Generation device, support system, generation method, and computer program

Abstract

We provide new technology to assist in analyzing customer characteristics. [Solution] The communication unit is configured to be able to communicate with the artificial intelligence. The control unit is configured to execute the following operations: acquire customer data regarding a customer set classified by a predetermined method, create shaped data by performing a shaping process on the customer data, send a request signal to the artificial intelligence requesting that one or more personas be generated based on the shaped data, and receive persona data, which is data representing the one or more personas, as output of the artificial intelligence based on the request signal. The shaping process includes creating explanatory phrases that are words that describe each of a plurality of numerical values ​​and associating the explanatory phrases with each of the plurality of numerical values ​​to create the shaped data.

Description

[Technical Field]

[0001] The present disclosure relates to an assistance system. [Background technology]

[0002] There are known technologies that support the analysis of customer characteristics in order to consider marketing measures for product promotions, new product development, etc. For example, Patent Document 1 discloses a marketing support system that can easily perform analysis of customer segments for appropriate marketing activities. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2020-149197 Summary of the Invention [Problem to be solved by the invention]

[0004] In the system disclosed in Patent Document 1, there is room for improvement in order to analyze customer characteristics in more detail. One aspect of the present disclosure is to provide new techniques to assist in analyzing customer characteristics. [Means for solving the problem]

[0005] One aspect of the present disclosure is a generation device including a control unit and a communication unit, the communication unit being configured to be able to communicate with artificial intelligence. The control unit is configured to acquire customer data regarding a customer set classified by a predetermined method. The control unit is configured to create shaped data by performing a shaping process on the customer data. The control unit is configured to send a request signal to the artificial intelligence requesting that one or more personas be generated based on the shaped data. The control unit is configured to receive persona data, which is data representing the one or more personas, as an output of the artificial intelligence based on the request signal.

[0006] The customer data includes a plurality of numerical values ​​related to a customer set. The formatting process includes creating explanatory phrases that are words that describe each of the plurality of numerical values ​​and associating the explanatory phrases with each of the plurality of numerical values ​​to create formatted data.

[0007] This configuration allows the AI ​​to generate an appropriate persona based on the shaping data, allowing the user of the generation device to analyze customer characteristics by referring to the persona data output by the AI.

[0008] In one aspect of the present disclosure, the customer data may include a plurality of items. Each of the plurality of numerical values ​​may be a value corresponding to each of the plurality of items. The descriptive phrase may include some or all of a phrase representing each of the plurality of items.

[0009] This configuration allows for the creation of explanatory phrases using words that represent multiple items, making it easy to create formatted data that allows AI to deeply understand the meaning of numerical values.

[0010] In one embodiment of the present disclosure, the descriptive phrase is a first phrase that represents each of a plurality of items. A second phrase that is common to a plurality of items and that explains the plurality of items may be added to at least one of the beginning and end of the above.

[0011] This configuration allows for the creation of explanatory phrases by adding words common to multiple items to supplement information that is insufficient when describing individual items alone. This allows for the AI ​​to appropriately generate personas based on the shaped data.

[0012] In one embodiment of the present disclosure, the plurality of items may be a plurality of items included in a specific attribute related to the customer. The second phrase may include a phrase for indicating that the item represented by the first phrase is an item related to the specific attribute.

[0013] In one aspect of the present disclosure, the specific attribute may relate to at least one of gender, age, residential area, family structure, occupation, educational background, and annual income.

[0014] This configuration allows the words representing each of the multiple items to be converted into a format that is easy for the artificial intelligence to analyze, thereby improving the accuracy of analyzing customer characteristics.

[0015] In one aspect of the present disclosure, the control unit may be configured to perform normalization of the plurality of numerical values, and the shaping data may be created by associating a descriptive phrase with each of the plurality of normalized numerical values.

[0016] This configuration can reduce the variation in multiple numerical values ​​in customer data, which helps AI generate appropriate personas.

[0017] In one aspect of the present disclosure, the plurality of numerical values ​​may include values ​​for each customer segment defined by classifying a set of customers based on their purchasing history.

[0018] With this configuration, it is possible to receive persona data corresponding to a predetermined purchasing category.

[0019] In one aspect of the present disclosure, the request signal may include a request to generate multiple personas as one or more personas. The persona data may be data representing the multiple personas.

[0020] With this configuration, multiple personas can be received from the artificial intelligence in response to one input.

[0021] One aspect of the present disclosure is a classification device comprising an acquisition unit and a classification unit. The acquisition unit is configured to acquire survey results, which are responses to a survey conducted for a plurality of customers, for each customer, and purchase data, which includes product purchase histories for each of the plurality of customers. The classification unit is configured to classify the plurality of customers based on the survey results and the purchase data. The survey includes questions requesting information about individual respondents.

[0022] This configuration allows multiple customers to be classified based on purchase data in addition to the survey results, thereby enabling multiple customers to be appropriately classified according to their purchasing characteristics and improving the accuracy of analyzing customer characteristics.

[0023] In one aspect of the present disclosure, the purchase history may include product identification information and the purchase date and time of the product. good.

[0024] In one aspect of the present disclosure, the questionnaire may include a question asking about product awareness, the purchase history may include information about product purchase frequency, and the classification unit may be configured to classify the plurality of customers based on product awareness and product purchase frequency.

[0025] In one aspect of the present disclosure, the information about the individual respondent may include at least one of the gender, age, occupation, preferences, and media usage of the individual respondent.

[0026] According to one aspect of the present disclosure, there is provided an assistance system including the above-described generating device and the above-described classifying device, wherein the generating device is configured to acquire, as customer data, data relating to a plurality of customers classified by the classifying device.

