Information processing system, program product, and information processing method

By inputting the overall virtual avatar information into the learning model, generating a collection of virtual avatars and conducting surveys, the problem of low efficiency in generating virtual avatars each time in marketing is solved, and efficient and accurate advertising information generation is achieved.

CN123071014APending Publication Date: 2026-11-10FUJIFILM BUSINESS INNOVATION CORP
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202511560903.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2025-05-09
Filing Date
2025-10-29
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

In marketing, each survey requires the generation of virtual avatars related to products and other resources, resulting in a large processing volume and low efficiency.

Method used

By inputting the information and resource information of the pre-generated overall virtual avatar into the learning model, a collection of virtual avatars is obtained, and the learning model is used to perform surveys and generate advertising information, avoiding the generation of new virtual avatars for each survey.

Benefits of technology

It improves the efficiency and reliability of the survey, reduces the processing volume, and can generate advertising information based on resource information, appeal information, and survey results, thereby improving the accuracy of advertising information.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123071014A_ABST
    Figure CN123071014A_ABST
Patent Text Reader

Abstract

An information processing system, a program product, and an information processing method, the information processing system including a processor that performs processing to acquire a collection of virtual personae extracted from a population by inputting information related to the virtual personae that make up the population and resource information related to a resource to a learning model, perform an investigation related to the resource on the collection, and output a result of the investigation.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to an information processing system, a program artifact, and an information processing method. This invention can also be applied to programs and program artifacts. Background Technology

[0002] One marketing method involves creating a virtual persona during product development and then conducting research on this virtual persona. Based on the research results, the advertising effectiveness of the product in the real world can be predicted.

[0003] In marketing, this virtual persona is called a "user profile." Creating a user profile involves not only defining gender or age group, but also, depending on the needs, research to define attributes such as upbringing, education, work experience, annual income, lifestyle, family structure, interests, skills, and values. It's preferable to set attributes that are relevant to the product being developed or the content of the survey. It's believed that the more clearly defined the attributes of the user profile—in other words, the more reliable the survey results based on that profile—the higher the reliability.

[0004] Patent Document 1: Japanese Patent Application Publication No. 2023-056557

[0005] Patent Document 2: Japanese Patent Application Publication No. 2023-000788

[0006] Patent Document 3: Japanese Patent Application Publication No. 2015-501990

[0007] Each product requires extensive processing when generating attributes for the numerous virtual avatars that become the subjects of the survey.

[0008] The purpose of this invention is to avoid generating virtual avatars of survey subjects related to resources such as goods for each survey. Summary of the Invention

[0009] The information processing system of the present invention is characterized by having a processor that performs the following processing: by inputting information related to virtual avatars constituting a pre-generated totality and resource information related to resources into a learning model, obtaining a collection of virtual avatars extracted from the totality; performing a survey related to the resources on the collection; and outputting the results of the survey.

[0010] Furthermore, the information processing system is characterized in that the processor inputs filtering information, which includes statistical information related to the virtual avatar, into the learning model in addition to information related to the virtual avatar and the resource information, to obtain a collection of virtual avatars extracted from the total.

[0011] Furthermore, the information processing system is characterized in that the processor adjusts the statistical information contained in the screening information by referring to statistical information related to potential customers of the resource.

[0012] Furthermore, the information processing system is characterized in that the processor is configured to adjust the importance of statistical information related to the potential customer in the statistical information contained in the screening information.

[0013] Furthermore, the information processing system is characterized in that the processor acquires information by inputting information related to the virtual avatars constituting the aggregate, the resource information, and the statistical information into a learning model, and outputs the information as appeal information representing the appeal of the resources in the advertisement.

[0014] Furthermore, the information processing system is characterized in that the processor inputs the search performance information of the potential customer into the learning model after including it in the resource information.

[0015] Furthermore, the information processing system is characterized in that the processor inputs the resource information and the appeal information into the learning model to execute the survey.

[0016] Furthermore, the information processing system is characterized in that the processor performs the following processing: acquiring appeal information representing the appeal power of the resource in the advertisement; and generating advertising information for the resource based on the resource information, the appeal power information, and the results of the survey.

[0017] Furthermore, the information processing system is characterized in that the processor performs the following processing: generating search keyword information based on the results of the survey; and generating advertising information for the resource with reference to the search keyword information.

[0018] Furthermore, the information processing system is characterized in that the processor performs the following processing: acquiring information output by inputting the resource information, the appeal information, and the survey results into a learning model as search intent information representing the search intent for the resource; and acquiring information output by inputting the search intent information into a learning model as search keyword information.

[0019] Furthermore, the information processing system is characterized in that the processor outputs comparative information that includes the statistical information used to generate the advertising information and the statistical information related to potential customers who viewed the advertising information in a comparable manner, and obtains information output by inputting the comparative information into a learning model as the analysis information of the advertising information.

[0020] Furthermore, the information processing system is characterized in that the processor also sets the number of virtual avatars input into the learning model.

[0021] The program artifact involved in this invention records a program for enabling a computer to perform the following functions: by inputting information related to virtual avatars constituting a pre-generated population and resource information related to resources into a learning model, obtaining a collection of virtual avatars extracted from the population; performing a survey related to the resources on the collection; and outputting the results of the survey.

[0022] The information processing method involved in this invention includes the following steps: by inputting information related to virtual avatars constituting a pre-generated totality and resource information related to resources into a learning model, a collection of virtual avatars extracted from the totality is obtained; an investigation related to the resources is performed on the collection; and the results of the investigation are output.

[0023] Invention Effects

[0024] According to the first aspect of the present invention, it is possible to avoid generating virtual avatars that are survey subjects related to resources such as goods for each survey.

[0025] According to the second aspect of the present invention, a collection of virtual avatars can be obtained by referring to statistical information related to virtual avatars.

[0026] According to the third aspect of the present invention, the reliability of statistical information can be improved.

[0027] According to the fourth aspect of the present invention, it is possible to set the degree to which the statistical information included in the filtering information is adjusted according to the importance.

[0028] According to the fifth aspect of the present invention, the learning model can generate resource appeal information.

[0029] According to the sixth aspect of the present invention, the learning model can further refer to search performance information to generate resource appeal information.

