information that shows the specific details of a service or campaign proposed by a user.

[0124] "Generative artificial intelligence" refers to artificial intelligence technology that uses user-inputted design proposals to check relevant regulations and guidelines, and conducts multifaceted A / B testing.

[0125] "Evaluation" refers to the process by which a generative artificial intelligence determines whether a design proposal has legal issues or complies with guidelines.

[0126] "Legal issues" refer to the question of whether the proposed design violates legal regulations.

[0127] "Regulations" refer to laws and guidelines that apply to the design of a service or campaign.

[0128] "Options" refers to multiple variations or different proposals with different conditions that are generated based on a design proposal.

[0129] "AB testing" refers to a testing method that presents multiple options to different user groups and compares and evaluates the performance of each option.

[0130] "Aggregation" refers to the act of statistically compiling the results data of an A / B test.

[0131] The "optimal choice" refers to the variation that shows the highest performance based on the results of the A / B test.

[0132] "Smart devices" refer to advanced devices such as smartphones, smart glasses, and head-mounted displays.

[0133] This invention involves several key steps to realize a system in which users input service and campaign design proposals using a smart device. These steps are described below.

[0134] First, the user uses a smart device such as a smartphone to input their service or campaign design proposal. The user enters the details of the specific design proposal and sends it from the device to the server.

[0135] Next, the server sends the received design proposal to the generative artificial intelligence (AI) and performs an evaluation based on relevant regulations and guidelines. The generative AI analyzes the design proposal and determines whether there are any legal issues. This legal evaluation includes consumer protection laws and the Premiums and Representations Act, among others. The evaluation results of the generative AI are returned to the server, which then sends the results to the user's terminal and displays them to the user.

[0136] Next, if the server determines the evaluation result is "no problems," it generates multiple options (variations) based on the design proposal. These variations have different settings and conditions, and may include options such as "15% cashback," "20% cashback," and "with a free coupon." These variations are then sent from the server to the terminal.

[0137] The device conducts multifaceted A / B testing based on the variations it receives. This test presents different variations to multiple user groups and collects performance data (click-through rates, registration rates, purchase rates, etc.) resulting from these tests. This data is then sent to the server.

[0138] The server aggregates the received performance data, analyzes it statistically, and selects the optimal option. The selected optimal option is then notified from the server to the terminal, informing the user of the best possible service design.

[0139] As a concrete example, a user designs a "new points reward campaign" and inputs a design proposal for a "20% cashback campaign." Generative artificial intelligence legally evaluates this proposal and confirms that it does not violate consumer protection laws or the Premiums and Representations Act. The server then generates variations such as "15% cashback," "20% cashback," and "with free coupon," and A / B testing is conducted on the terminal. The test results are sent to the server, which selects the "20% cashback" option, which showed the highest performance, as the optimal choice and notifies the user of the result.

[0140] Examples of prompt messages are as follows:

[0141] We are designing a new points reward campaign. Please perform legal checks and generate variations based on the details below.

[0142] Campaign type: Point rewards

[0143] Details: 20% cashback campaign

[0144] Thus, the system based on this invention maximizes marketing effectiveness by quickly and efficiently evaluating user-inputted design proposals and providing the optimal service design.

[0145] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0146] Step 1:

[0147] Users input service and campaign design proposals using a smart device. Specifically, users open a dedicated application on their smartphone, input their design proposal (e.g., "20% Cashback Campaign"), and press the submit button. Input data: Campaign details. Output data: Confirmation message for submitting the design proposal.

[0148] Step 2:

[0149] The server receives design proposals submitted by the user. The server then forwards the input to a generative artificial intelligence system, which conducts a legal evaluation based on relevant regulations and guidelines. This legal evaluation includes checks regarding consumer protection laws and the Premiums and Representations Act. Input data: Design proposals. Data processing: Evaluation based on legal regulations. Output data: Legal evaluation results.

[0150] Step 3:

[0151] Generative artificial intelligence analyzes received design proposals and determines whether they have any legal issues. Generative AI uses natural language processing and rule-based analysis to check which regulations or guidelines the design proposals violate. Input data: Design proposal. Data calculation: Comparison with legal regulations. Output data: Result of legal evaluation (e.g., "No issues").

[0152] Step 4:

[0153] The server receives evaluation results from the generative artificial intelligence, sends them to the user's terminal, and displays them to the user. The displayed content is the result of the legal evaluation and includes instructions for when there are no problems or when corrections are needed. Input data: Legal evaluation results. Output data: Content displayed to the user.

[0154] Step 5:

[0155] If the evaluation result is determined to be "no problems," the server generates multiple variations based on the design proposal. These variations include different reward rates and benefits. The generated variations are sent to the terminal. Input data: Evaluated design proposal. Data processing: Generation of variations. Output data: Multiple variations.

[0156] Step 6:

[0157] The device conducts multifaceted A / B testing based on the received variations. Specifically, it presents different variations to user groups and collects performance data such as click-through rates and purchase rates. Input data: Multiple variations. Data processing: Collection of user behavior. Output data: Performance data (e.g., click-through rate, purchase rate).

[0158] Step 7:

[0159] The terminal sends the collected performance data to the server. The server aggregates the received performance data, performs statistical analysis, and selects the optimal variation. Input data: Performance data. Data calculation: Statistical analysis. Output data: Result of the selection of the optimal variation.

[0160] Step 8:

[0161] The server selects the optimal variation and notifies the user of the result on their device. The user receives the notification and confirms the optimal campaign design. Input data: Selection result. Output data: Notification content to the user (e.g., "20% cashback is optimal").

[0162] By following these steps, users can efficiently design legally compliant and optimal advertising campaigns.

[0163] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0164] This invention provides a technology that combines an emotion engine with a system that supports the design of services and campaigns, enabling the design of optimal services and campaigns based on user emotion information. The operation of the system in each processing step is described in detail below.

[0165] User input

[0166] Users input service and campaign design proposals using a device. During this process, the device utilizes its built-in emotion engine to recognize emotions from the user's facial expressions, voice, and text, and collects that emotion data.

[0167] Checking laws and guidelines

[0168] The server transmits the design proposal and emotional data received from the user to a generative artificial intelligence (AI) system, which then performs an evaluation based on relevant laws and guidelines. The AI ​​analyzes the design proposal and conducts a legal evaluation, taking the emotional data into consideration. The evaluation results are returned to the server, which then transmits them to the terminal.

[0169] Displaying results and reflecting sentiment data

[0170] The device displays the evaluation results to the user. At the same time, it adjusts the design proposal as needed based on sentiment data and provides feedback in a way that is appropriate for the user. For example, if a campaign suggested by the user does not align with the user's preferences or expectations, the sentiment engine will use that information to suggest adjustments.

[0171] Generating Variations

[0172] The server generates multiple variations based on the refined design proposal. These variations include different settings and conditions, and are intended for multifaceted evaluation.

[0173] Conducting A / B testing

[0174] The device will conduct A / B testing using the generated variations. Multiple user groups will be presented with the variations, and performance data for each variation (click-through rate, registration rate, purchase rate, etc.) will be collected. During the test, an emotion engine will be used to evaluate how users feel about the variations.

[0175] Summary of results and selection of the optimal variation

[0176] After the test is complete, the device sends the collected performance and sentiment data to the server. The server aggregates and statistically analyzes this data to select the optimal variation. Sentiment data is considered an important indicator for evaluating user satisfaction and motivation.

[0177] Notification of optimal variation

[0178] Once the optimal variation is selected, the server sends the result to the terminal. The terminal notifies the user of the optimal service design and releases the final service or campaign based on it. An optimal design that reflects user sentiment information increases user satisfaction and reduces the risk of complaints.

[0179] Specific examples of operation

[0180] The following is a concrete example of how it works. The user designs a "new points reward campaign" and inputs it into the terminal. The emotion engine recognizes the user's emotions and sends it to the server along with the design proposal. A legal check is performed by generative artificial intelligence, and the evaluation results are notified to the user, while adjustments based on the emotion data are suggested. Subsequently, variations such as 15% reward, 20% reward, and with a free coupon are generated, and A / B testing is conducted. Based on the test results and emotion data, the server determines that "20% reward" is optimal, and this result is notified to the user.

[0181] As described above, the system based on the present invention can efficiently design optimal services and campaigns by combining evaluations of laws and guidelines with user sentiment information.

[0182] The following describes the processing flow.

[0183] Step 1:

[0184] The user inputs design proposals for services and campaigns using a device. Simultaneously with the user's input, the device activates an emotion engine, recognizing emotions from the user's facial expressions, voice, and text, and collecting emotion data.

[0185] Step 2:

[0186] The terminal sends the input design proposal and collected emotional data to the server. The server sends the received data to a generative artificial intelligence system and requests an evaluation based on relevant laws and guidelines.

[0187] Step 3:

[0188] The generative artificial intelligence analyzes design proposals and emotional data to perform legal evaluations. Simultaneously, it evaluates the user's emotional state, taking emotional data into consideration, and returns the results to the server.

[0189] Step 4:

[0190] The server sends the legal and emotional evaluation results received from the generative artificial intelligence to the terminal. The terminal then displays the results to the user.

[0191] Step 5:

[0192] The device adjusts the design proposal based on emotional data as needed. If the user's emotional state is negative, the device fine-tunes the design proposal based on suggestions from generative artificial intelligence and presents it to the user again.

[0193] Step 6:

[0194] The server generates multiple variations based on the refined design proposal. These generated variations contain different settings and conditions, and the server sends them to the terminal.

[0195] Step 7:

[0196] The device sets up test scenarios based on the generated variations and conducts A / B testing. Different variations are presented to multiple user groups, and performance and sentiment data for each variation are collected.

[0197] Step 8:

[0198] After the A / B test is complete, the device sends the collected performance and sentiment data to the server. The server aggregates the received data, performs statistical analysis, and selects the optimal variation.

[0199] Step 9:

[0200] The server selects the variation that elicited the most effective and emotionally positive response based on the aggregated data. Once the optimal variation is determined, it sends the result to the terminal.

[0201] Step 10:

[0202] The device notifies the user of the optimal service design. Based on the notified optimal design, the user releases the final service or campaign.

[0203] The above outlines the specific processing steps of this system, which incorporates an emotion engine. This enables the efficient design of optimal services and campaigns that comply with laws and guidelines and reflect user emotions.

[0204] (Example 2)

[0205] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0206] Traditional campaign and service design lacks the technology to incorporate user emotional information and select the optimal variations. Therefore, designing without considering user emotional satisfaction has failed to adequately improve actual customer satisfaction. Furthermore, the need to separately evaluate laws and guidelines complicates the design process. This invention aims to solve these problems.

[0207] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0208] In this invention, the server includes means for receiving service and campaign design proposals based on user input, means for transmitting the design proposals to a generative artificial intelligence and conducting evaluations based on relevant laws and guidelines, means for receiving evaluation results from the generative artificial intelligence and displaying them to the user, means for generating multiple variations based on the design proposals, means for conducting multifaceted A / B testing using the variations, means for aggregating the results of the A / B testing and selecting the optimal variation, means for aggregating user emotional information and reflecting it in the selection of the optimal variation, and means for recognizing user emotions using an emotion engine and providing feedback to the design proposals. This makes it possible to design optimal services and campaigns that reflect user emotional information, improve user satisfaction, and realize an efficient design process that also ensures legal compliance.

[0209] "User input" refers to the act of a user using a device to input design proposals for a service or campaign.

[0210] "Generative artificial intelligence" refers to artificial intelligence models that analyze design proposals and evaluate them based on relevant laws and guidelines.

[0211] A "variation" refers to multiple proposals with different settings and conditions based on a design proposal for a service or campaign.

[0212] A / B testing is an evaluation method that involves presenting multiple variations to user groups and collecting performance data.

[0213] "Emotional information" refers to data about a user's emotional state collected through their facial expressions, voice, text, etc.

[0214] An "emotion engine" is a system that recognizes a user's emotions and analyzes that emotional data.

[0215] "Legal evaluation" is the process by which generative artificial intelligence determines the suitability of a design proposal based on relevant laws and guidelines.

[0216] "Performance data" refers to experimental results such as click-through rates, registration rates, and purchase rates collected during A / B testing.

[0217] A "design proposal" refers to the specific suggestions a user has entered regarding a service or campaign.

[0218] The "optimal variation" is the variation that, based on aggregated performance and sentiment data, is judged to be the most effective and provide the highest user satisfaction.

[0219] This invention provides a technology that combines an emotion engine with a system that supports the design of services and campaigns, enabling the design of optimal services and campaigns based on user emotion information. The operation of the system in each processing step is described in detail below.

[0220] First, the user inputs their service or campaign design proposal using a device. This device utilizes a built-in emotion engine to recognize emotions from the user's facial expressions, voice, and text, and collects that emotion data. The emotion engine can utilize services such as "Emotion API" or "Azure Cognitive Services."

[0221] Next, the terminal sends the design proposals and sentiment data collected from the user to the server. The server sends this data to a generative artificial intelligence (e.g., GPT-4) which then performs an evaluation based on relevant laws and guidelines. The generative AI analyzes the design proposals and conducts a legal evaluation that takes the sentiment data into account.

[0222] Specifically, one could send the following prompt message to the generative artificial intelligence.

[0223] "Please evaluate the following campaign design proposal based on laws and guidelines, taking into account user sentiment data. Proposal: New points reward campaign. User sentiment data: 80% positive responses, 20% negative responses."

[0224] The evaluation results from the generative artificial intelligence are returned to the server, which then sends the results to the terminal. The terminal displays the evaluation results to the user and, at the same time, adjusts the design proposal as needed based on the sentiment data, providing feedback in a way that is appropriate for the user. For example, if a campaign proposed by the user does not align with the user's preferences or expectations, the sentiment engine will use that information to suggest adjustments.

[0225] Next, the server generates multiple variations based on the refined design proposal. The generated variations include different settings and conditions, and are intended for multifaceted evaluation. For example, variations such as "15% cashback," "20% cashback," and "with free coupon" are generated.

[0226] Subsequently, the device conducts A / B testing using the generated variations. The variations are presented to multiple user groups, and performance data (click-through rate, registration rate, purchase rate, etc.) for each variation is collected. The sentiment engine is also used during the test to evaluate how users feel about each variation.

[0227] Once the test is complete, the device sends the collected performance and sentiment data to the server. The server statistically analyzes this data and selects the optimal variation. Sentiment data is an important indicator for evaluating user satisfaction and motivation.

[0228] Once the optimal variation is selected, the server sends the result to the terminal. The terminal notifies the user of the optimal service design and releases the final service or campaign based on it. For example, the server selects "20% cashback" as the optimal variation and notifies the user of the result.

[0229] In this way, the system based on the present invention can efficiently design optimal services and campaigns by reflecting user emotional information and combining it with legal evaluations. This makes it possible to improve user satisfaction and reduce the risk of complaints.

[0230] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0231] Step 1:

[0232] User input

[0233] The user uses the device to input design proposals for services and campaigns. Specifically, the user designs a "new points reward campaign" and inputs it into the device. Based on this input, the device uses its built-in emotion engine to recognize the user's emotions from their facial expressions, voice, and text, and collects that emotion data. The input includes the text data of the design proposal and the user's emotion data (e.g., 80% positive reactions, 20% negative reactions). The device collects this data and sends it to the server.

[0234] Step 2:

[0235] Checking laws and guidelines

[0236] The server sends the design proposal and sentiment data received from the terminal to the generative artificial intelligence (AI). The input for this process consists of the text data of the design proposal and the sentiment data. The generative AI analyzes the design proposal and performs an evaluation based on relevant laws and guidelines. Specifically, an example of sending a prompt to the generative AI is as follows:

[0237] "Please evaluate the following campaign design proposal based on laws and guidelines, taking into account user sentiment data. Proposal: New points reward campaign. User sentiment data: 80% positive responses, 20% negative responses."

[0238] The output is a legal evaluation result. The evaluation result is returned to the server, which then sends the result to the terminal. This result may include information such as, "The point redemption rate for this campaign does not violate current laws."

[0239] Step 3:

[0240] Displaying results and reflecting sentiment data

[0241] The device displays the received evaluation results to the user. Inputs include evaluation results obtained from the server and user sentiment data. Based on these inputs, the device adjusts the design proposal as needed and provides appropriate feedback to the user. Specifically, if the proposed campaign does not align with the user's preferences or expectations, the sentiment engine uses this information to suggest adjustments. For example, it might output new variations such as 15% or 20% cashback offers.

[0242] Step 4:

[0243] Generating Variations

[0244] The server generates multiple variations based on the adjusted design proposal. The input is the text data of the adjusted design proposal. Based on this, the server generates variations with different settings and conditions. Specifically, it generates variations such as "15% cashback," "20% cashback," and "with free coupon," and outputs these variation proposals.

[0245] Step 5:

[0246] Conducting A / B testing

[0247] The device conducts A / B testing using the generated variations. The input is design data for the variations. The device presents each variation to multiple user groups and collects performance data such as click-through rates, registration rates, and purchase rates. It also uses an emotion engine to evaluate users' emotional responses. The output includes performance data and emotional data for each variation. For example, user group A is presented with "15% cashback" and user group B with "20% cashback," and the response and performance data for each group are collected.

[0248] Step 6:

[0249] Summary of results and selection of the optimal variation

[0250] After the test is complete, the device sends the collected performance and sentiment data to the server. The inputs are the A / B test results and sentiment data. The server statistically analyzes this data and selects the optimal variation. The output is the data for the selected optimal variation (e.g., "20% cashback" is optimal).

[0251] Step 7:

[0252] Notification of optimal variation

[0253] The server sends the result of the selected optimal variation to the terminal. The input includes the data of the optimal variation selected by the server. The terminal notifies the user of this, and the user uses this data to release the final service or campaign. For example, if the server selects "20% cashback" as the optimal variation and notifies the user of this result, the user can implement a "20% cashback campaign."

[0254] (Application Example 2)

[0255] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0256] Traditional service and campaign design support systems have the drawback of failing to take user emotions into account, thus hindering their ability to maximize user satisfaction. Furthermore, the difficulty in obtaining real-time feedback makes it challenging to quickly deliver campaigns tailored to user needs. Moreover, the lack of a means to adjust design proposals based on emotional data prevents them from addressing the diverse emotions of users.