[0027] According to this configuration, it is possible to receive one or more persona data corresponding to the plurality of customers classified by the classification device, thereby making it possible to analyze the characteristics of the plurality of customers classified by the classification device.

[0028] One aspect of the present disclosure is a computer-implemented generation method, the method including: communicating with an artificial intelligence; acquiring customer data regarding a customer set classified by a predetermined method; creating shaped data by performing a shaping process on the customer data; transmitting a request signal to the artificial intelligence requesting generation of one or more personas based on the shaped data; and receiving persona data representing the one or more personas as output of the artificial intelligence based on the request signal. The customer data includes a plurality of numerical values ​​regarding the customer set. The shaping process includes creating explanatory phrases that are words that describe each of the plurality of numerical values ​​and associating the explanatory phrases with each of the plurality of numerical values ​​to create the shaped data.

[0029] This generation method provides the same effects as the generation device described above.

[0030] In one aspect of the present disclosure, a computer program may be provided for causing a computer to execute, at least in part, the above-described generating method.

[0031] One aspect of the present disclosure is a classification method executed by a computer, which includes acquiring survey results, which are responses to a survey conducted for each of a plurality of customers, and purchasing data, which includes product purchasing histories for each of the plurality of customers, and classifying the plurality of customers based on the survey results and the purchasing data. The survey includes questions asking for information about individual respondents.

[0032] This classification method provides the same effects as the classification device described above.

[0033] In one aspect of the present disclosure, a computer program may be provided to cause a computer to at least partially execute the classification method described above. [Brief explanation of the drawings]

[0034] [Figure 1] FIG. 1 is a block diagram showing the configuration of a support system. [Figure 2] FIG. 10 is a diagram showing the structure of the survey results. [Figure 3] FIG. 2 is a diagram showing the configuration of purchase data. [Figure 4] FIG. 2 is a diagram showing the configuration of customer data. [Figure 5] 10 is a flowchart illustrating a classification process. [Figure 6] FIG. 1 is a diagram showing a customer pyramid. [Figure 7] FIG. 1 is a diagram showing a 9-segment map. [Figure 8] 10 is a flowchart showing a generation process. [Figure 9] FIG. 2 is a diagram illustrating a configuration of normalized data. [Figure 10] FIG. 2 is a diagram illustrating a configuration of shaped data. [Figure 11] FIG. 1 is a diagram showing persona data. [Figure 12] This diagram shows an interview conducted with an artificial intelligence acting as a persona. DETAILED DESCRIPTION OF THE INVENTION

[0035] Hereinafter, exemplary embodiments of the present disclosure will be described with reference to the drawings. [1. First embodiment] [1-1.Configuration] [1-1-1. Overall structure] The support system 100 shown in FIG. 1 is a system for supporting the analysis of customer characteristics.

[0036] The assistance system 100 includes an assistance device 1, an artificial intelligence 2, and a database group 3. The support device 1 includes a processor 11, a memory 12, a storage 13, a user interface 14, and a communication interface 15. The support device 1 is installed in an information terminal such as a personal computer.

[0037] The processor 11 is configured to execute processing in accordance with a computer program recorded in the storage 13 . The memory 12 is used as a work area when the processor 11 executes processing. Examples of the memory 12 include a random access memory (RAM), a read only memory (ROM), and a flash memory.

[0038] The storage 13 stores computer programs and data used when executing processes according to the computer programs. An example of the storage 13 is a hard disk drive (HDD). Hard disk drives (HDDs) and solid state drives (SSDs).

[0039] The user interface 14 is a general term for an interface for receiving various input operations from a user and an interface for outputting various information to the user. Examples of the user interface 14 include a keyboard, a mouse, a touch panel, a display, etc.

[0040] The communication interface 15 is an interface capable of communicating various types of data in accordance with a predetermined standard. The assistance device 1 is configured to be able to communicate with the artificial intelligence 2 and the database group 3 through the communication interface 15.

[0041] The artificial intelligence 2 is configured to generate text, images, etc. based on input information. An example of the artificial intelligence 2 is generative AI (artificial intelligence). Examples of generative AI include ChatGPT by OpenAI, Gemini by Google, Copilot by Microsoft, and Claude by Anthropic.

[0042] The database group 3 includes one or more databases. Each database is configured to be able to hold various types of data. The various types of data held in the database group 3 include survey results 40, purchase data 50, and customer data 60. There is no particular limitation on which database the various types of data held in the database group 3 are held in. In other words, Each piece of data may be held on the same server as the other data, or may be held on a server separate from the other data.

[0043] As shown in FIG. 2, the survey results 40 are data that represent the response results of a survey administered to multiple customers, for each customer ID that can identify the customer. A questionnaire includes questions that ask for various information about the individual respondent. The term "respondent" may be interpreted as "customer." For example, a questionnaire may include questions asking about the respondent's brand purchasing history, their attitude toward the brand, etc. For example, a questionnaire may include questions asking about the respondent's attributes, interests, values, media and service usage, etc.

[0044] As an example, the questionnaire may include questions about whether or not the customer is aware of the brand, past and current (e.g., last three months) purchasing experience with the brand, frequency of purchasing the brand, whether or not the customer intends to purchase the brand, what the customer considers important when purchasing the brand, usage of media and services, attitudes toward media and services, occupation, family composition, hobbies and preferences, daily behavior and attitudes, daily worries, and celebrities and talents that the customer likes.

