System, program, and method for generating persona information

Abstract

When generating persona information, the persona's age should be determined appropriately. [Solution] A system for generating persona information related to the target persona of a business acquires business information related to the business, and based on the business information, generates persona information for each of several personas of different ages, where the persona information includes the age of the persona, and outputs the persona information for each of the personas.

Description

Technical Field

[0007] ,

[0001] The present invention relates to a system, program, and method for generating persona information.

Background Art

[0002] In business strategy formulation, especially in marketing, it is important to determine which segments of the market to target (targeting). As one method of targeting, it is known to use personas (virtual customer personas). Patent Document 1 describes conducting sales activities using personas.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, the conventional technology has a problem that it is difficult to appropriately determine the age of a persona.

[0005] For example, in the technology of Patent Document 1, although a user is required to input an age as target information for consumers, in order to determine the age, in-depth knowledge of the market is necessary, and generally it is difficult to determine an appropriate age.

[0006] The present invention has been made to solve such problems, and an object thereof is to provide a technique for appropriately determining the age of a persona when generating persona information.

Means for Solving the Problems

[0007] An example of a system according to the present invention is a system that generates persona information relating to the target persona of a business, The aforementioned system, - Obtain business information related to the above business, - Based on the aforementioned business information, persona information is generated for each of several personas of different ages, where the persona information includes the age of the persona. - For each of the aforementioned personas, the persona information is output.

[0008] In one example, the system does not display a message prompting the user to enter age-related information when acquiring the business information. In one example, the business information includes text data representing the content of the business, text data representing the geographical scope of the business, and a numerical value representing the average customer spending for the business. In one example, the system displays the persona information in a way that allows switching on a persona-by-personal basis. In one example, the persona information includes text data that represents the reason for generating the persona. In one example, the persona information is: Text data representing the attributes of the persona, The statistical values ​​relating to the aforementioned attributes, Text data representing the basis for calculating the aforementioned statistical values, Includes. In one example, the system is It outputs one or more business information records acquired in the past, and accepts an operation to select any of the business information records. Output the persona information previously generated for the selected business information.

[0009] An example of a program according to the present invention causes a computer to function as the system described above.

[0010] One example of a method according to the present invention is a method for generating persona information relating to the target persona of a business, The aforementioned method, The step in which a computer acquires business information related to the business; The step in which a computer generates persona information regarding each of a plurality of personas of different ages based on the business information, where the persona information includes the age of the persona; The step in which a computer outputs the persona information for each of the personas; comprises.

Advantages of the Invention

[0011] According to the present invention, when generating persona information, the age of the persona can be appropriately determined.

[0012] Other effects than those described above are explained by the embodiments and modifications described in this specification and the drawings.

Brief Description of the Drawings

[0013] [Figure 1] A diagram showing a configuration including a persona generation system according to Embodiment 1. [Figure 2] An example of the hardware configuration of the persona generation server 10 in FIG. 1. [Figure 3] A flowchart showing the processing flow of the persona generation server 10 in FIG. 1. [Figure 4] An example of the GUI 40 used in relation to step S1 in FIG. 3. [Figure 5] An example of the configuration of the business information DB in the persona generation server 10 in FIG. 1. [Figure 6A] A part of an example of the format when persona information is displayed (first persona information). [Figure 6B] A part of an example of the format when persona information is displayed (first persona information). [Figure 6C] A part of an example of the format when persona information is displayed (first persona information). [Figure 7A] A part of an example of the format when persona information is displayed (second persona information). [Figure 7B] Part of an example of the format when persona information is displayed (second persona information). [Figure 7C] Part of an example of the format when persona information is displayed (second persona information). [Figure 8A] Part of an example of the format when persona information is displayed (third persona information). [Figure 8B] Part of an example of the format when persona information is displayed (third persona information). [Figure 8C] Part of an example of the format when persona information is displayed (third persona information). [Figure 9] An example of a switching operation. [Figure 10] An example of a history selection screen.