[0030] According to the seventh aspect of the present invention, the learning model is able to perform an investigation of the set.

[0031] According to the eighth aspect of the present invention, advertising information can be generated based on resource information, appeal information, and survey results.

[0032] According to the ninth aspect of the present invention, advertising information for resources can be generated by referring to search keywords.

[0033] According to the tenth aspect of the present invention, a learning model can be used to generate search keyword information for generating advertising information.

[0034] According to the eleventh aspect of the present invention, the learning model is able to generate analysis results of advertising information.

[0035] According to the 12th aspect of the present invention, it is possible to set the number of virtual human figures referenced when generating the aggregate.

[0036] According to the 13th aspect of the present invention, it is possible to avoid generating virtual avatars that are survey subjects related to resources such as goods for each survey.

[0037] According to the 14th aspect of the present invention, it is possible to avoid generating virtual avatars that are survey subjects related to resources such as goods for each survey. Attached Figure Description

[0038] The embodiments of the present invention will be described in detail with reference to the following figures.

[0039] Figure 1 This is a block diagram illustrating the information processing apparatus of this embodiment;

[0040] Figure 2 This is a flowchart illustrating the process of evaluating advertisements generated from virtual surveys in this embodiment;

[0041] Figure 3 This is a diagram illustrating an example of a screen display used to generate product concept information in this embodiment;

[0042] Figure 4 This is a diagram illustrating a display example where an example of external data used in generating subdivision information is displayed on the screen in this embodiment;

[0043] Figure 5 This is a diagram illustrating an example of a screen display where statistical information that does not reflect external data and statistical information that reflects external data are displayed in chart form in this embodiment.

[0044] Figure 6 This is a diagram showing an example of a confirmation screen for detailed item information in this embodiment;

[0045] Figure 7 This is a diagram showing an example of the final confirmation screen for the detailed item information in this embodiment;

[0046] Figure 8 This is a diagram showing an example of the settings screen when generating a user profile database in this embodiment;

[0047] Figure 9 This is a diagram illustrating an example of the display screen at the end of the user profile database generation in this embodiment;

[0048] Figure 10 This is a diagram showing an example of the display of the setting screen for the issue / claim axis in this embodiment;

[0049] Figure 11 This is a diagram showing an example of a screen displaying the results of a virtual survey in this embodiment;

[0050] Figure 12 This is a diagram illustrating an example of a screen display generated based on the results of the virtual survey in this embodiment;

[0051] Figure 13 This is another example of a screen display showing the results of a virtual survey in this embodiment;

[0052] Figure 14 It indicates and from Figure 13 The image shown is an example of a screen displaying personal information related to a selected user profile.

[0053] Figure 15 This is a diagram illustrating an example of a screen display used in this embodiment to allow the user to select search intent information;

[0054] Figure 16 This is a diagram illustrating an example of a screen display used in this embodiment to allow the user to select keyword information;

[0055] Figure 17 This is a diagram illustrating an example of a screen display used in this embodiment to allow a user to select the advertising information to be displayed;

[0056] Figure 18 This is a screenshot example showing a countermeasure analysis report that includes comparative information generated by analyzing information related to potential customers' responses to advertisements, as described in this embodiment.

[0057] Symbol Explanation

[0058] 2-User Interface (UI) Department, 4-Storage Department, 6-Information Processing Department, 8-Network, 10-Information Processing Device, 61-Basic Information Generation Department, 62-User Profile Database Generation Department, 63-Virtual Survey Execution Department, 64-Display Information Generation Department, 65-Advertising Information Processing Department, 66-Evaluation Department, 611-Product Information Generation Department, 612-Detailed Item Information Generation Department, 631-Appeal Strength Information Generation Department, 651-Search Intent Generation Department, 652-Keyword Generation Department, 653-Advertising Placement Processing Department. Detailed Implementation

[0059] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.

[0060] Figure 1This is a block diagram illustrating the information processing apparatus 10 in this embodiment. The information processing apparatus 10 is one embodiment of the information processing system according to the present invention. The information processing apparatus 10 in this embodiment can be implemented using a common hardware structure that has existed before, such as a personal computer (PC). That is, the information processing apparatus 10 has a processor, ROM, RAM, a hard disk drive as a storage device, a network interface provided as a communication device, and a user interface including an input device such as a mouse or keyboard and a display device such as a monitor.

[0061] In this embodiment, user profiles are used as the subjects of a survey related to resources, such as goods or advertisements. "Resources" include both tangible and intangible goods, and in addition to goods, also include resources not intended for sale, such as blog posts or videos uploaded to social media. "Goods" are an example of resources and can be defined as property (e.g., items) and services intended for sale. As mentioned above, one method in marketing involves virtually creating a specific persona of the primary target customer type, also known as the target customer (or "potential customer"), during the development or offering of a product. In this embodiment, the virtual persona that will be the subject of the survey is called a "user profile." Furthermore, the survey conducted on this virtual persona is called a "virtual survey." Sometimes, questionnaires are conducted on user profiles, which are called "virtual questionnaires." Since "virtual questionnaires" yield the same responses as "virtual surveys" from user profiles, in this embodiment, both "virtual surveys" and "virtual questionnaires" are collectively referred to as "virtual surveys."

[0062] exist Figure 1 The diagram shows a user interface (UI) unit 2, a storage unit 4, an information processing unit 6, and a network 8. The user interface (UI) unit 2 is implemented using the aforementioned user interface. The storage unit 4 is implemented using a storage device such as a hard disk drive. The information processing unit 6 is implemented through the coordinated operation of a processor and a program running on the processor. The information processing device 10 is connected to a network 8, such as the Internet or a LAN, via a network interface. Furthermore, the storage device is not limited to the storage unit 4 mounted on the information processing device 10; a storage device connected via the network 8 can also be used.

[0063] The user interface (UI) unit 2 has an operation receiving unit 21 that receives user operations on the input mechanism and a display unit 22 that displays various information on the display mechanism.

[0064] like Figure 1As illustrated, storage unit 4 stores various types of information. Among these, a preset user profile database, product information, and statistical information are used as inputs for the information processing described later. The product information and statistical information can be retrieved from external sources such as the internet when needed, without needing to be pre-stored in storage unit 4. The remaining information stored in storage unit 4 will be explained along with the processing instructions.