[0257] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0258] In this invention, the server includes means for receiving service and campaign design proposals based on user input; means for transmitting the design proposals to a generative artificial intelligence and conducting evaluations based on relevant laws and guidelines; means for receiving evaluation results from the generative artificial intelligence and displaying them to the user; means for collecting sentiment data based on the evaluation results; means for generating multiple variations based on the design proposals; means for conducting multifaceted A / B testing using the variations; means for aggregating the results of the A / B testing and selecting the optimal variation; means for notifying the user of the optimal variation; means for adjusting the design proposals in real time based on the user's sentiment data and providing feedback; and means for using smart glasses, a head-mounted display, or a similar device to collect the user's sentiment data. This makes it possible to reflect the user's emotions in real time and quickly provide more satisfying services and campaigns.

[0259] "User input" refers to the information that users provide to the system regarding their service or campaign design proposals.

[0260] "Service and campaign design proposals" refer to the marketing and promotional plans and their specific details that users envision.

[0261] "Generative artificial intelligence" refers to algorithms that use machine learning and natural language processing to analyze user design proposals and provide evaluations and suggestions based on relevant laws and guidelines.

[0262] "Relevant laws and guidelines" refer to the laws and regulations and guidelines that must be followed when conducting the service or campaign.

[0263] "Evaluation results" refer to feedback from a generative artificial intelligence system that evaluates the design proposal and identifies any legal issues or areas for improvement.

[0264] "Emotional data" refers to information about a user's emotions collected from their facial expressions, voice, text, and other sources.

[0265] "Variations" refer to different versions of a campaign or promotion that are generated based on the user's design proposal.

[0266] A / B testing is a method of presenting multiple variations to different user groups and comparing their performance data.

[0267] "Performance data" refers to metrics used to measure the effectiveness of a campaign or service (e.g., click-through rate, registration rate, purchase rate, etc.).

[0268] The "optimal variation" refers to the campaign or promotion that, based on A / B test results and sentiment data, is judged to be the most effective and provides the highest user satisfaction.

[0269] "Smart glasses" are wearable devices designed to collect the user's emotions in real time.

[0270] A "head-mounted display" is a wearable device that collects emotional data while displaying information in the user's field of vision.

[0271] "Adjusting and providing feedback in real time" is a process of immediately reflecting user sentiment data to revise the design proposal and returning the results to the user.

[0272] This invention combines an emotion engine and generative artificial intelligence in a system that assists in the design of services and campaigns, providing a technology that designs optimal services and campaigns based on user emotional information.

[0273] System Configuration

[0274] Hardware and software:

[0275] 1. User terminal: A device including smart glasses or a head-mounted display. This device collects the user's facial expressions, voice, and text in real time and generates emotion data through an emotion engine (e.g., Microsoft® Azure Emotion API).

[0276] 2. Server: A cloud platform (e.g., AWS®) will be used to run generative artificial intelligence (e.g., OpenAI GPT). The server will process design proposals and sentiment data received from users and perform legal evaluations and design proposal optimization.

[0277] Software processing and data computation:

[0278] 1. Receiving user input:

[0279] Users interact with touchpoints within the virtual store and input design proposals for services and campaigns.

[0280] 2. Collection and transmission of emotional data:

[0281] Capture the user's expressions, voice, and text through smart glasses or head-mounted displays, and analyze them as emotional data using an emotion engine.

[0282] Send the generated emotional data together with the design proposal to a cloud server.

[0283] 3. Evaluation and adjustment of the design proposal:

[0284] The generative artificial intelligence in the server analyzes the design proposal and conducts a legal evaluation based on relevant laws and guidelines.

[0285] Judge the user's satisfaction based on the emotional data and generate a design proposal with necessary adjustments.

[0286] 4. Display and feedback of the evaluation results:

[0287] Send the evaluation results from the server to the user terminal and display them in real time on a virtual display.

[0288] Feedback an optimized campaign plan based on the user's emotional data.

[0289] 5. Variation generation and A / B testing:

[0290] Generate multiple variations based on the adjusted design proposal and present them to different user groups.

[0291] Conduct an A / B test in real time and collect performance data and emotional data.

[0292] 6. Selection and notification of the optimal variation:

[0293] Perform statistical analysis on the collected data on the server and select the optimal variation.

[0294] The optimal variation is notified to the user's terminal and displayed on the virtual display.

[0295] Specific example

[0296] For example, a user designs a "Spring Bonus Points Campaign" and inputs their design proposal. At this time, an emotion engine is used to collect emotional data from the user's facial expressions and voice. A generative AI model is used to analyze "how to make the Spring Bonus Points Campaign legally compliant" and perform a legal evaluation. Based on this result, the optimal proposal (e.g., a 10% point increase) that reflects the user's emotional data is displayed in real time. Subsequently, A / B testing is conducted on the different variations (e.g., a 15% point increase, a 20% point increase, and one with a free coupon), and the "20% increase" with the highest performance is ultimately selected.

[0297] Example of a prompt:

[0298] "How can we make our spring bonus points campaign legally sound and offer exciting deals for users?"

[0299] Thus, the present invention enables the efficient design of optimal services and campaigns by combining evaluations of laws and guidelines with user sentiment information.

[0300] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0301] Step 1:

[0302] Receiving user input:

[0303] Users interact with interactive touchpoints within a virtual store to input service and campaign design proposals. This input data includes the campaign name, content, and terms and conditions. The terminal receives and stores this input data.

[0304] Step 2:

[0305] Collection of emotional data:

[0306] While the user is inputting the design plan, smart glasses or head-mounted displays capture the user's expressions and voices in real time. This data is sent to an emotion engine (e.g., Microsoft Azure Emotion API) for analysis and collection as emotional data. The input data is the user's expressions and voices, and the output data is the user's emotional state (e.g., joy, surprise, anxiety).

[0307] Step 3:

[0308] Transmission of the design plan and emotional data:

[0309] The terminal sends the design plan input by the user and the collected emotional data to the cloud server. The input data is the design plan and emotional data, and the output data is a request message containing these.

[0310] Step 4:

[0311] Legal evaluation of the design plan:

[0312] The server uses generative artificial intelligence (e.g., OpenAI GPT) to evaluate the design plan. For example, the prompt sentence "Does this design plan comply with laws and guidelines?" is input, and the AI model performs analysis. The output is the legal evaluation result, indicating whether there are problems.

[0313] Step 5:

[0314] Return and display of the evaluation result:

[0315] The server returns the legal evaluation result to the terminal. The terminal displays the evaluation result on the user's virtual display in real time. The input data is the legal evaluation result, and the output data is the information displayed to the user.

[0316] Step 6:

[0317] Generating adjustment proposals based on emotional data:

[0318] The server adjusts the design proposal based on emotional data. For example, it might input a prompt message to a generative artificial intelligence such as, "According to the user's emotional data, there are many concerns about the current design proposal, so I will suggest improvements." The output is an adjusted proposal, a specific suggestion that takes the user's emotions into consideration.

[0319] Step 7:

[0320] Generating multiple variations:

[0321] Based on the adjusted design proposal, the server generates multiple variations. These variations include different percentages of discounts and benefits. The input data is the adjusted proposal, and the output data is a set of variations.

[0322] Step 8:

[0323] Conducting A / B testing:

[0324] The device conducts A / B testing on different user groups using the generated variations. It collects performance data (e.g., click-through rate, purchase rate) and sentiment data for each variation. The input data consists of variation-to-user interaction data, and the output data consists of customized evaluation data.

[0325] Step 9:

[0326] Aggregation and optimization of test results:

[0327] The server statistically analyzes the collected performance and sentiment data to select the optimal variation. For example, it might input a prompt message to a generative artificial intelligence system such as, "Select the variation that shows the highest click-through rate and user satisfaction." The output data would then be the optimal variation.

[0328] Step 10:

[0329] Notification of optimal variation:

[0330] The server sends the optimal variation to the user's terminal and displays it on the virtual display. The input data is the optimal variation, and the output data is the optimal campaign proposal that is notified to the user.

[0331] This series of steps creates a system that reflects user emotions and delivers optimal services and campaigns in real time.

[0332] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0333] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0334] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0335] [Second Embodiment]

[0336] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0337] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0338] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0339] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0340] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0341] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0342] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0343] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0344] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0345] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0346] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0347] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. 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".

[0348] This invention provides a system in which users input service and campaign design proposals, and the system designs the optimal service through verification of laws and guidelines using generative artificial intelligence and multifaceted A / B testing. The operation of the system in each processing step is described in detail below.

[0349] User input

[0350] Users input service and campaign design proposals using their devices. The input design proposals are then sent from the device to the server.

[0351] Checking laws and guidelines

[0352] The server sends the design proposal received from the user to a generative artificial intelligence (AI) system, which then performs an evaluation based on relevant laws and guidelines. The AI ​​analyzes the design proposal and determines whether there are any legal issues. The evaluation results are returned to the server, which then sends them to the terminal. The terminal then displays the evaluation results to the user.

[0353] Generating Variations

[0354] If the evaluation result is determined to be "no problems," the server generates multiple variations based on the design proposal. These variations refer to multiple proposals containing different settings and conditions, and the server sends these to the terminal.

[0355] Conducting A / B testing

[0356] The device sets up test scenarios based on the received variations and conducts A / B testing from multiple angles. Different variations are presented to multiple user groups, and test results are collected. For example, variations with 15% cashback, 20% cashback, and a free coupon are set up, and performance data (click-through rate, registration rate, purchase rate, etc.) is collected for each.

[0357] Summary of results and selection of the optimal variation

[0358] After the test is complete, the terminal sends the obtained performance data to the server. The server aggregates this data, analyzes it statistically, and selects the optimal variation.

[0359] Notification of optimal variation

[0360] Once the optimal variation is selected, the server sends the result to the terminal. The terminal then notifies the user of the optimal service design and releases the final service based on it.

[0361] Specific examples of operation

[0362] The following is a concrete example of how it works. The user designs a "new points reward campaign" and inputs it into the terminal. This design proposal is sent to the server, where it is legally checked by generative artificial intelligence. It is confirmed that it does not violate consumer protection laws or the Premiums and Representations Act, and the results are notified to the user. The server then generates variations such as 15% reward, 20% reward, and with a free coupon, and A / B testing is conducted on the terminal. The test results are sent to the server, which selects the "20% reward" option that is most popular with the user, and notifies the user of the result.

[0363] As described above, the system based on the present invention can efficiently perform verification in accordance with laws and guidelines, and design optimal services that reflect user preferences through A / B testing.

[0364] The following describes the processing flow.

[0365] Step 1:

[0366] The user inputs a design proposal for a service or campaign into the terminal. The terminal then sends the input design proposal to the server.

[0367] Step 2:

[0368] The server sends the received design proposal to the generative artificial intelligence. The generative AI analyzes the design proposal and conducts a legal evaluation based on relevant laws and guidelines.

[0369] Step 3:

[0370] The evaluation results from the generative artificial intelligence are returned to the server. The server receives the evaluation results and sends them to the terminal.

[0371] Step 4:

[0372] The device displays the evaluation results to the user. The user reviews the evaluation results and understands that there are no legal issues.

[0373] Step 5:

[0374] The server generates multiple variations based on the design proposal that was deemed "problem-free" in the evaluation results. These generated variations include different settings and conditions.

[0375] Step 6:

[0376] The server sends the generated variations to the device. The device then prepares to use the variations to conduct an A / B test.

[0377] Step 7:

[0378] The device presents variations to multiple user groups. For example, it might offer variations such as 15% cashback, 20% cashback, or a free coupon, and collects performance data (click-through rate, registration rate, purchase rate, etc.) for each variation.

[0379] Step 8:

[0380] Once the A / B test is complete, the device sends the collected performance data to the server. The server then aggregates and statistically analyzes the received data.

[0381] Step 9:

[0382] The server selects the most effective variation based on the aggregated data. Once the optimal variation is determined, the server sends the result to the terminal.

[0383] Step 10:

[0384] The device notifies the user of the optimal service design. Based on the notified optimal design, the user releases the final service or campaign.

[0385] The above describes the specific actions of each processing step in the program. This system makes it possible to comply with laws and guidelines and efficiently provide the best possible service to users.

[0386] (Example 1)

[0387] Next, we will describe Example 1. 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."

[0388] In recent years, when designing services and campaigns that appeal to consumers, users are required to experiment with multiple design options while avoiding legal risks. However, checking legal issues and rationally testing various variations to select the optimal design is a multifaceted and complex process that requires considerable effort and time. This has created a challenge in that it is difficult for users to quickly and efficiently find the optimal design.

[0389] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0390] In this invention, the server includes means for receiving design proposals for services and campaigns based on user input; means for transmitting the design proposals to a generative artificial intelligence and conducting evaluations based on relevant laws and guidelines; means for receiving evaluation results from the generative artificial intelligence and displaying them to the user; means for generating multiple variations based on the design proposals; means for conducting comparative tests from multiple perspectives using the variations; means for aggregating the results of the comparative tests and selecting the optimal variation; and means for notifying the user of the optimal variation. This makes it possible to efficiently design optimal services and campaigns while avoiding legal risks.

[0391] A "user" refers to an individual or legal entity that inputs design proposals for services or campaigns and utilizes the system.

[0392] "Terminal" refers to an input device used by a user, and includes electronic devices such as personal computers, smartphones, and tablets.

[0393] A "server" is a computer system that performs the central processing of a system and provides multiple functions in an integrated manner.

[0394] A "design proposal" refers to a plan that describes the specific content and conditions of a service or campaign that users will input.

[0395] "Generative artificial intelligence" refers to an artificial intelligence system that performs advanced analytical processing, such as legal evaluation and variation generation, from given input data.

[0396] "Laws and guidelines" refer to the laws and industry standards that apply to the service or campaign, including consumer protection laws and the Premiums and Representations Act.

[0397] "Evaluation results" refer to the results obtained when a generative artificial intelligence analyzes a design proposal and determines whether there are any legal issues or whether it conforms to guidelines.

[0398] "Variations" refer to multiple design options with different conditions and settings, generated based on the original design proposal.

[0399] A "comparative test" refers to a testing method in which multiple variations are presented to different user groups, and their response data is collected and compared.

[0400] "Performance data" refers to data showing user responses to each variation, including click-through rates, registration rates, and purchase rates.

[0401] The "optimal variation" refers to the variation that showed the highest performance data in comparative tests.

[0402] "Notifying" refers to the process of transmitting information from a server to a user via a terminal.

[0403] This invention relates to a system in which a user inputs a design proposal for a service or campaign, and the system performs optimal service design through legal evaluation and multifaceted comparative testing using generative artificial intelligence. Specific embodiments for carrying out this invention are described in detail below.

[0404] Users input service and campaign design proposals using their devices. For example, for a "new points reward campaign," they would describe the campaign details (campaign period, reward rate, etc.) in an input form on their device. This design proposal is then sent from the device to the server.

[0405] The server sends the received design proposal to a generative artificial intelligence (e.g., OpenAI GPT-4). The generative AI analyzes the design proposal, which is input as a prompt: "Please check whether this design proposal violates consumer protection laws or the Premiums and Representations Act." Based on the analysis, it determines whether there are any legal issues and returns the evaluation result to the server. The evaluation result is resent from the server to the terminal and displayed to the user on the terminal. For example, the evaluation result displayed might be: "This campaign does not violate consumer protection laws."

[0406] If the evaluation result is determined to be "no problems," the server generates multiple variations based on the design proposal. Generative artificial intelligence and algorithms are used to create variations with different settings and conditions. For example, it might generate three variations: "15% cashback," "20% cashback," and "with a free coupon." The generated variations are then sent from the server to the terminal.

[0407] The device sets up a comparative test scenario based on the received variations. This scenario presents different variations to multiple user groups and collects user response data (click-through rate, registration rate, purchase rate, etc.) for each variation. For example, user group A is presented with "15% cashback," user group B with "20% cashback," and user group C with "free coupon."

[0408] Once the results of each comparison test are collected, the device sends this performance data to the server. The server aggregates the data and performs statistical analysis to select the most effective variation. For example, by comparing the click-through rate and purchase rate of each variation, the "20% cashback" variation may be found to have the best performance. In this process, the R or Python pandas library may be used.

[0409] Once the optimal variation is selected, the server notifies the terminal of the result. The terminal then notifies the user of the optimal variation, and the user ultimately releases the service based on that variation. For example, the user might see a notification that says, "Test results indicate that a 20% cashback has been selected as the most effective variation."

[0410] In this way, the system based on the present invention can generate design proposals based on user input and efficiently design optimal services and campaigns through multifaceted comparative testing while avoiding legal risks.

[0411] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0412] Step 1:

[0413] Users input service and campaign design proposals using their devices. For example, they might describe a "new points reward campaign," detailing specific conditions such as the campaign period and reward rate. This design proposal is sent from the device to the server as input data. Once the server receives the design proposal as input, processing begins on the server side.

[0414] Step 2:

[0415] The server sends the received design proposal to a generative artificial intelligence (e.g., OpenAI GPT-4). Specifically, the design proposal is used as input data with the prompt message "Please check whether this design proposal violates consumer protection laws or the Premiums and Representations Act." The generative AI model analyzes this input data and determines whether there are any legal issues. The evaluation result is sent back to the server as output, and this result serves as information to confirm whether there are any legal problems.

[0416] Step 3:

[0417] The server receives the evaluation results from the generative artificial intelligence and sends them to the user's terminal for them to review. The terminal displays the legal evaluation results to the user, for example, a message such as "This campaign does not violate consumer protection laws." Information to be notified to the user is generated based on the evaluation results as input.

[0418] Step 4:

[0419] If the evaluation result is determined to be "no problems," the server generates multiple variations based on the design proposal. Generative artificial intelligence and algorithms are used to create multiple variations with different settings and conditions. For example, variations such as "15% cashback," "20% cashback," and "with a free coupon" may be included. The generated variations are sent from the server to the terminal as output data, preparing it for the next test.

[0420] Step 5:

[0421] The device sets up a comparative test scenario based on the received variations. Specifically, it creates test cases that present different variations to multiple user groups. User group A is presented with "15% cashback," user group B with "20% cashback," and user group C with "free coupon." A multifaceted comparative test is performed based on the variations as input. The output is the collection of response data from each group.

[0422] Step 6:

[0423] Once the results of the comparative tests are collected, the device sends this performance data (e.g., click-through rate, registration rate, purchase rate) to the server. The server aggregates this data and performs statistical analysis. Based on the performance data as input, it selects the most effective variation. For example, by comparing the click-through rate and purchase rate of each variation, the results might show that "20% cashback" performs the best. As output, the optimal variation is selected.

[0424] Step 7:

[0425] Once the optimal variation is selected, the server notifies the terminal of the result. The terminal then notifies the user of the optimal variation, and the user ultimately releases the service based on that variation. For example, a notification might appear stating, "Based on the test results, a 20% cashback has been selected as the most effective variation." This provides the user with information to take their next action based on the test results as input.

