Information generation device, information generation method and program

Abstract

To allow information indicating a characteristic of an individual who has not participated in a survey to be obtained.SOLUTION: An information generation device includes characteristic information generation means for generating information indicating a characteristic of an individual by inputting information based on a product purchase history of the individual into a model that outputs information indicating a characteristic in response to input of the information based on the product purchase history.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to an information generation device, an information generation method, and a program.

Background Art

[0002] There are cases where a questionnaire is conducted on customers to acquire information on their values and used for marketing. For example, Patent Document 1 describes conducting a questionnaire on customers and classifying the customers into value clusters.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] It is preferable to obtain information indicating the characteristics of a person even for those who have not taken a questionnaire.

[0005] An example of the object of the present disclosure is to provide an information generation device, an information generation method, and a program capable of solving the above-described problems.

Means for Solving the Problems

[0006] According to a first aspect of the present disclosure, an information generation device includes characteristic information generation means for inputting information based on a purchase history of a product by a person into a model that receives an input of information based on the purchase history of the product and outputs information indicating characteristics, and generating information indicating the characteristics of the person.

[0007] According to a second aspect of the present disclosure, the information generation method includes a computer inputting information based on a person's purchase history of products into a model that receives an input of information based on the purchase history of products and outputs information indicating characteristics, and generating information indicating the characteristics of that person.

[0008] According to a third aspect of the present disclosure, the program is a program that causes a computer to input information based on a person's purchase history of products into a model that receives an input of information based on the purchase history of products and outputs information indicating characteristics, and generates information indicating the characteristics of that person.

Advantages of the Invention

[0009] According to one aspect of the present disclosure, information indicating the characteristics of a person can be obtained even for a person who has not conducted a questionnaire.

Brief Description of the Drawings

[0010]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Embodiments for Carrying Out the Invention

[0011] Hereinafter, embodiments of the present invention will be described. However, the following embodiments do not limit the invention according to the claims. Also, not all combinations of features described in the embodiments are essential for the solution means of the invention.

[0012] <First Embodiment> FIG. 1 is a diagram showing an example of the configuration of a sales strategy support system according to at least one embodiment. In the configuration shown in FIG. 1, the sales strategy support system 1 includes an information generation device 100, a POS system 910, a smartphone 921, and a point card 922. The POS system 910 is installed in a store. The smartphone 921 and the point card are owned by members of the point service. However, the people targeted by the sales strategy support system 1 are not limited to members of the point service, as long as each person can be identified. For example, the card used in the sales strategy support system 1 may be issued separately from the point card. And it may be possible to identify each person by the card number of that card.

[0013] The sales strategy support system 1 is a system that supports the sales strategy in stores. The products targeted by the sales strategy are not limited to specific ones. In particular, the products targeted by the sales strategy may be goods, services, or a combination of goods and services.

[0014] The point card 922 is a card that members present at the store to receive the point service. When obtaining the point card 922, or afterwards, the member may register their own information, but it is not limited to this. As long as the point card 922 is identified by an identification number (card number), and thereby the member who presents the point card 922 at the store to purchase a product can be identified. Also, the function of the point card 922 may be executed in a form other than a card, such as being implemented as one of the functions of the smartphone 921.

[0015] Hereinafter, the case where the information generation device 100 generates data regarding a member of the point card will be described as an example. However, as described above, the people targeted by the information generation device 100 for data generation are not limited to members of the point card, and any person who can be identified by some method is sufficient. For example, the card used in the sales strategy support system 1 may be issued separately from the point card. And it may be possible to identify each person by the card number of that card.

[0016] The smartphone 921 is used for a member to receive information such as product introductions. For example, the smartphone 921 and the card number of the point card 922 may be associated with each other by the member sending an email with the card number of the point card 922 described therein from the smartphone 921 to the information generation device 100.

[0017] However, the form in which a member receives information provision is not limited to a specific form. For example, a member may receive information provision by email on a personal computer (PC), or may receive information provision by direct mail. Also, a member may receive information provision at a store, such as by being handed a flyer at the cash register of the store.

[0018] The POS system 910 digitizes sales at a store. In particular, the POS system 910 generates product purchase data of a member who purchased a product by presenting the point card 922, and transmits the generated product purchase data to the information generation device 100.

[0019] The information generation device 100 generates information for supporting a sales strategy. The information generation device 100 may be configured using a computer. Also, the information generation device 100 may be configured using a plurality of devices, such as a combination of a computer for information generation and a database machine.

[0020] FIG. 2 is a diagram showing an example of the configuration of the information generation device 100. With the configuration of FIG. 2, the information generation device 100 includes a communication unit 110, a display unit 120, an operation input unit 130, a storage unit 180, and a processing unit 190. The processing unit 190 includes a learning unit 191, a characteristic information generation unit 192, and an action processing unit 193. The action processing unit 193 includes a target person determination unit 194, a target product determination unit 195, an action data generation unit 196, and a transmission processing unit 197.

[0021] The communication unit 110 communicates with other devices. For example, the communication unit 110 may receive product purchase data from the POS system 910. Also, the communication unit 110 may transmit information such as product introductions to the smartphone 921 for members.

[0022] The display unit 120 has a display screen such as a liquid crystal panel or an LED (Light Emitting Diode) panel, and acquires various images. For example, the display unit 120 may display various information generated as information for supporting sales strategies.

[0023] The operation input unit 130 includes input devices such as a keyboard and a mouse, and accepts user operations. For example, the operation input unit 130 may accept a user operation that instructs the generation of information for supporting sales strategies.

[0024] The storage unit 180 stores various data. The storage unit 180 is configured using a storage device included in the information generation device 100. The storage unit 180 may store various data used for generating information for supporting sales strategies, and information for supporting sales strategies generated by the information generation device 100. For example, the storage unit 180 may store product purchase data, questionnaire data, product data, awareness score data, product score data, product score aggregation data, awareness segment data, and segment ratio data.

[0025] FIG. 3 is a diagram showing an example of the data structure of product purchase data. The product purchase data is data related to the purchase of products by customers such as members generated by the POS system 910. The POS system 910 transmits the generated product purchase data to the information generation device 100.

[0026] In the example of FIG. 3, the product purchase data includes columns of "Purchase ID", "Member ID", "Purchase Date and Time", "Store", "Total Amount", and "Purchased Products". The "Store" column includes sub-item columns of "Store ID", "Store Name", "Business Type Name", and "Chain Name". The "Purchased Products" column includes sub-item columns of "JAN", "Item Number", "Product ID", "Product Name", "Unit Price", "Quantity", and "Amount".

[0027] The purchase ID is stored in the "Purchase ID" column. The purchase ID is identification information for identifying the product purchase data. The member ID of the purchaser is stored in the "Member ID" column. The member ID is identification information for identifying the member. As the member ID, the card number of the point card 922 held by the member can be used. In the product purchase data, the card number of the point card 922 presented by the purchaser at the time of purchase is stored in the "Member ID" column. The date and time of purchase are stored in the "Purchase Date and Time" column.

[0028] Information regarding the store where the purchase was made is stored in the "Store" column. The store ID of the store where the purchase was made is stored in the "Store ID" column. The store ID is identification information for identifying the store. The name (store name) of the store where the purchase was made is stored in the "Store Name" column. The business type name of the store where the purchase was made is stored in the "Business Type Name" column. When the store where the purchase was made is a chain store, the chain name is stored in the "Chain Name" column. Here, the chain store may be a corporate chain, and the corporate name may be stored in the "Chain Name" column.

[0029] However, the structure (data structure) of the data stored in the "Store" column is not limited to a specific structure, and can be various structures that can identify the store where the purchase was made. For example, the product purchase data may include only the store ID as the information in the "Store" column. The total amount of the purchased products is stored in the "Total Amount" column.

[0030] In the "Purchased Items" column, information regarding the purchased items is stored. If multiple types of items are purchased at once, in the item purchase data, a "Purchased Items" column is provided for each type of purchased item. In the "JAN" column, the JAN code (GTIN) of the purchased item is stored. In the "Branch Number" column, when there are variations for the item indicated by the JAN code stored in the "JAN" column, the branch number indicating the variation is stored.

[0031] In the "Item ID" column, the item ID of the purchased item is stored. The item ID is an identification number for identifying the item. Items with the same item ID shall be treated as the same item. The item identified by the item ID is also referred to as the type of item. That is, items with the same item ID are also referred to as items of the same type.

[0032] In the "Item Name" column, the name of the purchased item is stored. In the "Unit Price" column, the unit price of the purchased item is stored. In the "Quantity" column, the quantity of the item (the type of item identified by the item ID stored in the "Item ID" column) purchased is stored. In the "Amount" column, the purchase amount of the item is stored. If there is no discount or the like, the purchase amount of the item is calculated as the unit price of the item × quantity.

[0033] However, the composition of the item purchase data is not limited to a specific composition. For example, in the case of the purpose of generating awareness score data, which is data indicating the psychological characteristics of members, the item purchase data can be data of various compositions that can identify the member who purchased the item and the purchased item. In the case of the purpose of generating time-series data of member-style score data, the item purchase data can be data of various compositions that can identify the member who purchased the item and the purchased item and can specify the purchase date and time. In the case of the purpose of grasping the items purchased at a certain store, the item purchase data can be data of various compositions that can identify the store and the purchased item.

[0034] As in the example of FIG. 3, by showing the member ID and the purchase date and time in the product purchase data, the purchase history of products by individual members can be read from the product purchase data. Thus, the product purchase data corresponds to an example of data indicating the purchase history of products.

