Measure effect estimation device, measure effect estimation method, and measure effect estimation program

Abstract

The present invention provides a measure effect estimation device, a measure effect estimation method, and a measure effect estimation program capable of estimating an effect of a measure. A measure effect estimation device according to the present disclosure comprises: a customer image specification means for specifying a customer image on the basis of prescribed information on a customer; a measure reception unit for receiving prescribed measure information; and an output means for outputting the degree of interest of the customer to a measure estimated on the basis of the customer image and the prescribed measure information.

Description

Policy effect estimation device, policy effect estimation method, and policy effect estimation program

[0001] The present disclosure relates to a policy effect estimation device, a policy effect estimation method, and a policy effect estimation program.

[0002] In recent years, technologies have been developed to analyze customer consumption behavior data and utilize it for marketing. For example, Patent Literature 1 discloses a technology for scoring changes in customer interests and implementing measures such as campaigns by taking into account the fluctuations in the scores.

[0003] Japanese Patent Application Laid-Open No. 2023-068415

[0004] As disclosed in the above-mentioned Patent Document 1 and the like, various tools have been proposed for analyzing business data and extracting issues. However, it has not been easy to estimate the effectiveness, i.e., usefulness, of a campaign or other measure using existing tools before the measure is implemented.

[0005] In view of the above-mentioned problems, an object of the present disclosure is to provide a policy effect estimation device, a policy effect estimation method, and a policy effect estimation program that are capable of estimating the effects of policies.

[0006] The policy effect estimation device according to the present disclosure includes a customer profile identification means for identifying a customer profile based on predetermined information about the customer, a policy reception unit for receiving the predetermined policy information, and an output means for outputting the customer's level of interest in a policy estimated based on the customer profile and the predetermined policy information.

[0007] The method for estimating the effectiveness of a policy disclosed herein includes a computer identifying a customer profile based on specified information about the customer, accepting specified policy information, and outputting the customer's level of interest in a policy estimated based on the customer profile and the specified policy information.

[0008] The program for estimating the effectiveness of a policy disclosed herein causes a computer to perform the following processes: identifying a customer profile based on specified information about the customer; receiving specified policy information; and outputting the customer's level of interest in a policy estimated based on the customer profile and the specified policy information.

[0009] The present disclosure makes it possible to provide a policy effect estimation device, a policy effect estimation method, and a policy effect estimation program that are capable of estimating the effects of a policy.

[0010] 1 is a block diagram showing a configuration of a policy effect estimation device according to the present disclosure. FIG. 2 is a flowchart showing an example of the flow of a policy effect estimation method according to the present disclosure. FIG. 3 is a block diagram showing a configuration of a service request device capable of communicating with a policy effect estimation device according to the present disclosure. FIG. 4 is a block diagram showing a configuration of a policy effect estimation device according to the present disclosure. FIG. 5 is a sequence diagram showing an example of operation of the service request device and the policy effect estimation device when estimating policy effects. FIG. 6 is a diagram showing an example of customer profile information. FIG. 7 is a diagram showing an example of customer profile analysis. FIG. 8 is a diagram showing an example of policy information. FIG. 9 is a diagram showing an example of interest level information. FIG. 10 is a block diagram showing a configuration of a customer classification device. FIG. 11 is a block diagram showing a configuration of a policy effect estimation system according to the present disclosure. FIG. 12 is a block diagram showing configurations of a service request device and a customer information providing device capable of communicating with a policy effect estimation device according to the present disclosure. FIG. 13 is a sequence diagram showing an example of operation of each device when specifying a customer profile. FIG. 14 is a block diagram showing configurations of a service request device, a customer information providing device, and a policy generation device capable of communicating with a policy effect estimation device according to the present disclosure. FIG. 15 is a block diagram showing a configuration of a policy effect estimation system according to the present disclosure. FIG. 16 is a sequence diagram showing an example of operation of each device when generating policy information. FIG. 17 is a sequence diagram showing an example of operation of each device when estimating interest level information. FIG. 18 is a block diagram showing an example of hardware configuration of a computer.

[0011] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. In each drawing, the same or corresponding elements are designated by the same reference numerals, and for clarity of explanation, duplicate explanations will be omitted as necessary.