Embodiments of the Invention

[0014] Hereinafter, embodiments of the present invention will be described based on the accompanying drawings. [Embodiment 1] FIG. 1 is a diagram showing a configuration including a persona generation system according to Embodiment 1 of the present invention. The persona generation system generates information related to a persona by executing the method described in this specification. The persona generation system can be constituted by, for example, only the persona generation server 10, but may include components other than the persona generation server 10. The persona generation system generates information (persona information) related to a persona targeted by a business such as a company.

[0015] The meaning of "business" can be appropriately defined by those skilled in the art and can be interpreted broadly. As an example, it means that an organization (such as a company, a research institution, a public organization, a country, a local government, etc., but not limited to these) provides goods and / or services, etc. to a plurality of persons (such as general consumers, but not limited to this).

[0016] The meaning of "persona" can be defined as appropriate by those skilled in the art and is open to broad interpretation. For example, it refers to a fictional person set as the target of a product and / or service. For instance, it could be a user, customer, or general consumer, but it could also be a company executive, manager, or employee, or a person in charge or a member of another organization, or any other type of person.

[0017] The persona generation server 10 is configured to communicate with the LLM server 20 (LLM: Large Language Model), the administrator terminal 31, and the user terminal 32 via a communication network (including, for example, the Internet). As described above, the persona generation system can consist only of the persona generation server 10, but it may also include other components. For example, the persona generation system may include one or all of the LLM server 20, the administrator terminal 31, and the user terminal 32.

[0018] The administrator terminal 31 is, for example, a client terminal operated by an administrator who manages the persona generation server 10. The user terminal 32 is, for example, a client terminal operated by a user of the persona generation system. There may also be client terminals operated by other individuals.

[0019] In the example in Figure 1, the communication network to which the persona generation server 10 is connected is divided into a communication network for communicating with the LLM server 20 and a communication network for communicating with the administrator terminal 31 and user terminal 32. However, these may include the same communication network, or may include the internet, for example.

[0020] Figure 2 shows an example of the hardware configuration of the persona generation server 10. The persona generation server 10 has a hardware configuration similar to that of a known computer, and includes, for example, an arithmetic means 11, a storage means 12, and an input / output means 13. The arithmetic means 11 includes, for example, a processor. The processor can be manufactured using an integrated circuit, ASIC, FPGA, etc. The storage means 12 includes, for example, a storage medium such as a semiconductor memory device and a magnetic disk device. Some or all of the storage medium may be non-transitory storage medium.

[0021] The input / output means 13 includes, for example, a communication device such as a network interface. The communication device can function as either an input device or an output device. The communication device can perform wired communication and / or wireless communication. The input / output means 13 may further include input devices such as a keyboard and a mouse, and output devices such as a display and a printer.

[0022] The storage means 12 stores a business information database and a persona information database. The business information database contains information about the business, and the persona information database contains information about personas. The specific format and content of this information will be described later. Furthermore, the storage means 12 may store a program. By executing this program, the processor may perform the functions described in this embodiment. That is, this program causes the computer to function as a persona generation server 10 according to this embodiment.

[0023] Although a diagram of the hardware configuration of the LLM server 20 is omitted, the LLM server 20 has a hardware configuration similar to that of a well-known computer, for example, the same configuration as that of the persona generation server 10 described above.

[0024] The LLM server 20 stores an LLM and can input natural language data into the LLM and obtain its output. The LLM is a trained model that has been trained, for example, using a large amount of text data to output appropriate natural language responses to natural language inputs, and various known models can be used. For example, GPT, BERT, PaLM, Llama, etc. are known, but the LLM server 20 can input input data (including prompt and reference data) received from the outside into the LLM and send the output from the LLM to the outside. In this embodiment, the input and output to the LLM are in Japanese, but other languages ​​may be used instead of Japanese, or in addition to Japanese.

[0025] Although illustrations of the hardware configurations of the administrator terminal 31 and user terminal 32 are omitted, they also have a hardware configuration as a well-known computer, and for example, they have a configuration similar to that of the persona generation server 10 described above. In particular, the administrator terminal 31 and user terminal 32 can be configured as a personal computer (PC, for example, a desktop PC or notebook PC), a smartphone, a mobile information terminal, etc.