[0045] FIG. 2 shows an example of responses to various questions included in the questionnaire from a customer whose customer ID is "1234" in the questionnaire result 40. As shown in Fig. 3, the purchase data 50 is data including the purchase history of products for each customer ID. In the purchase data 50, the purchase date and information that can identify the product are associated with each customer ID. The customer ID in the purchase data 50 may be the same as the customer ID in the survey result 40.

[0046] The purchase date is the date on which the customer purchased the product. In addition to the purchase date, the purchase data 50 may also include the time when the product was purchased. The information that can identify a product is, for example, a JAN code (Japanese Article Number) or a product name.

[0047] The purchase data 50 may include the store where the customer purchased the product, the product category, the number of purchases, and the like. 3 shows, for each customer ID, the purchase date, purchase time, purchase store, category, JAN code, product name, and purchase quantity as an example of purchase data 50. For example, the purchase history shows that a customer with customer ID "14" purchased one product (JAN code = 49*********01) in the water category at a discount store at 4:00 PM on October 23, 2024.

[0048] As shown in Figure 4, customer data 60 is data relating to a set of customers classified by a predetermined method. The predetermined method is, for example, a classification process described below. The customer data 60 includes a plurality of items I and a plurality of statistical values ​​SV, which are numerical values ​​relating to the set of customers. Each of the plurality of statistical values ​​SV is a value corresponding to each of the plurality of items I.

[0049] In the example shown in FIG. 4, customer data 60 includes the number and percentage of people who fall into a predetermined occupation for a total of 50,000 customers. For example, the number and percentage of people who fall into the category of "food and beverage manufacturer" are "1020 (people)" and "2.0 (%)." In this case, the predetermined occupations, such as "food and beverage manufacturer" and "textile and clothing manufacturer," correspond to examples of multiple items I. The number and percentage of people who fall into a predetermined occupation, such as "1020" and "2.0," correspond to examples of multiple statistical values ​​SV.

[0050] The customer data 60 may include numbers and / or percentages of the entire customer population divided into categories based on purchasing categories. For example, purchasing categories may include "online purchasers" and "non-online purchasers." A category of "offline purchaser" and "offline non-purchaser" may be set as purchasing categories. An "online purchaser" and an "online non-purchaser" respectively represent customers who have purchased online via an e-commerce site, etc., and customers who have no purchasing experience. An "offline purchaser" and an "offline non-purchaser" respectively represent customers who have purchased in a physical store and customers who have no purchasing experience.

[0051] In the example shown in FIG. 4, the number and percentage of customers who belong to the "chemical and oil manufacturers" category who are "non-online purchasers" are "461 people" and "0.9%." The plurality of items I may correspond to options for various questions in the above-mentioned questionnaire, and the plurality of statistical values ​​SV may correspond to the number and / or percentage of people who selected the corresponding option.

[0052] For example, in a questionnaire, if the options for a question asking about the respondent's occupation include "food and beverage manufacturer" and "textile and clothing manufacturer," the multiple items I may include the options "food and beverage manufacturer" and "textile and clothing manufacturer," etc. The multiple statistical values ​​SV may include the number and / or percentage of people who selected "food and beverage manufacturer" in the questionnaire result 40.

[0053] That is, the customer data 60 may be created using the survey results 40. [1-2. Processing] A series of processes executed by the processor 11 will be described with reference to FIGS.

[0054] [1-2-1. Classification process] When an instruction is input from the user via the user interface 14, the processor 11 starts the classification process shown in FIG.

[0055] First, in S100, the processor 11 acquires the survey results 40 from the database group 3. Next, in S110, the processor 11 acquires the purchase data 50 from the database group 3.

[0056] Next, in S120, the processor 11 classifies the multiple customers into multiple segments based on the survey results 40 and the purchase data 50. Examples of multiple segments include five segments called a customer pyramid as shown in Fig. 6 and nine segments called a 9-segment map as shown in Fig. 7. The 9-segment map is a nickname for 9segs (registered trademark).

[0057] The customer pyramid classifies customers for a certain brand into five segments: "loyal customers," "general customers," "lost customers," "aware but non-purchasing customers," and "unaware customers."

[0058] Loyal customers are those who are aware of the brand and purchase frequently. General customers are those who are aware of the brand but purchase less frequently. A lapsed customer is a customer who is aware of the brand and has purchased from it in the past, but is not currently purchasing from it.

[0059] Aware / non-purchasers are customers who are aware of the brand but have no purchase experience. Unaware customers are customers who are not aware of the brand. The 9-segment map is a customer pyramid that divides the five segments into "loyal customers," "general customers," "loyal customers," and "unaware customers" according to whether they have a purchasing intention for the brand. Each of the four segments of "purchasing customers" is further divided into two, and together with "unaware customers" there are nine segments in total. In the example shown in Figure 7, customers who have the intention to purchase are referred to as "active" customers, and customers who have no intention to purchase are referred to as "passive" customers.

[0060] The processor 11 classifies a certain customer by determining which segment the customer belongs to based on the questionnaire results 40 and purchase data 50 relating to the customer. For example, the processor 11 can determine whether or not there is brand awareness based on the survey results 40. The processor 11 can determine whether or not there is purchase frequency and whether or not there is purchase experience based on the purchase data 50.

[0061] For example, if the customer responds in the survey results 40 that they are aware of a certain brand, and in the purchase data 50, they have purchased that brand two or more times within the last three months, the processor 11 may classify the customer as a loyal customer of that brand.

[0062] Thereafter, processor 11 ends the classification process shown in FIG. In this way, the support device 1 classifies each customer into one of a plurality of segments based on the brand purchase history and the presence or absence of purchase intention in the questionnaire results 40 and the purchase data 50.