[0012] First Embodiment An example of the configuration of a policy effect estimation device 100 will be described below with reference to Fig. 1. The policy effect estimation device 100 includes a customer information receiving unit 110, a customer image identification unit 120, a policy receiving unit 130, and an output unit 140.

[0013] The customer information receiving unit 110 receives predetermined information related to customers, i.e., first attribute information extracted from the customer information. The customer information is information related to customers of a target business for which a user is considering implementing a policy. Here, the user is a user of the policy effect estimation device 100. However, the user and the worker using the policy effect estimation device 100 do not necessarily have to be the same person. For example, the policy effect estimation device 100 may be operated by a worker who has heard about the user's goals, etc. The customer information is, for example, consumption behavior information related to customers of a retail store run by the user. The customer information is text information. The customer information may be diagram information such as graphs and photographs converted into text information.

[0014] The first attribute information is at least one word extracted from customer information, which is text information, and is information that may relate to the customer's consumption behavior. In other words, the first attribute information is at least one word that is associated with the customer's image. Note that consumption behavior refers to, for example, purchasing products, services, etc. Furthermore, such behavior is not limited to behavior performed in a physical store, but may also include online shopping performed on the Internet. Specifically, the first attribute information is, for example, information that can be extracted from customer consumption behavior information included in the customer information. Specifically, the first attribute information may include, for example, information related to a product purchased by a customer who is a registered member at a retail store, such as the product name and product attribute information previously assigned to the product.

[0015] Consumption behavior information includes the consumption behavior history of existing customers, etc., and specifically, is information linking a specific person to the consumption behavior history of that person. The consumption behavior information may be information linking a specific person to the consumption behavior history of that person directly or indirectly. Examples of consumption behavior information include information linking a customer who has registered as a member at a retail store to the customer's purchase history at the retail store. The purchase history included in the consumption behavior information includes, for example, the date and time of product purchase, the purchase amount, and the name of the purchased product. The consumption behavior information may also include basic attribute information of the customer. The basic attribute information is basic attribute information about the customer, such as demographic attributes, geographic attributes, and behavioral attributes. Demographic attributes are demographic attributes, such as the customer's age and gender. Geographic attributes are geographic attributes, such as the climate, culture, and economy specific to the region where the customer resides. Behavioral attributes are behavioral attributes, such as the frequency, purpose, purchase history, and range of activity of the customer using products and services.

[0016] The customer image identification unit 120 identifies a customer image including second attribute information expanded based on the first attribute information accepted by the customer information accepting unit 110. The second attribute information is character information expanded from the first attribute information and is a word indicating specific attributes related to the customer's consumption behavior. The second attribute information is more detailed information than the first attribute information. For example, the second attribute information expanded from the first attribute information "I like books" is "I like novels." The second attribute information is linked to the corresponding customer. The second attribute information is identified by, for example, performing one or a combination of expansion, conversion, and estimation processing on the first attribute information.

[0017] Specifically, the customer image identification unit 120 may identify, from the first attribute information, which is character information, character information having a concept or meaning similar to that of the first attribute information as the second attribute information. The customer image identification unit 120 may determine whether the character information identified as the second attribute information has a concept or meaning similar to that of the first attribute information by performing natural language processing. One aspect of such second attribute information is psychographic attributes expanded based on the first attribute information. If the first attribute information is the name of a product purchased by a customer, the second attribute information is, for example, a word or words that indicate the image of the customer, which are expanded from the name of the product purchased. Specifically, if the first attribute information is "pet supplies," the second attribute information may be "animal lover," "family-friendly," "safety measures," etc.

[0018] The policy receiving unit 130 receives policy information designed for the target business as text information. The policy information may be set by the user or generated using a specified device. The policy information may be, for example, a sentence such as "We will hold a XX fair," or one or more words such as "XX fair." The policy information may also be a policy name linked to policy details. The policy name is a word that succinctly indicates the content of the policy. The policy details are text information that constitutes the policy and indicate the content of the policy in detail. The policy information may also include information regarding the timing of the policy implementation, such as "We will hold a XX fair in mid-XX."

[0019] The output unit 140 outputs the customer's degree of interest in a policy estimated based on the customer profile identified by the customer profile identification unit 120 and the policy information received by the policy receiving unit 130. Here, the customer's degree of interest in a policy indicates the percentage of customers estimated to be interested in the policy. The degree of interest may be, for example, a numerical value obtained by estimating whether each customer is interested in the policy and then dividing the number of customers who are interested by the total number of customers. The presence or absence of each customer's interest in a policy may be determined by calculating the degree of association between the policy information and the customer profile. For example, each customer may be determined to be interested in the policy if the degree of association between second attribute information extending the policy information and the second attribute information included in the customer profile of the customer is equal to or greater than a predetermined value.