[0035] FIG. 4 is a diagram showing an example of the data structure of questionnaire data. The questionnaire data is data indicating the results of a questionnaire for members. The questionnaire data is conducted for a part of the members. Members who answered the questionnaire are also referred to as respondents or monitors.

[0036] In the example of FIG. 4, the questionnaire data includes columns of "questionnaire ID", "member ID", "questionnaire date and time", "demographic items", and "psychographic items". The "demographic items" column includes sub-item columns of "gender", "age", "occupation", and "residence". The "psychographic items" column includes sub-item columns of "interest in new products", "emphasis on value for money", and "emphasis on family".

[0037] The questionnaire ID is stored in the "questionnaire ID" column. The questionnaire ID is identification information for identifying the questionnaire. The member ID of the respondent is stored in the "member ID" column. In the questionnaire data, the card number of the point card 922 held by the respondent is stored in the "member ID" column. The date and time when the member answered the questionnaire is stored in the "questionnaire date and time" column.

[0038] Demographic information is stored in the "demographic items" column. Demographic items can be regarded as items whose information can be objectively observed. The demographic items may include geographic items.

[0039] In the "Gender" column, the gender of the respondent is stored. In the "Age" column, the age of the respondent is stored. In the "Occupation" column, the occupation of the respondent is stored. In the "Residence" column, the place of residence of the respondent is stored. However, the demographic items included in the questionnaire data are not limited to specific items and can be various objectively determinable items.

[0040] In the "Psychographic Items" column, psychographic (psychological) information is stored. Psychographic items can be regarded as items related to the psychology of the respondent. The questionnaire corresponds to an example of an investigation of the psychological characteristics of the questionnaire respondents.

[0041] Note that the questionnaire data may include answers to items related to objective facts (items other than psychographic items) such as the presence or absence of a private car. Items related to objective facts may be provided as part of the demographic items, or may be provided as items separate from the demographic items and psychographic items.

[0042] In "Interest in New Products", the degree of interest of the respondent in new products is stored. In the example of Figure 4, the degree of interest of the respondent in new products is shown in five levels by integers from 1 to 5. The larger the integer value, the greater the interest in new products. In the "Value for Money" column, the degree to which the respondent values value for money is stored. In the example of Figure 4, the degree to which the respondent values value for money is shown in five levels by integers from 1 to 5. The larger the integer value, the more the respondent values value for money. In the "Family Orientation" column, the degree to which the respondent values their family is stored. In the example of Figure 4, the degree to which the respondent values their family is shown in five levels by integers from 1 to 5. The larger the integer value, the more the respondent values their family.

[0043] However, the psychographic items included in the questionnaire data are not limited to specific items, and can be various items related to the subjective opinions of the respondents. In particular, the psychographic items included in the questionnaire data can be items corresponding to the items included in the awareness score data, which are data indicating the psychological characteristics of the members.

[0044] The composition of the questionnaire data is not limited to a specific composition. For example, in the case of the purpose of extracting the characteristics of the respondents, the questionnaire data can be data of various compositions that can identify the respondents and specify the characteristics of the respondents. In the case of the purpose of extracting the history of the characteristics of the respondents, the questionnaire data can be data of various compositions that can identify the respondents, the date and time when the respondents answered the questionnaire, and specify the characteristics of the respondents.

[0045] Based on the questionnaire results, the information generation device 100 estimates the characteristics of members who are not the targets of the questionnaire or who did not answer the questionnaire. Hereinafter, the case where the information generation device 100 estimates the psychological characteristics of the members will be described as an example. The psychological characteristics of a person here can be regarded as characteristics other than the objectively observable characteristics among the characteristics of that person. However, the information generation device 100 may estimate characteristics other than psychological characteristics in addition to or instead of psychological characteristics. For example, the information generation device 100 may estimate the age of a member based on the purchase history of products by the member. In this case, for example, when determining whether to inquire a member about a product for which the age range of the target customer layer is defined, the age of the member estimated by the information generation device 100 can be used.

[0046] Figure 5 is a diagram showing an example of the data structure of product data. Product data is data related to individual products. In the description of product data, the product targeted by the product data is also referred to as the target product. In the example of FIG. 5, the product data includes columns for "Product ID", "JAN", "Branch Number", "Category", "Manufacturer Name", "Product Name", and "Sales Start Date".

[0047] The "Product ID" column stores the product ID of the target product. The "JAN" column stores the JAN code of the target product. In the "Branch Number" column, if there are variations in the product indicated by the JAN code stored in the "JAN" column, the branch number indicating the variation is stored.

[0048] The "Category" column stores the category (the category of the target product) to which the target product belongs. The "Manufacturer Name" column stores the manufacturer name of the manufacturer of the target product. The "Product Name" column stores the sales start date and time of the target product.

[0049] However, the composition of the product data is not limited to a specific composition. The product data can be data of various compositions that are referred to for formulating a sales strategy according to the content of the sales strategy. Also, when the information generation device 100 generates awareness score data, the product data is not essential.

[0050] FIG. 6 is a diagram showing an example of the data structure of awareness score data. Awareness score data is data obtained by scoring (numericalizing, quantifying) the psychological characteristics of an individual for each item of the characteristics. The score for each item of the psychological characteristics in the awareness score data is also referred to as an awareness score. The item of the psychological characteristics is also referred to as a psychological characteristic item. The awareness score corresponds to an example of information indicating the psychological characteristics of a person. The awareness score shown in the awareness score data corresponds to an example of information indicating the degree of correspondence of each item of the psychological characteristic items set as the items of the psychological characteristics for one person. In the description of the awareness score data, the person who is the target of the awareness score data is also referred to as the target person.

[0051] The awareness score data of the monitors (questionnaire respondents) is also referred to as monitor awareness score data. The information generation device 100 generates monitor awareness score data for each monitor based on the questionnaire data. Alternatively, the information generation device 100 may acquire the monitor awareness score data generated by another device or person. The awareness score data of members, not limited to monitors, is also referred to as member awareness score data.

[0052] In the example of FIG. 6, the awareness score data includes columns of "member ID", "data creation date and time", "value-for-money emphasis", "thoughts about family", and "fashion sensitivity". The "member ID" column stores the member ID of the target person. The "data creation date and time" column stores the date and time when the awareness score data was generated.

[0053] Each of the items of "value-for-money emphasis", "thoughts about family", and "fashion sensitivity" corresponds to an example of an item of psychological characteristics. The "value-for-money emphasis" column stores a score value (score value of value-for-money emphasis) indicating the degree to which the target person emphasizes value for money. The larger the score value, the more the target person emphasizes value for money.

[0054] The "thoughts about family" column stores a score value (score value of thoughts about family) indicating the degree to which the target person emphasizes the target person's family. The larger the score value, the more the target person emphasizes the family. The "fashion sensitivity" column stores a score value (score value of fashion sensitivity) indicating the degree to which the target person is sensitive to fashion. The larger the score value, the more sensitive the target person is to fashion.

[0055] However, the configuration of the awareness score data is not limited to a specific configuration. In particular, the items of psychological characteristics included in the awareness score data can be various items referred to for formulating the sales strategy according to the content of the sales strategy.

[0056] FIG. 7 is a diagram showing an example of the data structure of product score data. Product score data is data obtained by scoring, for each item of characteristics, the psychological characteristics of the purchasers of the individual products. In the description of product score data, the product targeted by the product score data is also referred to as the target product. The score for each item of psychological characteristics in the product score data is also referred to as a product score. The product score corresponds to an example of information indicating the relevance between the target product and each of the psychological characteristic items.

[0057] For example, as the items of psychological characteristics in the product score data, the same items as those in the psychological characteristics in the awareness score data may be set. And the information generation device 100 may use, as the product score value, the average value of the awareness score values of the people who purchased the target product, for each item of psychological characteristics.

[0058] As described above, in addition to, or instead of, psychological characteristics, the information generation device 100 may also estimate characteristics other than psychological characteristics. In this case, the awareness score data may be generalized to data including items other than psychological characteristics. The generalized data in this case is also referred to as human characteristic score data. Also, psychological characteristics and characteristics other than psychological characteristics are collectively referred to as human characteristics. The awareness score (score of psychological characteristics) and the score of characteristics other than psychological characteristics are collectively referred to as human characteristic scores.

[0059] In the example of FIG. 7, the product score data includes columns of "product ID", "data creation date and time", "importance attached to value for money", "thoughts about family", and "sensitivity to fashion". In the "product ID" column, the product ID of the target product is stored. In the "data creation date and time" column, the date and time when the product score data was generated is stored.

[0060] In the "importance attached to value for money" column, the score value of the degree of importance attached to value for money of the purchaser of the target product is stored. In the "thoughts about family" column, the score value of the degree of thoughts about family of the purchaser of the target product is stored. In the "Fashion Sensitive" column, the score values of the fashion sensitivity of the purchasers of the target products are stored.

[0061] As described above, each of the items of "Value-for-Money Emphasis", "Family Consideration", and "Fashion Sensitive" corresponds to an example of an item of psychological characteristics. Each of the score values of the value-for-money emphasis, the family consideration degree, and the fashion sensitivity in the example of FIG. 7 corresponds to an example of the product score value.

[0062] However, the configuration of the product score data is not limited to a specific configuration. Also, as described above, the items of psychological characteristics included in the product score data may be the same items as the items of psychological characteristics included in the awareness score data.