[0063] [1-2-2. Generation process] When an instruction is input from the user via the user interface 14, the processor 11 starts the generation process shown in FIG.

[0064] First, in S200, the processor 11 acquires customer data 60 from the database group 3. The customer data 60 is data relating to a set of customers classified by a predetermined method. For example, the customer data 60 is data relating to a set of customers classified as loyal customers by classification processing.

[0065] Next, in S210, the processor 11 calculates normalized data 61 having a plurality of normalized values ​​NV, which are the normalized plurality of statistical values ​​SV, by normalizing the plurality of statistical values ​​SV in the customer data 60. An example of the normalized data 61 is shown in FIG.

[0066] Normalization is a process commonly used in fields such as machine learning and statistics, in which data is transformed according to certain rules to make it easier to use. Normalization can reduce the effects of data variability.

[0067] Examples of normalization include Min-Max normalization and Z-score normalization. Min-Max normalization is a method of scaling data so that the minimum value is 0 and the maximum value is 1. Z-score normalization is a method of scaling data so that the mean is 0 and the standard deviation is 1. Z-score normalization is also called standardization.

[0068] Processor 11 does not need to normalize numerical values ​​among the plurality of statistical values ​​SV that are not used in subsequent processing. 9, the normalized data 61 has a plurality of normalized values ​​NV, which are values ​​obtained by normalizing the numerical values ​​representing the number of people on a row-by-row basis, among the plurality of statistical values ​​SV in the customer data 60. The numerical values ​​representing the ratios are not used in subsequent processing, so they are not normalized and are left blank.

[0069] For example, among the multiple statistical values ​​SV in Figure 4, numerical values ​​such as "1020", "203", "462", etc. corresponding to "overall" are normalized to numerical values ​​such as "0.169", "0.004", "0.056", etc. corresponding to "overall" as multiple normalized values ​​NV in Figure 9.

[0070] For example, among the multiple statistical values ​​SV in Figure 4, numerical values ​​such as "0", "1", "2", etc. corresponding to "online buyer" are normalized to numerical values ​​such as "0", "0.111", "0.222", etc. corresponding to "online buyer" as multiple normalized values ​​NV in Figure 9.

[0071] Next, in S220, the processor 11 performs a shaping process on the normalized data 61 to create shaped data 62. Specifically, the processor 11 creates explanatory phrases that are words and phrases that describe each of the multiple normalized values ​​NV in the normalized data 61, and creates the shaped data 62 by associating the explanatory phrases with each of the multiple normalized values ​​NV.

[0072] In the formatted data 62 shown in Figure 10, the words in the "Customer Explanation" column represent explanatory words. The values ​​in the "Total Content Rate" column represent the values ​​corresponding to the "Total" row among the multiple normalized values ​​NV in Figure 9. The explanatory words explain the values ​​in adjacent cells among the multiple normalized values ​​NV.

[0073] For example, descriptive phrases are "people who work for a food or beverage manufacturer" or "people who live with their husband / wife." These descriptive phrases explain the values ​​in adjacent cells, such as "0.169" and "0.695." In other words, the value "0.169" represents the percentage of "people who work for a food or beverage manufacturer" among all customers. The value "0.695" represents the percentage of "people who live with their husband / wife" among all customers.

[0074] The descriptive phrase may include some or all of the phrases representing each of the multiple items I in the customer data 60. For example, the descriptive phrase "a person who works for a food and beverage manufacturer" includes the phrase "food and beverage manufacturer" in the multiple items I in FIG. 4.

[0075] In other words, the explanatory phrase may be a phrase in which a first phrase, which is a phrase representing each of the multiple items I included in the customer data 60, is added to at least one of the beginning and end of the first phrase, and a second phrase, which is common to at least some of the multiple items I and describes the item, is added.

[0076] The plurality of items I may be a plurality of items included in a specific attribute related to the customer. The second phrase may include a phrase indicating that the item represented by the first phrase is an item related to the specific attribute.

[0077] The specific attributes are, for example, gender, age, residential area, family structure, occupation, educational background, annual income, etc. Alternatively, the specific attributes may be customer attributes categorized based on purchase history, interests, values, media and service usage, etc.

[0078] As an example, the explanatory phrase may be a phrase in which the first phrase "food and beverage manufacturer" in multiple items I in Figure 4 has the second phrase "place of employment" added to the beginning and the second phrase "person who is" added to the end.

[0079] In this case, the first phrase in multiple items I, "food and beverage manufacturer," is a specific It is included in the occupation as an attribute. The second phrases "place of employment" and "person who is" are phrases used to indicate that "food and beverage manufacturer" is an item related to occupation. In particular, the phrase "place of employment" is a phrase used to explain the item, which is common to multiple items I, such as "food and beverage manufacturer," "textile and clothing manufacturer," and "chemical and oil manufacturer."

[0080] In other words, by adding the second phrases "workplace is" and "person who is" to the first phrase "food and beverage manufacturer," the explanatory phrase "person who works at a food and beverage manufacturer" is created.

[0081] In another example, the explanatory phrase "people who live with children" in Figure 10 may be a phrase in which the phrase "people who live with" is added to the end of the phrase "children" if the omitted part of multiple items I in Figure 4 includes the item "children" as a cohabitant.

[0082] In this case, the first term "children" in the multiple items I is included in the family structure as a specific attribute. The second term "people living with" is a term used to indicate that "children" is an item related to family structure. In particular, the term "people living with" is a term used to explain the item, which is common to items included in family structure such as "husband / wife," "father," "mother," and "children" in the multiple items I.

[0083] That is, by adding the second phrase "person living with" to the first phrase "child," the explanatory phrase "person living with child" is created.