[0026] The storage devices of the LLM server 20, administrator terminal 31, and user terminal 32 may store programs not shown. By having the computer's processor execute these programs, each computer may perform the functions of the LLM server 20, administrator terminal 31, and user terminal 32 as described in this embodiment. In other words, these programs may cause the computer to function as the LLM server 20, administrator terminal 31, and user terminal 32 according to this embodiment.

[0027] The operation of persona generation server 10 is described below.

[0028] Figure 3 is a flowchart illustrating the processing flow of the persona generation server 10. This flowchart shows an example of a method for generating persona information. The flowchart mainly describes the operation of the persona generation server 10, and the operation of other computers (LLM server 20, administrator terminal 31, and user terminal 32, etc.) is partially omitted, but the operation of other computers can be appropriately understood by those skilled in the art based on the description in this specification.

[0029] In the process shown in Figure 3, the persona generation server 10 first acquires business information (step S1).

[0030] Figure 4 shows an example of a GUI 40 used in relation to step S1. The GUI 40 is displayed, for example, on a user terminal 32. The user terminal 32 accepts business information input via the GUI 40 and sends the input business information to the persona generation server 10. The persona generation server 10 may also send information (for example, HTML data, but not limited to this) to the user terminal 32 to display the GUI 40.

[0031] In this embodiment, the business information includes text data representing the content of the business ("business content"), text data representing the geographical scope of the business ("business area"), and a numerical value representing the average customer spending for the business ("business average customer spending"). The business average customer spending may also include data other than numerical values ​​(for example, symbols or strings representing currency units).

[0032] To input this information, the GUI 40 includes a business content field 41, a business area field 42, and a business customer average spending field 43. Through this GUI 40 and the user terminal 32, the persona generation server 10 obtains business content, business area, and business customer average spending as business information. Note that the business information may also include other information related to the business.

[0033] Note that the example in Figure 4 shows the state before the input process begins, and each field displays a message prompting input (for example, "Please enter your business details"). However, these messages may not be displayed, or they may be automatically removed when the input process begins.

[0034] Furthermore, in the example shown in Figure 4, GUI40 includes a descriptive text 44 about the operation of the persona generation server 10. For example, it explains the process before and after persona generation, stating that generation takes some time (e.g., 5 minutes), that clearing GUI40 on the user terminal 32 during generation does not affect the process, that an email notification is sent when generation is complete, that the email contains a URL to view the persona generation results, and that the persona generation results can also be viewed via another GUI (e.g., a menu screen). Including such a descriptive text in GUI40 improves user convenience. Note that the specific content of the descriptive text can be changed as appropriate. Also, GUI40 may not include such a descriptive text.

[0035] Figure 5 shows an example of the configuration of the business information database in the persona generation server 10. After step S1, the persona generation server 10 can save the entered business information in this format.

[0036] The User ID is information that identifies the user using the Persona Generation Server 10. In the Business Information Database, information about the user is associated with the User ID and stored accordingly. The user information includes the email address, Business ID, and Business Information. The email address is the email address used by the user. The Persona Generation Server 10 may accept email address input for each user in advance and store it in the Business Information Database.

[0037] The business ID is information that identifies the entered business information. If the same user enters business information multiple times (i.e., if the process in Figure 3 is repeated for the same user based on different business information), a unique business ID will be assigned to each entry. As mentioned above, the business information includes the business content, business area, and average customer spending.

[0038] Returning to Figure 3, the persona generation server 10 then obtains a prompt (first prompt) that constitutes part of the instructions for the LLM (step S2). The first prompt is pre-stored in the storage means 12 of the persona generation server 10, for example, but may also be stored on another computer (for example, the administrator terminal 31), in which case the persona generation server 10 may obtain the first prompt by receiving it from that other computer. An example of a first prompt is as follows.