[0426] (Application Example 1)

[0427] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0428] Traditional advertising campaign design often involved manual processes for checking legal issues, generating multiple variations, testing, and selecting the optimal variation, resulting in inefficiencies and time-consuming tasks. Furthermore, there was a lack of systems providing user-friendly interfaces using smart devices. This made it difficult to quickly and effectively maximize marketing effectiveness.

[0429] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0430] In this invention, the server includes means for receiving service and campaign design proposals based on user input; means for transmitting the design proposals to a generative artificial intelligence and conducting evaluations based on relevant regulations and guidelines; means for receiving evaluation results from the generative artificial intelligence and displaying them to the user; means for generating multiple options based on the design proposals; means for conducting multifaceted A / B testing using the options; means for aggregating the results of the A / B testing and selecting the optimal option; means for notifying the user of the optimal option; and means including an application installed on a smart device. This enables efficient advertising campaign design, legal review, variation generation, testing, and optimization, and maximizes marketing effectiveness with a user-friendly interface.

[0431] A "user" refers to a person who uses the system to design services and campaigns.

[0432] "Input" refers to the act of a user providing design proposals and related information to the system.

[0433] A "design proposal" refers to information that shows the specific details of a service or campaign proposed by a user.

[0434] "Generative artificial intelligence" refers to artificial intelligence technology that uses user-inputted design proposals to check relevant regulations and guidelines, and conducts multifaceted A / B testing.

[0435] "Evaluation" refers to the process by which a generative artificial intelligence determines whether a design proposal has legal issues or complies with guidelines.

[0436] "Legal issues" refer to the question of whether the proposed design violates legal regulations.

[0437] "Regulations" refer to laws and guidelines that apply to the design of a service or campaign.

[0438] "Options" refers to multiple variations or different proposals with different conditions that are generated based on a design proposal.

[0439] "AB testing" refers to a testing method that presents multiple options to different user groups and compares and evaluates the performance of each option.

[0440] "Aggregation" refers to the act of statistically compiling the results data of an A / B test.

[0441] The "optimal choice" refers to the variation that shows the highest performance based on the results of the A / B test.

[0442] "Smart devices" refer to advanced devices such as smartphones, smart glasses, and head-mounted displays.

[0443] This invention involves several key steps to realize a system in which users input service and campaign design proposals using a smart device. These steps are described below.

[0444] First, the user uses a smart device such as a smartphone to input their service or campaign design proposal. The user enters the details of the specific design proposal and sends it from the device to the server.

[0445] Next, the server sends the received design proposal to the generative artificial intelligence (AI) and performs an evaluation based on relevant regulations and guidelines. The generative AI analyzes the design proposal and determines whether there are any legal issues. This legal evaluation includes consumer protection laws and the Premiums and Representations Act, among others. The evaluation results of the generative AI are returned to the server, which then sends the results to the user's terminal and displays them to the user.

[0446] Next, if the server determines the evaluation result is "no problems," it generates multiple options (variations) based on the design proposal. These variations have different settings and conditions, and may include options such as "15% cashback," "20% cashback," and "with a free coupon." These variations are then sent from the server to the terminal.

[0447] The device conducts multifaceted A / B testing based on the variations it receives. This test presents different variations to multiple user groups and collects performance data (click-through rates, registration rates, purchase rates, etc.) resulting from these tests. This data is then sent to the server.

[0448] The server aggregates the received performance data, analyzes it statistically, and selects the optimal option. The selected optimal option is then notified from the server to the terminal, informing the user of the best possible service design.

[0449] As a concrete example, a user designs a "new points reward campaign" and inputs a design proposal for a "20% cashback campaign." Generative artificial intelligence legally evaluates this proposal and confirms that it does not violate consumer protection laws or the Premiums and Representations Act. The server then generates variations such as "15% cashback," "20% cashback," and "with free coupon," and A / B testing is conducted on the terminal. The test results are sent to the server, which selects the "20% cashback" option, which showed the highest performance, as the optimal choice and notifies the user of the result.

[0450] Examples of prompt messages are as follows:

[0451] We are designing a new points reward campaign. Please perform legal checks and generate variations based on the details below.

[0452] Campaign type: Point rewards

[0453] Details: 20% cashback campaign

[0454] Thus, the system based on this invention maximizes marketing effectiveness by quickly and efficiently evaluating user-inputted design proposals and providing the optimal service design.

[0455] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0456] Step 1:

[0457] Users input service and campaign design proposals using a smart device. Specifically, users open a dedicated application on their smartphone, input their design proposal (e.g., "20% Cashback Campaign"), and press the submit button. Input data: Campaign details. Output data: Confirmation message for submitting the design proposal.

[0458] Step 2:

[0459] The server receives design proposals submitted by the user. The server then forwards the input to a generative artificial intelligence system, which conducts a legal evaluation based on relevant regulations and guidelines. This legal evaluation includes checks regarding consumer protection laws and the Premiums and Representations Act. Input data: Design proposals. Data processing: Evaluation based on legal regulations. Output data: Legal evaluation results.

[0460] Step 3:

[0461] Generative artificial intelligence analyzes received design proposals and determines whether they have any legal issues. Generative AI uses natural language processing and rule-based analysis to check which regulations or guidelines the design proposals violate. Input data: Design proposal. Data calculation: Comparison with legal regulations. Output data: Result of legal evaluation (e.g., "No issues").

[0462] Step 4:

[0463] The server receives evaluation results from the generative artificial intelligence, sends them to the user's terminal, and displays them to the user. The displayed content is the result of the legal evaluation and includes instructions for when there are no problems or when corrections are needed. Input data: Legal evaluation results. Output data: Content displayed to the user.

[0464] Step 5:

[0465] If the evaluation result is determined to be "no problems," the server generates multiple variations based on the design proposal. These variations include different reward rates and benefits. The generated variations are sent to the terminal. Input data: Evaluated design proposal. Data processing: Generation of variations. Output data: Multiple variations.

[0466] Step 6:

[0467] The device conducts multifaceted A / B testing based on the received variations. Specifically, it presents different variations to user groups and collects performance data such as click-through rates and purchase rates. Input data: Multiple variations. Data processing: Collection of user behavior. Output data: Performance data (e.g., click-through rate, purchase rate).

[0468] Step 7:

[0469] The terminal sends the collected performance data to the server. The server aggregates the received performance data, performs statistical analysis, and selects the optimal variation. Input data: Performance data. Data calculation: Statistical analysis. Output data: Result of the selection of the optimal variation.

[0470] Step 8:

[0471] The server selects the optimal variation and notifies the user of the result on their device. The user receives the notification and confirms the optimal campaign design. Input data: Selection result. Output data: Notification content to the user (e.g., "20% cashback is optimal").

[0472] By following these steps, users can efficiently design legally compliant and optimal advertising campaigns.

[0473] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0474] This invention provides a technology that combines an emotion engine with a system that supports the design of services and campaigns, enabling the design of optimal services and campaigns based on user emotion information. The operation of the system in each processing step is described in detail below.

[0475] User input

[0476] Users input service and campaign design proposals using a device. During this process, the device utilizes its built-in emotion engine to recognize emotions from the user's facial expressions, voice, and text, and collects that emotion data.

[0477] Checking laws and guidelines

[0478] The server transmits the design proposal and emotional data received from the user to a generative artificial intelligence (AI) system, which then performs an evaluation based on relevant laws and guidelines. The AI ​​analyzes the design proposal and conducts a legal evaluation, taking the emotional data into consideration. The evaluation results are returned to the server, which then transmits them to the terminal.

[0479] Displaying results and reflecting sentiment data

[0480] The device displays the evaluation results to the user. At the same time, it adjusts the design proposal as needed based on sentiment data and provides feedback in a way that is appropriate for the user. For example, if a campaign suggested by the user does not align with the user's preferences or expectations, the sentiment engine will use that information to suggest adjustments.

[0481] Generating Variations

[0482] The server generates multiple variations based on the refined design proposal. These variations include different settings and conditions, and are intended for multifaceted evaluation.

[0483] Conducting A / B testing

[0484] The device will conduct A / B testing using the generated variations. Multiple user groups will be presented with the variations, and performance data for each variation (click-through rate, registration rate, purchase rate, etc.) will be collected. During the test, an emotion engine will be used to evaluate how users feel about the variations.

[0485] Summary of results and selection of the optimal variation

[0486] After the test is complete, the device sends the collected performance and sentiment data to the server. The server aggregates and statistically analyzes this data to select the optimal variation. Sentiment data is considered an important indicator for evaluating user satisfaction and motivation.

[0487] Notification of optimal variation

[0488] Once the optimal variation is selected, the server sends the result to the terminal. The terminal notifies the user of the optimal service design and releases the final service or campaign based on it. An optimal design that reflects user sentiment information increases user satisfaction and reduces the risk of complaints.

[0489] Specific examples of operation

[0490] The following is a concrete example of how it works. The user designs a "new points reward campaign" and inputs it into the terminal. The emotion engine recognizes the user's emotions and sends it to the server along with the design proposal. A legal check is performed by generative artificial intelligence, and the evaluation results are notified to the user, while adjustments based on the emotion data are suggested. Subsequently, variations such as 15% reward, 20% reward, and with a free coupon are generated, and A / B testing is conducted. Based on the test results and emotion data, the server determines that "20% reward" is optimal, and this result is notified to the user.

[0491] As described above, the system based on the present invention can efficiently design optimal services and campaigns by combining evaluations of laws and guidelines with user sentiment information.

[0492] The following describes the processing flow.

[0493] Step 1:

[0494] The user inputs design proposals for services and campaigns using a device. Simultaneously with the user's input, the device activates an emotion engine, recognizing emotions from the user's facial expressions, voice, and text, and collecting emotion data.

[0495] Step 2:

[0496] The terminal sends the input design proposal and collected emotional data to the server. The server sends the received data to a generative artificial intelligence system and requests an evaluation based on relevant laws and guidelines.

[0497] Step 3:

[0498] The generative artificial intelligence analyzes design proposals and emotional data to perform legal evaluations. Simultaneously, it evaluates the user's emotional state, taking emotional data into consideration, and returns the results to the server.

[0499] Step 4:

[0500] The server sends the legal and emotional evaluation results received from the generative artificial intelligence to the terminal. The terminal then displays the results to the user.

[0501] Step 5:

[0502] The device adjusts the design proposal based on emotional data as needed. If the user's emotional state is negative, the device fine-tunes the design proposal based on suggestions from generative artificial intelligence and presents it to the user again.

[0503] Step 6:

[0504] The server generates multiple variations based on the refined design proposal. These generated variations contain different settings and conditions, and the server sends them to the terminal.

[0505] Step 7:

[0506] The device sets up test scenarios based on the generated variations and conducts A / B testing. Different variations are presented to multiple user groups, and performance and sentiment data for each variation are collected.

[0507] Step 8:

[0508] After the A / B test is complete, the device sends the collected performance and sentiment data to the server. The server aggregates the received data, performs statistical analysis, and selects the optimal variation.

[0509] Step 9:

[0510] The server selects the variation that elicited the most effective and emotionally positive response based on the aggregated data. Once the optimal variation is determined, it sends the result to the terminal.

[0511] Step 10:

[0512] The device notifies the user of the optimal service design. Based on the notified optimal design, the user releases the final service or campaign.

[0513] The above outlines the specific processing steps of this system, which incorporates an emotion engine. This enables the efficient design of optimal services and campaigns that comply with laws and guidelines and reflect user emotions.

[0514] (Example 2)

[0515] Next, we will describe Example 2. 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".

[0516] Traditional campaign and service design lacks the technology to incorporate user emotional information and select the optimal variations. Therefore, designing without considering user emotional satisfaction has failed to adequately improve actual customer satisfaction. Furthermore, the need to separately evaluate laws and guidelines complicates the design process. This invention aims to solve these problems.

[0517] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0518] In this invention, the server includes means for receiving service and campaign design proposals based on user input, means for transmitting the design proposals to a generative artificial intelligence and conducting evaluations based on relevant laws and guidelines, means for receiving evaluation results from the generative artificial intelligence and displaying them to the user, means for generating multiple variations based on the design proposals, means for conducting multifaceted A / B testing using the variations, means for aggregating the results of the A / B testing and selecting the optimal variation, means for aggregating user emotional information and reflecting it in the selection of the optimal variation, and means for recognizing user emotions using an emotion engine and providing feedback to the design proposals. This makes it possible to design optimal services and campaigns that reflect user emotional information, improve user satisfaction, and realize an efficient design process that also ensures legal compliance.

[0519] "User input" refers to the act of a user using a device to input design proposals for a service or campaign.

[0520] "Generative artificial intelligence" refers to artificial intelligence models that analyze design proposals and evaluate them based on relevant laws and guidelines.

[0521] A "variation" refers to multiple proposals with different settings and conditions based on a design proposal for a service or campaign.

[0522] A / B testing is an evaluation method that involves presenting multiple variations to user groups and collecting performance data.

[0523] "Emotional information" refers to data about a user's emotional state collected through their facial expressions, voice, text, etc.

[0524] An "emotion engine" is a system that recognizes a user's emotions and analyzes that emotional data.

[0525] "Legal evaluation" is the process by which generative artificial intelligence determines the suitability of a design proposal based on relevant laws and guidelines.

[0526] "Performance data" refers to experimental results such as click-through rates, registration rates, and purchase rates collected during A / B testing.

[0527] A "design proposal" refers to the specific suggestions a user has entered regarding a service or campaign.

[0528] The "optimal variation" is the variation that, based on aggregated performance and sentiment data, is judged to be the most effective and provide the highest user satisfaction.

[0529] This invention provides a technology that combines an emotion engine with a system that supports the design of services and campaigns, enabling the design of optimal services and campaigns based on user emotion information. The operation of the system in each processing step is described in detail below.

[0530] First, the user inputs their service or campaign design proposal using a device. This device utilizes a built-in emotion engine to recognize emotions from the user's facial expressions, voice, and text, and collects that emotion data. The emotion engine can utilize either the "Emotion API" or "Azure Cognitive Services."

[0531] Next, the terminal sends the design proposals and sentiment data collected from the user to the server. The server sends this data to a generative artificial intelligence (e.g., GPT-4) which then performs an evaluation based on relevant laws and guidelines. The generative AI analyzes the design proposals and conducts a legal evaluation that takes the sentiment data into account.

[0532] Specifically, one could send the following prompt message to the generative artificial intelligence.

[0533] "Please evaluate the following campaign design proposal based on laws and guidelines, taking into account user sentiment data. Proposal: New points reward campaign. User sentiment data: 80% positive responses, 20% negative responses."

[0534] The evaluation results from the generative artificial intelligence are returned to the server, which then sends the results to the terminal. The terminal displays the evaluation results to the user and, at the same time, adjusts the design proposal as needed based on the sentiment data, providing feedback in a way that is appropriate for the user. For example, if a campaign proposed by the user does not align with the user's preferences or expectations, the sentiment engine will use that information to suggest adjustments.

[0535] Next, the server generates multiple variations based on the refined design proposal. The generated variations include different settings and conditions, and are intended for multifaceted evaluation. For example, variations such as "15% cashback," "20% cashback," and "with free coupon" are generated.

[0536] Subsequently, the device conducts A / B testing using the generated variations. The variations are presented to multiple user groups, and performance data (click-through rate, registration rate, purchase rate, etc.) for each variation is collected. The sentiment engine is also used during the test to evaluate how users feel about each variation.

[0537] Once the test is complete, the device sends the collected performance and sentiment data to the server. The server statistically analyzes this data and selects the optimal variation. Sentiment data is an important indicator for evaluating user satisfaction and motivation.

[0538] Once the optimal variation is selected, the server sends the result to the terminal. The terminal notifies the user of the optimal service design and releases the final service or campaign based on it. For example, the server selects "20% cashback" as the optimal variation and notifies the user of the result.

[0539] In this way, the system based on the present invention can efficiently design optimal services and campaigns by reflecting user emotional information and combining it with legal evaluations. This makes it possible to improve user satisfaction and reduce the risk of complaints.

[0540] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0541] Step 1:

[0542] User input

[0543] The user uses the device to input design proposals for services and campaigns. Specifically, the user designs a "new points reward campaign" and inputs it into the device. Based on this input, the device uses its built-in emotion engine to recognize the user's emotions from their facial expressions, voice, and text, and collects that emotion data. The input includes the text data of the design proposal and the user's emotion data (e.g., 80% positive reactions, 20% negative reactions). The device collects this data and sends it to the server.

[0544] Step 2:

[0545] Checking laws and guidelines

[0546] The server sends the design proposal and sentiment data received from the terminal to the generative artificial intelligence (AI). The input for this process consists of the text data of the design proposal and the sentiment data. The generative AI analyzes the design proposal and performs an evaluation based on relevant laws and guidelines. Specifically, an example of sending a prompt to the generative AI is as follows:

[0547] "Please evaluate the following campaign design proposal based on laws and guidelines, taking into account user sentiment data. Proposal: New points reward campaign. User sentiment data: 80% positive responses, 20% negative responses."

[0548] The output is a legal evaluation result. The evaluation result is returned to the server, which then sends the result to the terminal. This result may include information such as, "The point redemption rate for this campaign does not violate current laws."

[0549] Step 3:

[0550] Displaying results and reflecting sentiment data

[0551] The device displays the received evaluation results to the user. Inputs include evaluation results obtained from the server and user sentiment data. Based on these inputs, the device adjusts the design proposal as needed and provides appropriate feedback to the user. Specifically, if the proposed campaign does not align with the user's preferences or expectations, the sentiment engine uses this information to suggest adjustments. For example, it might output new variations such as 15% or 20% cashback offers.

[0552] Step 4:

[0553] Generating Variations

[0554] The server generates multiple variations based on the adjusted design proposal. The input is the text data of the adjusted design proposal. Based on this, the server generates variations with different settings and conditions. Specifically, it generates variations such as "15% cashback," "20% cashback," and "with free coupon," and outputs these variation proposals.

[0555] Step 5:

[0556] Conducting A / B testing

[0557] The device conducts A / B testing using the generated variations. The input is design data for the variations. The device presents each variation to multiple user groups and collects performance data such as click-through rates, registration rates, and purchase rates. It also uses an emotion engine to evaluate users' emotional responses. The output includes performance data and emotional data for each variation. For example, user group A is presented with "15% cashback" and user group B with "20% cashback," and the response and performance data for each group are collected.