[0065] The information processing unit 6 includes a basic information generation unit 61, a user profile database generation unit 62, a virtual survey execution unit 63, a display information generation unit 64, an advertising information processing unit 65, and an evaluation unit 66. In this embodiment, the components other than the display information generation unit 64 use generative AI such as LLM (Large-Scale Language Model) or SLM (Small-Scale Language Model) as the prescribed learning model.

[0066] The survey typically specifies the target product and the object of the survey, i.e., the product and the target. The basic information generation unit 61 determines the product and the target person, i.e., creates a user profile in this embodiment. That is, as basic information for generating a user profile database, the basic information generation unit 61 generates product information through the product information generation unit 611 and generates sub-item information through the sub-item information generation unit 612. In this embodiment, the product information generated by the product information generation unit 611 is also referred to as "product concept information," and the relationship between product information and product concept information will be explained later.

[0067] The user profile database generation unit 62 acquires a collection of user profiles output by inputting information related to user profiles set in a preset user profile database and product information related to products into a predetermined first learning model. In this embodiment, the collection of user profiles extracted from the preset user profile database is called the "user profile database".

[0068] Here, "pre-generated user profile library" refers to the total number of user profiles generated in advance. For example, the pre-generated user profile library is a collection of user profiles pre-created based on the population distribution of Japan, and is defined as the total number of user profiles. For instance, the pre-generated user profile library may consist of user profiles of approximately 30,000 people. In this embodiment, age, gender, presence or absence of children, and annual household income are specifically used as attributes representing population distribution, but other attribute items may also be used. In the user profile, the specific values ​​of these attribute items become the attributes representing that user profile.

[0069] Furthermore, in this embodiment, based on research into advertisements for domestic Japanese products, a pre-defined user profile database generated using actual statistical information on Japan's population distribution is used as the overall user profile. However, the statistical information obtained is not limited to Japan. For example, depending on the product being investigated, a pre-defined user profile database can be generated using statistical information limited to the domestic area, such as Tokyo or several prefectures or the Kanto region, or it can be generated using statistical information obtained from a broader scope, such as Asia including Japan. Additionally, the pre-defined user profile database used can also be the overall user profile database generated without utilizing statistical information.

[0070] Furthermore, in this embodiment, as described later, the prescribed learning models 1 through 6 are utilized. Moreover, the learning models are also utilized when generating certain information. These learning models can utilize the same generative AI, or different generative AIs can be used depending on the characteristics of the information being processed. In this embodiment, generative AI is utilized, but the specific type of generative AI used is arbitrary.

[0071] The Virtual Research Execution Department 63 conducts virtual research on the products targeted at the survey as a survey of the user profile database. Incidentally, the subjects of this "virtual research" are virtual avatars, i.e., user profiles. "Products targeted at the survey" refers to the products that are the subject of inquiry during the survey of the user profiles of the survey targets.

[0072] The virtual survey execution unit 63 includes an appeal information generation unit 631. The appeal information generation unit 631 acquires information representing the appeal of the product in the advertisement by inputting information related to user profiles constituting the user profile database, product information of the surveyed product, and the detailed item information described later into a prescribed second learning model. In this embodiment, the appeal information processed includes the issue and the appeal axis. The "issue" refers to the issue that the product aims to solve for consumers. The "appeal axis" refers to the selling points or advantages that are highly effective in stimulating users' desire to purchase when advertising the product.

[0073] The virtual research execution unit 63 performs a virtual research by inputting product information and appeal information of the product being investigated into a prescribed third learning model. Furthermore, the virtual research execution unit 63 obtains the output from the third learning model as the result of the virtual research.

[0074] The display information generation unit 64 generates the execution results of the virtual survey for display in various display modes; in other words, the display information of the survey results. The display unit 22 displays the display information generated by the display information generation unit 64 on the monitor. Furthermore, the output of the survey results is not limited to the monitor; it can be saved in a file format in the storage unit 4, or provided externally via the network 8.

[0075] The advertising information processing unit 65 performs advertising-related processing. Specifically, the advertising information processing unit 65 obtains the appeal information of the product in the advertisement, and generates advertising information for the product based on the product information, appeal information, and survey results. Furthermore, the advertising information processing unit 65 delivers advertisements based on the generated advertising information. The advertising information processing unit 65 includes a search intent generation unit 651, a keyword generation unit 652, and an advertisement delivery processing unit 653.

[0076] The search intent generation unit 651 acquires search intent information representing the search intent for a product by inputting product-related information, product appeal information, and the results of a virtual survey into a predetermined fourth learning model. The keyword generation unit 652 acquires search keyword information by inputting the search intent information generated by the search intent generation unit 651 into a predetermined fifth learning model. The advertising delivery processing unit 653 processes the actual placement of advertisements in media, etc., based on the generated advertising information.

[0077] Evaluation unit 66 evaluates the advertisements generated by advertising information processing unit 65. Specifically, evaluation unit 66 outputs comparative information that includes statistical information used to generate the advertising information, namely segmentation information, and statistical information related to potential customers who viewed the advertising information, in a comparable manner. Furthermore, evaluation unit 66 acquires information output by inputting the comparative information into a prescribed sixth learning model as analytical information for the advertising information.

[0078] Next, the operation of this embodiment will be explained. In this embodiment, the following is used: Figure 2 The flowchart shown and Figures 3 to 18 The example shown illustrates the process of conducting a virtual survey on user profiles that constitute a user profile library, and generating, delivering, and evaluating advertisements based on the results.

[0079] Furthermore, the explanation is based on the premise that the aforementioned preset user profile database has already been generated. Also, as described above, a learning hierarchy including learning models 1 through 6 is utilized; however, for convenience, these will be referred to as "generative AI" in the following explanation.

[0080] First, the product information generation unit 611 generates product information that is input into the generative AI (step S110).

[0081] In this embodiment, information related to the product is referred to as "product information." Two types of product information are processed: one is the product information input into the generative AI that generates the user profile database, and the other is the product information used to generate the product information input into the generative AI. In the following description, the former type of product information will be referred to as "product concept information," or simply "concept information." Figure 1 In this context, the former type of product information is illustrated as "product concept information," and the latter type is illustrated as "product information." The latter type of product information is related to the product used to generate the product concept information, such as information describing the product's features, specifications, or functions, such as those found in product catalogs or landing pages. Figure 1 The data is stored in storage unit 4, but can also be accessed from outside via network 8.