[0039] [Specific example of the first prompt] Instructions: Based on the business information entered by the user, please generate a persona that includes the following information. # Instructions 1. Users will enter the following information regarding their business, service area, and average customer spending. 2. Based on the entered information, please generate the answer following the answer generation process. # the purpose • Personas are generated based on the business information entered by the user. • All data is generated based on each individual persona that is created. # Flow of response generation 1. Please identify the age groups that are likely to use the entered service. Please provide the minimum and maximum age range for that age group. The minimum age should be 15 years old or older, and the maximum age should be 80 years old or younger. 2. After providing three age ranges, please generate an answer according to the following guidelines. Do not write your own original text. The important point here is that we want to exclude people who are unlikely to use our service, and the personas generated here are important because they will continue to be used afterward. Therefore, please keep in mind that these are personas who are likely to use our service. *Important Note* There is no single correct answer, but please think through everything thoroughly, without omitting anything, and explore all possibilities in a way that feels natural. Please think through it thoroughly, even if it means loop checking. You need to consider the consistency to the point where you can answer the question, "Why did you choose this persona?" Furthermore, if the entered business is an online service, please analyze it in depth, as this is especially important. For example, if you use Twitter, you might have multiple accounts, such as a secondary account or an account for real-life friends, so please include aspects such as your level of engagement with the service, whether you are a lurker or a poster, and what genres of Netflix you watch.

[0040] The first prompt is configured to cause the LLM to output multiple different ages. In the example above, it states, "The minimum age should be 15 years or older, and the maximum age should be 80 years or younger. ...Once you have given three ages, generate an answer according to the following." Therefore, the LLM can determine three different ages within the age range of 15 to 80 years old. The first prompt also anticipates that the user will input business information and includes expressions that refer to that information (for example, "Business information entered by the user," "Business content, business area, average customer spending," etc.).

[0041] The first prompt may include instructions that specify the output format from the LLM. For example, any known format such as HTML, XML, CSV, or JSON can be specified.

[0042] The first prompt may include instructions to suppress prompt injection. For example, it may include instructions that limit the output information to a specific output format, instructions that prohibit changing the prompt, etc.

[0043] Next, the persona generation server 10 combines the business information obtained in step S1 and the first prompt obtained in step S2 and sends it to the LLM server 20, thereby determining the persona's age (step S3). Here, the business information specifies the business content, business area, and average customer spending, and the first prompt includes a sentence that refers to this information. Therefore, the input data to the LLM includes information for determining the persona's age based on the business information, and the LLM can appropriately determine the persona's age accordingly.

[0044] In step S3, the LLM server 20 receives this input data and inputs it into the LLM, obtaining the corresponding persona age from the LLM. The LLM server 20 then sends the persona age to the persona generation server 10. The persona generation server 10 receives the persona age from the LLM server 20. In this way, the persona generation server 10 uses the LLM server 20 to determine the persona age.

[0045] Next, the persona generation server 10 obtains a prompt (second prompt) that constitutes another part of the instructions for the LLM (step S4). The second prompt is pre-stored in the storage means 12 of the persona generation server 10, for example, but may also be stored on another computer (for example, the administrator terminal 31), in which case the persona generation server 10 may obtain the second prompt by receiving it from that other computer. An example of a second prompt is as follows:

[0046] [Specific examples of the second prompt] Instructions: Based on the business details entered by the user, please generate a persona that includes the following information. Please enter values ​​in all fields. # the purpose • Generates personas based on the business details entered by the user. • All data will be generated based on the created personas. # Instructions 1. Users enter their business details, service area, and average customer spending. 2. Based on the input, please generate the answer following the answer generation process. Points to note: Please exclude personas who are unlikely to use the service. • Since the generated personas will be used continuously in the future, it is important that they are personas who are likely to use the business. There is no single correct answer, but explore the possibilities, don't omit anything, and think through everything deeply without any sense of incongruity. Please ensure consistency to a level where you can answer the question, "Why did you choose this persona?" This is especially important if your business involves internet services, so please analyze it in detail. • Example: In the case of Twitter, consider the possibility of having multiple accounts, such as a secret account or an account for real-life friends. Consider your level of engagement with the service, such as whether you'll be a lurker or a poster. • In the case of Netflix, we also consider what genres you watch. - profile - Name (Common names like Sato, Tanaka, or Suzuki are not allowed. Since we will be handling multiple personas after generating the information, duplicate names are unacceptable, so please randomize the names. Also, please consider age and gender when selecting names.) - age - Gender - Place of residence (prefecture, city, town, or village) - Annual income - Occupation - post - Commuting area (prefecture, city, ward, town, or village) - Educational background - hobby - Family composition - Number of people - Marital status - Spouse's age - Do you have pets? - Pet information - Human relationships - Number of friends - How to interact with friends - Lifestyle - Wake up time - Bedtime - Commuting time - Commuting methods - Working hours - holiday - How to spend your holidays - Internet-related - Owned devices - Usage status - Hours of use - Timing of use - Web services used - In-depth analysis of web services and apps used (Please provide specific names of existing services and apps that are likely to be used by someone who fits the persona's profile, rather than general services or apps.) - Entered business details related items - The original reason for using it - What is the best timing for using it? - What is the cost of using it? 3. Please generate one persona for each of the three age groups that you will enter separately. 4. Summarize the reason for creating the persona in 250-600 characters and output it as the reason for persona creation. Points to note: There is no single correct answer, but please create a well-structured and convincing argument that is easy to understand and persuasive. Please focus on individual personas, rather than age groups, and describe them from the perspective of "this person." Avoid grouping them together as a target zone. 5. Generate estimated values ​​for the created personas according to the following items. Points to note: Please avoid unnecessary embellishments and display the information clearly using only numbers. • Answers should be based on reliable data (referencing multiple sources as needed) or pre-trained data, and Fermi estimation should be used if the answer is unclear. • For the calculation of the population and target group (totalPopulationReason), please formulate a hypothesis based on the factual data in the source and summarize the reasoning behind the calculation of the number of relevant individuals. This is extremely important. It is acceptable for this section to be lengthy. We want a convincing explanation from both a quantitative and qualitative perspective. The logic used to calculate the target group from the population must not be flawed. Example output: - Age - Number of target individuals (targetCount) - Total Population - Reasons for calculating the population and target group (totalPopulationReason) - Source (URL required) (sourceURL) - Gender - Number of target individuals (targetCount) - Total Population - Reasons for calculating the population and target group (totalPopulationReason) - Source (URL required) (sourceURL) - Education - Number of target individuals (targetCount) - Total Population - Hypothesis - Fact - Reasons for calculating the population and target group (totalPopulationReason) - Source (URL required) (sourceURL) - Occupation - Number of target individuals (targetCount) - Total Population - Reasons for calculating the population and target group (totalPopulationReason) - Source (URL required) (sourceURL) - Age x Gender (agegender) (Considering the content generated in the previous stage) - Number of target individuals (targetCount) - Total Population - Reasons for calculating the population and target group (totalPopulationReason) - Age x Occupation (ageoccupation) (Considering the content generated in the previous stage) - Number of target individuals (targetCount) - Total Population - Reasons for calculating the population and target group (totalPopulationReason) - Gender x Occupation (considering the content generated in the previous stage) - Number of target individuals (targetCount) - Total Population - Reasons for calculating the population and target group (totalPopulationReason) 6. Finally, summarize the content you have generated so far in 250-600 characters and output it to the summary. # Additional information for the whole There is no single correct answer when it comes to personas. Display only numbers; no unnecessary decorations or repetitions are needed. • Please write your answer in Japanese and complete the entire process without stopping midway. Please provide all the listed items and include proper sources. Please be sure to include your city / ward / town / village level when specifying your place of residence.

[0047] The second prompt is configured to cause LLM to output persona information for multiple personas of different ages. In the example above, it states, "Generate one persona for each of the three ages that will be entered separately," so LLM generates persona information for each age. Also, like the first prompt, the second prompt anticipates that the user will input business information and includes expressions that refer to that information (for example, "User-inputted business details," "Business details, business area, average customer spending," etc.). Furthermore, the "Profile" section in the second prompt includes an "Age" field, which ensures that the persona information clearly includes the persona's age.

[0048] The second prompt may include instructions specifying the output format from the LLM. For example, any known format such as HTML, XML, CSV, or JSON can be specified.