[0084] Next, at S230, the processor 11 sends a request signal to the artificial intelligence 2 requesting that one or more personas be generated based on the shaping data 62. In addition, the request signal may be a signal requesting that detailed profiles and person images for each persona be further generated based on the shaping data 62.

[0085] For example, the request signal may be a signal that includes the shaped data 62 and requests, "Please read the shaped data below and generate personas for three people in descending order of inclusion rate in the shaped data."

[0086] Next, at S240, the processor 11 receives persona data 70, which is data representing one or more personas, as an output of the artificial intelligence 2 based on the request signal. For example, the processor 11 receives persona data 70 representing three personas. The persona data 70 may include a detailed profile and a person image of each persona.

[0087] Next, in S250, the processor 11 displays one or more personas represented by the persona data 70, for example, through a display in the user interface 14. The processor 11 may further display a detailed profile and a person image for each persona. Figure 11 shows an example of persona data 70 including a persona profile and a person image.

[0088] Thereafter, the processor 11 ends the generation process shown in FIG. In this way, the support device 1 acquires the customer data 60 and calculates normalized data 61 by normalizing a plurality of statistical values ​​SV in the customer data 60. The support device 1 performs a shaping process on the normalized data 61 to create shaping data 62. The support device 1 receives a request to generate one or more personas based on the shaping data 62. The support device 1 transmits a signal to the artificial intelligence 2 and receives persona data 70, which is data representing one or more personas, as an output of the artificial intelligence 2 based on the request signal. The support device 1 displays the one or more personas represented by the persona data 70 through the user interface 14.

[0089] [1-3.Effects] According to the embodiment described above, the following actions and effects can be obtained. (1a) In the classification process, the processor 11 classifies a plurality of customers into a plurality of segments based on the survey results 40 and the purchase data 50.

[0090] According to this process, multiple customers can be classified based on the purchase data 50, which is objective data, in addition to the survey results 40, which are dependent on the subjective opinions of the survey respondents. This improves the accuracy of analyzing customer characteristics.

[0091] (1b) In the classification process, questions in a survey administered to multiple customers do not need to include information that can be obtained from the purchase data 50. For example, information such as purchase history indicating whether a certain product has been purchased or purchase frequency within the last three months can be obtained from the purchase data 50, so questions asking about such information do not need to be included in the survey.

[0092] With this configuration, the number of questions in the questionnaire can be reduced, thereby reducing the burden on the person answering the questionnaire. (1c) In the generation process, the processor 11 sends a request signal to the artificial intelligence 2 requesting that one or more personas be generated based on the shaping data 62, and receives persona data 70, which is data representing one or more personas, as an output of the artificial intelligence 2 based on the request signal.

[0093] According to this process, by referring to the persona data 70 output by the artificial intelligence 2, the characteristics of the customer can be easily analyzed. (1d) By inputting the persona data 70 obtained in the generation process to the artificial intelligence 2, the user can assign the artificial intelligence 2 the role of one of the one or more personas represented by the persona data 70. The user can conduct an interview as shown in FIG. 12 with the artificial intelligence 2 behaving as a persona.

[0094] In the example shown in Figure 12, the user requests AI 2 to list five important points when buying a vacuum cleaner. Upon receiving the instruction, AI 2 outputs an answer to the request as a persona.

[0095] This type of processing allows for the analysis of customer characteristics for each persona through interviews with the personas, making it possible to analyze the purchasing trends, needs, behavioral characteristics, etc. of customers when, for example, developing products, considering marketing strategies, considering promotion strategies, or designing sales channels.

[0096] (1e) In the generation process, the processor 11 obtains persona data 70 based on the customer data 60. The customer data 60 is data relating to a set of customers classified by a predetermined method.

[0097] By performing such processing, persona data 70 relating to the customer group can be obtained. For example, if the customer data 60 is a customer group classified as loyal customers by classification processing, the persona represented by the persona data 70 will have the purchasing trends, needs, behavioral characteristics, etc. of loyal customers. Therefore, it is possible to perform analysis on a customer group classified by a predetermined method.

[0098] (1f) The request signal requests the artificial intelligence 2 to generate one or more personas. The request signal may request the generation of multiple personas. According to this processing, multiple personas can be received from the artificial intelligence 2 in response to one input to the artificial intelligence 2. This reduces the number of times that requests are made to the artificial intelligence 2 to analyze multiple personas.

[0099] (1g) The descriptive phrase may include some or all of the phrases representing each of the multiple items I in the customer data 60. According to this process, it is possible to create explanatory phrases using words that represent multiple items I. That is, it is possible to create explanatory phrases using the customer data 60. Therefore, it is possible to easily create the formatted data 62.

[0100] (1h) In the generation process, the processor 11 normalizes the plurality of statistical values ​​SV in the customer data 60 to calculate normalized data 61 having a plurality of normalized values ​​NV, which are the normalized plurality of statistical values ​​SV.

[0101] This process can reduce the variation in the multiple statistical values ​​SV in the customer data 60. Therefore, the variation in the numerical values ​​input to the artificial intelligence 2 can be reduced.

[0102] (1i) A descriptive phrase is a phrase in which a first phrase, which is a phrase representing each of the multiple items I included in the customer data 60, is added to at least one of the beginning and end of the first phrase, and a second phrase, which is a phrase common to at least some of the multiple items I and describes the item, is added.

[0103] The plurality of items I may be a plurality of items included in a specific attribute related to the customer. The second phrase may include a phrase indicating that the item represented by the first phrase is an item related to the specific attribute.