[0082] Incidentally, in this embodiment, the former type of product information is input into the generative AI, and its output is used to obtain the latter type of product information, namely, product concept information that explicitly expresses the product concept. Furthermore, the generative AI that generates the product concept information can utilize generative AI such as LLM, which is the same as the first learning model.

[0083] Figure 3 This diagram illustrates an example of a screen display used to generate product concept information in this embodiment. The user specifies the filename of a PDF file or image file related to the previous product information (hereinafter, for ease of explanation, also referred to as "original information") used to generate the product concept information in designated areas 31 and 32. Then, by selecting the upload button 33, the file specified in designated areas 31 and 32 is uploaded. Then, if the generate button 34 is selected, the product information generation unit 611 inputs the uploaded file into the generative AI. Furthermore, the product information generation unit 611 obtains the output from the generative AI as concept information. That is, the product information generation unit 611 generates concept information using the generative AI. The generated concept information is displayed in display area 35. Additionally, the user can edit the concept information displayed in display area 35. Alternatively, the user can manually input concept information in display area 35 without using the original information.

[0084] As described above, conceptual information is generated. If the user determines that the content of the conceptual information is feasible, they select button 36 to proceed to the processing of input sub-item information. Alternatively, the user can stop the process by selecting the "Stop" button 37. The following screens displaying the "Stop" button (...) Figure 4 In addition to the above, users can also stop the process as needed.

[0085] Incidentally, in this embodiment, the prompts input into the generative AI to generate conceptual information indicate the output of the product name, 3 to 5 key product information, target market / expected users, price range, the product's main USP (Unique Selling Position), and information on activities or promotions.

[0086] In addition, in this embodiment, conceptual information is generated using generative AI, but it is also possible to process the information by directly inputting the original information into the generative AI.

[0087] Next, the subdivision information generation unit 612 generates subdivision information input to the generative AI (step S120). Using... Figure 4 and Figure 5 This section provides an explanation of the sub-item information.

[0088] Figure 4 This is a diagram illustrating an example of external data used in generating detailed information. Furthermore, Figure 5 The graph consists of a chart 51 on the left that shows statistical information that does not reflect external data and a chart 52 on the right that shows statistical information that reflects external data.

[0089] When generating a preset user profile database, statistical information related to Japan's population distribution is utilized. By processing the preset user profile database as the processing object, statistical information about the user profiles constituting the preset user profile database can be obtained, that is, statistical information related to the population distribution of the user profiles. In this embodiment, in Figure 5 In this context, the statistical information is represented by Chart 51 on the left, which shows "Density of Target Market Segments Generated Based on Statistical Information". Furthermore, as mentioned above, in this embodiment, age, gender, presence or absence of children, and annual household income are used as examples of attributes representing population distribution. In the following description, the statistical information related to the population distribution that constitutes the user profiles in the preset user profile database will also be referred to as "raw statistical information".

[0090] As statistical information, raw statistical information can be directly used in subsequent processing. However, when the product being investigated is already available in the market, the company can obtain access information related to publicly available information about the actual product (hereinafter referred to as "Web data"), access information about product advertisements (hereinafter referred to as "advertising data"), and information related to CRM (Customer Relationship Management) (hereinafter referred to as "CRM data"). Furthermore, Web data, advertising data, and CRM data are information that becomes available after the product is supplied externally, and therefore are collectively referred to as "external data" in the following description. "External data" can essentially be statistical information related to so-called potential customers who are interested in or pay attention to the product being investigated and access information such as advertisements. In addition, customers who purchase the actual product may become repeat customers; therefore, in this embodiment, potential customers plus repeat customers are broadly understood as "potential customers." It is believed that by referencing external data and reflecting it in the raw statistical information, the accuracy of the statistical information used in this embodiment can be further improved. In this embodiment, in Figure 5 In this embodiment, the statistical information reflecting the external data is represented by Chart 52 on the right, which reads "Target Market Segment Density Enhanced by External Data". In this embodiment, to distinguish between the statistical information reflecting the external data and the original statistical information, the statistical information used in subsequent processing is specifically referred to as "segment information". The segment information generation unit 612 generates segment information by adjusting the statistical information included in the filtering information input into the generative AI in subsequent processing, based on statistical information related to potential customers of the product, i.e., external data.

[0091] exist Figure 4 The example demonstrates three types of external data. Furthermore, it allows you to assign importance to each type of external data. For example, you can select one of the three types of external data. Figure 4 In the chart, colors are used to indicate the state of CRM data selected based on user actions. Furthermore, by sliding slider 41 located below the chart, the importance of each external data point can be set. Incidentally, "importance" indicates the degree of importance of the external data used to adjust the raw statistical information. The higher the importance, the greater the impact of the external data on the generation of sub-item information. That is, the value of the external data is more easily reflected in the sub-item information. The sub-item information generation unit 612 sets the importance of the external data used to adjust the sub-item information included in the filter information input to the generative AI in subsequent processing, based on the settings made by the user using slider 41.

[0092] After the user sets the importance of external data, the subdivision information generation unit 612, according to... Figure 4When button 42, which enhances statistical information, is selected, detailed item information is generated.

[0093] Furthermore, the statistical information processed in this embodiment is expressed in absolute value (in... Figure 4 The statistical result is represented by the number of people (in Chinese), but this is just one example. Other types of data, such as percentages (%), can also be used.

[0094] As mentioned above, users who refer to the detailed information generated based on external data sometimes want to make further adjustments. Therefore, in this embodiment, a function is provided that allows for further adjustment of the detailed information.

[0095] Figure 6 This is an example of a confirmation screen displaying detailed item information. In Figure 6 The display area 613 shows detailed item information. Furthermore, because the detailed item information filters the original statistical information, therefore... Figure 6 For convenience, this is displayed as "Detailed Item Filtering Information". Furthermore, the detailed item information generation unit 612 generates detailed item information using generative AI, while simultaneously displaying the reasons for setting this detailed item. When the user wants to change the detailed item information based on the reasons displayed in the display area 613, the user checks or dechecks the checkboxes corresponding to the desired detailed item from the setting area 614. Then, if the user selects the final confirmation button 615, the detailed item information generation unit 612 generates detailed item information reflecting the settings based on the user's selections and displays it on the final confirmation screen. An example of displaying this final confirmation screen is shown below. Figure 7 .