[0049] The second prompt may include instructions to suppress prompt injection. For example, it may include instructions that limit the output information to a specific output format, instructions that prohibit changing the prompt, etc.

[0050] Next, the persona generation server 10 combines the business information obtained in step S1, the age determined in step S3, and the second prompt obtained in step S4 and sends them to the LLM server 20, thereby generating persona information (step S5). Here, the business information specifies the business content, business area, and average customer spending, and also specifies three different ages, and the second prompt includes a sentence that refers to this information. Therefore, the input data to the LLM includes information for generating persona information based on the business information and the persona's age, and the LLM can appropriately generate persona information accordingly.

[0051] In step S5, the LLM server 20 receives this input data and inputs it into the LLM, obtaining the corresponding persona information from the LLM. The LLM server 20 then sends the persona information to the persona generation server 10. The persona generation server 10 receives the persona information from the LLM server 20. In this way, the persona generation server 10 generates persona information using the LLM server 20.

[0052] In this way, the persona generation server 10 generates persona information for each of several personas of different ages based on business information. As mentioned above, the persona information includes the persona's age. For example, according to the second prompt, the persona information includes name, age, gender, place of residence, annual income, occupation, position, commuting area, educational background, hobbies, family structure, relationships, lifestyle, internet-related information, items related to business content, etc. Family structure includes the number of people, marital status, spouse's age, pet ownership, pet information, etc. Relationships include the number of friends, how they interact with friends, etc. Lifestyle includes wake-up time, bedtime, commuting time, means of commuting, working hours, holidays, how holidays are spent, etc. Internet-related information includes owned devices, usage status, usage time, timing of usage, web services used, application in-depth information (detailed information on application usage), etc. Items related to business content include reasons for use, timing of use, amount of money spent, etc.

[0053] Persona information can be useful for targeting as long as it includes at least age, but it is more beneficial to include a variety of information other than age, as described above. The persona information described above is just an example, and variations may be made, such as omitting some of these information items or including other information. The structure of the persona information database that stores such persona information is not specifically illustrated, but a person skilled in the art can design it as appropriate.

[0054] Next, the persona generation server 10 stores the persona information in the storage means 12 (step S6). This allows the persona information to be output at any time thereafter.

[0055] Next, the persona generation server 10 outputs the saved persona information for each generated persona (step S7). Output can be performed at any time. For example, output may be performed by sending the persona information to the user terminal 32 immediately after step S6. Alternatively, after step S6, if the user terminal 32 requests the persona generation server 10 to output the persona information, output may be performed by sending the persona information to the user terminal 32 in response.

[0056] After step S7, the persona information is further output and made available for the user to view. For example, the user terminal 32 may output the received persona information by displaying it on a display device (for example, an LCD display device, but not limited to that) or by printing it from a printing device.

[0057] The persona generation server 10 may perform processing to facilitate viewing of persona information on the user terminal 32. For example, upon completion of step S6, an email may be sent to the relevant user to notify them that the generation of persona information is complete. This email may contain information (such as a unique URL) that allows the persona generation server 10 to identify the persona information. The parts of the explanation in Figure 4 that say "2. After generation, a notification email of generation completion will be sent to the logged-in email address" and "3. You can view the generation results from the URL in the email" are based on the assumption that such an email will be sent.

[0058] Figures 6A and 6C show examples of the format in which persona information may be displayed. The output examples in Figures 6A and 6C show the same screen divided vertically, with Figure 6A representing the top of the screen, Figure 6C representing the bottom of the screen, and Figure 6B representing the portion between Figures 6A and 6C. The entire content shown in Figures 6A and 6C may be displayed on a single screen, on a page-by-page basis, or in a scrollable format.

[0059] As shown in Figure 6A, business information is displayed along with persona information. "Input Value" indicates business information, and "Input ID" corresponds to the business ID. In this embodiment, three personas are generated, but Figures 6A to 6C display the persona information (first persona information) of one of them, "Shinya Miyamoto" (first persona). For the sake of explanation, the output examples in Figures 6A to 6C are not output by the second prompt described above, but rather output by a prompt with some modifications to the second prompt. There are some differences in some items, but it still serves as a concrete example of persona information.