[0558] Step 6:

[0559] Summary of results and selection of the optimal variation

[0560] After the test is complete, the device sends the collected performance and sentiment data to the server. The inputs are the A / B test results and sentiment data. The server statistically analyzes this data and selects the optimal variation. The output is the data for the selected optimal variation (e.g., "20% cashback" is optimal).

[0561] Step 7:

[0562] Notification of optimal variation

[0563] The server sends the result of the selected optimal variation to the terminal. The input includes the data of the optimal variation selected by the server. The terminal notifies the user of this, and the user uses this data to release the final service or campaign. For example, if the server selects "20% cashback" as the optimal variation and notifies the user of this result, the user can implement a "20% cashback campaign."

[0564] (Application Example 2)

[0565] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0566] Traditional service and campaign design support systems have the drawback of failing to take user emotions into account, thus hindering their ability to maximize user satisfaction. Furthermore, the difficulty in obtaining real-time feedback makes it challenging to quickly deliver campaigns tailored to user needs. Moreover, the lack of a means to adjust design proposals based on emotional data prevents them from addressing the diverse emotions of users.

[0567] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0568] In this invention, the server includes means for receiving service and campaign design proposals based on user input; means for transmitting the design proposals to a generative artificial intelligence and conducting evaluations based on relevant laws and guidelines; means for receiving evaluation results from the generative artificial intelligence and displaying them to the user; means for collecting sentiment data based on the evaluation results; means for generating multiple variations based on the design proposals; means for conducting multifaceted A / B testing using the variations; means for aggregating the results of the A / B testing and selecting the optimal variation; means for notifying the user of the optimal variation; means for adjusting the design proposals in real time based on the user's sentiment data and providing feedback; and means for using smart glasses, a head-mounted display, or a similar device to collect the user's sentiment data. This makes it possible to reflect the user's emotions in real time and quickly provide more satisfying services and campaigns.

[0569] "User input" refers to the information that users provide to the system regarding their service or campaign design proposals.

[0570] "Service and campaign design proposals" refer to the marketing and promotional plans and their specific details that users envision.

[0571] "Generative artificial intelligence" refers to algorithms that use machine learning and natural language processing to analyze user design proposals and provide evaluations and suggestions based on relevant laws and guidelines.

[0572] "Relevant laws and guidelines" refer to the laws and regulations and guidelines that must be followed when conducting the service or campaign.

[0573] "Evaluation results" refer to feedback from a generative artificial intelligence system that evaluates the design proposal and identifies any legal issues or areas for improvement.

[0574] "Emotional data" refers to information about a user's emotions collected from their facial expressions, voice, text, and other sources.

[0575] "Variations" refer to different versions of a campaign or promotion that are generated based on the user's design proposal.

[0576] A / B testing is a method of presenting multiple variations to different user groups and comparing their performance data.

[0577] "Performance data" refers to metrics used to measure the effectiveness of a campaign or service (e.g., click-through rate, registration rate, purchase rate, etc.).

[0578] The "optimal variation" refers to the campaign or promotion that, based on A / B test results and sentiment data, is judged to be the most effective and provides the highest user satisfaction.

[0579] "Smart glasses" are wearable devices designed to collect the user's emotions in real time.

[0580] A "head-mounted display" is a wearable device that collects emotional data while displaying information in the user's field of vision.

[0581] "Adjusting and providing feedback in real time" is a process of immediately reflecting user sentiment data to revise the design proposal and returning the results to the user.

[0582] This invention combines an emotion engine and generative artificial intelligence in a system that assists in the design of services and campaigns, providing a technology that designs optimal services and campaigns based on user emotional information.

[0583] System Configuration

[0584] Hardware and software:

[0585] 1. User terminal: A device including smart glasses or a head-mounted display. This device collects the user's facial expressions, voice, and text in real time and generates emotion data through an emotion engine (e.g., Microsoft Azure Emotion API).

[0586] 2. Server: A cloud platform (e.g., AWS) will be used to run generative artificial intelligence (e.g., OpenAI GPT). The server will process design proposals and sentiment data received from users and perform legal evaluations and design proposal optimization.

[0587] Software processing and data computation:

[0588] 1. Receiving user input:

[0589] Users interact with touchpoints within the virtual store and input design proposals for services and campaigns.

[0590] 2. Collection and transmission of emotional data:

[0591] Smart glasses and head-mounted displays capture the user's facial expressions, voice, and text, which are then analyzed as emotional data by an emotion engine.

[0592] The generated emotional data is sent to the cloud server along with the design proposal.

[0593] 3. Evaluation and adjustment of the design proposal:

[0594] The generative artificial intelligence within the server analyzes the design proposal and conducts a legal evaluation based on relevant laws and guidelines.

[0595] Based on emotional data, user satisfaction is assessed, and a design proposal with necessary adjustments is generated.

[0596] 4. Display of evaluation results and feedback:

[0597] The evaluation results from the server are sent to the user's terminal and displayed in real time on a virtual display.

[0598] We provide feedback on optimized campaign proposals based on user sentiment data.

[0599] 5. Variation generation and A / B testing:

[0600] Based on the refined design proposal, multiple variations are generated and presented to different user groups.

[0601] Conduct A / B testing in real time and collect performance and sentiment data.

[0602] 6. Selection and notification of the optimal variation:

[0603] Based on the collected data, statistical analysis is performed on the server to select the optimal variation.

[0604] The optimal variation is notified to the user's terminal and displayed on the virtual display.

[0605] Specific example

[0606] For example, a user designs a "Spring Bonus Points Campaign" and inputs their design proposal. At this time, an emotion engine is used to collect emotional data from the user's facial expressions and voice. A generative AI model is used to analyze "how to make the Spring Bonus Points Campaign legally compliant" and perform a legal evaluation. Based on this result, the optimal proposal (e.g., a 10% point increase) that reflects the user's emotional data is displayed in real time. Subsequently, A / B testing is conducted on the different variations (e.g., a 15% point increase, a 20% point increase, and one with a free coupon), and the "20% increase" with the highest performance is ultimately selected.

[0607] Example of a prompt:

[0608] "How can we make our spring bonus points campaign legally sound and offer exciting deals for users?"

[0609] Thus, the present invention enables the efficient design of optimal services and campaigns by combining evaluations of laws and guidelines with user sentiment information.

[0610] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0611] Step 1:

[0612] Receiving user input:

[0613] Users interact with interactive touchpoints within a virtual store to input service and campaign design proposals. This input data includes the campaign name, content, and terms and conditions. The terminal receives and stores this input data.

[0614] Step 2:

[0615] Collection of emotional data:

[0616] While the user inputs design proposals, smart glasses or a head-mounted display captures the user's facial expressions and voice in real time. This data is sent to an emotion engine (e.g., Microsoft Azure Emotion API) for analysis and collection as emotion data. The input data consists of the user's facial expressions and voice, while the output data represents the user's emotional state (e.g., joy, surprise, anxiety).

[0617] Step 3:

[0618] Sending design proposals and emotional data:

[0619] The terminal sends the user's input design proposal and collected sentiment data to the cloud server. The input data consists of the design proposal and sentiment data, while the output data is a request message containing these.

[0620] Step 4:

[0621] Legal evaluation of the design proposal:

[0622] The server uses generative artificial intelligence (e.g., OpenAI GPT) to evaluate the design proposal. For example, it takes the prompt "Does this design proposal comply with laws and guidelines?" as input, and the AI ​​model performs the analysis. The output is a legal evaluation result, indicating whether or not there are any problems.

[0623] Step 5:

[0624] Return and display of evaluation results:

[0625] The server returns the legal assessment results to the terminal. The terminal displays the assessment results in real time on the user's virtual display. The input data is the legal assessment results, and the output data is the information displayed to the user.

[0626] Step 6:

[0627] Generating adjustment proposals based on emotional data:

[0628] The server adjusts the design proposal based on emotional data. For example, it might input a prompt message to a generative artificial intelligence such as, "According to the user's emotional data, there are many concerns about the current design proposal, so I will suggest improvements." The output is an adjusted proposal, a specific suggestion that takes the user's emotions into consideration.

[0629] Step 7:

[0630] Generating multiple variations:

[0631] Based on the adjusted design proposal, the server generates multiple variations. These variations include different percentages of discounts and benefits. The input data is the adjusted proposal, and the output data is a set of variations.

[0632] Step 8:

[0633] Conducting A / B testing:

[0634] The device conducts A / B testing on different user groups using the generated variations. It collects performance data (e.g., click-through rate, purchase rate) and sentiment data for each variation. The input data consists of variation-to-user interaction data, and the output data consists of customized evaluation data.

[0635] Step 9:

[0636] Aggregation and optimization of test results:

[0637] The server statistically analyzes the collected performance and sentiment data to select the optimal variation. For example, it might input a prompt message to a generative artificial intelligence system such as, "Select the variation that shows the highest click-through rate and user satisfaction." The output data would then be the optimal variation.

[0638] Step 10:

[0639] Notification of optimal variation:

[0640] The server sends the optimal variation to the user's terminal and displays it on the virtual display. The input data is the optimal variation, and the output data is the optimal campaign proposal that is notified to the user.

[0641] This series of steps creates a system that reflects user emotions and delivers optimal services and campaigns in real time.

[0642] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0643] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0644] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0645] [Third Embodiment]

[0646] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0647] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0648] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0649] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0650] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0651] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0652] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0653] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0654] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0655] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0656] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0657] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0658] This invention provides a system in which users input service and campaign design proposals, and the system designs the optimal service through verification of laws and guidelines using generative artificial intelligence and multifaceted A / B testing. The operation of the system in each processing step is described in detail below.

[0659] User input

[0660] Users input service and campaign design proposals using their devices. The input design proposals are then sent from the device to the server.

[0661] Checking laws and guidelines

[0662] The server sends the design proposal received from the user to a generative artificial intelligence (AI) system, which then performs an evaluation based on relevant laws and guidelines. The AI ​​analyzes the design proposal and determines whether there are any legal issues. The evaluation results are returned to the server, which then sends them to the terminal. The terminal then displays the evaluation results to the user.

[0663] Generating Variations

[0664] If the evaluation result is determined to be "no problems," the server generates multiple variations based on the design proposal. These variations refer to multiple proposals containing different settings and conditions, and the server sends these to the terminal.

[0665] Conducting A / B testing

[0666] The device sets up test scenarios based on the received variations and conducts A / B testing from multiple angles. Different variations are presented to multiple user groups, and test results are collected. For example, variations with 15% cashback, 20% cashback, and a free coupon are set up, and performance data (click-through rate, registration rate, purchase rate, etc.) is collected for each.

[0667] Summary of results and selection of the optimal variation

[0668] After the test is complete, the terminal sends the obtained performance data to the server. The server aggregates this data, analyzes it statistically, and selects the optimal variation.

[0669] Notification of optimal variation

[0670] Once the optimal variation is selected, the server sends the result to the terminal. The terminal then notifies the user of the optimal service design and releases the final service based on it.

[0671] Specific examples of operation

[0672] The following is a concrete example of how it works. The user designs a "new points reward campaign" and inputs it into the terminal. This design proposal is sent to the server, where it is legally checked by generative artificial intelligence. It is confirmed that it does not violate consumer protection laws or the Premiums and Representations Act, and the results are notified to the user. The server then generates variations such as 15% reward, 20% reward, and with a free coupon, and A / B testing is conducted on the terminal. The test results are sent to the server, which selects the "20% reward" option that is most popular with the user, and notifies the user of the result.

[0673] As described above, the system based on the present invention can efficiently perform verification in accordance with laws and guidelines, and design optimal services that reflect user preferences through A / B testing.

[0674] The following describes the processing flow.

[0675] Step 1:

[0676] The user inputs a design proposal for a service or campaign into the terminal. The terminal then sends the input design proposal to the server.

[0677] Step 2:

[0678] The server sends the received design proposal to the generative artificial intelligence. The generative AI analyzes the design proposal and conducts a legal evaluation based on relevant laws and guidelines.

[0679] Step 3:

[0680] The evaluation results from the generative artificial intelligence are returned to the server. The server receives the evaluation results and sends them to the terminal.

[0681] Step 4:

[0682] The device displays the evaluation results to the user. The user reviews the evaluation results and understands that there are no legal issues.

[0683] Step 5:

[0684] The server generates multiple variations based on the design proposal that was deemed "problem-free" in the evaluation results. These generated variations include different settings and conditions.

[0685] Step 6:

[0686] The server sends the generated variations to the device. The device then prepares to use the variations to conduct an A / B test.

[0687] Step 7:

[0688] The device presents variations to multiple user groups. For example, it might offer variations such as 15% cashback, 20% cashback, or a free coupon, and collects performance data (click-through rate, registration rate, purchase rate, etc.) for each variation.

[0689] Step 8:

[0690] Once the A / B test is complete, the device sends the collected performance data to the server. The server then aggregates and statistically analyzes the received data.

[0691] Step 9:

[0692] The server selects the most effective variation based on the aggregated data. Once the optimal variation is determined, the server sends the result to the terminal.

[0693] Step 10:

[0694] The device notifies the user of the optimal service design. Based on the notified optimal design, the user releases the final service or campaign.

[0695] The above describes the specific actions of each processing step in the program. This system makes it possible to comply with laws and guidelines and efficiently provide the best possible service to users.

[0696] (Example 1)

[0697] Next, we will describe Example 1. 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."

[0698] In recent years, when designing services and campaigns that appeal to consumers, users are required to experiment with multiple design options while avoiding legal risks. However, checking legal issues and rationally testing various variations to select the optimal design is a multifaceted and complex process that requires considerable effort and time. This has created a challenge in that it is difficult for users to quickly and efficiently find the optimal design.

[0699] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0700] In this invention, the server includes means for receiving design proposals for services and campaigns based on user input; means for transmitting the design proposals to a generative artificial intelligence and conducting evaluations based on relevant laws and guidelines; means for receiving evaluation results from the generative artificial intelligence and displaying them to the user; means for generating multiple variations based on the design proposals; means for conducting comparative tests from multiple perspectives using the variations; means for aggregating the results of the comparative tests and selecting the optimal variation; and means for notifying the user of the optimal variation. This makes it possible to efficiently design optimal services and campaigns while avoiding legal risks.

[0701] A "user" refers to an individual or legal entity that inputs design proposals for services or campaigns and utilizes the system.

[0702] "Terminal" refers to an input device used by a user, and includes electronic devices such as personal computers, smartphones, and tablets.

[0703] A "server" is a computer system that performs the central processing of a system and provides multiple functions in an integrated manner.

[0704] A "design proposal" refers to a plan that describes the specific content and conditions of a service or campaign that users will input.

[0705] "Generative artificial intelligence" refers to an artificial intelligence system that performs advanced analytical processing, such as legal evaluation and variation generation, from given input data.

[0706] "Laws and guidelines" refer to the laws and industry standards that apply to the service or campaign, including consumer protection laws and the Premiums and Representations Act.

[0707] "Evaluation results" refer to the results obtained when a generative artificial intelligence analyzes a design proposal and determines whether there are any legal issues or whether it conforms to guidelines.

[0708] "Variations" refer to multiple design options with different conditions and settings, generated based on the original design proposal.

[0709] A "comparative test" refers to a testing method in which multiple variations are presented to different user groups, and their response data is collected and compared.

[0710] "Performance data" refers to data showing user responses to each variation, including click-through rates, registration rates, and purchase rates.

[0711] The "optimal variation" refers to the variation that showed the highest performance data in comparative tests.

[0712] "Notifying" refers to the process of transmitting information from a server to a user via a terminal.

[0713] This invention relates to a system in which a user inputs a design proposal for a service or campaign, and the system performs optimal service design through legal evaluation and multifaceted comparative testing using generative artificial intelligence. Specific embodiments for carrying out this invention are described in detail below.

[0714] Users input service and campaign design proposals using their devices. For example, for a "new points reward campaign," they would describe the campaign details (campaign period, reward rate, etc.) in an input form on their device. This design proposal is then sent from the device to the server.

[0715] The server sends the received design proposal to a generative artificial intelligence (e.g., OpenAI GPT-4). The generative AI analyzes the design proposal, which is input as a prompt: "Please check whether this design proposal violates consumer protection laws or the Premiums and Representations Act." Based on the analysis, it determines whether there are any legal issues and returns the evaluation result to the server. The evaluation result is resent from the server to the terminal and displayed to the user on the terminal. For example, the evaluation result displayed might be: "This campaign does not violate consumer protection laws."

[0716] If the evaluation result is determined to be "no problems," the server generates multiple variations based on the design proposal. Generative artificial intelligence and algorithms are used to create variations with different settings and conditions. For example, it might generate three variations: "15% cashback," "20% cashback," and "with a free coupon." The generated variations are then sent from the server to the terminal.

[0717] The device sets up a comparative test scenario based on the received variations. This scenario presents different variations to multiple user groups and collects user response data (click-through rate, registration rate, purchase rate, etc.) for each variation. For example, user group A is presented with "15% cashback," user group B with "20% cashback," and user group C with "free coupon."

[0718] Once the results of each comparison test are collected, the device sends this performance data to the server. The server aggregates the data and performs statistical analysis to select the most effective variation. For example, by comparing the click-through rate and purchase rate of each variation, the "20% cashback" variation may be found to have the best performance. In this process, the R or Python pandas library may be used.

[0719] Once the optimal variation is selected, the server notifies the terminal of the result. The terminal then notifies the user of the optimal variation, and the user ultimately releases the service based on that variation. For example, the user might see a notification that says, "Test results indicate that a 20% cashback has been selected as the most effective variation."

[0720] In this way, the system based on the present invention can generate design proposals based on user input and efficiently design optimal services and campaigns through multifaceted comparative testing while avoiding legal risks.

[0721] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0722] Step 1:

[0723] Users input service and campaign design proposals using their devices. For example, they might describe a "new points reward campaign," detailing specific conditions such as the campaign period and reward rate. This design proposal is sent from the device to the server as input data. Once the server receives the design proposal as input, processing begins on the server side.

[0724] Step 2:

[0725] The server sends the received design proposal to a generative artificial intelligence (e.g., OpenAI GPT-4). Specifically, the design proposal is used as input data with the prompt message "Please check whether this design proposal violates consumer protection laws or the Premiums and Representations Act." The generative AI model analyzes this input data and determines whether there are any legal issues. The evaluation result is sent back to the server as output, and this result serves as information to confirm whether there are any legal problems.

[0726] Step 3:

[0727] The server receives the evaluation results from the generative artificial intelligence and sends them to the user's terminal for them to review. The terminal displays the legal evaluation results to the user, for example, a message such as "This campaign does not violate consumer protection laws." Information to be notified to the user is generated based on the evaluation results as input.