[0096] In addition, strictly speaking, the information used for display on the monitor may also be generated by the display information generation unit 64 and displayed on the display unit 22, but for the sake of simplicity, it is described as being displayed by the subdivision information generation unit 612. The same applies in the following description.

[0097] If the user from Figure 7 If the user profile database settings button 71 is selected on the screen shown, the user profile database generation unit 62 will begin the user profile database generation process (step S130). If the generation process begins, the user profile database generation unit 62 will first display... Figure 8 The settings screen shown. Users can access it from... Figure 8 The settings screen shown specifies the number of users whose profiles constitute the preset user profile library. That is, the user profile library is created by extracting the number of users specified by the user, up to the number specified by the user. Figure 8 The example shown is an example of specifying XS (720 people), and an example of displaying the notification to the user that the user profile library consists of user profiles of 5 to 720 people.

[0098] Furthermore, in this embodiment, the number of user profiles constituting the preset user profile database is considered the maximum value, and the maximum number of user profiles extracted from the preset user profile database is preset. Figure 8 The example shown illustrates a user selecting XS (720 people) from multiple stages. Besides XS (720 people), other stages such as S (1440 people), M (2880 people), L (5960 people), and XL (11920 people) are also available. Furthermore, the number of stages and the number of people in each level can be arbitrarily determined. Alternatively, the number of people can be specified by the user. For example, in... Figure 8 The screen displayed shows information indicating that the processing load required for virtual surveys and other user profile database operations varies depending on the number of people specified on the screen. Specifically, in the case of a virtual survey, the more people participate, the more user profiles are generated, resulting in a higher processing load but also higher reliability of the survey results. Conversely, fewer people participate, resulting in fewer user profiles, leading to a lower processing load but also lower reliability of the survey results. Furthermore, in this embodiment, a structure is provided that allows setting the number of users constituting the preset user profile database; however, in other embodiments, a structure can also be provided that allows setting the number of users extracted from the preset user profile database.

[0099] If the user selects the "Generate User Profile Library" button 81, the User Profile Library Generation Unit 62 inputs information related to the user profiles constituting the preset user profile library, product concept information, sub-item information, and filtering information including the maximum number of people specified in the settings screen into the generative AI, thereby outputting user profiles of up to 720 people. That is, the User Profile Library Generation Unit 62 obtains a collection of virtual avatars extracted from the preset user profile library by the generative AI as the user profile library. Here, the information input into the generative AI is collectively referred to as "filtering information" to give it the meaning of information used to filter user profiles extracted from the preset user profile library for the purpose of generating the user profile library.

[0100] Incidentally, in this embodiment, the prompts input into the generative AI for generating a user profile library instruct the reference product concept information to select the most suitable target segment from the segment information (e.g., male or female in the case of gender), and output the reason for the selection.

[0101] As described above, when the user profile database generation unit 62 generates a user profile database, it displays an example of... Figure 9 The user profile database generates the final screen. Users can then access it from... Figure 9 The display area 91 of the screen shows that the user profiles that make up the user profile database consist of 41 people.

[0102] Next, if the user from Figure 9 Selecting the input button 92 for the topic / appeal axis in the shown screen will display... Figure 10 The screen showing the settings for the topic / appeal axis.

[0103] The virtual survey execution unit 63 uses the appeal information generation unit 631 to generate appeal information to effectively conduct virtual surveys (step S140). As described above, in this embodiment, the topic and appeal axis are set as appeal information. The topic and appeal axis are set because, in virtual surveys, instead of simply asking users (here, user profiles) about the attractiveness of a product based on product concept information, the attractiveness of the product is clearly defined (i.e., the topic and appeal axis) to ask users if they are interested in products with such attractiveness. As a method for setting appeal information, in this embodiment, it can be set in three ways. This allows the user to choose what to display. Figure 10 One of the setting method selection buttons in the button display area 101 of the screen shown.

[0104] The first option is for the user to select "Manually Add" and manually input the appeal information in the display area 102. The second option is for the user to select "Automatically Generate Based on Product Concept," causing the generative AI to generate the appeal information. That is, the appeal information generation unit 631 acquires information related to user profiles constituting the user profile library, product concept information, and sub-item information, and outputs this information as appeal information by inputting these into the generative AI. The appeal information thus acquired is displayed in the display area 102. The third option is for the user to select "Automatically Generate Based on Product Concept + External Data," causing the generative AI to generate the appeal information. That is, similar to the second option, the appeal information generation unit 631 causes the generative AI to generate the appeal information, but in the third option, external data is appended to the further input information. For example, based on segmentation information, the personal attributes of potential customers can be determined, and the actual search performance of potential customers for product-related information (e.g., web or advertisements) can be obtained. Therefore, the external data input into the generative AI is the actual search performance information of potential customers for advertisements, etc., that matches the target segmentation, such as objective external data like the number of clicks on advertisements, etc. The appeal information obtained in this way is displayed in display area 102.

[0105] Incidentally, in this embodiment, the prompts input into the generative AI to generate appeal information are groups that instruct the output of topics and appeal axes when selling by target segments, taking into account five psychological factors: "extroversion," "honesty," "harmony," "openness," and "neurotic tendencies."

[0106] As described above, appeal information is generated. If the user selects the "Start Virtual Survey" button 103, the virtual survey execution unit 63 begins to execute the virtual survey (step S150). Specifically, the virtual survey execution unit 63 inputs product concept information and appeal information into the generative AI to enable it to execute the virtual survey, and obtains the output from the generative AI as the execution result of the virtual survey. Furthermore, the execution result of the virtual survey has the same meaning as "survey results".

[0107] Incidentally, in this embodiment, the prompts input into the generative AI that performs the virtual survey indicate which user profile felt attracted and the reasons for feeling so, as well as concerns about purchasing, in daily life during the period (e.g., within 3 months from now) and in what context.