[0063] FIG. 8 is a diagram showing an example of the data structure of the product score aggregation data. The product score aggregation data is data obtained by aggregating, for each individual, the product score values of the products purchased by that person for each item of the product score. The value obtained by aggregating the product score values is also referred to as the product score aggregation value. The product score aggregation value corresponds to an example of information based on the purchase history of a certain product by a certain person. Furthermore, the product score aggregation value corresponds to an example of information obtained by aggregating the product score values of the products purchased by one person for that person. As described above, the product score corresponds to an example of information indicating the relevance of each of the products to the items of psychological characteristics. In the description of the product score aggregation data, the person targeted by the product score aggregation data is also referred to as the target person.

[0064] For example, the product score aggregation value may be the average value of the product score values of the products purchased by the target person within a predetermined period for each item of the product score. In this case, it can be said that the aggregation method for calculating the product score aggregation value from the product score values is to take the average.

[0065] However, the aggregation method for calculating the product score aggregated value from the product score values is not limited to taking the average. For example, the aggregation method for calculating the product score aggregated value from the product score values may be any one of taking the sum, obtaining the median, obtaining the mode, obtaining the maximum value, or obtaining the minimum value. That is, the product score aggregated value may be any one of the sum value, median value, mode value, maximum value, or minimum value of the product score values of the products purchased by the target person for each item of the product score within a predetermined period.

[0066] In the example of FIG. 8, the product score aggregated data includes columns of "member ID", "data creation date", "emphasis on value for money", "thoughts for family", and "sensitivity to fashion". The "member ID" column stores the member ID of the target person. The "data creation date and time" column stores the date and time when the product score aggregated data was generated.

[0067] The "emphasis on value for money" column stores the score value of the degree of emphasis on value for money of the purchaser of the target product. The "thoughts for family" column stores the score value of the degree of thoughts for family of the purchaser of the target product. The "sensitivity to fashion" column stores the score value of the degree of sensitivity to fashion of the purchaser of the target product.

[0068] As described above, each of the items of "emphasis on value for money", "thoughts for family", and "sensitivity to fashion" corresponds to an example of an item of psychological characteristics. Each of the score value of the degree of emphasis on value for money, the score value of the degree of thoughts for family, and the score value of the degree of sensitivity to fashion in the example of FIG. 8 corresponds to an example of the product score aggregated value.

[0069] The data structure of the product score aggregated data may include data for identifying the target person instead of the data for identifying the target product from the data structure of the product score data, and include the product score aggregated value instead of the product score value as the score value. The items of psychological characteristics included in the product score aggregated data may be the same items as the items of psychological characteristics included in the product score data.

[0070] FIG. 9 is a diagram showing an example of the data structure of consciousness segment data. The consciousness segment referred to here is a class in which the psychological characteristics of an individual are classified. Consciousness segment data is data indicating the classification result of the psychological characteristics of an individual shown by consciousness score data. The classification of the psychological characteristics of an individual can also be regarded as the classification of an individual based on the psychological characteristics. The classes in the classification of psychological characteristics are also referred to as psychological characteristic classes. In the description of consciousness segment data, the person targeted by the consciousness segment data is also referred to as the target person.

[0071] In the example of FIG. 9, the consciousness segment data includes columns of "member ID", "data creation date", and "consciousness segment". The "member ID" column stores the member ID of the target person. The "data creation date" column stores the date and time when the consciousness segment data was generated. The "consciousness segment" column stores the class name in which the psychological characteristics are classified. "Health orientation" shown in FIG. 9 corresponds to an example of the class name.

[0072] However, the configuration of the consciousness segment data is not limited to a specific configuration, and can be various configurations that can show the classification result of the psychological characteristics of an individual. Also, the number of classes and classes in the classification of the psychological characteristics of an individual are not limited to specific numbers of classes and classes, and can be various numbers of classes and classes according to the content of the sales strategy and the items of psychological characteristics in the consciousness score data.

[0073] FIG. 10 is a diagram showing an example of the data structure of segment ratio data. Segment ratio data is data indicating the ratio of the people classified into each class in the classification result of an individual based on psychological characteristics. In this case, the ratio of the people classified into each class is also referred to as the segment ratio.

[0074] Figure 10 shows an example of the configuration of segment ratio data for individual stores. In this case, the store targeted by the segment ratio data is also referred to as the corresponding store. In the example of Figure 10, the segment ratio data includes items such as "store ID", "data creation date", and "segment ratio".

[0075] The store ID of the corresponding store is stored in the "store ID" column. The date and time when the segment ratio data was generated are stored in the "data creation date and time" column. In the "segment ratio" column, for each psychological characteristic class, the ratio of people classified into that class is shown. Each of "personality emphasis", "simple thinking", "authentic orientation", "health orientation", "family efficiency", and "family enthusiasm" shown in Figure 10 corresponds to an example of the class name of a psychological characteristic class.

[0076] However, the configuration of the segment ratio data is not limited to a specific configuration. Also, the information generation device 100 may generate segment ratio data for each chain (of a chain store). In this case, instead of the "store ID" column, a column for identifying the chain may be provided in the segment ratio data.

[0077] The processing unit 190 controls each part of the information generation device 100 to perform various processes. The functions of the processing unit 190 are executed, for example, by a CPU (Central Processing Unit) provided in the information generation device 100 reading a program from the storage unit 180 and executing it.

[0078] The learning unit 191 performs machine learning of a machine learning model used for generating awareness score data of members other than monitors. The learning unit 191 corresponds to an example of a learning means.

[0079] The machine learning model here is not limited to a specific type. For example, a neural network (NN) may be used as the machine learning model, but it is not limited to this. Machine learning is also simply referred to as learning. The machine learning model is also simply referred to as the model. The learning of the model here is to adjust the parameter values of the model using training data. The learning of the model can also be referred to as the training of the model.

[0080] The learning unit 191 uses, as the input to the model, the vector of the aggregated product score values shown in the aggregated product score data of each monitor, and uses, as the correct answer, the vector of the awareness score values shown in the awareness score data of that monitor, to perform the learning of the model. The learning unit 191 performs the learning of the model so that the difference between the vector of the awareness score values obtained as the output of the model when the vector of the aggregated product score values is input and the vector of the awareness score values shown as the correct answer becomes smaller. Thereby, the learning unit 191 generates a model that receives the input of the aggregated product score value and outputs the awareness score value.

[0081] The model used for generating the awareness score data is also referred to as the awareness score data generation model. The awareness score data generation model corresponds to an example of a model that receives the input of information based on the purchase history of a product (for example, the aggregated product score value) and outputs information indicating characteristics (for example, the human characteristic score value or the awareness score value).

[0082] The input to the awareness score data generation model is not limited to the aggregated product score data, and can be various data based on the purchase history of the product. For example, the awareness score data generation model may be configured to receive the input of the product data of the most recent predetermined number of purchased products in the purchase history of the person for whom the awareness score data is targeted.

[0083] The learning method used by the learning unit 191 is not limited to a specific method. For example, the learning unit 191 may perform model learning using the backpropagation method, but it is not limited to this. Also, the index of the magnitude of the difference between vectors here is not limited to a specific one. For example, the learning unit 191 may use the mean squared error as the index of the magnitude of the difference between vectors, but it is not limited to this.

[0084] The learning unit 191 may perform re-learning of the model. For example, each time the information generation device 100 obtains new questionnaire data, it may generate product score aggregation data and awareness score data for the subjects of the questionnaire. Then, the learning unit 191 may perform model learning using a training data set including the training data based on the product score aggregation data and the awareness score data. The learning unit 191 may perform re-learning of the model by fine-tuning or by transfer learning.

[0085] Also, one model may be provided for one store. Then, the learning unit 191 may perform model learning for each member using the product score aggregation value obtained by aggregating the product score values for the products handled by the store targeted by the model among the products purchased by the member.

[0086] Among the products purchased by a certain member, the product score aggregation value obtained by aggregating the product score values for the products handled by the store targeted by the model to be learned corresponds to an example of the information aggregated for each purchaser for each psychological characteristic item of the information indicating the relevance between each of the products and each of the psychological characteristic items in the store targeted by the model.

[0087] The learning unit 191 may perform model learning using the aggregated product score values of all members who have purchase histories of products handled at the stores targeted by the model to be learned. Alternatively, the learning unit 191 may perform model learning using the aggregated product score values of members who have purchase histories at the stores targeted by the model to be learned.

[0088] The learning unit 191 performs model learning by using information based on the purchase histories of products handled at the stores targeted by the model to be learned, and thus performs model learning using training data excluding products not handled at those stores. Performing model learning using training data excluding products not handled at the target stores can be regarded as performing model learning using training data with noise removed, and in this regard, it is expected that the learning unit 191 can perform model learning with relatively high accuracy.

[0089] Alternatively, the model may be provided for each chain of (chain stores). And the learning unit 191 may perform model learning using the aggregated product score values obtained by aggregating the product score values for the products handled at the chain targeted by the model among the products purchased by each member.

[0090] The learning unit 191 may perform model learning using the aggregated product score values of all members who have purchase histories of products handled at the chain targeted by the model. Alternatively, the learning unit 191 may perform model learning using the aggregated product score values of members who have purchase histories at the chain targeted by the model to be learned.

[0091] The learning unit 191 performs learning of the model by using the aggregated product score value for the products handled in the chain targeted by the model to be learned, so as to perform learning of the model using the training data excluding the products not handled in that chain. Performing learning of the model using the training data excluding the products not handled in the targeted chain can be regarded as performing learning of the model using the training data with noise removed. In this regard, it is expected that the learning unit 191 can perform learning of the model with relatively high accuracy.