[0020] Furthermore, if the policy information includes information regarding the implementation timing of the policy, the relevance between the policy information and the customer profile may be calculated taking into account the implementation timing. For example, consider a case where the policy information is "A Christmas fair will be held in mid-December." Even a customer who does not usually purchase sweets may want to purchase sweets that are only sold during the Christmas season. Therefore, the relevance between the policy information and the customer profile may be calculated taking into account the fact that the customer is more likely to purchase sweets than at other times.

[0021] Next, an example of a campaign effect estimation method according to the present disclosure will be described with reference to FIG. 2 . First, the customer information receiving unit 110 receives first attribute information extracted from predetermined information related to a customer (step S101). Next, the customer image identification unit 120 identifies a customer image including second attribute information expanded based on the first attribute information received in step S101 (step S102). Next, the campaign receiving unit 130 receives predetermined campaign information (step S103). Next, the output unit 140 outputs the customer's level of interest in the campaign, estimated based on the customer image identified in step S102 and the campaign information received in step S103 (step S104).

[0022] In this way, the policy effect estimation method using the policy effect estimation device 100 identifies a customer profile including attribute-extended information, and calculates the customer's level of interest in the policy based on the customer profile and the policy information. Therefore, the customer's level of interest in the accepted policy, i.e., the effect of the policy, can be provided to the user.

[0023] The policy effect estimation device 100 includes a processor, memory, and storage device (not shown). The storage device stores a computer program that implements the processing of the policy effect estimation method according to this embodiment. The processor then loads the computer program from the storage device into the memory and executes the computer program. As a result, the processor realizes the functions of the customer information receiving unit 110, customer profile identification unit 120, policy receiving unit 130, and output unit 140.

[0024] Alternatively, each component of the policy effect estimation device 100 may be realized by dedicated hardware. Furthermore, some or all of the components of each device may be realized by general-purpose or dedicated circuits, processors, etc., or a combination of these. These may be configured by a single chip, or by multiple chips connected via a bus. Some or all of the components of each device may be realized by a combination of the above-mentioned circuits, etc., and a program. Furthermore, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), an FPGA (Field-Programmable Gate Array), a quantum processor (quantum computer control chip), etc., may be used as the processor.

[0025] Furthermore, when some or all of the components of the policy effect estimation device 100 are realized by multiple information processing devices, circuits, etc., the multiple information processing devices, circuits, etc. may be centrally or decentralized. For example, the information processing devices, circuits, etc. may be realized as a client-server system, a cloud computing system, or the like, in a form in which each is connected via a communication network. Furthermore, the functions of the policy effect estimation device 100 may be provided in a SaaS (Software as a Service) format.

[0026] Second Embodiment The policy effect estimation device 400 shown in FIG. 3 is an example of the policy effect estimation device 100 described above. A configuration example of a service request device 600 capable of communicating with the policy effect estimation device 400 will be described with reference to FIG. 3 . As shown in FIG. 3 , the policy effect estimation device 400 is connected to the service request device 600. The service request device 600 is an information processing device that requests the policy effect estimation device 400 to estimate a policy effect. For example, the service request device 600 is owned by a user requesting the policy effect estimation, i.e., a business operator operating a target business. As will be described in detail later, the service request device 600 transmits customer information of a target business for which a policy is to be implemented to the policy effect estimation device 400 and receives a customer profile of the customer from the policy effect estimation device 400. The service request device 600 then transmits policy information of a policy to be implemented to the policy effect estimation device 400 and receives interest level information related to the policy from the policy effect estimation device 400.

[0027] The service request device 600 includes a customer information database 610, a customer information transmission unit 620, a customer profile information reception unit 630, a policy information transmission unit 640, and an interest level information reception unit 650. The customer information database 610 stores customer information related to customers of the target business. The customer information database 610 may store the customer information as text data, i.e., character information, or may store the customer information in other formats such as graphs or photographs. The customer information stored in the customer information database 610 is specifically, for example, consumer behavior information including basic attribute information and purchase history data of each customer at a retail store run by the user.