[0108] Figure 11 This is a diagram showing an example of a screen displaying the results of a virtual survey. The survey results are displayed in the designated display area 111. The display information generation unit 64 generates information based on the results of the virtual survey. Figure 11 The information displayed on the screen shown, and the information used for performing Figures 12-14 The screen shown displays the information. In this embodiment, when generating appeal strength information, the generative AI is instructed to generate three appeal strength information points, therefore, as shown... Figure 11 The display example shows three groups of appeal information: the topic and the appeal axis. For example... Figure 11 As shown in the example, in a virtual survey, the number of user profiles who find the content of the appealing information attractive can be determined, and therefore this number is displayed corresponding to each appealing information. Figure 11 In the process, a virtual survey identified 41 individuals who would form the user profile database (reference). Figure 9 The number of people listed as "27", "27", and "21" in the user profiles cited in the example are attracted to the appeal information. Incidentally, in this embodiment, the collection of user profiles that are attracted to the appeal information from the user profile database is called a "user profile group". The "user profile group" is generated for each appeal information.

[0109] When a user selects the desired checkboxes corresponding to each appeal information on the left side of the display area 111 and then selects the save button 112 for the user profile group, the virtual survey execution unit 63 saves the information related to the appeal information whose checkboxes are checked, specifically the information displayed in the display area 111 or the information related to the user profile belonging to that user profile group, in the storage unit 4.

[0110] Figure 12 This is an example of a screen display generated based on the results of a virtual survey. Figure 12China is through the Figure 11 The display example is shown when the user selects any saved appeal information according to the specified user operation. Figure 12 Two display areas, 121 and 122, are set up to display the following information: Display area 121 displays appeal information (topic / attraction axis); Display area 122 displays a summary of the virtual survey's response results. Furthermore, "Execution Results" and "Response Results" have the same meaning.

[0111] Figure 13 This is another example of a screen displaying the results of a virtual survey. Figure 13 In the middle, by selecting to display in Figure 11 The user profile group ID displayed on the screen shows information related to the selected user profile group. Figure 13 The screen shown has two display areas, 131 and 132, displaying the following information: Display area 131 displays the gender ratio, average age, number of users, and age group as member attribute information for each user profile. Display area 132 displays a list of user profiles that make up the user profile group. An example of the screen display when selecting a user profile from this list through a specified user operation is shown below. Figure 14 .

[0112] exist Figure 14 The system displays the user profile's personal information. The upper display area 141 shows the user profile's attribute information. The lower display area 142 shows the responses obtained from the user profile through a virtual survey. Users can conduct an interview with the user profile by selecting the "Conduct an Interview" button 143.

[0113] In addition, in this embodiment, information related to user profile groups is divided into... Figures 12-14 It can be shown in the diagram, but it can also be summarized and displayed on one screen.

[0114] return Figure 12 If the user selects the "Advertising Support" button 123 on the screen, the search intent generation unit 651 in the advertising information processing unit 65 generates search intent information (step S160). Specifically, the search intent generation unit 651 acquires the information output by inputting product concept information, appeal information, and the results of virtual surveys into the generative AI as the product's search intent information for the appeal information input into the generative AI. "Search intent" is defined as the specific purpose or need held by a user when using a search engine to obtain product information or to seek solutions.

[0115] Incidentally, in this embodiment, the prompts input into the generative AI to generate search intent information are instructed to output a group of 10 reasons that are relevant to the search intent.

[0116] Figure 15 This is an example of a screen display used to allow users to select search intent information. Figure 15 The appeal strength information is displayed in the display area 151 shown. Figure 12 The display area 121 is used for searching intentions. Furthermore, the display area 152 displays search intention information generated by the search intention generation unit 651. Search intentions and their rationale are grouped and included in the search intention information in this embodiment. The display area 152 displays search intention information generated based on appeal information generated from external data input into the generative AI, and search intention information generated based on appeal information not reflecting external data. Figure 15 In the example shown, the user can select up to three appeal information items from the appeal information displayed in the display area 152 by checking the checkboxes.

[0117] After selecting the search intent, if the user selects Figure 15 When the "Generate Keywords" button 153 is shown, the advertising information processing unit 65 receives the search intent information selected by the user (step S170).

[0118] Next, the keyword generation unit 652 generates keyword information (step S180). Specifically, the keyword generation unit 652 acquires the information output by inputting the received search intent information into the generative AI as keyword information.

[0119] Incidentally, in this embodiment, among the prompts input into the generative AI for generating keyword information, the output is instructed to include 10 keywords representing the broad category (hereinafter referred to as "primary keywords") and 10 keywords within that category used to further narrow down the scope (hereinafter referred to as "auxiliary keywords").

[0120] Figure 16 This is an example of a screen display used to allow users to select keyword information. Figure 16 The display areas 161a and 161b shown can scroll to display 10 main keywords. Furthermore, display areas 162a and 162b can scroll to display 10 auxiliary keywords. In this embodiment, similar to the search intent information, display areas 161a and 161b display keyword information generated from appeal information not reflecting external data, and display areas 162a and 162b display keyword information generated from appeal information generated from external data input into generative AI, arranged vertically. Figure 16 In the example screen shown, the user can select up to 10 main keywords from the main keywords displayed in display areas 161a and 161b, and up to 10 auxiliary keywords from the auxiliary keywords displayed in display areas 162a and 162b by checking the checkboxes.

[0121] After selecting keywords, users can choose to set... Figure 16 When the button 163 for generating advertising information is pressed on the screen shown, the advertising information processing unit 65 receives the main keywords and auxiliary keywords selected by the user (step S190).

[0122] As described above, if a keyword selected by the user is received, the advertising information processing unit 65 generates advertising information (step S200). In this embodiment, an advertising title and advertising copy are generated as advertising information. Therefore, the advertising information processing unit 65 acquires the information output by inputting product concept information, sub-item information, results of virtual survey execution, search intent information, and keyword information into the generative AI as advertising information (step S200).

[0123] Incidentally, in this embodiment, when inputting prompts into the generative AI to generate advertising information, the system instructs the output of advertising headlines and copy (e.g., 5 options each) within a specified word range and in a prescribed output format, which are considered likely to be of interest to users who are targets that can be identified based on segmentation information.