[0092] The characteristic information generation unit 192 generates awareness score data. In particular, the characteristic information generation unit 192 uses the machine learning model obtained by the machine learning by the learning unit 191 to generate awareness score data of members other than the monitors. The characteristic information generation unit 192 corresponds to an example of the characteristic information generation means.

[0093] Specifically, the characteristic information generation unit 192 generates aggregated product score data of the members for whom the awareness score data is to be calculated. Then, the characteristic information generation unit 192 inputs the generated aggregated product score data into the machine learning model obtained by the machine learning by the learning unit 191 to calculate the awareness score data of that member.

[0094] In the generation of the aggregated product score data, the characteristic information generation unit 192 may calculate the aggregated product score value of the member by calculating the average value for each item of the product score values of the products purchased by the member within a predetermined period based on the purchase history of the products by the member for whom the awareness score data is to be calculated.

[0095] Also, the characteristic information generation unit 192 may generate the data used by the learning unit 191 for performing learning of the model. For example, the characteristic information generation unit 192 may generate monitor awareness score data based on questionnaire data indicating the results of the questionnaire for the monitors (the responses of the monitors to the questionnaire).

[0096] The method by which the characteristic information generation unit 192 generates monitor awareness score data based on the questionnaire data is not limited to a specific method. For example, the psychographic items of the questionnaire may include the same items as the items of the monitor awareness score. Then, the characteristic information generation unit 192 may convert the answers to the psychographic items of the questionnaire into scores to generate monitor awareness score data. For example, the characteristic information generation unit 192 may calculate the score by multiplying the answers to the five-level questionnaire with integers from 1 to 5 by 20.

[0097] And the characteristic information generation unit 192 may generate product score data based on the monitor awareness score data. The method by which the characteristic information generation unit 192 generates product score data based on the monitor awareness score data is not limited to a specific method. For example, the items of the product score data may be the same as the items of the awareness score data. Then, the characteristic information generation unit 192 may calculate the average value of the awareness score values for each item of the monitors who purchased a certain product among the monitors as the product score value of that product.

[0098] In this case, the product score corresponds to an example of information obtained by aggregating the awareness scores for each person who purchased the target product (the product for which the product score is calculated) and for each psychographic characteristic item, for each psychographic characteristic item. The awareness score corresponds to an example of information indicating the degree of correspondence of a certain psychographic characteristic item for a certain person.

[0099] And the characteristic information generation unit 192 may generate product score aggregation data based on the product score data. For example, as described above, the product score aggregation value may be the average value for each item of the product score values of the products purchased by the target person within a predetermined period. Then, the characteristic information generation unit 192 may calculate the product score aggregation value by calculating the average value for each item of the product score values.

[0100] Alternatively, as described above, the product score aggregation value may be any one of the total value, median value, mode value, maximum value, or minimum value of the product score values of the products purchased by the target person within a predetermined period for each item of the product score. Then, the characteristic information generation unit 192 may calculate the product score aggregation value by calculating any one of the total value, median value, mode value, maximum value, or minimum value of the product score values for each item.

[0101] However, it is not essential for the characteristic information generation unit 192 to generate the monitor awareness score data, product score data, and product score aggregation data. For example, a device other than the information generation device 100 may generate one or more of these data.

[0102] Also, the characteristic information generation unit 192 may generate awareness segment data and segment ratio data. Regarding the generation of the awareness segment data, for each psychological characteristic class (class of awareness segments), an equation may be defined to calculate the score value (e.g., likelihood of class classification) of that class by weighted summing the values for each item of the awareness score. Then, the characteristic information generation unit 192 may calculate the score value for each class for each member and determine the class with the largest score value as the psychological characteristic class of that member.

[0103] Alternatively, rules for determining the class of the awareness segment from the values of each item of the awareness score may be modeled. Then, the characteristic information generation unit 192 may input the values of each item of the awareness score into the model for each member and determine the class of the awareness segment of that member.

[0104] Regarding the generation of the segment ratio data, the characteristic information generation unit 192 may calculate the ratio of the members belonging to that class among the members targeted for the calculation of the segment ratio for each psychological characteristic class. For example, when the characteristic information generation unit 192 generates the segment ratio data for a certain store, the segment ratio may be calculated for the members with a purchase history at that store.

[0105] However, it is not essential for the characteristic information generation unit 192 to generate the awareness segment data and the segment ratio data. For example, a device other than the information generation device 100 may generate one or more of these data. Alternatively, when a sales strategy is formulated based on the member awareness score data, and the awareness segment data and the segment ratio data are not necessary for formulating the sales strategy, the characteristic information generation unit 192 may not generate these data.

[0106] When an action in the sales strategy is performed based on the member awareness score data, the action processing unit 193 executes the action or a part of the action. In particular, the action processing unit 193 generates data for a notification to the member, such as an introduction of a product to the member, and transmits the notification to the member's smartphone 921 via the communication unit 110. Transmitting a notification to the member's smartphone 921 is also referred to as transmitting a notification to the member.

[0107] The target determination unit 194 determines a member who is the target of the action. In particular, when the action processing unit 193 transmits a notification to a member, the target determination unit 194 determines a target person to whom the notification is to be transmitted. The method by which the target determination unit 194 determines a member who is the target of the action is not limited to a specific method.

[0108] For example, when a certain store is the target of the sales strategy, the target determination unit 194 may determine, based on the member awareness score data, a member who is determined to have a relatively high possibility of using the store as the member who is the target of the action.

[0109] For example, a calculation method may be predetermined for calculating the likelihood of a certain member using a certain store based on the degree of match between the characteristics of the customers targeted by the store and the awareness score of the member. Then, the target person determination unit 194 may apply this calculation method to the member awareness score data for each member to calculate the likelihood of that member using the store targeted by the sales strategy. And the target person determination unit 194 may compare the calculated likelihood with a predetermined threshold value, and when the likelihood is higher than the threshold value, determine that the likelihood of that member using the store is relatively high.

[0110] Alternatively, the target person determination unit 194 may determine members with a purchase history at the store as the members to be the target of the action. Alternatively, the target person determination unit 194 may determine all members as the members to be the target of the action.

[0111] Also, when a certain product is targeted by the sales strategy, the target person determination unit 194 may, based on the member awareness score data, determine members who are determined to have a relatively high likelihood of purchasing the product as the members to be the target of the action.

[0112] For example, a calculation method may be predetermined for calculating the likelihood of a certain member purchasing a certain product based on the degree of match between the product score of the product and the awareness score of the member. Instead of the product score of the product, the characteristics of the customers targeted by the product may be used.

[0113] Then, the target person determination unit 194 may apply this calculation method to the member awareness score data for each member to calculate the likelihood of that member purchasing the product targeted by the sales strategy. And the target person determination unit 194 may compare the calculated likelihood with a predetermined threshold value, and when the likelihood is higher than the threshold value, determine that the likelihood of that member purchasing the product is relatively high.

[0114] Alternatively, the target determination unit 194 may determine a member with a purchase history of the product as the member to be the target of the action. Alternatively, the target determination unit 194 may determine all members as the members to be the targets of the action.

[0115] The target product determination unit 195 determines the product to be the target of the action. In particular, when the action processing unit 193 sends a notification to a member, the target product determination unit 195 determines the product to be the target of the notification. The method by which the target product determination unit 195 determines the product to be the target of the action is not limited to a specific method.

[0116] For example, when a certain store is the target of a sales strategy, the target product determination unit 195 may determine, based on the awareness score data of members with a purchase history at that store, products that are determined to have a relatively high likelihood of being purchased by those members as the products to be the targets of the action.

[0117] For example, as described above, a calculation method may be predetermined for calculating the likelihood of a certain member purchasing a certain product based on the degree of match between the product score of the product and the awareness score of the member. Instead of the product score of the product, the characteristics of the customers targeted by the product may be used.

[0118] Then, for each of the products handled at the store that is the target of the sales strategy and the products that are planned to be handled at that store, the target determination unit 194 may calculate the average value of the likelihood that a member with a purchase history at that store will purchase that product. Then, the target determination unit 194 may compare the calculated average value with a predetermined threshold value, and if the average value is higher than the threshold value, determine that those members have a relatively high likelihood of purchasing that product. Alternatively, the target product determination unit 195 may determine the product with the largest calculated average value as the product to be the target of the action.

[0119] Alternatively, the target product determination unit 195 may determine, as the product to be the target of the action, a product among the products sold in the store that is determined to have a low sales volume compared to other stores. When a certain product is the target of the sales strategy, the target product determination unit 195 may determine that product as the product to be the target of the action.

[0120] The action data generation unit 196 generates data to be used in the action. For example, in the case of an action to guide a product to a member, the action data generation unit 196 generates transmission data for guiding the product to the member. The transmission data for guiding the product to the member is also referred to as guidance data. The action data generation unit 196 corresponds to an example of the proposal information generation means.

[0121] The action data generation unit 196 may generate the guidance data by attaching the product name of the product to be guided and the image of the product to be guided to the template of the guidance data.

[0122] In the case of an action for a member, the transmission processing unit 197 transmits the data to be used in the action generated by the action data generation unit 196 to the smartphone 921 of the member determined by the target person determination unit 194 via the communication unit 110. For example, in the case of an action to guide a product to a member, the transmission processing unit 197 transmits the guidance data to the smartphone 921 of the member determined by the target person determination unit 194 via the communication unit 110. The transmission processing unit 197 corresponds to an example of the guidance information transmission means.