[0060] In the example in Figure 6A, the persona information includes the reason for persona creation (i.e., text data representing the reason for creating the persona based on business information). Users can use this as a reference for targeting.

[0061] In the example in Figure 6B, the persona information includes text data representing the attributes of the persona. For example, the persona information includes information such as age (47 years old), gender, place of residence, annual income, occupation, position, educational background, and hobbies, making it useful for specifically targeting customers. Furthermore, the persona information includes information about family structure, which is particularly useful for businesses targeting families as customers. In addition, it includes information such as wake-up time, bedtime, commute time, mode of transportation, start time, end time, and holidays, which is particularly useful when determining the hours during which the business is offered.

[0062] Furthermore, persona information is particularly useful for businesses targeting customer groups of friends, as it includes information about the number of friends a person has and how they interact with those friends. Persona information is also particularly useful for businesses that utilize computers (especially mobile devices), as it includes information about devices owned, their usage, usage time, timing of use, web services or applications used, and in-depth information about those services or applications. Additionally, persona information is particularly useful when determining the content of a business, timing of service provision, average customer spending, etc., as it includes reasons for use, timing (reasons for using the service at a particular time), and price (reasons for using the service at a particular price point).

[0063] In the example in Figure 6C, the persona information includes statistical values ​​related to the attributes in Figure 6B, and text data representing the basis for calculating those statistics. For example, the number of people matching the persona's age (47 in this example), gender (male in this example), education level (university graduate in this example), and occupation (company employee in this example) is displayed, along with the basis for that estimation. This is particularly useful when trying to understand the target number. Similarly, the number and the basis for that estimation are also displayed for combinations of these conditions (age and gender combination, age and occupation combination, gender and occupation combination, etc.), allowing you to understand the target number based on more complex conditions.

[0064] Figures 6A to 6C, described above, show the persona information for one of the three personas, but the same format of persona information is displayed for the other two. Figures 7A to 7C show the persona information for the second persona (age 15), and Figures 8A to 8C show the persona information for the third persona (age 80).

[0065] The display format for the persona information of the three personas can be designed as appropriate, but for example, the persona generation server 10 may display the persona information in a switchable format on a persona-by-persona basis. As a specific example, it can be displayed as follows: When displaying persona information for the first time, the information of the first persona is displayed, and the information of the second and third personas is not displayed. Depending on the user's display switching operation, the information of the second or third persona is displayed, and the information of the first persona is cleared.

[0066] This type of display switching allows for efficient use of the display screen. Specifically, it allows for the clear display of information for a particular persona using the entire screen, while simultaneously displaying information for all personas through screen switching.

[0067] Figure 9 shows an example of such a switching operation. When the user terminal 32 is equipped with a touch panel display device, the second persona information (Figure 7C) is displayed by swiping the screen to the left while the first persona information (Figure 6C) is displayed. Figure 9 shows the state in the middle of this switching operation. Similarly, the third persona information is displayed by swiping the screen to the left while the second persona information is displayed. Conversely, the second persona information is displayed by swiping the screen to the right while the third persona information is displayed, and the first persona information is displayed by swiping the screen to the right while the second persona information is displayed.

[0068] The switching operation is not limited to a swipe operation; it may also be a drag operation using a mouse. It may also be a button operation. For example, the user terminal 32 may display buttons corresponding to the first to third persona information (first button to third button), and display the corresponding persona information in response to the operation of each button. That is, the first persona information may be displayed in response to the operation of the first button, the second persona information in response to the operation of the second button, and the third persona information in response to the operation of the third button. Furthermore, other switching operations can also be designed.

[0069] As described above, the persona generation server 10 according to Embodiment 1 automatically generates persona information in response to the input of business information. As is clear from the above description, the persona generation server 10 does not request the input of age when acquiring business information, nor does it display a message prompting the input of age-related information (age, age group, age group of birth date, etc.). Therefore, even if the user does not determine the persona's age when inputting business information, persona generation is possible. As a result, even for users who do not have knowledge about the target group's age, the persona's age is automatically and appropriately determined.