[0124] Figure 17 This is an example of a screen display used to allow users to select which advertising information to display. Figure 17 Display area 171 shows a list of ad titles generated by generative AI. Display area 172 shows a list of ad copy generated by generative AI. Users can select the ad title and ad copy to be displayed in the advertisement from display areas 171 and 172 respectively by checking the checkboxes.

[0125] After selecting the ad title and ad copy, if the user chooses to set it in Figure 17 When the "Start Ad Serving" button 173 is displayed on the screen, the advertising information processing unit 65 receives the advertising title and copy selected by the user (step S210). Furthermore, the advertising serving processing unit 653 serves the ad by actually publishing it on the company's webpage or advertising page using the received advertising title and copy. It also obtains information related to the user's reaction to the served ad (step S220).

[0126] Figure 18This is an example of a screen display showing a strategy analysis report that includes comparative information generated by analyzing information related to potential customers' responses to advertising. Comparative information refers to information that includes, in a comparable manner, the expected statistical information at the time of advertising and the statistical information obtained from the actual responses to that advertising. Figure 18 In the example above, the comparison information displays detailed item information. That is, in... Figure 18 Display area 181 displays the segmentation information expected to be obtained by delivering the advertisements generated through the above processing. On the other hand, display area 182 displays segmentation information generated by statistical processing based on the actual reactions of users (i.e., "potential customers") after delivering the advertisements generated through the above processing (e.g., click-through rate, conversion rate, cost efficiency, etc.). In marketing, for example, it is preferable to obtain the desired results, but... Figure 18 As the comparison information displayed on the screen shows, there are actually quite a few cases where discrepancies occur.

[0127] Therefore, the evaluation unit 66 acquires the information output by inputting comparison information into the generative AI as analytical information for advertising. Incidentally, in this embodiment, the prompts input into the generative AI to generate analytical information instruct the user to specify analytical viewpoints and output suggestions regarding overall effectiveness evaluation or improvement points as analytical information. Furthermore, the analytical information acquired by the evaluation unit 66 can also be included in a countermeasure analysis report and displayed on a screen.

[0128] According to this embodiment, as explained above, a virtual survey is performed, advertisements are generated and delivered based on the results, and the advertisements can be evaluated. The above processing loop can be appropriately reviewed and feedback provided. For example, users can easily improve the advertisements by reviewing product information, sub-item information, and appeal information selection. In particular, in this embodiment, a user profile library for virtual survey subjects can be easily generated by extracting from a preset user profile library that is a total of user profiles pre-generated based on actual Japanese statistical information. That is, it is not necessary to generate a user profile library for each product as a virtual survey subject each time.

[0129] In this embodiment, an example is shown of implementing a virtual questionnaire for product advertising and generating / delivering advertisements based on the results. However, if the resource is a blog post or something similar instead of a product, the content of the resource itself, i.e., the blog post, can also be generated based on the results of the virtual questionnaire.

[0130] In this embodiment, each process is executed by any computer. Furthermore, any computer can execute these processes via a processor, a program, or a combination thereof. Any computer can also be a general-purpose computer, a special-purpose computer, a workstation, or other hardware components that can execute programs.

[0131] The processor can be composed of one or more hardware components, and the type of hardware is not limited. For example, the processor can be composed of programmable logic devices such as CPUs (Central Processing Units), MPUs (Micro Processing Units), FPGAs (Field Programmable Gate Arrays), dedicated circuits for performing specific processes such as ASICs (Application Specific Integrated Circuits), GPUs (Graphics Processing Units), or NPUs (Neural Processing Units). Furthermore, the processor has units or means that execute the various processes described in this embodiment. The type of hardware can be a combination of different types of hardware. When multiple hardware components are configured to execute one or more processes of a processor, these multiple hardware components can exist in physically separate devices or in the same device. Furthermore, in any embodiment, the order of the processes performed by the processor is not limited to the above order and can be appropriately varied. Additionally, the hardware is composed of circuits, such as those combining semiconductor elements.

[0132] Furthermore, this embodiment can also be implemented using hardware, software, firmware, microcode, or a combination thereof. Software, firmware, and microcode consist of programs. Furthermore, a program can be, for example, a group of program modules, each of whose functions can be implemented by a processor configured to execute those functions. A program can be program code or multiple code segments stored on one or more non-transitory computer-readable media (e.g., storage media or other storage devices). A program can be divided and stored on multiple non-transitory computer-readable media existing in physically separate devices. Program code or code segments can represent any combination of steps, functions, subroutines, routines, subroutines, modules, software packages, classes, commands, data structures, or program statements. Program code or code segments can also be connected to other code segments or hardware circuits by sending and receiving information, data, arguments, parameters, or memory contents.

[0133] (Postscript)

[0134] (1) An information processing system, comprising a processor,

[0135] The processor performs the following processing:

[0136] By inputting information related to the virtual avatars constituting the pre-generated totality and resource information related to resources into the learning model, a collection of virtual avatars extracted from the totality is obtained;

[0137] Perform an investigation related to the resources on the collection; and

[0138] Output the results of the survey.

[0139] (2) According to the information processing system described in (1),

[0140] The processor inputs filtering information, including statistical information related to the virtual avatars, into the learning model in addition to information related to the virtual avatars and the resource information, to obtain a collection of virtual avatars extracted from the overall population.

[0141] (3) According to the information processing system described in (2),

[0142] The processor adjusts the statistical information included in the filtering information by referencing statistical information related to potential customers of the resource.

[0143] (4) The information processing system described in (3),

[0144] The processor is configured to adjust the importance of statistical information related to the potential customer within the filtering information.

[0145] (5) The information processing system according to (3) or (4),

[0146] The processor acquires information by inputting information related to the virtual avatars constituting the aggregate, the resource information, and the statistical information into a learning model, and outputs this information as appeal information representing the appeal of the resources in the advertisement.

[0147] (6) The information processing system according to (5),

[0148] The processor inputs the search performance information of the potential customers into the resource information and then feeds it into the learning model.

[0149] (7) According to the information processing system described in (5) or (6), the processor inputs the resource information and the appeal information into the learning model to perform the survey.

[0150] (8) The information processing system according to (2),

[0151] The processor performs the following processing:

[0152] Obtain appeal information representing the appeal of the resource in the advertisement; and

[0153] The advertising information for the resource is generated based on the resource information, the appeal information, and the survey results.