[0028] The customer information sending unit 620 is a communication means that sends customer information stored in the customer information database 610 to the policy effect estimation device 400. When customer information is stored in a format other than character information, the customer information sending unit 620 may convert the business information into character information using an external service and send the character information to the policy effect estimation device 400. Here, the external service is a service that converts data such as graphs into character information, and may be provided as a cloud service, for example. Customer information stored in a format other than character information typically has a larger data volume than customer information stored as character information. Therefore, converting customer information stored in a format other than character information into character information before transmitting it can reduce the amount of data transmitted. Note that the process of converting customer information stored in a format other than character information into character information may be performed on the policy effect estimation device 400 side. In this case, the customer information sending unit 620 sends customer information stored in a format other than character information to the policy effect estimation device 400 without converting it into character information.

[0029] The policy effect estimation device 400 is configured to, upon receiving customer information, identify and output customer profile information of the customer included in the customer information. The customer profile information receiving unit 630 is a communication means for receiving customer profile information from the policy effect estimation device 400. A user views the received customer profile information and devises and inputs a policy. The policy information sending unit 640 is a communication means for sending the policy information input by the user to the policy effect estimation device 400. Upon receiving policy information, the policy effect estimation device 400 is configured to estimate and output customer interest level information for the policy information. The interest level information receiving unit 650 is a communication means for receiving interest level information from the policy effect estimation device 400. A user views the received interest level information to confirm the customer's interest in the policy, thereby judging the usefulness of the policy.

[0030] Next, an example of the configuration of the policy effect estimation device 400 will be described with reference to Fig. 4. The policy effect estimation device 400 is an information processing device that performs policy effect estimation processing, etc., and is, for example, a server device realized by a computer. The policy effect estimation device 400 may be redundantly configured with multiple servers, and each functional block may be realized by multiple computers. The policy effect estimation device 400 includes a memory 410, a communication unit 420, a storage unit 430, and a control unit 440.

[0031] The memory 410 is a storage area that temporarily stores the processing contents of the control unit 440, and is a volatile storage device such as a RAM (Random Access Memory). The communication unit 420 is an interface that communicates with the outside of the policy effect estimation device 400. The storage unit 430 is a storage device that stores a program 431 and the like. The program 431 is a computer program that implements the policy effect estimation process according to the present disclosure.

[0032] The control unit 440 includes a customer information receiving unit 441, a customer image identification unit 442, a classification unit 443, a policy receiving unit 444, an estimation unit 445, and an output unit 446. The control unit 440 is a control device that controls the operation of the policy effect estimation device 400, and is, for example, a processor such as a CPU. The control unit 440 loads the program 431 from the storage unit 430 into the memory 410 and executes it. In this way, the control unit 440 realizes the functions of the customer information receiving unit 441, the customer image identification unit 442, the classification unit 443, the policy receiving unit 444, the estimation unit 445, and the output unit 446.

[0033] When requesting an estimation of a policy effect, the service request device 600 first transmits customer information for a target business in which the policy is to be implemented to the policy effect estimation device 400. Upon receiving the customer information, the customer information receiving unit 441 accepts input of the customer information. If the received customer information is information other than text information, the customer information receiving unit 441 converts the received customer information into text information using the external service described above before accepting the information. Note that the customer information received by the customer information receiving unit 441 may be information about one customer, or information about each of multiple customers.

[0034] The customer image identification unit 442 identifies second attribute information of a customer by attribute-expanding the customer information, i.e., the customer's first attribute information, and links the second attribute information to the customer. Attribute expansion is a technique for expanding attributes contained in information. Customer image information is information that includes the customer's second attribute information. When customer information about multiple customers is received, the customer image identification unit 442 expands the customer information, i.e., identifies a customer image, for each customer.

[0035] When the customer information received by the customer information receiving unit 441 relates to a large number of customers, simply listing the customer images identified for each customer may make it difficult for the user to understand the attributes of all customers who use the store. Therefore, the classification unit 443 classifies the multiple customer images into predetermined clusters. Here, a cluster is a persona captured from the multiple customer images. The classification unit 443 classifies, for example, tens to thousands of customer images into approximately 10 to 30 clusters, and assigns text information to each cluster indicating the specific type of person each cluster represents.