[0104] According to this process, the words and phrases representing the plurality of items I can be converted into a format that is easy for the artificial intelligence 2 to analyze. This improves the accuracy of analyzing customer characteristics.

[0105] [1-4. Correspondence between terms] In the above embodiment, the processor 11 corresponds to an example of a control unit, and the communication interface 15 corresponds to an example of a communication unit. The plurality of statistical values ​​SV and the plurality of normalized values ​​NV correspond to an example of a plurality of numerical values.

[0106] The processes of S100 and S110 correspond to an example of a process executed by an acquisition unit, and the process of S130 corresponds to an example of a process executed by a classification unit. The processing of S200 corresponds to an example of obtaining customer data, the processing of S220 corresponds to an example of creating formatting data, the processing of S230 corresponds to an example of sending a request signal to artificial intelligence, and the processing of S240 corresponds to an example of receiving persona data.

[0107] 2. Other Embodiments Although the embodiments of the present disclosure have been described above, it goes without saying that the present disclosure is not limited to the above-described embodiments and can take on various forms.

[0108] (2a) In the above embodiment, the classification process and the generation process are both realized by the assistance device 1. However, the classification process and the generation process may be realized by separate devices. The classification process and the generation process may be performed independently. The generation process may be performed without performing the classification process. After the classification process is performed, the generation process does not have to be performed.

[0109] (2b) In the above embodiment, the shaped data 62 shown in FIG. 10 indicates the content rate for all customers as the "total content rate." However, the content rate included in the formatted data 62 is not limited to the content rate for all customers. For example, the content rate may be the content rate for a predetermined purchase category, such as the "content rate for online purchasers." The content rate is not limited to the purchase category, and may be the content rate for a predetermined category.

[0110] With this configuration, it is possible to receive persona data 70 corresponding to a predetermined purchase category. (2c) In the above embodiment, the explanatory phrase is a phrase that explains each of the plurality of normalized values ​​NV. The shaping data 62 is created by associating an explanatory phrase with each of the plurality of normalized values ​​NV.

[0111] However, the descriptive phrases are not limited to the plurality of normalized values ​​NV, and may be phrases that describe each of the plurality of numerical values ​​related to the customer set included in the customer data 60. The shaped data 62 may be created by associating each of the plurality of numerical values ​​with a descriptive phrase. The plurality of numerical values ​​may be, for example, a plurality of statistical values ​​SV, or may be values ​​obtained by processing the plurality of statistical values ​​SV.

[0112] (2d) In the above embodiment, in the classification process, processor 11 classifies multiple customers into multiple segments. At this time, processor 11 may further classify each segment into multiple segments based on survey results 40 and purchase data 50.

[0113] For example, after categorizing multiple customers based on the brand's purchasing history and whether or not they have a purchasing intention, they may be further categorized based on the media usage status included in the survey results 40. According to this type of processing, for example, multiple customers may be categorized based on a customer pyramid, and based on the media usage status of multiple customers in the two segments of "aware / non-purchasing customers" and "unaware customers" in the customer pyramid, the multiple customers may be further categorized into "customers who watch television frequently" or "customers who watch SNS frequently," etc. With this configuration, multiple customers can be classified in more detail.

[0114] (2e) The classification process in the above embodiment may be performed periodically. This allows tracking of customers who move from one segment to another, allowing for analysis of the characteristics of customers who move between segments.

[0115] (2f) In the above embodiment, the customer data 60 is data related to a set of customers classified by classification processing. However, the customer data 60 may be data related to a set of customers classified by other methods than classification processing. Other methods may include, for example, RFM (Recency, Frequency, Monetary) analysis, LTV (Life Time Value) analysis, cohort analysis, etc. The customer data 60 may also be data related to an unclassified set of customers.

[0116] (2g) Multiple functions possessed by one component in the above embodiments may be realized by multiple components, or one function possessed by one component may be realized by multiple components. Multiple functions possessed by multiple components may be realized by one component, or one function realized by multiple components may be realized by one component. Part of the configuration of the above embodiments may be omitted. At least part of the configuration of the above embodiments may be added to or substituted for the configuration of another of the above embodiments.

[0117] (2h) The present disclosure may be realized in various forms other than the above-described assistance system, such as a system including the assistance system as a component, a computer program for causing a computer to function as the assistance system, a non-transitory tangible recording medium such as a semiconductor memory on which the computer program is recorded, an assistance method, etc.

[0118] [Technical idea disclosed in this specification] [Item 1] A generating device, comprising: A control unit; a communication unit configured to be able to communicate with the artificial intelligence; Equipped with The control unit acquiring customer data relating to a set of customers categorized in a predetermined manner; generating shaped data by performing shaping processing on the customer data; sending a request signal to the artificial intelligence requesting generation of one or more personas based on the shaping data; receiving persona data representing the one or more personas as an output of the artificial intelligence based on the request signal; configured to run the customer data includes a plurality of values ​​related to the customer set; the formatting process includes creating explanatory phrases that are words and phrases that explain each of the plurality of numerical values, and associating the explanatory phrases with each of the plurality of numerical values ​​to create the formatted data; generator.

[0119] [Item 2] The generating device according to item 1, The customer data includes a plurality of items, each of the plurality of numerical values ​​is a value corresponding to each of the plurality of items, The descriptive phrase includes a part or all of a phrase representing each of the plurality of items. generator.

[0120] [Item 3] The generating device according to item 2, The explanatory phrase is a phrase in which a first phrase representing each of the plurality of items is added to at least one of the beginning and end of the first phrase and a second phrase that is common to the plurality of items and that explains the plurality of items is added. generator.