[0123] The psychological score data, the awareness segment data, or the segment ratio data can be used, for example, as reference materials when planning a campaign in a store, a chain, or a company. For example, by referring to the segment ratio data of a certain store, it is possible to grasp the psychological tendencies of the customers of that store, such as that there are many health-conscious people, many people interested in new products, and many family-centered people. When considering what kind of campaigns are effective in that store, it is conceivable to take into account the psychological tendencies of the customers of that store.

[0124] Also, the psychological score data, awareness segment data, or segment ratio data can be used, for example, as reference materials when planning products or creating private brands. For example, by referring to the segment ratio data of a certain company, it is possible to grasp the psychological tendencies of the customers of that company. When the company is planning products or creating private brands, when considering the types of products, product concepts or brand concepts, prices, or marketing methods, it is conceivable to take into account the psychological tendencies of the customers of that company.

[0125] Also, as described above, when the action processing unit 193 sends product guidance to members, it can refer to the awareness score data to determine the target customers or target products, or both. The action processing unit 193 may send product advertisements to members. Alternatively, the action processing unit 193 may send data containing benefits for members, such as coupons, to members.

[0126] Also, in addition to or instead of sending product guidance, the action processing unit 193 may send guidance on stores, chains, companies, or brands to members. In that case as well, the action processing unit 193 may refer to the awareness score data to determine the target customers or the content of the guidance, or both. Also, the action processing unit 193 may send advertisements to members, or may send data containing benefits for members, such as coupons, to members.

[0127] Psychological score data, consciousness segment data, or segment ratio data may be used as a reference for teleshopping. The teleshopping referred to here may be, but is not limited to, electronic commerce. For example, catalog sales or television shopping may be conducted as a communication medium.

[0128] For example, the consciousness score of customers in a physical store of a certain company may be used as reference information for determining the target audience for teleshopping guidance, determining the medium or means for guiding teleshopping, determining the medium or means of teleshopping, or determining the language or image for guiding teleshopping.

[0129] For example, the target determination unit 194 may determine the target audience for teleshopping guidance based on the consciousness score data. Then, the transmission processing unit 197 may transmit the teleshopping guidance information to each member's smartphone 921 determined as the target audience for teleshopping via the communication unit 110. The target determination unit 194 corresponds to an example of a target determination means. As described above, the transmission processing unit 197 corresponds to an example of a guidance information transmission means.

[0130] During the RFM analysis (Recency Frequency Monetary Analysis) of customers, the consciousness score data and product score data may be referred to. For example, consider the case where customers of a certain company or store are classified by RFM analysis, and actions for promotion (increasing purchases) or preventing churn (ceasing to use that store) are considered based on the RFM analysis results or historical information of the RFM analysis results.

[0131] In this case, it is conceivable to propose to the company or store products that show a strong correlation with the consciousness score of customers who have been promoted by RFM analysis as the target for the product lineup or enhanced campaigns of the company or store. In addition, it may be considered to propose to an enterprise or a store products that exhibit a product score strongly correlated with the awareness score of churned customers in RFM analysis as the target for store assortments or enhanced campaigns.

[0132] The action data generation unit 196 may generate data for making a proposal to an enterprise or a store. For example, the action data generation unit 196 may generate information indicating products to be proposed as products sold in one or more stores based on the awareness scores of customers in one or more stores and the product scores of each product. As described above, the action data generation unit 196 corresponds to an example of proposal information generation means.

[0133] In addition, the action data generation unit 196 may generate information indicating products to be proposed as products sold in one or more stores for products that exhibit a product score with a correlation with the awareness score of customers selected as customers affecting the change in store sales in RFM analysis that is stronger than a predetermined condition.

[0134] For example, in a certain store, if the awareness segment data of customers who are rapidly ranking up in RFM analysis indicates an interest in new products, referring to the product score data, it may be considered to propose products purchased by people interested in new products as products sold in that store. Alternatively, proposals such as increasing the new products sold in that store, issuing coupons for new products, arranging products on the shelves so that new products are more prominent, and periodically setting up a corner for introducing new products in the store may be made to that store. Even when the awareness segment data of customers identified as churned customers in RFM analysis indicates an interest in new products, similar proposals may be made to that store.

[0135] Also, if the awareness segment data of customers whose rankings have improved in the RFM analysis indicates a health orientation, proposals such as increasing the handling of supplements, issuing coupons for health-related products, arranging products on the shelves so that health-related products are more prominent, and regularly setting up corners for introducing health-related products in the store can be considered for that store.

[0136] The time-series information of the awareness score may also be used in the sales strategy. For example, the characteristic information generation unit 192 may generate awareness score data for each member at regular intervals such as every six months. And the storage unit 180 may store the time-series data of the awareness score for each member. Alternatively, the storage unit 180 may store, for each member, the time-series data of the awareness segment in addition to or instead of the time-series data of the awareness score.

[0137] For example, it is conceivable to use the time change of a certain member's awareness score as reference information for determining an action for that member. Furthermore, the time change of that member's awareness score may be used as reference information for determining whether to take an action for that member.

[0138] The target product determination unit 195 may determine the product to be guided to a certain member based on the time change of the awareness score of that member and the product score. And the transmission processing unit 197 may transmit the guidance of the determined product to the member's smartphone 921 via the communication unit 110. The target product determination unit 195 corresponds to an example of the guidance product determination means. The awareness score corresponds to an example of information indicating a person's psychological characteristics. The product score corresponds to an example of information indicating the relationship between a product and the psychological characteristics of the person who purchased that product.

[0139] For example, when the score value of "thinking about family" in a certain member's awareness score increases, coupons for products with a score value of "thinking about family" greater than or equal to a predetermined condition in the product score may be issued to that member. The issuance of coupons may be performed by the information generation device 100, or may be performed by other devices or people.

[0140] FIG. 11 is a diagram showing an example of a procedure for the information generation device 100 to perform preprocessing. FIG. 11 shows an example of a processing procedure when the characteristic information generation unit 192 generates training data for model learning and the learning unit 191 performs model learning. In the process of FIG. 11, the information generation device 100 acquires questionnaire data and purchase data (step S11).

[0141] For example, the smartphone 921 may receive input of answers to questionnaires by a monitor and transmit questionnaire data. Then, the communication unit 110 may receive the questionnaire data from the smartphone 921, and the storage unit 180 may store the questionnaire data. Also, the POS system 910 may generate and transmit purchase data in response to reading a point card and performing a cash register operation at the time of purchasing a product. Then, the communication unit 110 may receive the purchase data from the POS system 910, and the storage unit 180 may store the purchase data.

[0142] Next, the characteristic information generation unit 192 generates monitor awareness score data based on the questionnaire data (step S12). For example, as described above, the psychographic items of the questionnaire may include the same items as the items of the monitor awareness score. Then, the characteristic information generation unit 192 may convert the answers to the psychographic items of the questionnaire into scores to generate monitor awareness score data.

[0143] Next, the characteristic information generation unit 192 generates product score data based on the monitor awareness score data and the purchase data (step S13). For example, as described above, the items of the product score data may be the same as the items of the awareness score data. Then, the characteristic information generation unit 192 may calculate, for each item of the awareness score values of the monitors who have purchased a certain product among the monitors, the average value as the product score value of that product.

[0144] Next, the characteristic information generation unit 192 generates product score aggregation data for each monitor based on the purchase data (purchase history data) for each monitor and the product score data (step S14). For example, as described above, the product score aggregation value may be the average value for each item of the product scores of the products purchased by the target person within a predetermined period. Then, the characteristic information generation unit 192 may calculate the product score aggregation value by calculating the average value for each item of the product score values.

[0145] Next, the learning unit 191 performs model learning using the purchase data (purchase history data) for each monitor and the product score aggregation data (step S15). For example, as described above, the learning unit 191 may use, as the input to the model, the vector of the product score aggregation values shown in the product score aggregation data of each individual monitor, and perform model learning using the training data with the vector of the awareness score values shown in the awareness score data of that monitor as the correct answer. After step S15, the information generation device 100 ends the process of FIG. 11.

[0146] FIG. 12 is a diagram showing an example of a procedure for the information generation device 100 to perform processing using the product purchase data of members. FIG. 12 shows an example of a procedure for the characteristic information generation unit 192 to generate segment ratio data for a certain store.

[0147] In the process of FIG. 12, the information generation device 100 acquires purchase data (step S21). For example, as described above, the POS system 910 may generate and transmit purchase data in response to the reading of the point card and the cash register operation at the time of purchasing a product. Then, the communication unit 110 may receive the purchase data from the POS system 910, and the storage unit 180 may store the purchase data.

[0148] Next, the characteristic information generation unit 192 generates product score aggregation data for each member based on the purchase data (purchase history data) for each member and the product score data (step S22). For example, as described above, the product score aggregation value may be the average value for each item of the product score of the product score values of the products purchased by the target person within a predetermined period. Then, the characteristic information generation unit 192 may calculate the product score aggregation value by calculating the average value for each item of the product score value.

[0149] Next, the characteristic information generation unit 192 inputs the product score aggregation data for each member into the model to generate member awareness score data (step S23). In particular, the characteristic information generation unit 192 generates awareness score data for members other than the monitors (members other than those who answered the questionnaire). Regarding the awareness score data of the monitors, the awareness score data generated in step S12 of FIG. 11 can be used. Alternatively, in step S23, the characteristic information generation unit 192 may also generate monitor awareness score data.