[0036] The method by which the classification unit 443 classifies each customer image into a cluster is not particularly limited, but for example, the classification unit 443 may summarize the second attribute information contained in each customer image into multiple words using so-called LLMs (Large Language Models), and classify customer images in which multiple words are similar into the same cluster.

[0037] Here, the LLM is realized by, for example, a neural network. The neural network includes a plurality of artificial neurons and has synapses connecting the artificial neurons. Each synapse has a weight. When such a neural network receives an input, it performs a calculation using the weight associated with each synapse and produces an output according to the input.

[0038] A model representing the connection relationship between neurons and synapses is stored in memory, for example, in the form of software. Alternatively, the model may be realized as a dedicated circuit. Similarly, the weights of each synapse are stored in memory, for example, in the form of software. Alternatively, a circuit representing the weights may be implemented in a dedicated circuit. Note that when configuring an LLM using multiple models, all of the models do not necessarily need to be stored in the same memory.

[0039] There are a variety of models using such neural networks, and LLM may be realized by adopting or replacing a wide variety of models, such as a Transformer, a Convolutional Neural Network (CNN), or a Recurrent Neural Network (RNN).

[0040] When using LLM, the classification unit 443 determines whether multiple words associated with each customer image are similar to each other. For example, when determining whether customer images A and B are similar, the classification unit 443 determines whether multiple words associated with customer image B are similar to multiple words associated with customer image A. In this case, similar words are assigned a "1" and dissimilar words are assigned a "0." Next, the classification unit 443 adds up the numerical values ​​assigned to multiple words associated with customer image B for each of the multiple words associated with customer image A and divides the sum by the number of words in customer image B. This calculates the relevance of customer image B to each of the multiple words associated with customer image A. Next, the classification unit 443 calculates the similarity between customer image A and customer image B by calculating the average value of the relevance between customer image B and the multiple words associated with customer image A. The classification unit 443 performs this process for each customer image, and classifies those with high similarity into the same cluster.

[0041] The output unit 446 may transmit the clusters classified by the classification unit 443 as customer image information to the service request device 600. The output unit 446 may also transmit the customer image information identified by the customer image identification unit 442 to the service request device 600. The output unit 446 may transmit the customer image information as text data, or may transmit the results of analyzing the customer image information as data other than text data, such as graph data.

[0042] The user sets a policy and inputs it into the service request device 600. The user may set the policy while viewing the received customer profile information. When receiving the policy information from the service request device 600, the policy reception unit 444 accepts the policy information as text information. The estimation unit 445 estimates the customer's level of interest in the policy indicated by the policy information based on the customer profile identified by the customer profile identification unit 442 and the policy information accepted by the policy reception unit 444. Specifically, the estimation unit 445 identifies second attribute information in the policy information by expanding words extracted from the policy information (text information), i.e., the first attribute information. The estimation unit 445 then estimates each customer's level of interest in the policy by calculating the degree of association between the second attribute information included in each customer profile and the second attribute information expanded from the policy information. Furthermore, when estimating each customer's level of interest in the policy, the estimation unit 445 may calculate the degree of association between the customer profile information or first attribute information included in each customer profile and the customer profile information or first attribute information expanded from the policy information. The method of attribute expansion and relevance calculation is the same as described above, and therefore will not be described here. The output unit 446 transmits the interest level estimated by the estimation unit 445 as interest level information to the service request device 600. The user can check whether the measure is likely to interest many customers by looking at the received interest level information.

[0043] When the classification unit 443 classifies each customer image into a cluster, the estimation unit 445 may estimate the degree of interest in the measure for each customer, and then generate statistical information by statistically processing the degree of interest for each cluster. In this case, the output unit 446 transmits the generated statistical information to the service request device 600 as interest level information.

[0044] Next, an example of the operation of the service request device 600 and the policy effect estimation device 400 will be described with reference to FIG. 5 . First, when the service request device 600 transmits customer information (step S201), the customer information receiving unit 441 receives the customer information. Next, the customer image identification unit 442 identifies a customer image including second attribute information that is an extension of the customer information received in step S201 (step S202). Next, the classification unit 443 classifies each customer image into a predetermined cluster based on the second attribute information identified in step S202 (step S203). Next, the output unit 446 transmits the clusters classified in step S203 as customer image information (step S204).