[0121] [Item 4] Item 3. The generating device according to item 3, the plurality of items are a plurality of items included in specific attributes related to the customer, the second term includes a term indicating that the item represented by the first term is an item related to the specific attribute; generator.

[0122] [Item 5] Item 4. The generating device according to item 4, The specific attributes relate to at least one of gender, age, residential area, family structure, occupation, educational background, and annual income. generator.

[0123] [Item 6] The generating device according to any one of items 1 to 5, the control unit is configured to perform normalization of the plurality of numerical values; the formatted data is created by associating the descriptive phrase with each of the normalized numerical values; generator.

[0124] [Item 7] The generating device according to any one of items 1 to 6, The plurality of numerical values ​​include values ​​for each customer category defined by classifying the customer group based on purchase history. generator.

[0125] [Item 8] The generating device according to any one of items 1 to 7, the request signal includes a request to generate a plurality of personas as the one or more personas; The persona data is data representing the plurality of personas. generator.

[0126] [Item 9] 1. A classification device, comprising: an acquisition unit configured to acquire survey results, which are responses to a survey conducted for a plurality of customers, for each customer, and purchase data including a product purchase history for each of the plurality of customers; a classification unit configured to classify the plurality of customers based on the survey results and the purchase data; Equipped with The survey includes questions requesting information about the individual respondent; Classification device.

[0127] [Item 10] Item 9. The classification device according to item 9, The purchase history includes identification information of the product and a purchase date and time of the product. Classification device.

[0128] [Item 11] Item 9 or Item 10, the classification device The questionnaire includes a question asking whether or not the customer is aware of the product, The purchase history includes information regarding the frequency of purchase of the product, the classification unit is configured to classify the plurality of customers based on whether or not they are aware of the product and their purchasing frequency of the product. Classification device.

[0129] [Item 12] Item 12. The classification device according to any one of items 9 to 11, The information about the individual respondents includes at least one of the gender, age, occupation, preferences, and media usage status of the individual respondents; Classification device.

[0130] [Item 13] An assistance system, comprising: A generating device according to any one of items 1 to 8; A classification device according to any one of items 9 to 12; Equipped with the generation device is configured to acquire, as the customer data, data regarding the plurality of customers classified by the classification device; Support system.

[0131] [Item 14] 1. A computer-implemented generation method comprising: Communicating with artificial intelligence, acquiring customer data relating to a set of customers categorized in a predetermined manner; generating shaped data by performing shaping processing on the customer data; sending a request signal to the artificial intelligence requesting generation of one or more personas based on the shaping data; receiving persona data representing the one or more personas as an output of the artificial intelligence based on the request signal; Including, the customer data includes a plurality of values ​​related to the customer set; the formatting process includes creating explanatory phrases that are words and phrases that explain each of the plurality of numerical values, and associating the explanatory phrases with each of the plurality of numerical values ​​to create the formatted data; Generation method.

[0132] [Item 15] Item 15. A computer program for causing a computer to execute the generating method according to Item 14. [Item 16] 1. A computer-implemented classification method comprising: Acquiring survey results, which are responses to a survey conducted for a plurality of customers, and purchasing data including product purchasing histories for each of the plurality of customers; classifying the plurality of customers based on the survey results and the purchase data; Including, The survey includes questions requesting information about the individual respondent; Classification method.

[0133] [Item 17] Item 17. A computer program for causing a computer to execute the classification method according to Item 16. [Explanation of symbols]

[0134] 1...Assistive device, 2...Artificial intelligence, 11...Processor, 15...Communication interface, 40...Survey results, 50...Purchase data, 60...Customer data, 62...Format data, 70...Performance data Sonadata, 100...support system, SV...multiple statistical values, NV...multiple normalized values.

Claims

1. A generating device, comprising: A control unit; a communication unit configured to be able to communicate with the artificial intelligence; Equipped with The control unit acquiring customer data relating to a set of customers categorized in a predetermined manner; generating shaped data by performing shaping processing on the customer data; sending a request signal to the artificial intelligence requesting generation of one or more personas based on the shaping data; receiving persona data representing the one or more personas as an output of the artificial intelligence based on the request signal; configured to run the customer data includes a plurality of values ​​related to the customer set; the formatting process includes creating explanatory phrases that are words and phrases that explain each of the plurality of numerical values, and associating the explanatory phrases with each of the plurality of numerical values ​​to create the formatted data; the control unit is configured to perform normalization of the plurality of numerical values; the formatted data is created by associating the descriptive phrase with each of the normalized numerical values; generator.

2. A generating device, comprising: A control unit; a communication unit configured to be able to communicate with the artificial intelligence; Equipped with The control unit acquiring customer data relating to a set of customers categorized in a predetermined manner; generating shaped data by performing shaping processing on the customer data; sending a request signal to the artificial intelligence requesting generation of one or more personas based on the shaping data; receiving persona data representing the one or more personas as an output of the artificial intelligence based on the request signal; configured to run the customer data includes a plurality of values ​​related to the customer set; each of the plurality of numerical values ​​includes a percentage of a customer group that satisfies a predetermined condition in the customer set; the formatting process includes creating explanatory phrases that are words and phrases that explain each of the plurality of numerical values, and associating the explanatory phrases with each of the plurality of numerical values ​​to create the formatted data; generator.

3. 3. The generating device according to claim 1 or claim 2, The customer data includes a plurality of items, each of the plurality of numerical values ​​is a value corresponding to each of the plurality of items, The descriptive phrase includes a part or all of a phrase representing each of the plurality of items. generator.

4. The generating device according to claim 3, The explanatory phrase is a phrase obtained by adding a second phrase, which is common to the plurality of items and is used to describe the plurality of items, to at least one of the beginning and end of a first phrase which is a phrase representing each of the plurality of items. generator.