[0150] Next, the characteristic information generation unit 192 generates awareness segment data based on the awareness score data (step S24). For example, as described above, for each psychological characteristic class, an equation may be defined in which the values for each item of the awareness score are weighted and summed to calculate the score value for that class. Then, the characteristic information generation unit 192 may calculate the score value for each class for each member, and determine the class with the highest score value as the psychological characteristic class of that member.

[0151] Next, the characteristic information generation unit 192 generates segment ratio data based on the awareness segment data (step S25). For example, as described above, the characteristic information generation unit 192 may calculate the ratio of members belonging to a certain class among the members with a purchase history at the store for which the segment ratio data is to be generated, for each psychological characteristic class. After step S25, the information generation device 100 ends the process of FIG. 12.

[0152] FIG. 13 is a diagram showing an example of a processing procedure when the information generation device 100 guides a product to a member. In the process of FIG. 13, the action processing unit 193 starts a loop L11 for performing processing for each member (step S31). The member who is the target of the processing in loop L11 is also referred to as the target member.

[0153] Next, the target determination unit 194 determines whether to guide a product to the target member (step S32). For example, as described above, when a certain store is the target of the sales strategy, the target determination unit 194 may determine, based on the member awareness score data, a member who is determined to have a relatively high possibility of using the store as the member to be the target of the action. Also, when a certain product is the target of the sales strategy, the target determination unit 194 may determine, based on the member awareness score data, a member who is determined to have a relatively high possibility of purchasing the product as the member to be the target of the action.

[0154] When the target person determination unit 194 decides to guide a product to the target member (step S32: YES), the target product determination unit 195 determines the product to be guided to the target member (step S33). For example, as described above, when a certain store is the target of the sales strategy, the target product determination unit 195 may determine, based on the awareness score data of the members who have a purchase history at that store, the products that are determined to have a relatively high possibility of being purchased by those members as the products to be the target of the action. Also, when a certain product is the target of the sales strategy, the target product determination unit 195 may determine that product as the product to be the target of the action.

[0155] Next, the action data generation unit 196 generates guidance data for guiding a product to the target member (step S34). For example, as described above, the action data generation unit 196 may generate the guidance data by pasting the product name of the product to be the target of the guidance and the image of the product to be the target of the guidance on the template of the guidance data.

[0156] Next, the transmission processing unit 197 transmits the guidance data to the smartphone 921 of the target member via the communication unit 110 (step S35).

[0157] Next, the action processing unit 193 performs the end processing of loop L11 (step S36). Specifically, the action processing unit 193 determines whether the processing of loop L11 has been performed for all the registered members. If it is determined that there are still members for whom the processing of loop L11 has not been performed, the action processing unit 193 continues to perform the processing of loop L11 for the unprocessed members.

[0158] On the other hand, if it is determined that the processing of loop L11 has been performed for all the registered members, the action processing unit 193 ends loop L11. When the action processing unit 193 ends loop L11 in step S36, the information generation device 100 ends the processing of FIG. 13. Also, in step S32, when the target determination unit 194 determines not to guide the product to the target member (step S32: NO), the process proceeds to step S37.

[0159] As described above, the characteristic information generation unit 192 inputs information based on a person's purchase history of products into a model that receives input of information based on the purchase history of products and outputs information indicating characteristics, and generates information indicating the characteristics of that person. According to the information generation device 100, information indicating the characteristics of a person (for example, a member) who has not conducted a questionnaire can also be obtained. The information indicating a person's characteristics can be used for formulating a sales strategy. Also, according to the information generation device 100, in terms of generating information indicating a person's characteristics using a model, it is expected that information can be generated with a relatively light computational load and in a relatively short time.

[0160] In addition, the characteristic information generation unit 192 inputs information based on a person's purchase history of products into a model (consciousness score data generation model) that receives input of information based on the purchase history of products and outputs information indicating psychological characteristics, and generates information indicating the psychological characteristics of that person. According to the information generation device 100, information indicating the psychological characteristics of a customer, which cannot be obtained through objective observation, can be obtained. The information indicating the psychological characteristics of a customer can be used for formulating a sales strategy.

[0161] Also, the information indicating psychological characteristics is information indicating the degree of correspondence of each item for each psychological characteristic item set as a psychological characteristic item for one person. The information based on the purchase history of products is information aggregated for one person, indicating the relevance between each product purchased by that person and each of the psychological characteristic items. According to the information generation device 100, the psychological characteristics of individual members can be quantitatively estimated. In this regard, according to the information generation device 100, it is expected that a more detailed sales strategy can be formulated than when qualitatively (in terms of YES / NO) estimating the psychological characteristics of individual members. For members, it is expected that they can receive more appropriate information that better matches their own psychological characteristics.

[0162] In addition, the information indicating the relevance between each product and the psychological characteristic items is information obtained by aggregating, for each psychological characteristic item, the degree of correspondence of that psychological characteristic item to a person for each person who purchased the product. According to the information generation device 100, the relevance between each product and the psychological characteristic items can be statistically calculated using the information indicating the psychological characteristics of the purchasers of the product. In this regard, according to the information generation device 100, it is expected that the relevance between the product and the psychological characteristic items can be calculated with relatively high accuracy.

[0163] In addition, the learning unit 191 performs learning of the model so that the difference between the information indicating the psychological characteristics of a person obtained by inputting information based on the purchase history of a product by a certain person into the model (consciousness score data generation model) and the information indicating the psychological characteristics of that person obtained from the investigation of the psychological characteristics of that person becomes smaller.

[0164] According to the information generation device 100, by constructing a consciousness score data generation model using a differentiable function, a gradient method such as the error backpropagation method can be used for learning the model. In this regard, according to the information generation device 100, it is expected that the learning of the model can be efficiently performed. In addition, according to the information generation device 100, the re-learning of the consciousness score data generation model can be performed, and in this regard, the accuracy of the model can be improved.

[0165] In addition, one consciousness score data generation model is provided for one store. The learning unit 191 performs learning of the consciousness score data generation model for each product handled in the store targeted by the model, using information indicating the relevance between each product and each psychological characteristic item, and information aggregated for each psychological characteristic item for each purchaser.

[0166] The information generation device 100 performs learning of the model using information based on the purchase history of products handled in the store targeted by the model to be learned, thereby performing learning of the model using training data excluding products not handled in that store. Performing learning of the model using training data excluding products not handled in the target store can be regarded as performing learning of the model using training data with noise removed. According to the information generation device 100, in this regard, it is expected that the model can be learned with relatively high accuracy.

[0167] In addition, the learning unit 191 performs learning of the consciousness score data generation model for each product handled in the store targeted by the model, using information indicating the relevance between each product and each psychological characteristic item, and information aggregated for each psychological characteristic item for each person with a purchase history of products in that store.

[0168] The information generation device 100 performs learning of the model using information of people with a purchase history of products in the store targeted by the model to be learned, thereby performing learning of the model using training data excluding people without a purchase history of products in that store. Performing learning of the model using training data excluding people without a purchase history of products in the target store can be regarded as performing learning of the model using training data with noise removed. According to the information generation device 100, in this regard, it is further expected that the model can be learned with relatively high accuracy.

[0169] In addition, the target person determination unit 194 determines a target person to whom telesales guidance is to be provided based on the consciousness score data. The transmission processing unit 197 transmits telesales guidance information to the person determined as the target person to whom telesales guidance is to be provided. According to the information generation device 100, it is possible to determine a target person to whom telesales guidance is to be provided based on the psychological characteristics of that person. According to the information generation device 100, it is expected that telesales guidance can be efficiently provided in this regard.

[0170] In addition, the action data generation unit 196 generates information indicating a product to be proposed as a product sold in one or more stores based on information indicating the psychological characteristics of customers in one or more stores and information indicating the relevance between each product and each psychological characteristic item. As described above, the consciousness score corresponds to an example of information indicating the psychological characteristics of customers. The product score corresponds to an example of information indicating the relevance between each product and each psychological characteristic item. According to the information generation device 100, it is possible to determine a product to be proposed as a product sold in a store based on the psychological characteristics of customers. According to the information generation device 100, it is expected that effective proposals can be made in this regard.

[0171] In addition, the action data generation unit 196 generates information indicating a product to be proposed as a product sold in one or more stores for a product whose correlation with the value of the psychological characteristics of a customer selected as a customer affecting the change in store sales is stronger than a predetermined condition, for each of the relevance values between the product and each psychological characteristic item. According to the information generation device 100, it is possible to determine a product to be proposed as a product sold in a store based on the psychological characteristics of customers who affect the change in sales of that store. According to the information generation device 100, it is expected that effective proposals can be made in this regard.

[0172] In addition, the target product determination unit 195 determines a product to be guided to a person based on the temporal change of information indicating the psychological characteristics of a certain person and information indicating the relationship between the product and the psychological characteristics of the person who purchased the product. The transmission processing unit 197 transmits the guidance of the determined product to the person. According to the information generation device 100, it is expected that appropriate products can be guided to a person at an appropriate timing in terms of determining products to guide the person based on the temporal change of the psychological characteristics of the person.

[0173] <Second Embodiment> FIG. 14 is a diagram showing another example of the configuration of an information generation device according to at least one embodiment. In the configuration shown in FIG. 14, the information generation device 610 includes a characteristic information generation unit 611.