[0728] Step 4:

[0729] If the evaluation result is determined to be "no problems," the server generates multiple variations based on the design proposal. Generative artificial intelligence and algorithms are used to create multiple variations with different settings and conditions. For example, variations such as "15% cashback," "20% cashback," and "with a free coupon" may be included. The generated variations are sent from the server to the terminal as output data, preparing it for the next test.

[0730] Step 5:

[0731] The device sets up a comparative test scenario based on the received variations. Specifically, it creates test cases that present different variations to multiple user groups. User group A is presented with "15% cashback," user group B with "20% cashback," and user group C with "free coupon." A multifaceted comparative test is performed based on the variations as input. The output is the collection of response data from each group.

[0732] Step 6:

[0733] Once the results of the comparative tests are collected, the device sends this performance data (e.g., click-through rate, registration rate, purchase rate) to the server. The server aggregates this data and performs statistical analysis. Based on the performance data as input, it selects the most effective variation. For example, by comparing the click-through rate and purchase rate of each variation, the results might show that "20% cashback" performs the best. As output, the optimal variation is selected.

[0734] Step 7:

[0735] Once the optimal variation is selected, the server notifies the terminal of the result. The terminal then notifies the user of the optimal variation, and the user ultimately releases the service based on that variation. For example, a notification might appear stating, "Based on the test results, a 20% cashback has been selected as the most effective variation." This provides the user with information to take their next action based on the test results as input.

[0736] (Application Example 1)

[0737] Next, we will explain Application Example 1. In the following explanation, 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."

[0738] Traditional advertising campaign design often involved manual processes for checking legal issues, generating multiple variations, testing, and selecting the optimal variation, resulting in inefficiencies and time-consuming tasks. Furthermore, there was a lack of systems providing user-friendly interfaces using smart devices. This made it difficult to quickly and effectively maximize marketing effectiveness.

[0739] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0740] In this invention, the server includes means for receiving service and campaign design proposals based on user input; means for transmitting the design proposals to a generative artificial intelligence and conducting evaluations based on relevant regulations and guidelines; means for receiving evaluation results from the generative artificial intelligence and displaying them to the user; means for generating multiple options based on the design proposals; means for conducting multifaceted A / B testing using the options; means for aggregating the results of the A / B testing and selecting the optimal option; means for notifying the user of the optimal option; and means including an application installed on a smart device. This enables efficient advertising campaign design, legal review, variation generation, testing, and optimization, and maximizes marketing effectiveness with a user-friendly interface.

[0741] A "user" refers to a person who uses the system to design services and campaigns.

[0742] "Input" refers to the act of a user providing design proposals and related information to the system.

[0743] A "design proposal" refers to information that shows the specific details of a service or campaign proposed by a user.

[0744] "Generative artificial intelligence" refers to artificial intelligence technology that uses user-inputted design proposals to check relevant regulations and guidelines, and conducts multifaceted A / B testing.

[0745] "Evaluation" refers to the process by which a generative artificial intelligence determines whether a design proposal has legal issues or complies with guidelines.

[0746] "Legal issues" refer to the question of whether the proposed design violates legal regulations.

[0747] "Regulations" refer to laws and guidelines that apply to the design of a service or campaign.

[0748] "Options" refers to multiple variations or different proposals with different conditions that are generated based on a design proposal.

[0749] "AB testing" refers to a testing method that presents multiple options to different user groups and compares and evaluates the performance of each option.

[0750] "Aggregation" refers to the act of statistically compiling the results data of an A / B test.

[0751] The "optimal choice" refers to the variation that shows the highest performance based on the results of the A / B test.

[0752] "Smart devices" refer to advanced devices such as smartphones, smart glasses, and head-mounted displays.

[0753] This invention involves several key steps to realize a system in which users input service and campaign design proposals using a smart device. These steps are described below.

[0754] First, the user uses a smart device such as a smartphone to input their service or campaign design proposal. The user enters the details of the specific design proposal and sends it from the device to the server.

[0755] Next, the server sends the received design proposal to the generative artificial intelligence (AI) and performs an evaluation based on relevant regulations and guidelines. The generative AI analyzes the design proposal and determines whether there are any legal issues. This legal evaluation includes consumer protection laws and the Premiums and Representations Act, among others. The evaluation results of the generative AI are returned to the server, which then sends the results to the user's terminal and displays them to the user.

[0756] Next, if the server determines the evaluation result is "no problems," it generates multiple options (variations) based on the design proposal. These variations have different settings and conditions, and may include options such as "15% cashback," "20% cashback," and "with a free coupon." These variations are then sent from the server to the terminal.

[0757] The device conducts multifaceted A / B testing based on the variations it receives. This test presents different variations to multiple user groups and collects performance data (click-through rates, registration rates, purchase rates, etc.) resulting from these tests. This data is then sent to the server.

[0758] The server aggregates the received performance data, analyzes it statistically, and selects the optimal option. The selected optimal option is then notified from the server to the terminal, informing the user of the best possible service design.

[0759] As a concrete example, a user designs a "new points reward campaign" and inputs a design proposal for a "20% cashback campaign." Generative artificial intelligence legally evaluates this proposal and confirms that it does not violate consumer protection laws or the Premiums and Representations Act. The server then generates variations such as "15% cashback," "20% cashback," and "with free coupon," and A / B testing is conducted on the terminal. The test results are sent to the server, which selects the "20% cashback" option, which showed the highest performance, as the optimal choice and notifies the user of the result.

[0760] Examples of prompt messages are as follows:

[0761] We are designing a new points reward campaign. Please perform legal checks and generate variations based on the details below.

[0762] Campaign type: Point rewards

[0763] Details: 20% cashback campaign

[0764] Thus, the system based on this invention maximizes marketing effectiveness by quickly and efficiently evaluating user-inputted design proposals and providing the optimal service design.

[0765] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0766] Step 1:

[0767] Users input service and campaign design proposals using a smart device. Specifically, users open a dedicated application on their smartphone, input their design proposal (e.g., "20% Cashback Campaign"), and press the submit button. Input data: Campaign details. Output data: Confirmation message for submitting the design proposal.

[0768] Step 2:

[0769] The server receives design proposals submitted by the user. The server then forwards the input to a generative artificial intelligence system, which conducts a legal evaluation based on relevant regulations and guidelines. This legal evaluation includes checks regarding consumer protection laws and the Premiums and Representations Act. Input data: Design proposals. Data processing: Evaluation based on legal regulations. Output data: Legal evaluation results.

[0770] Step 3:

[0771] Generative artificial intelligence analyzes received design proposals and determines whether they have any legal issues. Generative AI uses natural language processing and rule-based analysis to check which regulations or guidelines the design proposals violate. Input data: Design proposal. Data calculation: Comparison with legal regulations. Output data: Result of legal evaluation (e.g., "No issues").

[0772] Step 4:

[0773] The server receives evaluation results from the generative artificial intelligence, sends them to the user's terminal, and displays them to the user. The displayed content is the result of the legal evaluation and includes instructions for when there are no problems or when corrections are needed. Input data: Legal evaluation results. Output data: Content displayed to the user.

[0774] Step 5:

[0775] If the evaluation result is determined to be "no problems," the server generates multiple variations based on the design proposal. These variations include different reward rates and benefits. The generated variations are sent to the terminal. Input data: Evaluated design proposal. Data processing: Generation of variations. Output data: Multiple variations.

[0776] Step 6:

[0777] The device conducts multifaceted A / B testing based on the received variations. Specifically, it presents different variations to user groups and collects performance data such as click-through rates and purchase rates. Input data: Multiple variations. Data processing: Collection of user behavior. Output data: Performance data (e.g., click-through rate, purchase rate).

[0778] Step 7:

[0779] The terminal sends the collected performance data to the server. The server aggregates the received performance data, performs statistical analysis, and selects the optimal variation. Input data: Performance data. Data calculation: Statistical analysis. Output data: Result of the selection of the optimal variation.

[0780] Step 8:

[0781] The server selects the optimal variation and notifies the user of the result on their device. The user receives the notification and confirms the optimal campaign design. Input data: Selection result. Output data: Notification content to the user (e.g., "20% cashback is optimal").

[0782] By following these steps, users can efficiently design legally compliant and optimal advertising campaigns.

[0783] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0784] This invention provides a technology that combines an emotion engine with a system that supports the design of services and campaigns, enabling the design of optimal services and campaigns based on user emotion information. The operation of the system in each processing step is described in detail below.

[0785] User input

[0786] Users input service and campaign design proposals using a device. During this process, the device utilizes its built-in emotion engine to recognize emotions from the user's facial expressions, voice, and text, and collects that emotion data.

[0787] Checking laws and guidelines

[0788] The server transmits the design proposal and emotional data received from the user to a generative artificial intelligence (AI) system, which then performs an evaluation based on relevant laws and guidelines. The AI ​​analyzes the design proposal and conducts a legal evaluation, taking the emotional data into consideration. The evaluation results are returned to the server, which then transmits them to the terminal.

[0789] Displaying results and reflecting sentiment data

[0790] The device displays the evaluation results to the user. At the same time, it adjusts the design proposal as needed based on sentiment data and provides feedback in a way that is appropriate for the user. For example, if a campaign suggested by the user does not align with the user's preferences or expectations, the sentiment engine will use that information to suggest adjustments.

[0791] Generating Variations

[0792] The server generates multiple variations based on the refined design proposal. These variations include different settings and conditions, and are intended for multifaceted evaluation.

[0793] Conducting A / B testing

[0794] The device will conduct A / B testing using the generated variations. Multiple user groups will be presented with the variations, and performance data for each variation (click-through rate, registration rate, purchase rate, etc.) will be collected. During the test, an emotion engine will be used to evaluate how users feel about the variations.

[0795] Summary of results and selection of the optimal variation

[0796] After the test is complete, the device sends the collected performance and sentiment data to the server. The server aggregates and statistically analyzes this data to select the optimal variation. Sentiment data is considered an important indicator for evaluating user satisfaction and motivation.

[0797] Notification of optimal variation

[0798] Once the optimal variation is selected, the server sends the result to the terminal. The terminal notifies the user of the optimal service design and releases the final service or campaign based on it. An optimal design that reflects user sentiment information increases user satisfaction and reduces the risk of complaints.

[0799] Specific examples of operation

[0800] The following is a concrete example of how it works. The user designs a "new points reward campaign" and inputs it into the terminal. The emotion engine recognizes the user's emotions and sends it to the server along with the design proposal. A legal check is performed by generative artificial intelligence, and the evaluation results are notified to the user, while adjustments based on the emotion data are suggested. Subsequently, variations such as 15% reward, 20% reward, and with a free coupon are generated, and A / B testing is conducted. Based on the test results and emotion data, the server determines that "20% reward" is optimal, and this result is notified to the user.

[0801] As described above, the system based on the present invention can efficiently design optimal services and campaigns by combining evaluations of laws and guidelines with user sentiment information.

[0802] The following describes the processing flow.

[0803] Step 1:

[0804] The user inputs design proposals for services and campaigns using a device. Simultaneously with the user's input, the device activates an emotion engine, recognizing emotions from the user's facial expressions, voice, and text, and collecting emotion data.

[0805] Step 2:

[0806] The terminal sends the input design proposal and collected emotional data to the server. The server sends the received data to a generative artificial intelligence system and requests an evaluation based on relevant laws and guidelines.

[0807] Step 3:

[0808] The generative artificial intelligence analyzes design proposals and emotional data to perform legal evaluations. Simultaneously, it evaluates the user's emotional state, taking emotional data into consideration, and returns the results to the server.

[0809] Step 4:

[0810] The server sends the legal and emotional evaluation results received from the generative artificial intelligence to the terminal. The terminal then displays the results to the user.

[0811] Step 5:

[0812] The device adjusts the design proposal based on emotional data as needed. If the user's emotional state is negative, the device fine-tunes the design proposal based on suggestions from generative artificial intelligence and presents it to the user again.

[0813] Step 6:

[0814] The server generates multiple variations based on the refined design proposal. These generated variations contain different settings and conditions, and the server sends them to the terminal.

[0815] Step 7:

[0816] The device sets up test scenarios based on the generated variations and conducts A / B testing. Different variations are presented to multiple user groups, and performance and sentiment data for each variation are collected.

[0817] Step 8:

[0818] After the A / B test is complete, the device sends the collected performance and sentiment data to the server. The server aggregates the received data, performs statistical analysis, and selects the optimal variation.

[0819] Step 9:

[0820] The server selects the variation that elicited the most effective and emotionally positive response based on the aggregated data. Once the optimal variation is determined, it sends the result to the terminal.

[0821] Step 10:

[0822] The device notifies the user of the optimal service design. Based on the notified optimal design, the user releases the final service or campaign.

[0823] The above outlines the specific processing steps of this system, which incorporates an emotion engine. This enables the efficient design of optimal services and campaigns that comply with laws and guidelines and reflect user emotions.

[0824] (Example 2)

[0825] Next, we will describe Example 2. 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."

[0826] Traditional campaign and service design lacks the technology to incorporate user emotional information and select the optimal variations. Therefore, designing without considering user emotional satisfaction has failed to adequately improve actual customer satisfaction. Furthermore, the need to separately evaluate laws and guidelines complicates the design process. This invention aims to solve these problems.

[0827] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0828] In this invention, the server includes means for receiving service and campaign design proposals based on user input, means for transmitting the design proposals to a generative artificial intelligence and conducting evaluations based on relevant laws and guidelines, means for receiving evaluation results from the generative artificial intelligence and displaying them to the user, means for generating multiple variations based on the design proposals, means for conducting multifaceted A / B testing using the variations, means for aggregating the results of the A / B testing and selecting the optimal variation, means for aggregating user emotional information and reflecting it in the selection of the optimal variation, and means for recognizing user emotions using an emotion engine and providing feedback to the design proposals. This makes it possible to design optimal services and campaigns that reflect user emotional information, improve user satisfaction, and realize an efficient design process that also ensures legal compliance.

[0829] "User input" refers to the act of a user using a device to input design proposals for a service or campaign.

[0830] "Generative artificial intelligence" refers to artificial intelligence models that analyze design proposals and evaluate them based on relevant laws and guidelines.

[0831] A "variation" refers to multiple proposals with different settings and conditions based on a design proposal for a service or campaign.

[0832] A / B testing is an evaluation method that involves presenting multiple variations to user groups and collecting performance data.

[0833] "Emotional information" refers to data about a user's emotional state collected through their facial expressions, voice, text, etc.

[0834] An "emotion engine" is a system that recognizes a user's emotions and analyzes that emotional data.

[0835] "Legal evaluation" is the process by which generative artificial intelligence determines the suitability of a design proposal based on relevant laws and guidelines.

[0836] "Performance data" refers to experimental results such as click-through rates, registration rates, and purchase rates collected during A / B testing.

[0837] A "design proposal" refers to the specific suggestions a user has entered regarding a service or campaign.

[0838] The "optimal variation" is the variation that, based on aggregated performance and sentiment data, is judged to be the most effective and provide the highest user satisfaction.

[0839] This invention provides a technology that combines an emotion engine with a system that supports the design of services and campaigns, enabling the design of optimal services and campaigns based on user emotion information. The operation of the system in each processing step is described in detail below.

[0840] First, the user inputs their service or campaign design proposal using a device. This device utilizes a built-in emotion engine to recognize emotions from the user's facial expressions, voice, and text, and collects that emotion data. The emotion engine can utilize either the "Emotion API" or "Azure Cognitive Services."

[0841] Next, the terminal sends the design proposals and sentiment data collected from the user to the server. The server sends this data to a generative artificial intelligence (e.g., GPT-4) which then performs an evaluation based on relevant laws and guidelines. The generative AI analyzes the design proposals and conducts a legal evaluation that takes the sentiment data into account.

[0842] Specifically, one could send the following prompt message to the generative artificial intelligence.

[0843] "Please evaluate the following campaign design proposal based on laws and guidelines, taking into account user sentiment data. Proposal: New points reward campaign. User sentiment data: 80% positive responses, 20% negative responses."

[0844] The evaluation results from the generative artificial intelligence are returned to the server, which then sends the results to the terminal. The terminal displays the evaluation results to the user and, at the same time, adjusts the design proposal as needed based on the sentiment data, providing feedback in a way that is appropriate for the user. For example, if a campaign proposed by the user does not align with the user's preferences or expectations, the sentiment engine will use that information to suggest adjustments.

[0845] Next, the server generates multiple variations based on the refined design proposal. The generated variations include different settings and conditions, and are intended for multifaceted evaluation. For example, variations such as "15% cashback," "20% cashback," and "with free coupon" are generated.

[0846] Subsequently, the device conducts A / B testing using the generated variations. The variations are presented to multiple user groups, and performance data (click-through rate, registration rate, purchase rate, etc.) for each variation is collected. The sentiment engine is also used during the test to evaluate how users feel about each variation.

[0847] Once the test is complete, the device sends the collected performance and sentiment data to the server. The server statistically analyzes this data and selects the optimal variation. Sentiment data is an important indicator for evaluating user satisfaction and motivation.

[0848] Once the optimal variation is selected, the server sends the result to the terminal. The terminal notifies the user of the optimal service design and releases the final service or campaign based on it. For example, the server selects "20% cashback" as the optimal variation and notifies the user of the result.

[0849] In this way, the system based on the present invention can efficiently design optimal services and campaigns by reflecting user emotional information and combining it with legal evaluations. This makes it possible to improve user satisfaction and reduce the risk of complaints.

[0850] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0851] Step 1:

[0852] User input

[0853] The user uses the device to input design proposals for services and campaigns. Specifically, the user designs a "new points reward campaign" and inputs it into the device. Based on this input, the device uses its built-in emotion engine to recognize the user's emotions from their facial expressions, voice, and text, and collects that emotion data. The input includes the text data of the design proposal and the user's emotion data (e.g., 80% positive reactions, 20% negative reactions). The device collects this data and sends it to the server.

[0854] Step 2:

[0855] Checking laws and guidelines

[0856] The server sends the design proposal and sentiment data received from the terminal to the generative artificial intelligence (AI). The input for this process consists of the text data of the design proposal and the sentiment data. The generative AI analyzes the design proposal and performs an evaluation based on relevant laws and guidelines. Specifically, an example of sending a prompt to the generative AI is as follows:

[0857] "Please evaluate the following campaign design proposal based on laws and guidelines, taking into account user sentiment data. Proposal: New points reward campaign. User sentiment data: 80% positive responses, 20% negative responses."