[0070] Furthermore, users can also enter their age themselves as part of their business information. In addition, as a variation, the persona generation server 10 may request input of information representing the persona's age, or it may display a message prompting the user to enter age-related information.

[0071] In addition to the processes described above, the persona generation server 10 may perform processes to enhance user convenience. For example, it may associate and store persona information generated based on the entered business information for each business ID, allowing users to freely refer to the history later.

[0072] Figure 10 shows an example of a history selection screen. Immediately after a user logs in to the user terminal 32, the persona generation server 10 outputs the business information (which may be multiple) previously acquired from that user, along with a business ID (referred to as "Input ID" in Figure 10) that identifies the business information, and displays it on the user terminal 32. The server then accepts an operation to select one of the business information via the user terminal 32. If any of the business information is selected, the persona generation server 10 outputs the persona information previously generated for the selected business information (for example, sending the persona information to the user terminal 32 in the same manner as in step S7, and the user terminal 32 displays the persona information).

[0073] This approach allows users to refer to previously generated persona information at any time, improving convenience. Furthermore, while LLM (Limited Licensing Management) has the characteristic of sometimes producing different output even with the same input, using the history allows for the display of previously generated persona information. Users can also obtain different persona information based on the same business information by re-entering the business information without using the history.

[0074] [Other variations]

[0075] In Embodiment 1, the persona generation server 10 uses an LLM to generate persona information, but it may use other pre-trained models instead of or in addition to the LLM, and it is also possible to generate persona information using a system other than a pre-trained model. As a specific example, it is also possible to configure the system to output multiple ages according to a deterministic algorithm based on business information.

[0076] The transmission and reception of information between the persona generation server 10 and the LLM server 20 is not limited to the content and format described in Embodiment 1. For example, three or more prompts may be used, or a single prompt (for example, one that substantially includes the content of the first and second prompts) may be used.

[0077] Those skilled in the art can add, modify, or delete components as they see fit. For example, the persona generation server 10 can be composed of multiple computers connected via a communication network.

[0078] The order of the steps shown in Figure 3 can be changed as appropriate. For example, the order of steps S1 and S2 in Figure 3 can be swapped, and step S4 can be executed at any time before step S5. [Explanation of Symbols]

[0079] 10... Persona generation server 11...Arithmetic means 12...Memory means 13…Input / output means 20...LLM Server 31…Administrator terminal 32... User terminal 40…GUI 41…Business Description Field 42…Business Area Field 43… Business Customer Average Price Field

Claims

1. A system for generating persona information about the target personas of a business, The aforementioned system, - Obtain business information related to the above business, - Based on the aforementioned business information, persona information is generated for each of several personas of different ages, where the persona information includes the age of the persona. - For each of the aforementioned personas, output the aforementioned persona information. system.

2. The aforementioned system does not display a message prompting the input of age-related information when acquiring the aforementioned business information. The system according to claim 1.

3. The aforementioned business information includes text data representing the content of the business, text data representing the geographical scope of the business, and numerical data representing the average customer spending for the business. The system according to claim 1.

4. The system displays the persona information in a way that can be switched on a persona-by-person basis. The system according to claim 1.

5. The aforementioned persona information includes text data that represents the reason for generating the persona. The system according to claim 1.

6. The aforementioned persona information is, Text data representing the attributes of the persona, The statistical values ​​relating to the aforementioned attributes, Text data representing the basis for calculating the aforementioned statistical values, including, The system according to claim 1.

7. The aforementioned system, It outputs one or more business information records acquired in the past, and accepts an operation to select any of the business information records. Output the persona information previously generated for the selected business information. The system according to claim 1.

8. A program that causes a computer to function as the system described in claim 1.

9. A method for generating persona information about the target persona of a business, The aforementioned method, The computer acquires business information relating to the said business, The computer generates persona information for each of several personas of different ages based on the business information, where the persona information includes the persona's age, in the following steps: The computer outputs the persona information for each persona, A method that includes [a certain feature].

Images