[0174] With such a configuration, the characteristic information generation unit 611 inputs information based on the purchase history of products by a certain person into a model that receives an input of information based on the purchase history of products and outputs information indicating characteristics, and generates information indicating the characteristics of that person. The characteristic information generation unit 611 corresponds to an example of a characteristic information generation means.

[0175] According to the information generation device 610, even for a person who has not conducted a questionnaire, information indicating the characteristics of that person can be obtained. The information indicating the characteristics of a person can be used for formulating a sales strategy. Also, according to the information generation device 610, it is expected that information can be generated with a relatively light computational load and in a relatively short time in terms of generating information indicating the characteristics of a certain person using a model. The characteristic information generation unit 611 can be realized, for example, by using the functions of the characteristic information generation unit 192 in FIG. 2 and the like.

[0176] <Third Embodiment> FIG. 15 is a diagram showing an example of a processing procedure in an information generation method according to at least one embodiment. The information generation method shown in FIG. 15 includes generating information (step S611). In generating information (step S611), a computer inputs information based on the purchase history of products by a certain person into a model that receives an input of information based on the purchase history of products and outputs information indicating characteristics, and generates information indicating the characteristics of that person.

[0177] According to the information generation method shown in FIG. 15, even for a person who has not conducted a questionnaire, information indicating the characteristics of that person can be obtained. The information indicating the characteristics of a person can be used for formulating a sales strategy. Also, according to the information generation method shown in FIG. 15, it is expected that information indicating the characteristics of a certain person can be generated using a model with a relatively light computational load and in a relatively short time.

[0178] FIG. 16 is a diagram showing an example of the configuration of a computer according to at least one embodiment. In the configuration shown in FIG. 16, the computer 700 includes a CPU 710, a main storage device 720, an auxiliary storage device 730, an interface 740, and a non-volatile recording medium 750.

[0179] One or more or a part of the above information generation device 100 and the information generation device 610 may be implemented in the computer 700. In that case, the operations of each of the above-described processing units are stored in the auxiliary storage device 730 in the form of a program. The CPU 710 reads the program from the auxiliary storage device 730 and expands it in the main storage device 720, and executes the above processing according to the program. Also, the CPU 710 secures a storage area corresponding to each of the above-described storage units in the main storage device 720 according to the program. Communication between each device and other devices is executed by the interface 740 having a communication function and performing communication under the control of the CPU 710. Also, the interface 740 has a port for the non-volatile recording medium 750, and reads information from the non-volatile recording medium 750 and writes information to the non-volatile recording medium 750.

[0180] When the information generation device 100 is implemented in the computer 700, the operations of the processing unit 190 and each of its parts are stored in the auxiliary storage device 730 in the form of a program. The CPU 710 reads the program from the auxiliary storage device 730 and expands it in the main storage device 720, and executes the above processing according to the program.

[0181] Also, the CPU 710 secures a storage area for the storage unit 180 in the main storage device 720 according to a program. Communication with other devices by the communication unit 110 is executed by the interface 740 having a communication function and operating according to the control of the CPU 710. Display of an image by the display unit 120 is executed by the interface 740 including a display device and displaying various images according to the control of the CPU 710. Reception of a user operation by the operation input unit 130 is executed by the interface 740 including an input device and receiving a user operation according to the control of the CPU 710.

[0182] When the information generation device 610 is implemented in the computer 700, the operation of the characteristic information generation unit 611 is stored in the auxiliary storage device 730 in the form of a program. The CPU 710 reads the program from the auxiliary storage device 730 and expands it in the main storage device 720, and executes the above processing according to the program.

[0183] Also, the CPU 710 secures a storage area for the information generation device 610 to perform processing in the main storage device 720 according to a program. Communication between the information generation device 610 and other devices is executed by the interface 740 having a communication function and operating according to the control of the CPU 710. Interaction between the information generation device 610 and the user is executed by the interface 740 having an input device and an output device, presenting information to the user at the output device according to the control of the CPU 710, and receiving a user operation at the input device.

[0184] Any one or more of the above-described programs may be recorded on the non-volatile recording medium 750. In this case, the interface 740 may be configured to read a program from the non-volatile recording medium 750. Then, the CPU 710 may directly execute the program read by the interface 740, or may temporarily store it in the main storage device 720 or the auxiliary storage device 730 and then execute it.

[0185] Note that a program for executing all or part of the processes performed by the information generation device 100 and the information generation device 610 may be recorded on a computer-readable recording medium, and the program recorded on this recording medium may be read into a computer system and executed to perform the processes of each part. Here, the "computer system" is assumed to include hardware such as an OS (Operating System) and peripheral devices. Also, the "computer-readable recording medium" refers to a portable medium such as a flexible disk, a magneto-optical disk, a ROM (Read Only Memory), a CD-ROM (Compact Disc Read Only Memory), or a storage device such as a hard disk built into a computer system. Further, the above program may be for realizing a part of the aforementioned functions, or may be for realizing the aforementioned functions in combination with a program already recorded in the computer system.

[0186] As described above, the embodiments of the present invention have been described in detail with reference to the drawings. However, the specific configuration is not limited to this embodiment, and designs and the like within the scope not departing from the gist of the present invention are also included. Also, the above-described embodiments may be combined with other embodiments as appropriate.

[0187] Some or all of the above embodiments may be described as follows in the appended claims, but are not limited thereto.

[0188] (Appended Claim 1) Characteristic information generation means for generating information indicating the characteristics of a person by inputting information based on the purchase history of a product by a person into a model that receives input of information based on the purchase history of a product and outputs information indicating characteristics. An information generation device comprising the same.

[0189] (Appended Claim 2) The characteristic information generation means inputs information based on the purchase history of products by a certain person into a model that receives input of information based on the purchase history of products and outputs information indicating psychological characteristics, and generates information indicating the psychological characteristics of that person. The information generation device according to Supplementary Note 1.

[0190] (Supplementary Note 3) The information indicating psychological characteristics is information indicating the degree of applicability of each item of psychological characteristics, which is set as an item of psychological characteristics, for one person. The information based on the purchase history of the product is information obtained by aggregating, for each item of psychological characteristics of that person, information indicating the relevance between each of the products purchased by one person and each of the items of psychological characteristics. The information generation device according to Supplementary Note 2.

[0191] (Supplementary Note 4) The information indicating the relevance between each of the products and each of the items of psychological characteristics is information obtained by aggregating, for each item of psychological characteristics, the degree of applicability of that item of psychological characteristics to that person, for each person who purchased that product and for each item of psychological characteristics. The information generation device according to Supplementary Note 3.

[0192] (Supplementary Note 5) Learning means for performing learning of the model so that the difference between the information indicating the psychological characteristics of a person obtained by inputting information based on the purchase history of products by a certain person into the model and the information indicating the psychological characteristics of that person obtained by investigating the psychological characteristics of that person becomes smaller. The information generation device according to any one of Supplementary Notes 2 to 4, comprising the above.

[0193] (Supplementary Note 6) One of the models is provided for one store. The learning means performs the learning of the model for each product handled in the store targeted by the model, using information aggregated for each of the psychological characteristic items for each purchaser, which indicates the relevance between the product and each of the psychological characteristic items set as items of psychological characteristics. The information generation device according to Supplementary Note 5.

[0194] (Supplementary Note 7) The learning means performs the learning of the model for each product handled in the store targeted by the model, using information aggregated for each of the psychological characteristic items for each person with a purchase history of the product in that store, which indicates the relevance between the product and each of the psychological characteristic items. The information generation device according to Supplementary Note 6.

[0195] (Supplementary Note 8) Target person determination means for determining a target person to whom to provide a catalog sales guide based on information indicating the psychological characteristics of each person, generated by the characteristic information generation means for each person; Guide information transmission means for transmitting the catalog sales guide information to the person determined as the target person for the catalog sales guide; The information generation device according to any one of Supplementary Notes 2 to 7, comprising:

[0196] (Supplementary Note 9) Proposal information generation means for generating information indicating a product to be proposed as a product to be sold in the one or more stores, based on information indicating the psychological characteristics of customers of the one or more stores and information indicating the relevance between each of the products and psychological characteristic items set as items of psychological characteristics; The information generation device according to any one of Supplementary Notes 2 to 8, comprising:

[0197] (Supplementary Note 10) The proposal information generation means generates information indicating a product to be proposed as a product to be sold in the one or more stores, for a product that shows a value of relevance between the product and each of the psychological characteristic items, where the correlation with the value of the psychological characteristics of a customer selected as a customer affecting the change in the sales of the store is strong enough to meet a predetermined condition. The information generation device described in Supplementary Note 9.

[0198] (Supplementary Note 11) Based on the time change of information indicating a certain person's psychological characteristics and information indicating the relationship between a product and the psychological characteristics of the person who purchased the product, guidance product determination means for determining a product to guide that person, Guidance information transmission means for transmitting the guidance of the determined product to that person, The information generation device according to any one of Supplementary Notes 2 to 10, comprising:

[0199] (Supplementary Note 12) A computer Inputs information based on a certain person's purchase history of products into a model that receives input of information based on the purchase history of products and outputs information indicating characteristics, and generates information indicating that person's characteristics An information generation method including:

[0200] (Supplementary Note 13) Generating the information indicating the characteristics of the person means that the computer inputs information based on a certain person's purchase history of products into a model that receives input of information based on the purchase history of products and outputs information indicating psychological characteristics, and generates information indicating the psychological characteristics of that person The information generation method according to Supplementary Note 12, including:

[0201] (Supplementary Note 14) The information indicating the psychological characteristics is information indicating the degree of correspondence of each item for each psychological characteristic item set as an item of psychological characteristics for one person, The information based on the purchase history of the product is information obtained by aggregating, for each psychological characteristic item for that person, information indicating the relevance between each of the products purchased by one person and each of the psychological characteristic items, The information generation method according to Supplementary Note 13.