[0858] The output is a legal evaluation result. The evaluation result is returned to the server, which then sends the result to the terminal. This result may include information such as, "The point redemption rate for this campaign does not violate current laws."

[0859] Step 3:

[0860] Displaying results and reflecting sentiment data

[0861] The device displays the received evaluation results to the user. Inputs include evaluation results obtained from the server and user sentiment data. Based on these inputs, the device adjusts the design proposal as needed and provides appropriate feedback to the user. Specifically, if the proposed campaign does not align with the user's preferences or expectations, the sentiment engine uses this information to suggest adjustments. For example, it might output new variations such as 15% or 20% cashback offers.

[0862] Step 4:

[0863] Generating Variations

[0864] The server generates multiple variations based on the adjusted design proposal. The input is the text data of the adjusted design proposal. Based on this, the server generates variations with different settings and conditions. Specifically, it generates variations such as "15% cashback," "20% cashback," and "with free coupon," and outputs these variation proposals.

[0865] Step 5:

[0866] Conducting A / B testing

[0867] The device conducts A / B testing using the generated variations. The input is design data for the variations. The device presents each variation to multiple user groups and collects performance data such as click-through rates, registration rates, and purchase rates. It also uses an emotion engine to evaluate users' emotional responses. The output includes performance data and emotional data for each variation. For example, user group A is presented with "15% cashback" and user group B with "20% cashback," and the response and performance data for each group are collected.

[0868] Step 6:

[0869] Summary of results and selection of the optimal variation

[0870] After the test is complete, the device sends the collected performance and sentiment data to the server. The inputs are the A / B test results and sentiment data. The server statistically analyzes this data and selects the optimal variation. The output is the data for the selected optimal variation (e.g., "20% cashback" is optimal).

[0871] Step 7:

[0872] Notification of optimal variation

[0873] The server sends the result of the selected optimal variation to the terminal. The input includes the data of the optimal variation selected by the server. The terminal notifies the user of this, and the user uses this data to release the final service or campaign. For example, if the server selects "20% cashback" as the optimal variation and notifies the user of this result, the user can implement a "20% cashback campaign."

[0874] (Application Example 2)

[0875] Next, we will explain application example 2. In the following explanation, 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."

[0876] Traditional service and campaign design support systems have the drawback of failing to take user emotions into account, thus hindering their ability to maximize user satisfaction. Furthermore, the difficulty in obtaining real-time feedback makes it challenging to quickly deliver campaigns tailored to user needs. Moreover, the lack of a means to adjust design proposals based on emotional data prevents them from addressing the diverse emotions of users.

[0877] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0878] In this invention, the server includes means for receiving service and campaign design proposals based on user input; means for transmitting the design proposals to a generative artificial intelligence and conducting evaluations based on relevant laws and guidelines; means for receiving evaluation results from the generative artificial intelligence and displaying them to the user; means for collecting sentiment data based on the evaluation results; means for generating multiple variations based on the design proposals; means for conducting multifaceted A / B testing using the variations; means for aggregating the results of the A / B testing and selecting the optimal variation; means for notifying the user of the optimal variation; means for adjusting the design proposals in real time based on the user's sentiment data and providing feedback; and means for using smart glasses, a head-mounted display, or a similar device to collect the user's sentiment data. This makes it possible to reflect the user's emotions in real time and quickly provide more satisfying services and campaigns.

[0879] "User input" refers to the information that users provide to the system regarding their service or campaign design proposals.

[0880] "Service and campaign design proposals" refer to the marketing and promotional plans and their specific details that users envision.

[0881] "Generative artificial intelligence" refers to algorithms that use machine learning and natural language processing to analyze user design proposals and provide evaluations and suggestions based on relevant laws and guidelines.

[0882] "Relevant laws and guidelines" refer to the laws and regulations and guidelines that must be followed when conducting the service or campaign.

[0883] "Evaluation results" refer to feedback from a generative artificial intelligence system that evaluates the design proposal and identifies any legal issues or areas for improvement.

[0884] "Emotional data" refers to information about a user's emotions collected from their facial expressions, voice, text, and other sources.

[0885] "Variations" refer to different versions of a campaign or promotion that are generated based on the user's design proposal.

[0886] A / B testing is a method of presenting multiple variations to different user groups and comparing their performance data.

[0887] "Performance data" refers to metrics used to measure the effectiveness of a campaign or service (e.g., click-through rate, registration rate, purchase rate, etc.).

[0888] The "optimal variation" refers to the campaign or promotion that, based on A / B test results and sentiment data, is judged to be the most effective and provides the highest user satisfaction.

[0889] "Smart glasses" are wearable devices designed to collect the user's emotions in real time.

[0890] A "head-mounted display" is a wearable device that collects emotional data while displaying information in the user's field of vision.

[0891] "Adjusting and providing feedback in real time" is a process of immediately reflecting user sentiment data to revise the design proposal and returning the results to the user.

[0892] This invention combines an emotion engine and generative artificial intelligence in a system that assists in the design of services and campaigns, providing a technology that designs optimal services and campaigns based on user emotional information.

[0893] System Configuration

[0894] Hardware and software:

[0895] 1. User terminal: A device including smart glasses or a head-mounted display. This device collects the user's facial expressions, voice, and text in real time and generates emotion data through an emotion engine (e.g., Microsoft Azure Emotion API).

[0896] 2. Server: A cloud platform (e.g., AWS) will be used to run generative artificial intelligence (e.g., OpenAI GPT). The server will process design proposals and sentiment data received from users and perform legal evaluations and design proposal optimization.

[0897] Software processing and data computation:

[0898] 1. Receiving user input:

[0899] Users interact with touchpoints within the virtual store and input design proposals for services and campaigns.

[0900] 2. Collection and transmission of emotional data:

[0901] Smart glasses and head-mounted displays capture the user's facial expressions, voice, and text, which are then analyzed as emotional data by an emotion engine.

[0902] The generated emotional data is sent to the cloud server along with the design proposal.

[0903] 3. Evaluation and adjustment of the design proposal:

[0904] The generative artificial intelligence within the server analyzes the design proposal and conducts a legal evaluation based on relevant laws and guidelines.

[0905] Based on emotional data, user satisfaction is assessed, and a design proposal with necessary adjustments is generated.

[0906] 4. Display of evaluation results and feedback:

[0907] The evaluation results from the server are sent to the user's terminal and displayed in real time on a virtual display.

[0908] We provide feedback on optimized campaign proposals based on user sentiment data.

[0909] 5. Variation generation and A / B testing:

[0910] Based on the refined design proposal, multiple variations are generated and presented to different user groups.

[0911] Conduct A / B testing in real time and collect performance and sentiment data.

[0912] 6. Selection and notification of the optimal variation:

[0913] Based on the collected data, statistical analysis is performed on the server to select the optimal variation.

[0914] The optimal variation is notified to the user's terminal and displayed on the virtual display.

[0915] Specific example

[0916] For example, a user designs a "Spring Bonus Points Campaign" and inputs their design proposal. At this time, an emotion engine is used to collect emotional data from the user's facial expressions and voice. A generative AI model is used to analyze "how to make the Spring Bonus Points Campaign legally compliant" and perform a legal evaluation. Based on this result, the optimal proposal (e.g., a 10% point increase) that reflects the user's emotional data is displayed in real time. Subsequently, A / B testing is conducted on the different variations (e.g., a 15% point increase, a 20% point increase, and one with a free coupon), and the "20% increase" with the highest performance is ultimately selected.

[0917] Example of a prompt:

[0918] "How can we make our spring bonus points campaign legally sound and offer exciting deals for users?"

[0919] Thus, the present invention enables the efficient design of optimal services and campaigns by combining evaluations of laws and guidelines with user sentiment information.

[0920] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0921] Step 1:

[0922] Receiving user input:

[0923] Users interact with interactive touchpoints within a virtual store to input service and campaign design proposals. This input data includes the campaign name, content, and terms and conditions. The terminal receives and stores this input data.

[0924] Step 2:

[0925] Collection of emotional data:

[0926] While the user inputs design proposals, smart glasses or a head-mounted display captures the user's facial expressions and voice in real time. This data is sent to an emotion engine (e.g., Microsoft Azure Emotion API) for analysis and collection as emotion data. The input data consists of the user's facial expressions and voice, while the output data represents the user's emotional state (e.g., joy, surprise, anxiety).

[0927] Step 3:

[0928] Sending design proposals and emotional data:

[0929] The terminal sends the user's input design proposal and collected sentiment data to the cloud server. The input data consists of the design proposal and sentiment data, while the output data is a request message containing these.

[0930] Step 4:

[0931] Legal evaluation of the design proposal:

[0932] The server uses generative artificial intelligence (e.g., OpenAI GPT) to evaluate the design proposal. For example, it takes the prompt "Does this design proposal comply with laws and guidelines?" as input, and the AI ​​model performs the analysis. The output is a legal evaluation result, indicating whether or not there are any problems.

[0933] Step 5:

[0934] Return and display of evaluation results:

[0935] The server returns the legal assessment results to the terminal. The terminal displays the assessment results in real time on the user's virtual display. The input data is the legal assessment results, and the output data is the information displayed to the user.

[0936] Step 6:

[0937] Generating adjustment proposals based on emotional data:

[0938] The server adjusts the design proposal based on emotional data. For example, it might input a prompt message to a generative artificial intelligence such as, "According to the user's emotional data, there are many concerns about the current design proposal, so I will suggest improvements." The output is an adjusted proposal, a specific suggestion that takes the user's emotions into consideration.

[0939] Step 7:

[0940] Generating multiple variations:

[0941] Based on the adjusted design proposal, the server generates multiple variations. These variations include different percentages of discounts and benefits. The input data is the adjusted proposal, and the output data is a set of variations.

[0942] Step 8:

[0943] Conducting A / B testing:

[0944] The device conducts A / B testing on different user groups using the generated variations. It collects performance data (e.g., click-through rate, purchase rate) and sentiment data for each variation. The input data consists of variation-to-user interaction data, and the output data consists of customized evaluation data.

[0945] Step 9:

[0946] Aggregation and optimization of test results:

[0947] The server statistically analyzes the collected performance and sentiment data to select the optimal variation. For example, it might input a prompt message to a generative artificial intelligence system such as, "Select the variation that shows the highest click-through rate and user satisfaction." The output data would then be the optimal variation.

[0948] Step 10:

[0949] Notification of optimal variation:

[0950] The server sends the optimal variation to the user's terminal and displays it on the virtual display. The input data is the optimal variation, and the output data is the optimal campaign proposal that is notified to the user.

[0951] This series of steps creates a system that reflects user emotions and delivers optimal services and campaigns in real time.

[0952] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0953] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0954] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0955] [Fourth Embodiment]

[0956] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0957] As shown in Figure 7, the 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.

[0958] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0959] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0960] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0961] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0962] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0963] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive 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 robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0964] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0965] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0966] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0967] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0968] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0969] This invention provides a system in which users input service and campaign design proposals, and the system designs the optimal service through verification of laws and guidelines using generative artificial intelligence and multifaceted A / B testing. The operation of the system in each processing step is described in detail below.

[0970] User input

[0971] Users input service and campaign design proposals using their devices. The input design proposals are then sent from the device to the server.

[0972] Checking laws and guidelines

[0973] The server sends the design proposal received from the user to a generative artificial intelligence (AI) system, which then performs an evaluation based on relevant laws and guidelines. The AI ​​analyzes the design proposal and determines whether there are any legal issues. The evaluation results are returned to the server, which then sends them to the terminal. The terminal then displays the evaluation results to the user.

[0974] Generating Variations

[0975] If the evaluation result is determined to be "no problems," the server generates multiple variations based on the design proposal. These variations refer to multiple proposals containing different settings and conditions, and the server sends these to the terminal.

[0976] Conducting A / B testing

[0977] The device sets up test scenarios based on the received variations and conducts A / B testing from multiple angles. Different variations are presented to multiple user groups, and test results are collected. For example, variations with 15% cashback, 20% cashback, and a free coupon are set up, and performance data (click-through rate, registration rate, purchase rate, etc.) is collected for each.

[0978] Summary of results and selection of the optimal variation

[0979] After the test is complete, the terminal sends the obtained performance data to the server. The server aggregates this data, analyzes it statistically, and selects the optimal variation.

[0980] Notification of optimal variation

[0981] Once the optimal variation is selected, the server sends the result to the terminal. The terminal then notifies the user of the optimal service design and releases the final service based on it.

[0982] Specific examples of operation

[0983] The following is a concrete example of how it works. The user designs a "new points reward campaign" and inputs it into the terminal. This design proposal is sent to the server, where it is legally checked by generative artificial intelligence. It is confirmed that it does not violate consumer protection laws or the Premiums and Representations Act, and the results are notified to the user. The server then generates variations such as 15% reward, 20% reward, and with a free coupon, and A / B testing is conducted on the terminal. The test results are sent to the server, which selects the "20% reward" option that is most popular with the user, and notifies the user of the result.

[0984] As described above, the system based on the present invention can efficiently perform verification in accordance with laws and guidelines, and design optimal services that reflect user preferences through A / B testing.

[0985] The following describes the processing flow.

[0986] Step 1:

[0987] The user inputs a design proposal for a service or campaign into the terminal. The terminal then sends the input design proposal to the server.

[0988] Step 2:

[0989] The server sends the received design proposal to the generative artificial intelligence. The generative AI analyzes the design proposal and conducts a legal evaluation based on relevant laws and guidelines.

[0990] Step 3:

[0991] The evaluation results from the generative artificial intelligence are returned to the server. The server receives the evaluation results and sends them to the terminal.

[0992] Step 4:

[0993] The device displays the evaluation results to the user. The user reviews the evaluation results and understands that there are no legal issues.

[0994] Step 5:

[0995] The server generates multiple variations based on the design proposal that was deemed "problem-free" in the evaluation results. These generated variations include different settings and conditions.

[0996] Step 6:

[0997] The server sends the generated variations to the device. The device then prepares to use the variations to conduct an A / B test.

[0998] Step 7:

[0999] The device presents variations to multiple user groups. For example, it might offer variations such as 15% cashback, 20% cashback, or a free coupon, and collects performance data (click-through rate, registration rate, purchase rate, etc.) for each variation.

[1000] Step 8:

[1001] Once the A / B test is complete, the device sends the collected performance data to the server. The server then aggregates and statistically analyzes the received data.

[1002] Step 9:

[1003] The server selects the most effective variation based on the aggregated data. Once the optimal variation is determined, the server sends the result to the terminal.

[1004] Step 10:

[1005] The device notifies the user of the optimal service design. Based on the notified optimal design, the user releases the final service or campaign.

[1006] The above describes the specific actions of each processing step in the program. This system makes it possible to comply with laws and guidelines and efficiently provide the best possible service to users.

[1007] (Example 1)

[1008] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1009] In recent years, when designing services and campaigns that appeal to consumers, users are required to experiment with multiple design options while avoiding legal risks. However, checking legal issues and rationally testing various variations to select the optimal design is a multifaceted and complex process that requires considerable effort and time. This has created a challenge in that it is difficult for users to quickly and efficiently find the optimal design.

[1010] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1011] In this invention, the server includes means for receiving design proposals for services and campaigns based on user input; means for transmitting the design proposals to a generative artificial intelligence and conducting evaluations based on relevant laws and guidelines; means for receiving evaluation results from the generative artificial intelligence and displaying them to the user; means for generating multiple variations based on the design proposals; means for conducting comparative tests from multiple perspectives using the variations; means for aggregating the results of the comparative tests and selecting the optimal variation; and means for notifying the user of the optimal variation. This makes it possible to efficiently design optimal services and campaigns while avoiding legal risks.

[1012] A "user" refers to an individual or legal entity that inputs design proposals for services or campaigns and utilizes the system.

[1013] "Terminal" refers to an input device used by a user, and includes electronic devices such as personal computers, smartphones, and tablets.

[1014] A "server" is a computer system that performs the central processing of a system and provides multiple functions in an integrated manner.

[1015] A "design proposal" refers to a plan that describes the specific content and conditions of a service or campaign that users will input.

[1016] "Generative artificial intelligence" refers to an artificial intelligence system that performs advanced analytical processing, such as legal evaluation and variation generation, from given input data.

[1017] "Laws and guidelines" refer to the laws and industry standards that apply to the service or campaign, including consumer protection laws and the Premiums and Representations Act.

[1018] "Evaluation results" refer to the results obtained when a generative artificial intelligence analyzes a design proposal and determines whether there are any legal issues or whether it conforms to guidelines.

[1019] "Variations" refer to multiple design options with different conditions and settings, generated based on the original design proposal.

[1020] A "comparative test" refers to a testing method in which multiple variations are presented to different user groups, and their response data is collected and compared.

[1021] "Performance data" refers to data showing user responses to each variation, including click-through rates, registration rates, and purchase rates.

[1022] The "optimal variation" refers to the variation that showed the highest performance data in comparative tests.

[1023] "Notifying" refers to the process of transmitting information from a server to a user via a terminal.

[1024] This invention relates to a system in which a user inputs a design proposal for a service or campaign, and the system performs optimal service design through legal evaluation and multifaceted comparative testing using generative artificial intelligence. Specific embodiments for carrying out this invention are described in detail below.

[1025] Users input service and campaign design proposals using their devices. For example, for a "new points reward campaign," they would describe the campaign details (campaign period, reward rate, etc.) in an input form on their device. This design proposal is then sent from the device to the server.

[1026] The server sends the received design proposal to a generative artificial intelligence (e.g., OpenAI GPT-4). The generative AI analyzes the design proposal, which is input as a prompt: "Please check whether this design proposal violates consumer protection laws or the Premiums and Representations Act." Based on the analysis, it determines whether there are any legal issues and returns the evaluation result to the server. The evaluation result is resent from the server to the terminal and displayed to the user on the terminal. For example, the evaluation result displayed might be: "This campaign does not violate consumer protection laws."

[1027] If the evaluation result is determined to be "no problems," the server generates multiple variations based on the design proposal. Generative artificial intelligence and algorithms are used to create variations with different settings and conditions. For example, it might generate three variations: "15% cashback," "20% cashback," and "with a free coupon." The generated variations are then sent from the server to the terminal.

[1028] The device sets up a comparative test scenario based on the received variations. This scenario presents different variations to multiple user groups and collects user response data (click-through rate, registration rate, purchase rate, etc.) for each variation. For example, user group A is presented with "15% cashback," user group B with "20% cashback," and user group C with "free coupon."