5. 5. The generating device of claim 4, the plurality of items are a plurality of items included in specific attributes related to the customer, the second term includes a term indicating that the item represented by the first term is an item related to the specific attribute; generator.

6. A generating device, comprising: A control unit; a communication unit configured to be able to communicate with the artificial intelligence; Equipped with The control unit acquiring customer data relating to a set of customers categorized in a predetermined manner; generating shaped data by performing shaping processing on the customer data; sending a request signal to the artificial intelligence requesting generation of one or more personas based on the shaping data; receiving persona data representing the one or more personas as an output of the artificial intelligence based on the request signal; configured to run the customer data includes a plurality of numerical values ​​and a plurality of items related to the customer set; each of the plurality of numerical values ​​is a value corresponding to each of the plurality of items, the formatting process includes creating explanatory phrases that are words and phrases that explain each of the plurality of numerical values, and associating the explanatory phrases with each of the plurality of numerical values ​​to create the formatted data; the explanatory phrase includes a part or all of a phrase representing each of the plurality of items, and is a phrase in which a second phrase that is common to the plurality of items and that describes the plurality of items is added to at least one of the beginning and end of a first phrase that is a phrase representing each of the plurality of items, the plurality of items are a plurality of items included in specific attributes related to the customer, the second term includes a term indicating that the item represented by the first term is an item related to the specific attribute; generator.

7. 10. The generating device according to claim 1, claim 2 or claim 6, The plurality of numerical values ​​include values ​​for each customer category defined by classifying the customer group based on purchase history. generator.

8. 10. The generating device according to claim 1, claim 2 or claim 6, the request signal includes a request to generate a plurality of personas as the one or more personas; The persona data is data representing the plurality of personas. generator.

9. An assistance system, comprising: A generating device according to claim 1, claim 2 or claim 6; a classification device; Equipped with The classification device an acquisition unit configured to acquire survey results, which are responses to a survey conducted for a plurality of customers, for each customer, and purchase data including a product purchase history for each of the plurality of customers; a classification unit configured to classify the plurality of customers based on the survey results and the purchase data; Equipped with The questionnaire includes questions requesting information about the individual respondent; the generation device is configured to acquire, as the customer data, data regarding the plurality of customers classified by the classification device; Support system.

10. 10. The assistance system according to claim 9, The purchase history includes identification information of the product and a purchase date and time of the product. Support system.

11. 10. The assistance system according to claim 9, The questionnaire includes a question asking whether or not the customer is aware of the product, The purchase history includes information regarding the frequency of purchase of the product, the classification unit is configured to classify the plurality of customers based on whether or not they are aware of the product and their purchasing frequency of the product. Support system.

12. 10. The assistance system according to claim 9, The information about the individual respondents includes at least one of the gender, age, occupation, preferences, and media usage status of the individual respondents. Support system.

13. 1. A computer-implemented generation method comprising: Communicating with artificial intelligence, acquiring customer data relating to a set of customers categorized in a predetermined manner; generating shaped data by performing shaping processing on the customer data; sending a request signal to the artificial intelligence requesting generation of one or more personas based on the shaping data; receiving persona data representing the one or more personas as an output of the artificial intelligence based on the request signal; Including, the customer data includes a plurality of values ​​related to the customer set; the formatting process includes creating explanatory phrases that are words and phrases that explain each of the plurality of numerical values, and associating the explanatory phrases with each of the plurality of numerical values ​​to create the formatted data; The generating method further includes normalizing the plurality of numerical values; the formatted data is created by associating the descriptive phrase with each of the normalized numerical values; Generation method.

14. 1. A computer-implemented generation method comprising: Communicating with artificial intelligence, acquiring customer data relating to a set of customers categorized in a predetermined manner; generating shaped data by performing shaping processing on the customer data; sending a request signal to the artificial intelligence requesting generation of one or more personas based on the shaping data; receiving persona data representing the one or more personas as an output of the artificial intelligence based on the request signal; Including, the customer data includes a plurality of values ​​related to the customer set; each of the plurality of numerical values ​​includes a percentage of a customer group that satisfies a predetermined condition in the customer set; the formatting process includes creating explanatory phrases that are words and phrases that explain each of the plurality of numerical values, and associating the explanatory phrases with each of the plurality of numerical values ​​to create the formatted data; Generation method.

15. 1. A computer-implemented generation method comprising: Communicating with artificial intelligence, acquiring customer data relating to a set of customers categorized in a predetermined manner; generating shaped data by performing shaping processing on the customer data; sending a request signal to the artificial intelligence requesting generation of one or more personas based on the shaping data; receiving persona data representing the one or more personas as an output of the artificial intelligence based on the request signal; Including, the customer data includes a plurality of numerical values ​​and a plurality of items related to the customer set; each of the plurality of numerical values ​​is a value corresponding to each of the plurality of items, the formatting process includes creating explanatory phrases that are words and phrases that explain each of the plurality of numerical values, and associating the explanatory phrases with each of the plurality of numerical values ​​to create the formatted data; the explanatory phrase includes a part or all of a phrase representing each of the plurality of items, and is a phrase in which a second phrase that is common to the plurality of items and that describes the plurality of items is added to at least one of the beginning and end of a first phrase that is a phrase representing each of the plurality of items, the plurality of items are a plurality of items included in specific attributes related to the customer, the second term includes a term indicating that the item represented by the first term is an item related to the specific attribute; Generation method.

16. A computer program for causing a computer to execute the generating method according to claim 13, 14 or 15.

Images