[0154] (9) The information processing system according to (8),

[0155] The processor performs the following processing:

[0156] Generate search keyword information based on the results of the survey; and

[0157] Based on the search keyword information, advertising information for the resource is generated.

[0158] (10) The information processing system according to (9),

[0159] The processor performs the following processing:

[0160] The information output by inputting the resource information, the appeal information, and the survey results into a learning model is used as search intent information representing the search intent for the resource; and

[0161] The information output by inputting the search intent information into the learning model is used as the search keyword information.

[0162] (11) According to the information processing system of (8), the processor outputs comparative information that includes the statistical information used to generate the advertising information and the statistical information related to potential customers who viewed the advertising information in a comparable manner, and obtains information output by inputting the comparative information into a learning model as the analysis information of the advertising information.

[0163] (12) The information processing system according to any one of (1) to (11),

[0164] The processor also sets the number of virtual avatars input into the learning model.

[0165] (13) A program article containing a program for enabling a computer to perform the following functions:

[0166] By inputting information related to the virtual avatars constituting the pre-generated totality and resource information related to resources into the learning model, a collection of virtual avatars extracted from the totality is obtained;

[0167] Perform an investigation related to the resources on the collection; and

[0168] Output the results of the survey.

[0169] According to the invention described in (1), it is possible to avoid generating virtual avatars of survey subjects related to resources such as goods for each survey.

[0170] According to the invention described in (2), a collection of virtual avatars can be obtained by referring to statistical information related to virtual avatars.

[0171] According to the invention described in (3), the reliability of statistical information can be improved.

[0172] According to the invention described in (4), it is possible to set the degree of statistical information included in the filtering information to be adjusted according to the importance.

[0173] According to the invention described in (5), the learning model can generate resource appeal information.

[0174] According to the invention described in (6), the learning model can further refer to search performance information to generate resource appeal information.

[0175] According to the invention described in (7), the learning model is able to perform an investigation of a collection.

[0176] According to the invention described in (8), advertising information can be generated based on resource information, appeal information and survey results.

[0177] According to the invention described in (9), it is possible to generate advertising information for resources by referring to search keywords.

[0178] According to the invention described in (10), a learning model can be used to generate search keyword information for generating advertising information.

[0179] According to the invention described in (11), the learning model can generate analysis results of advertising information.

[0180] According to the invention described in (12), it is possible to set the number of virtual human figures referenced when generating a collection.

[0181] According to the invention described in (13), it is possible to avoid generating virtual avatars of survey subjects related to resources such as goods for each survey.

[0182] The embodiments of the present invention described above are provided for illustrative purposes. Furthermore, these embodiments do not exhaustively encompass the present invention, nor do they limit the invention to the disclosed methods. It will be apparent to those skilled in the art that various modifications and variations will be readily understood. These embodiments were chosen and described to most readily explain the principles and applications of the invention. Thus, those skilled in the art can understand the invention through various modifications that are assumed to be optimized for specific uses of various embodiments. The scope of the invention is defined by the foregoing claims and their equivalents.

Claims

1. An information processing system, characterized in that, Equipped with a processor The processor performs the following processing: By inputting information related to the virtual avatars constituting the pre-generated totality and resource information related to resources into the learning model, a collection of virtual avatars extracted from the totality is obtained; Perform an investigation related to the resources on the collection; and Output the results of the survey.

2. The information processing system according to claim 1, characterized in that, The processor inputs filtering information, including statistical information related to the virtual avatars, into the learning model in addition to information related to the virtual avatars and the resource information, to obtain a collection of virtual avatars extracted from the overall population.

3. The information processing system according to claim 2, characterized in that, The processor adjusts the statistical information included in the filtering information by referencing statistical information related to potential customers of the resource.

4. The information processing system according to claim 3, characterized in that, The processor is configured to adjust the importance of statistical information related to the potential customer within the filtering information.

5. The information processing system according to claim 3 or 4, characterized in that, The processor acquires information by inputting information related to the virtual avatars constituting the aggregate, the resource information, and the statistical information into a learning model, and outputs this information as appeal information representing the appeal of the resources in the advertisement.

6. The information processing system according to claim 5, characterized in that, The processor inputs the search performance information of the potential customers into the resource information and then feeds it into the learning model.

7. The information processing system according to claim 5 or 6, characterized in that, The processor inputs the resource information and the appeal information into the learning model to execute the survey.

8. The information processing system according to claim 2, characterized in that, The processor performs the following processing: Obtain appeal information representing the appeal of the resource in the advertisement; and The advertising information for the resource is generated based on the resource information, the appeal information, and the survey results.

9. The information processing system according to claim 8, characterized in that, The processor performs the following processing: Generate search keyword information based on the results of the survey; and Based on the search keyword information, advertising information for the resource is generated.

10. The information processing system according to claim 9, characterized in that, The processor performs the following processing: The information output by inputting the resource information, the appeal information, and the survey results into the learning model is used as search intent information representing the search intent for the resource. and The information output by inputting the search intent information into the learning model is used as the search keyword information.

11. The information processing system according to claim 8, characterized in that, The processor outputs comparative information that includes the statistical information used to generate the advertising information and the statistical information related to potential customers who viewed the advertising information in a comparable manner, and obtains information output by inputting the comparative information into a learning model as the analysis information of the advertising information.

12. The information processing system according to any one of claims 1 to 11, characterized in that, The processor also sets the number of virtual avatars input into the learning model.

13. A program article containing a program for enabling a computer to perform the following functions: By inputting information related to the virtual avatars constituting the pre-generated totality and resource information related to resources into the learning model, a collection of virtual avatars extracted from the totality is obtained; Perform an investigation related to the resources on the collection; and Output the results of the survey.

14. An information processing method, comprising the following steps: By inputting information related to the virtual avatars constituting the pre-generated totality and resource information related to resources into the learning model, a collection of virtual avatars extracted from the totality is obtained; Perform an investigation related to the resources on the collection; and Output the results of the survey.

Citation Information

Patent Citations

  • Support and coordination for advertising campaigns

    JP2015501990A

  • Information processing device, information processing method and information processing program

    JP2023000788A

  • Intelligence generation system, method, and program

    JP2023056557A