[1029] Once the results of each comparison test are collected, the device sends this performance data to the server. The server aggregates the data and performs statistical analysis to select the most effective variation. For example, by comparing the click-through rate and purchase rate of each variation, the "20% cashback" variation may be found to have the best performance. In this process, the R or Python pandas library may be used.

[1030] Once the optimal variation is selected, the server notifies the terminal of the result. The terminal then notifies the user of the optimal variation, and the user ultimately releases the service based on that variation. For example, the user might see a notification that says, "Test results indicate that a 20% cashback has been selected as the most effective variation."

[1031] In this way, the system based on the present invention can generate design proposals based on user input and efficiently design optimal services and campaigns through multifaceted comparative testing while avoiding legal risks.

[1032] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1033] Step 1:

[1034] Users input service and campaign design proposals using their devices. For example, they might describe a "new points reward campaign," detailing specific conditions such as the campaign period and reward rate. This design proposal is sent from the device to the server as input data. Once the server receives the design proposal as input, processing begins on the server side.

[1035] Step 2:

[1036] The server sends the received design proposal to a generative artificial intelligence (e.g., OpenAI GPT-4). Specifically, the design proposal is used as input data with the prompt message "Please check whether this design proposal violates consumer protection laws or the Premiums and Representations Act." The generative AI model analyzes this input data and determines whether there are any legal issues. The evaluation result is sent back to the server as output, and this result serves as information to confirm whether there are any legal problems.

[1037] Step 3:

[1038] The server receives the evaluation results from the generative artificial intelligence and sends them to the user's terminal for them to review. The terminal displays the legal evaluation results to the user, for example, a message such as "This campaign does not violate consumer protection laws." Information to be notified to the user is generated based on the evaluation results as input.

[1039] Step 4:

[1040] If the evaluation result is determined to be "no problems," the server generates multiple variations based on the design proposal. Generative artificial intelligence and algorithms are used to create multiple variations with different settings and conditions. For example, variations such as "15% cashback," "20% cashback," and "with a free coupon" may be included. The generated variations are sent from the server to the terminal as output data, preparing it for the next test.

[1041] Step 5:

[1042] The device sets up a comparative test scenario based on the received variations. Specifically, it creates test cases that present different variations to multiple user groups. User group A is presented with "15% cashback," user group B with "20% cashback," and user group C with "free coupon." A multifaceted comparative test is performed based on the variations as input. The output is the collection of response data from each group.

[1043] Step 6:

[1044] Once the results of the comparative tests are collected, the device sends this performance data (e.g., click-through rate, registration rate, purchase rate) to the server. The server aggregates this data and performs statistical analysis. Based on the performance data as input, it selects the most effective variation. For example, by comparing the click-through rate and purchase rate of each variation, the results might show that "20% cashback" performs the best. As output, the optimal variation is selected.

[1045] Step 7:

[1046] Once the optimal variation is selected, the server notifies the terminal of the result. The terminal then notifies the user of the optimal variation, and the user ultimately releases the service based on that variation. For example, a notification might appear stating, "Based on the test results, a 20% cashback has been selected as the most effective variation." This provides the user with information to take their next action based on the test results as input.

[1047] (Application Example 1)

[1048] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1049] Traditional advertising campaign design often involved manual processes for checking legal issues, generating multiple variations, testing, and selecting the optimal variation, resulting in inefficiencies and time-consuming tasks. Furthermore, there was a lack of systems providing user-friendly interfaces using smart devices. This made it difficult to quickly and effectively maximize marketing effectiveness.

[1050] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1051] In this invention, the server includes means for receiving service and campaign design proposals based on user input; means for transmitting the design proposals to a generative artificial intelligence and conducting evaluations based on relevant regulations and guidelines; means for receiving evaluation results from the generative artificial intelligence and displaying them to the user; means for generating multiple options based on the design proposals; means for conducting multifaceted A / B testing using the options; means for aggregating the results of the A / B testing and selecting the optimal option; means for notifying the user of the optimal option; and means including an application installed on a smart device. This enables efficient advertising campaign design, legal review, variation generation, testing, and optimization, and maximizes marketing effectiveness with a user-friendly interface.

[1052] A "user" refers to a person who uses the system to design services and campaigns.

[1053] "Input" refers to the act of a user providing design proposals and related information to the system.

[1054] A "design proposal" refers to information that shows the specific details of a service or campaign proposed by a user.

[1055] "Generative artificial intelligence" refers to artificial intelligence technology that uses user-inputted design proposals to check relevant regulations and guidelines, and conducts multifaceted A / B testing.

[1056] "Evaluation" refers to the process by which a generative artificial intelligence determines whether a design proposal has legal issues or complies with guidelines.

[1057] "Legal issues" refer to the question of whether the proposed design violates legal regulations.

[1058] "Regulations" refer to laws and guidelines that apply to the design of a service or campaign.

[1059] "Options" refers to multiple variations or different proposals with different conditions that are generated based on a design proposal.

[1060] "AB testing" refers to a testing method that presents multiple options to different user groups and compares and evaluates the performance of each option.

[1061] "Aggregation" refers to the act of statistically compiling the results data of an A / B test.

[1062] The "optimal choice" refers to the variation that shows the highest performance based on the results of the A / B test.

[1063] "Smart devices" refer to advanced devices such as smartphones, smart glasses, and head-mounted displays.

[1064] This invention involves several key steps to realize a system in which users input service and campaign design proposals using a smart device. These steps are described below.

[1065] First, the user uses a smart device such as a smartphone to input their service or campaign design proposal. The user enters the details of the specific design proposal and sends it from the device to the server.

[1066] Next, the server sends the received design proposal to the generative artificial intelligence (AI) and performs an evaluation based on relevant regulations and guidelines. The generative AI analyzes the design proposal and determines whether there are any legal issues. This legal evaluation includes consumer protection laws and the Premiums and Representations Act, among others. The evaluation results of the generative AI are returned to the server, which then sends the results to the user's terminal and displays them to the user.

[1067] Next, if the server determines the evaluation result is "no problems," it generates multiple options (variations) based on the design proposal. These variations have different settings and conditions, and may include options such as "15% cashback," "20% cashback," and "with a free coupon." These variations are then sent from the server to the terminal.

[1068] The device conducts multifaceted A / B testing based on the variations it receives. This test presents different variations to multiple user groups and collects performance data (click-through rates, registration rates, purchase rates, etc.) resulting from these tests. This data is then sent to the server.

[1069] The server aggregates the received performance data, analyzes it statistically, and selects the optimal option. The selected optimal option is then notified from the server to the terminal, informing the user of the best possible service design.

[1070] As a concrete example, a user designs a "new points reward campaign" and inputs a design proposal for a "20% cashback campaign." Generative artificial intelligence legally evaluates this proposal and confirms that it does not violate consumer protection laws or the Premiums and Representations Act. The server then generates variations such as "15% cashback," "20% cashback," and "with free coupon," and A / B testing is conducted on the terminal. The test results are sent to the server, which selects the "20% cashback" option, which showed the highest performance, as the optimal choice and notifies the user of the result.

[1071] Examples of prompt messages are as follows:

[1072] We are designing a new points reward campaign. Please perform legal checks and generate variations based on the details below.

[1073] Campaign type: Point rewards

[1074] Details: 20% cashback campaign

[1075] Thus, the system based on this invention maximizes marketing effectiveness by quickly and efficiently evaluating user-inputted design proposals and providing the optimal service design.

[1076] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1077] Step 1:

[1078] Users input service and campaign design proposals using a smart device. Specifically, users open a dedicated application on their smartphone, input their design proposal (e.g., "20% Cashback Campaign"), and press the submit button. Input data: Campaign details. Output data: Confirmation message for submitting the design proposal.

[1079] Step 2:

[1080] The server receives design proposals submitted by the user. The server then forwards the input to a generative artificial intelligence system, which conducts a legal evaluation based on relevant regulations and guidelines. This legal evaluation includes checks regarding consumer protection laws and the Premiums and Representations Act. Input data: Design proposals. Data processing: Evaluation based on legal regulations. Output data: Legal evaluation results.

[1081] Step 3:

[1082] Generative artificial intelligence analyzes received design proposals and determines whether they have any legal issues. Generative AI uses natural language processing and rule-based analysis to check which regulations or guidelines the design proposals violate. Input data: Design proposal. Data calculation: Comparison with legal regulations. Output data: Result of legal evaluation (e.g., "No issues").

[1083] Step 4:

[1084] The server receives evaluation results from the generative artificial intelligence, sends them to the user's terminal, and displays them to the user. The displayed content is the result of the legal evaluation and includes instructions for when there are no problems or when corrections are needed. Input data: Legal evaluation results. Output data: Content displayed to the user.

[1085] Step 5:

[1086] If the evaluation result is determined to be "no problems," the server generates multiple variations based on the design proposal. These variations include different reward rates and benefits. The generated variations are sent to the terminal. Input data: Evaluated design proposal. Data processing: Generation of variations. Output data: Multiple variations.

[1087] Step 6:

[1088] The device conducts multifaceted A / B testing based on the received variations. Specifically, it presents different variations to user groups and collects performance data such as click-through rates and purchase rates. Input data: Multiple variations. Data processing: Collection of user behavior. Output data: Performance data (e.g., click-through rate, purchase rate).

[1089] Step 7:

[1090] The terminal sends the collected performance data to the server. The server aggregates the received performance data, performs statistical analysis, and selects the optimal variation. Input data: Performance data. Data calculation: Statistical analysis. Output data: Result of the selection of the optimal variation.

[1091] Step 8:

[1092] The server selects the optimal variation and notifies the user of the result on their device. The user receives the notification and confirms the optimal campaign design. Input data: Selection result. Output data: Notification content to the user (e.g., "20% cashback is optimal").

[1093] By following these steps, users can efficiently design legally compliant and optimal advertising campaigns.

[1094] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1095] This invention provides a technology that combines an emotion engine with a system that supports the design of services and campaigns, enabling the design of optimal services and campaigns based on user emotion information. The operation of the system in each processing step is described in detail below.

[1096] User input

[1097] Users input service and campaign design proposals using a device. During this process, the device utilizes its built-in emotion engine to recognize emotions from the user's facial expressions, voice, and text, and collects that emotion data.

[1098] Checking laws and guidelines

[1099] The server transmits the design proposal and emotional data received from the user to a generative artificial intelligence (AI) system, which then performs an evaluation based on relevant laws and guidelines. The AI ​​analyzes the design proposal and conducts a legal evaluation, taking the emotional data into consideration. The evaluation results are returned to the server, which then transmits them to the terminal.

[1100] Displaying results and reflecting sentiment data

[1101] The device displays the evaluation results to the user. At the same time, it adjusts the design proposal as needed based on sentiment data and provides feedback in a way that is appropriate for the user. For example, if a campaign suggested by the user does not align with the user's preferences or expectations, the sentiment engine will use that information to suggest adjustments.

[1102] Generating Variations

[1103] The server generates multiple variations based on the refined design proposal. These variations include different settings and conditions, and are intended for multifaceted evaluation.

[1104] Conducting A / B testing

[1105] The device will conduct A / B testing using the generated variations. Multiple user groups will be presented with the variations, and performance data for each variation (click-through rate, registration rate, purchase rate, etc.) will be collected. During the test, an emotion engine will be used to evaluate how users feel about the variations.

[1106] Summary of results and selection of the optimal variation

[1107] After the test is complete, the device sends the collected performance and sentiment data to the server. The server aggregates and statistically analyzes this data to select the optimal variation. Sentiment data is considered an important indicator for evaluating user satisfaction and motivation.

[1108] Notification of optimal variation

[1109] Once the optimal variation is selected, the server sends the result to the terminal. The terminal notifies the user of the optimal service design and releases the final service or campaign based on it. An optimal design that reflects user sentiment information increases user satisfaction and reduces the risk of complaints.

[1110] Specific examples of operation

[1111] The following is a concrete example of how it works. The user designs a "new points reward campaign" and inputs it into the terminal. The emotion engine recognizes the user's emotions and sends it to the server along with the design proposal. A legal check is performed by generative artificial intelligence, and the evaluation results are notified to the user, while adjustments based on the emotion data are suggested. Subsequently, variations such as 15% reward, 20% reward, and with a free coupon are generated, and A / B testing is conducted. Based on the test results and emotion data, the server determines that "20% reward" is optimal, and this result is notified to the user.

[1112] As described above, the system based on the present invention can efficiently design optimal services and campaigns by combining evaluations of laws and guidelines with user sentiment information.

[1113] The following describes the processing flow.

[1114] Step 1:

[1115] The user inputs design proposals for services and campaigns using a device. Simultaneously with the user's input, the device activates an emotion engine, recognizing emotions from the user's facial expressions, voice, and text, and collecting emotion data.

[1116] Step 2:

[1117] The terminal sends the input design proposal and collected emotional data to the server. The server sends the received data to a generative artificial intelligence system and requests an evaluation based on relevant laws and guidelines.

[1118] Step 3:

[1119] The generative artificial intelligence analyzes design proposals and emotional data to perform legal evaluations. Simultaneously, it evaluates the user's emotional state, taking emotional data into consideration, and returns the results to the server.

[1120] Step 4:

[1121] The server sends the legal and emotional evaluation results received from the generative artificial intelligence to the terminal. The terminal then displays the results to the user.

[1122] Step 5:

[1123] The device adjusts the design proposal based on emotional data as needed. If the user's emotional state is negative, the device fine-tunes the design proposal based on suggestions from generative artificial intelligence and presents it to the user again.

[1124] Step 6:

[1125] The server generates multiple variations based on the refined design proposal. These generated variations contain different settings and conditions, and the server sends them to the terminal.

[1126] Step 7:

[1127] The device sets up test scenarios based on the generated variations and conducts A / B testing. Different variations are presented to multiple user groups, and performance and sentiment data for each variation are collected.

[1128] Step 8:

[1129] After the A / B test is complete, the device sends the collected performance and sentiment data to the server. The server aggregates the received data, performs statistical analysis, and selects the optimal variation.

[1130] Step 9:

[1131] The server selects the variation that elicited the most effective and emotionally positive response based on the aggregated data. Once the optimal variation is determined, it sends the result to the terminal.

[1132] Step 10:

[1133] The device notifies the user of the optimal service design. Based on the notified optimal design, the user releases the final service or campaign.

[1134] The above outlines the specific processing steps of this system, which incorporates an emotion engine. This enables the efficient design of optimal services and campaigns that comply with laws and guidelines and reflect user emotions.

[1135] (Example 2)

[1136] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1137] Traditional campaign and service design lacks the technology to incorporate user emotional information and select the optimal variations. Therefore, designing without considering user emotional satisfaction has failed to adequately improve actual customer satisfaction. Furthermore, the need to separately evaluate laws and guidelines complicates the design process. This invention aims to solve these problems.

[1138] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1139] In this invention, the server includes means for receiving service and campaign design proposals based on user input, means for transmitting the design proposals to a generative artificial intelligence and conducting evaluations based on relevant laws and guidelines, means for receiving evaluation results from the generative artificial intelligence and displaying them to the user, means for generating multiple variations based on the design proposals, means for conducting multifaceted A / B testing using the variations, means for aggregating the results of the A / B testing and selecting the optimal variation, means for aggregating user emotional information and reflecting it in the selection of the optimal variation, and means for recognizing user emotions using an emotion engine and providing feedback to the design proposals. This makes it possible to design optimal services and campaigns that reflect user emotional information, improve user satisfaction, and realize an efficient design process that also ensures legal compliance.

[1140] "User input" refers to the act of a user using a device to input design proposals for a service or campaign.

[1141] "Generative artificial intelligence" refers to artificial intelligence models that analyze design proposals and evaluate them based on relevant laws and guidelines.

[1142] A "variation" refers to multiple proposals with different settings and conditions based on a design proposal for a service or campaign.

[1143] A / B testing is an evaluation method that involves presenting multiple variations to user groups and collecting performance data.

[1144] "Emotional information" refers to data about a user's emotional state collected through their facial expressions, voice, text, etc.

[1145] An "emotion engine" is a system that recognizes a user's emotions and analyzes that emotional data.

[1146] "Legal evaluation" is the process by which generative artificial intelligence determines the suitability of a design proposal based on relevant laws and guidelines.

[1147] "Performance data" refers to experimental results such as click-through rates, registration rates, and purchase rates collected during A / B testing.

[1148] A "design proposal" refers to the specific suggestions a user has entered regarding a service or campaign.

[1149] The "optimal variation" is the variation that, based on aggregated performance and sentiment data, is judged to be the most effective and provide the highest user satisfaction.

[1150] This invention provides a technology that combines an emotion engine with a system that supports the design of services and campaigns, enabling the design of optimal services and campaigns based on user emotion information. The operation of the system in each processing step is described in detail below.

[1151] First, the user inputs their service or campaign design proposal using a device. This device utilizes a built-in emotion engine to recognize emotions from the user's facial expressions, voice, and text, and collects that emotion data. The emotion engine can utilize either the "Emotion API" or "Azure Cognitive Services."

[1152] Next, the terminal sends the design proposals and sentiment data collected from the user to the server. The server sends this data to a generative artificial intelligence (e.g., GPT-4) which then performs an evaluation based on relevant laws and guidelines. The generative AI analyzes the design proposals and conducts a legal evaluation that takes the sentiment data into account.

[1153] Specifically, one could send the following prompt message to the generative artificial intelligence.

[1154] "Please evaluate the following campaign design proposal based on laws and guidelines, taking into account user sentiment data. Proposal: New points reward campaign. User sentiment data: 80% positive responses, 20% negative responses."

[1155] The evaluation results from the generative artificial intelligence are returned to the server, which then sends the results to the terminal. The terminal displays the evaluation results to the user and, at the same time, adjusts the design proposal as needed based on the sentiment data, providing feedback in a way that is appropriate for the user. For example, if a campaign proposed by the user does not align with the user's preferences or expectations, the sentiment engine will use that information to suggest adjustments.

[1156] Next, the server generates multiple variations based on the refined design proposal. The generated variations include different settings and conditions, and are intended for multifaceted evaluation. For example, variations such as "15% cashback," "20% cashback," and "with free coupon" are generated.

[1157] Subsequently, the device conducts A / B testing using the generated variations. The variations are presented to multiple user groups, and performance data (click-through rate, registration rate, purchase rate, etc.) for each variation is collected. The sentiment engine is also used during the test to evaluate how users feel about each variation.

[1158] Once the test is complete, the device sends the collected performance and sentiment data to the server. The server statistically analyzes this data and selects the optimal variation. Sentiment data is an important indicator for evaluating user satisfaction and motivation.