[0202] (Supplementary Note 15) The information indicating the relevance between each of the products and the psychological characteristic items is information obtained by aggregating, for each psychological characteristic item, the degree of applicability of that psychological characteristic item to that person for each person who purchased the product. The information generation method described in Supplementary Note 14.

[0203] (Supplementary Note 16) The computer performs learning of the model so that the difference between the information indicating the psychological characteristics of the person obtained by inputting information based on the purchase history of the product by a certain person into the model and the information indicating the psychological characteristics of the person obtained by investigating the psychological characteristics of the person becomes smaller. The information generation method according to any one of Supplementary Notes 13 to 15, including this.

[0204] (Supplementary Note 17) One of the models is provided for one store. Performing the learning means that the computer performs learning of the model for each product handled at the store targeted by the model, using information indicating the relevance between each of the products and the psychological characteristic items set as items of psychological characteristics, aggregated for each psychological characteristic item for each purchaser. The information generation method according to Supplementary Note 16, including this.

[0205] (Supplementary Note 18) Performing the learning means that the computer performs learning of the model for each product handled at the store targeted by the model, using information indicating the relevance between each of the products and the psychological characteristic items, aggregated for each psychological characteristic item for each person who has a purchase history of the product at that store. The information generation method according to Supplementary Note 17, including this.

[0206] (Supplementary Note 19) The computer Based on the information indicating the psychological characteristics of the person generated for each person, determining the target person to whom to offer mail-order sales. Transmitting the direct marketing guidance information to the person determined to be the target of the direct marketing guidance; An information generation method according to any one of Appendices 13 to 18, including this.

[0207] (Appendix 20) The computer, Based on information indicating the psychological characteristics of customers in one or more stores and information indicating the relevance between each of the psychological characteristic items set as items of the product and the psychological characteristics, generating information indicating a product to be proposed as a product sold in the one or more stores An information generation method according to any one of Appendices 13 to 19, including this.

[0208] (Appendix 21) Generating the information indicating the product to be proposed includes the computer generating information indicating a product to be proposed as a product sold in the one or more stores, where the value of the relevance between each of the product and the psychological characteristic items is such that the correlation with the value of the psychological characteristics of the customers selected as the customers affecting the change in the sales of the store is stronger than a predetermined condition. An information generation method according to Appendix 20, including this.

[0209] (Appendix 22) The computer, Based on the time change of information indicating the psychological characteristics of a person and information indicating the relationship between the product and the psychological characteristics of the person who purchased the product, determining a product to guide the person, Transmitting the guidance of the determined product to the person, An information generation method according to any one of Appendices 13 to 21, including this.

[0210] (Appendix 23) A program for causing a computer to Input information based on the purchase history of a product by a person into a model that receives an input of information based on the purchase history of a product and outputs information indicating characteristics, and generate information indicating the characteristics of the person. Execute this.

[0211] (Appendix 24) In generating information indicating the characteristics of the person, the computer is caused to input information based on the purchase history of goods by a certain person into a model that receives input of information based on the purchase history of goods and outputs information indicating psychological characteristics, and generate information indicating the psychological characteristics of that person. The program described in Appendix 23.

[0212] (Appendix 25) The information indicating the psychological characteristics is information indicating the degree of applicability of each item of psychological characteristics set as items of psychological characteristics for one person. The information based on the purchase history of the goods is information obtained by aggregating, for each item of the psychological characteristics for that person, information indicating the relevance between each of the goods purchased by one person and each of the items of the psychological characteristics. The program described in Appendix 24.

[0213] (Appendix 26) The information indicating the relevance between each of the goods and each of the items of the psychological characteristics is information obtained by aggregating, for each item of the psychological characteristics, the degree of applicability of that item of the psychological characteristics to that person for each person who purchased that good. The program described in Appendix 25.

[0214] (Appendix 27) For the computer, Performing learning of the model so that the difference between the information indicating the psychological characteristics of a person obtained by inputting information based on the purchase history of goods by a certain person into the model and the information indicating the psychological characteristics of that person obtained from an investigation of the psychological characteristics of that person becomes smaller. The program according to any one of Appendices 24 to 26, which causes the above to be executed.

[0215] (Appendix 28) One of the models is provided for one store. By performing the learning, the computer performs the learning of the model for each product handled at the store targeted by the model, using information aggregated for each purchaser for each of the psychological characteristic items indicating the relevance between the product and each of the psychological characteristic items set as items of psychological characteristics. The program according to appended note 27 that causes the above to be executed.

[0216] (Appended note 29) By performing the learning, the computer performs the learning of the model for each product handled at the store targeted by the model, using information aggregated for each of the psychological characteristic items indicating the relevance between the product and each of the psychological characteristic items for each person having a purchase history of the product at that store. The program according to appended note 28 that causes the above to be executed.

[0217] (Appended note 30) For the computer, Based on the information indicating the psychological characteristics of each person, determining a person to be targeted for direct marketing guidance; Sending the direct marketing guidance information to the person determined to be the person targeted for direct marketing guidance; The program according to any one of appended notes 24 to 29 that causes the above to be executed.

[0218] (Appended note 31) For the computer, Based on the information indicating the psychological characteristics of the customers of one or more stores and the information indicating the relevance between each of the psychological characteristic items set as items of psychological characteristics and the product, generating information indicating a product to be proposed as a product to be sold at the one or more stores The program according to any one of appended notes 24 to 30 that causes the above to be executed.

[0219] (Appended note 32) In generating information indicating the proposed product, the computer generates information indicating a product to be proposed as a product sold at the one or more stores, which has a strong correlation with each of the products and psychological characteristic items such that the correlation with the value of the psychological characteristics of the customers selected as the customers affecting the change in the sales of the store is equal to or greater than a predetermined condition. The program according to Supplementary Note 31, which causes the above to be executed.

[0220] (Supplementary Note 33) To the computer, Based on the time change of the information indicating the psychological characteristics of a person and the information indicating the relationship between the product and the psychological characteristics of the person who purchased the product, determining a product to guide the person, Sending a guide for the determined product to the person, The program according to any one of Supplementary Notes 24 to 32, which causes the above to be executed.

Explanation of Signs

[0221] 1 Sales Strategy Support System 100, 610 Information Generation Device 110 Communication Unit 120 Display Unit 130 Operation Input Unit 180 Storage Unit 190 Processing Unit 191 Learning Unit 192, 611 Characteristic Information Generation Unit 193 Action Processing Unit 194 Target Person Determination Unit 195 Target Product Determination Unit 196 Action Data Generation Unit 197 Transmission Processing Unit 910 POS System 921 Smartphone 922 Point Card

Claims

1. Characteristic information generation means for inputting information based on the purchase history of a person into a model that receives input of information based on the purchase history of a product and outputs information indicating characteristics, and generating information indicating the characteristics of that person An information generation device comprising the same.

2. The characteristic information generation means inputs information based on the purchase history of a person into a model that receives input of information based on the purchase history of a product and outputs information indicating psychological characteristics, and generates information indicating the psychological characteristics of that person The information generation device according to claim 1.

3. The information indicating the psychological characteristics is information indicating the degree of correspondence of each item of the psychological characteristics set as items of psychological characteristics for one person, The information based on the purchase history of the product is information obtained by aggregating, for each item of the psychological characteristics for that person, information indicating the relevance of each of the products purchased by one person to each of the psychological characteristics items. The information generation device according to claim 2.

4. The information indicating the relevance of each of the products to each of the psychological characteristics items is information obtained by aggregating, for each item of the psychological characteristics, the degree of correspondence of that psychological characteristics item to that person for each person who purchased the product. The information generation device according to claim 3.

5. Learning means for performing learning of the model so that the difference between the information indicating the psychological characteristics of a person obtained by inputting information based on the purchase history of a product by that person into the model and the information indicating the psychological characteristics of that person obtained by investigating the psychological characteristics of that person becomes smaller The information generation device according to claim 2, comprising the same.

6. Target person determination means for determining a target person to whom to provide a mail order sales guide based on the information indicating the psychological characteristics of each person generated by the characteristic information generation means, Guide information transmission means for transmitting the mail order sales guide information to the person determined as the target person to whom to provide the mail order sales guide, The information generation device according to claim 2, comprising the same.

7. Proposal information generation means for generating information indicating a product to be proposed as a product to be sold in the one or more stores based on information indicating the psychological characteristics of customers of the one or more stores and information indicating the relevance of each of the products to each of the psychological characteristics items set as items of psychological characteristics The information generation device according to claim 2, comprising the same.

8. Based on the time change of information indicating certain person's psychological characteristics and information indicating the relationship between a product and the psychological characteristics of the person who purchased the product, a guidance product determination means for determining a product to guide the person, a guidance information transmission means for transmitting the guidance of the determined product to the person, The information generation device according to claim 2, comprising the above.

9. A computer, inputs information based on a person's purchase history of a product into a model that receives an input of information based on the purchase history of the product and outputs information indicating characteristics, and generates information indicating the characteristics of the person An information generation method including this.

10. To a computer, input information based on a person's purchase history of a product into a model that receives an input of information based on the purchase history of the product and outputs information indicating characteristics, and generate information indicating the characteristics of the person A program for causing the above to be executed.

Images