[1159] Once the optimal variation is selected, the server sends the result to the terminal. The terminal notifies the user of the optimal service design and releases the final service or campaign based on it. For example, the server selects "20% cashback" as the optimal variation and notifies the user of the result.

[1160] In this way, the system based on the present invention can efficiently design optimal services and campaigns by reflecting user emotional information and combining it with legal evaluations. This makes it possible to improve user satisfaction and reduce the risk of complaints.

[1161] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1162] Step 1:

[1163] User input

[1164] The user uses the device to input design proposals for services and campaigns. Specifically, the user designs a "new points reward campaign" and inputs it into the device. Based on this input, the device uses its built-in emotion engine to recognize the user's emotions from their facial expressions, voice, and text, and collects that emotion data. The input includes the text data of the design proposal and the user's emotion data (e.g., 80% positive reactions, 20% negative reactions). The device collects this data and sends it to the server.

[1165] Step 2:

[1166] Checking laws and guidelines

[1167] The server sends the design proposal and sentiment data received from the terminal to the generative artificial intelligence (AI). The input for this process consists of the text data of the design proposal and the sentiment data. The generative AI analyzes the design proposal and performs an evaluation based on relevant laws and guidelines. Specifically, an example of sending a prompt to the generative AI is as follows:

[1168] "Please evaluate the following campaign design proposal based on laws and guidelines, taking into account user sentiment data. Proposal: New points reward campaign. User sentiment data: 80% positive responses, 20% negative responses."

[1169] The output is a legal evaluation result. The evaluation result is returned to the server, which then sends the result to the terminal. This result may include information such as, "The point redemption rate for this campaign does not violate current laws."

[1170] Step 3:

[1171] Displaying results and reflecting sentiment data

[1172] The device displays the received evaluation results to the user. Inputs include evaluation results obtained from the server and user sentiment data. Based on these inputs, the device adjusts the design proposal as needed and provides appropriate feedback to the user. Specifically, if the proposed campaign does not align with the user's preferences or expectations, the sentiment engine uses this information to suggest adjustments. For example, it might output new variations such as 15% or 20% cashback offers.

[1173] Step 4:

[1174] Generating Variations

[1175] The server generates multiple variations based on the adjusted design proposal. The input is the text data of the adjusted design proposal. Based on this, the server generates variations with different settings and conditions. Specifically, it generates variations such as "15% cashback," "20% cashback," and "with free coupon," and outputs these variation proposals.

[1176] Step 5:

[1177] Conducting A / B testing

[1178] The device conducts A / B testing using the generated variations. The input is design data for the variations. The device presents each variation to multiple user groups and collects performance data such as click-through rates, registration rates, and purchase rates. It also uses an emotion engine to evaluate users' emotional responses. The output includes performance data and emotional data for each variation. For example, user group A is presented with "15% cashback" and user group B with "20% cashback," and the response and performance data for each group are collected.

[1179] Step 6:

[1180] Summary of results and selection of the optimal variation

[1181] After the test is complete, the device sends the collected performance and sentiment data to the server. The inputs are the A / B test results and sentiment data. The server statistically analyzes this data and selects the optimal variation. The output is the data for the selected optimal variation (e.g., "20% cashback" is optimal).

[1182] Step 7:

[1183] Notification of optimal variation

[1184] The server sends the result of the selected optimal variation to the terminal. The input includes the data of the optimal variation selected by the server. The terminal notifies the user of this, and the user uses this data to release the final service or campaign. For example, if the server selects "20% cashback" as the optimal variation and notifies the user of this result, the user can implement a "20% cashback campaign."

[1185] (Application Example 2)

[1186] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1187] Traditional service and campaign design support systems have the drawback of failing to take user emotions into account, thus hindering their ability to maximize user satisfaction. Furthermore, the difficulty in obtaining real-time feedback makes it challenging to quickly deliver campaigns tailored to user needs. Moreover, the lack of a means to adjust design proposals based on emotional data prevents them from addressing the diverse emotions of users.

[1188] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1189] In this invention, the server includes means for receiving service and campaign design proposals based on user input; means for transmitting the design proposals to a generative artificial intelligence and conducting evaluations based on relevant laws and guidelines; means for receiving evaluation results from the generative artificial intelligence and displaying them to the user; means for collecting sentiment data based on the evaluation results; means for generating multiple variations based on the design proposals; means for conducting multifaceted A / B testing using the variations; means for aggregating the results of the A / B testing and selecting the optimal variation; means for notifying the user of the optimal variation; means for adjusting the design proposals in real time based on the user's sentiment data and providing feedback; and means for using smart glasses, a head-mounted display, or a similar device to collect the user's sentiment data. This makes it possible to reflect the user's emotions in real time and quickly provide more satisfying services and campaigns.

[1190] "User input" refers to the information that users provide to the system regarding their service or campaign design proposals.

[1191] "Service and campaign design proposals" refer to the marketing and promotional plans and their specific details that users envision.

[1192] "Generative artificial intelligence" refers to algorithms that use machine learning and natural language processing to analyze user design proposals and provide evaluations and suggestions based on relevant laws and guidelines.

[1193] "Relevant laws and guidelines" refer to the laws and regulations and guidelines that must be followed when conducting the service or campaign.

[1194] "Evaluation results" refer to feedback from a generative artificial intelligence system that evaluates the design proposal and identifies any legal issues or areas for improvement.

[1195] "Emotional data" refers to information about a user's emotions collected from their facial expressions, voice, text, and other sources.

[1196] "Variations" refer to different versions of a campaign or promotion that are generated based on the user's design proposal.

[1197] A / B testing is a method of presenting multiple variations to different user groups and comparing their performance data.

[1198] "Performance data" refers to metrics used to measure the effectiveness of a campaign or service (e.g., click-through rate, registration rate, purchase rate, etc.).

[1199] The "optimal variation" refers to the campaign or promotion that, based on A / B test results and sentiment data, is judged to be the most effective and provides the highest user satisfaction.

[1200] "Smart glasses" are wearable devices designed to collect the user's emotions in real time.

[1201] A "head-mounted display" is a wearable device that collects emotional data while displaying information in the user's field of vision.

[1202] "Adjusting and providing feedback in real time" is a process of immediately reflecting user sentiment data to revise the design proposal and returning the results to the user.

[1203] This invention combines an emotion engine and generative artificial intelligence in a system that assists in the design of services and campaigns, providing a technology that designs optimal services and campaigns based on user emotional information.

[1204] System Configuration

[1205] Hardware and software:

[1206] 1. User terminal: A device including smart glasses or a head-mounted display. This device collects the user's facial expressions, voice, and text in real time and generates emotion data through an emotion engine (e.g., Microsoft Azure Emotion API).

[1207] 2. Server: A cloud platform (e.g., AWS) will be used to run generative artificial intelligence (e.g., OpenAI GPT). The server will process design proposals and sentiment data received from users and perform legal evaluations and design proposal optimization.

[1208] Software processing and data computation:

[1209] 1. Receiving user input:

[1210] Users interact with touchpoints within the virtual store and input design proposals for services and campaigns.

[1211] 2. Collection and transmission of emotional data:

[1212] Smart glasses and head-mounted displays capture the user's facial expressions, voice, and text, which are then analyzed as emotional data by an emotion engine.

[1213] The generated emotional data is sent to the cloud server along with the design proposal.

[1214] 3. Evaluation and adjustment of the design proposal:

[1215] The generative artificial intelligence within the server analyzes the design proposal and conducts a legal evaluation based on relevant laws and guidelines.

[1216] Based on emotional data, user satisfaction is assessed, and a design proposal with necessary adjustments is generated.

[1217] 4. Display of evaluation results and feedback:

[1218] The evaluation results from the server are sent to the user's terminal and displayed in real time on a virtual display.

[1219] We provide feedback on optimized campaign proposals based on user sentiment data.

[1220] 5. Variation generation and A / B testing:

[1221] Based on the refined design proposal, multiple variations are generated and presented to different user groups.

[1222] Conduct A / B testing in real time and collect performance and sentiment data.

[1223] 6. Selection and notification of the optimal variation:

[1224] Based on the collected data, statistical analysis is performed on the server to select the optimal variation.

[1225] The optimal variation is notified to the user's terminal and displayed on the virtual display.

[1226] Specific example

[1227] For example, a user designs a "Spring Bonus Points Campaign" and inputs their design proposal. At this time, an emotion engine is used to collect emotional data from the user's facial expressions and voice. A generative AI model is used to analyze "how to make the Spring Bonus Points Campaign legally compliant" and perform a legal evaluation. Based on this result, the optimal proposal (e.g., a 10% point increase) that reflects the user's emotional data is displayed in real time. Subsequently, A / B testing is conducted on the different variations (e.g., a 15% point increase, a 20% point increase, and one with a free coupon), and the "20% increase" with the highest performance is ultimately selected.

[1228] Example of a prompt:

[1229] "How can we make our spring bonus points campaign legally sound and offer exciting deals for users?"

[1230] Thus, the present invention enables the efficient design of optimal services and campaigns by combining evaluations of laws and guidelines with user sentiment information.

[1231] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1232] Step 1:

[1233] Receiving user input:

[1234] Users interact with interactive touchpoints within a virtual store to input service and campaign design proposals. This input data includes the campaign name, content, and terms and conditions. The terminal receives and stores this input data.

[1235] Step 2:

[1236] Collection of emotional data:

[1237] While the user inputs design proposals, smart glasses or a head-mounted display captures the user's facial expressions and voice in real time. This data is sent to an emotion engine (e.g., Microsoft Azure Emotion API) for analysis and collection as emotion data. The input data consists of the user's facial expressions and voice, while the output data represents the user's emotional state (e.g., joy, surprise, anxiety).

[1238] Step 3:

[1239] Sending design proposals and emotional data:

[1240] The terminal sends the user's input design proposal and collected sentiment data to the cloud server. The input data consists of the design proposal and sentiment data, while the output data is a request message containing these.

[1241] Step 4:

[1242] Legal evaluation of the design proposal:

[1243] The server uses generative artificial intelligence (e.g., OpenAI GPT) to evaluate the design proposal. For example, it takes the prompt "Does this design proposal comply with laws and guidelines?" as input, and the AI ​​model performs the analysis. The output is a legal evaluation result, indicating whether or not there are any problems.

[1244] Step 5:

[1245] Return and display of evaluation results:

[1246] The server returns the legal assessment results to the terminal. The terminal displays the assessment results in real time on the user's virtual display. The input data is the legal assessment results, and the output data is the information displayed to the user.

[1247] Step 6:

[1248] Generating adjustment proposals based on emotional data:

[1249] The server adjusts the design proposal based on emotional data. For example, it might input a prompt message to a generative artificial intelligence such as, "According to the user's emotional data, there are many concerns about the current design proposal, so I will suggest improvements." The output is an adjusted proposal, a specific suggestion that takes the user's emotions into consideration.

[1250] Step 7:

[1251] Generating multiple variations:

[1252] Based on the adjusted design proposal, the server generates multiple variations. These variations include different percentages of discounts and benefits. The input data is the adjusted proposal, and the output data is a set of variations.

[1253] Step 8:

[1254] Conducting A / B testing:

[1255] The device conducts A / B testing on different user groups using the generated variations. It collects performance data (e.g., click-through rate, purchase rate) and sentiment data for each variation. The input data consists of variation-to-user interaction data, and the output data consists of customized evaluation data.

[1256] Step 9:

[1257] Aggregation and optimization of test results:

[1258] The server statistically analyzes the collected performance and sentiment data to select the optimal variation. For example, it might input a prompt message to a generative artificial intelligence system such as, "Select the variation that shows the highest click-through rate and user satisfaction." The output data would then be the optimal variation.

[1259] Step 10:

[1260] Notification of optimal variation:

[1261] The server sends the optimal variation to the user's terminal and displays it on the virtual display. The input data is the optimal variation, and the output data is the optimal campaign proposal that is notified to the user.

[1262] This series of steps creates a system that reflects user emotions and delivers optimal services and campaigns in real time.

[1263] 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 controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1264] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1265] 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 this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[1266] Furthermore, the emotion identification model 59, acting 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 a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1267] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[1268] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[1269] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[1270] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[1271] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is 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 the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[1272] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[1273] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[1274] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[1275] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[1276] 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.

[1277] Furthermore, it is not necessary to store the entirety 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 the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[1278] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[1279] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of 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). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[1280] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[1281] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[1282] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[1283] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted as being incorporated by reference.

[1284] The following is further disclosed regarding the embodiments described above.

[1285] (Claim 1)

[1286] A means of receiving service and campaign design proposals based on user input,

[1287] A means of transmitting the aforementioned design proposal to a generative artificial intelligence and conducting an evaluation based on relevant laws and guidelines,

[1288] A means of receiving evaluation results from a generative artificial intelligence and displaying them to the user,

[1289] Means for generating multiple variations based on the aforementioned design proposal,

[1290] A means for conducting A / B testing from multiple angles using the aforementioned variations,

[1291] A means for aggregating the results of the aforementioned A / B test and selecting the optimal variation,

[1292] A system that includes this.

[1293] (Claim 2)

[1294] The system according to claim 1, in which a generative artificial intelligence evaluates proposed service or campaign designs based on relevant laws and guidelines and determines whether or not there are any legal issues.

[1295] (Claim 3)

[1296] The system according to claim 1, which aggregates the results of an A / B test and notifies the user of the optimal variation.

[1297] "Example 1"

[1298] (Claim 1)

[1299] A means of receiving service and campaign design proposals based on user input,

[1300] A means of transmitting the aforementioned design proposal to a generative artificial intelligence and conducting an evaluation based on relevant laws and guidelines,

[1301] A means of receiving evaluation results from a generative artificial intelligence and displaying them to the user,

[1302] Means for generating multiple variations based on the aforementioned design proposal,

[1303] A means for conducting comparative tests from multiple perspectives using the aforementioned variations,

[1304] A means for compiling the results of the aforementioned comparative tests and selecting the optimal variation,

[1305] A means of notifying the user of the optimal variation,

[1306] A system that includes this.

[1307] (Claim 2)

[1308] The system according to claim 1, in which a generative artificial intelligence evaluates proposed service or campaign designs based on relevant laws and guidelines and determines whether or not there are any legal issues.

[1309] (Claim 3)

[1310] The system according to claim 1, which aggregates the results of comparative tests and notifies the user of the optimal variation.

[1311] "Application Example 1"

[1312] New Claim

[1313] (Claim 1)

[1314] A means of receiving service and campaign design proposals based on user input,

[1315] A means for transmitting the aforementioned design proposal to a generative artificial intelligence and conducting an evaluation based on relevant regulations and guidelines,

[1316] A means of receiving evaluation results from a generative artificial intelligence and displaying them to the user,

[1317] Means for generating multiple options based on the aforementioned design proposal,

[1318] A means of conducting A / B testing from multiple perspectives using the aforementioned options,

[1319] A means for aggregating the results of the aforementioned A / B test and selecting the optimal option,

[1320] A means of notifying the user of the optimal option,

[1321] Means including an application installed on a smart device,

[1322] A system that includes this.

[1323] (Claim 2)

[1324] The system according to claim 1, in which a generative artificial intelligence evaluates proposed service or campaign designs based on relevant regulations and guidelines and determines whether or not there are any legal issues.

[1325] (Claim 3)

[1326] The system according to claim 1, which aggregates the results of an A / B test and notifies the user of the optimal choice.

[1327] "Example 2 of combining an emotion engine"

[1328] (Claim 1)

[1329] A means of receiving service and campaign design proposals based on user input,

[1330] A means of transmitting the aforementioned design proposal to a generative artificial intelligence and conducting an evaluation based on relevant laws and guidelines,

[1331] A means of receiving evaluation results from a generative artificial intelligence and displaying them to the user,

[1332] Means for generating multiple variations based on the aforementioned design proposal,

[1333] A means for conducting A / B testing from multiple angles using the aforementioned variations,

[1334] A means for aggregating the results of the aforementioned A / B test and selecting the optimal variation,

[1335] A means of aggregating user sentiment information and reflecting it in the selection of the optimal variation,

[1336] A means of recognizing user emotions using an emotion engine and providing feedback to design proposals,

[1337] A system that includes this.

[1338] (Claim 2)

[1339] The system according to claim 1, in which a generative artificial intelligence evaluates proposed service or campaign designs based on relevant laws and guidelines and determines whether or not there are any legal issues.

[1340] (Claim 3)

[1341] The system according to claim 1, which aggregates the results of an A / B test and sentiment data, and notifies the user of the optimal variation.

[1342] "Application example 2 of combining emotional engines"

[1343] (Claim 1)

[1344] A means of receiving service and campaign design proposals based on user input,

[1345] A means of transmitting the aforementioned design proposal to a generative artificial intelligence and conducting an evaluation based on relevant laws and guidelines,

[1346] A means of receiving evaluation results from a generative artificial intelligence and displaying them to the user,

[1347] A means for collecting emotional data based on the aforementioned evaluation results,

[1348] Means for generating multiple variations based on the aforementioned design proposal,

[1349] A means for conducting A / B testing from multiple angles using the aforementioned variations,

[1350] A means for aggregating the results of the aforementioned A / B test and selecting the optimal variation,

[1351] A means for notifying the user of the aforementioned optimal variation,

[1352] A means of adjusting the design proposal in real time based on the aforementioned user sentiment data and providing feedback,

[1353] Means of using smart glasses, head-mounted displays, or similar devices to collect user emotional data,

[1354] A system that includes this.

[1355] (Claim 2)

[1356] The system according to claim 1, in which a generative artificial intelligence evaluates proposed service or campaign designs based on relevant laws and guidelines and determines whether or not there are any legal issues.

[1357] (Claim 3)

[1358] The system according to claim 1, which aggregates the results of an A / B test and notifies the user of the optimal variation. [Explanation of symbols]

[1359] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of receiving service and campaign design proposals based on user input, A means of transmitting the aforementioned design proposal to a generative artificial intelligence and conducting an evaluation based on relevant laws and guidelines, A means of receiving evaluation results from a generative artificial intelligence and displaying them to the user, Means for generating multiple variations based on the aforementioned design proposal, A means for conducting A / B testing from multiple angles using the aforementioned variations, A means for aggregating the results of the aforementioned A / B test and selecting the optimal variation, A system that includes this.

2. The system according to claim 1, in which a generative artificial intelligence evaluates service and campaign design proposals based on relevant laws and guidelines and determines whether or not there are any legal issues.

3. The system according to claim 1, which aggregates the results of an A / B test and notifies the user of the optimal variation.

Images