[0045] FIG. 6 shows an example of customer profile information transmitted to the service request device 600 in step S204. In FIG. 6, the letters a to e assigned to each cluster are codes for identifying the cluster. Each cluster a to e is assigned character information as a persona, which indicates a specific example of the customer profile classified into the cluster. By viewing the persona assigned to each cluster, the user can specifically understand the types of customers who use the store. Furthermore, as shown in FIG. 6, the customer profile information may include various data related to the customers who make up each cluster, such as the total purchase amount at the store (i.e., total amount spent), the total number of visits, and the number of customers.

[0046] The output unit 446 may transmit the analysis results of the customer profile information to the service request device 600. FIG. 7 shows another example of the customer profile information transmitted to the service request device 600 in step S204. As shown in FIG. 7, the output unit 446 may output the analysis results of the customer profile information as a graph. The graph shown in FIG. 7 is a graph showing an example of an RFM analysis performed using the various data shown in FIG. 6. Here, RFM analysis is a customer analysis method that analyzes customers using three indices: "Recency (purchase date)," "Frequency (purchase frequency)," and "Monetary (purchase amount)."

[0047] Furthermore, the output unit 446 may output text data explaining the graph data in addition to the graph data shown in Fig. 7. The text data explaining the graph data may be generated using an external service such as LLM. An example of the text data explaining the graph in Fig. 7 is "In clusters a and d, the amount of usage per person and the number of usages per person are high, and measures targeting these clusters are expected to be highly effective."

[0048] Returning to FIG. 5 , the explanation will be continued. Upon receiving the customer profile information, the service request device 600 presents the customer profile information to the user (step S205). The user confirms the presented customer profile information, devises a measure, and inputs the measure into the service request device 600. The service request device 600 may present the user with a query display prompting the user to input measure information. The query display is not particularly limited, and may, for example, include an input field that allows the user to input the measure name and measure details associated with each measure. FIG. 8 shows an example of input measure information. In the above example, the user may have devised, for example, a "seasonal food fair" as a measure for cluster a, i.e., people who prefer stores with a good selection of groceries, and a "product placement next to the cash register" as a measure for cluster d, i.e., people who are interested in pet products.

[0049] Returning to FIG. 5 , the explanation continues. The service request device 600 transmits the input policy information to the policy effect estimation device 400 (step S206). Upon receiving the policy information, the estimation unit 445 estimates the customer's level of interest in the policy indicated by the policy information based on the customer profile identified in step S202 and the policy information received in step S206 (step S207). In step S207, the estimation unit 445 estimates each customer's level of interest in the policy and then generates statistical information by statistically processing the level of interest for each cluster. The output unit 446 transmits the statistical information generated in step S207 to the service request device 600 as interest level information (step S208). The service request device 600 presents the received interest level information to the user (step S209).

[0050] FIG. 9 is an example of interest level information presented to the user. The interest level information is information regarding each customer's level of interest in the campaign and may include statistical information generated in step S207. FIG. 9 is a graph showing the level of interest of each cluster in the campaign "Seasonal Food Fair." In addition to the interest level information, the graph shown in FIG. 9 may also present customer profile information, such as personas, for each cluster to the user. By calculating and displaying the interest level for each cluster as shown in FIG. 9, the user can understand the extent to which each cluster is likely to be interested in the campaign they have devised. In this way, by using the campaign effect estimation device 400, the user can estimate the effectiveness of the campaign and implement more effective campaigns.

[0051] <Embodiment 3> In the above-described embodiment 2, customer profile classification is performed within the policy effect estimation device, but customer profile classification may also be performed in a separate device, i.e., a customer classification device. Below, an example configuration of the customer classification device 403 will be described with reference to FIG. 10 . The customer classification device 403 is an information processing device that performs customer classification processing, etc., and is, for example, a server device implemented by a computer. The customer classification device 403 may be redundantly configured with multiple servers, and each functional block may be implemented by multiple computers. The policy effect estimation device 400 includes a memory 450, a communication unit 460, a storage unit 470, and a control unit 480.

[0052] The memory 450 is a storage area that temporarily stores the processing contents of the control unit 480, and is a volatile storage device such as a RAM (Random Access Memory). The communication unit 460 is an interface that communicates with the outside of the customer classification device 403. The storage unit 470 is a storage device that stores the program 471 and the like. The program 471 is a computer program that implements the customer classification processing according to the present disclosure.

[0053] The control unit 480 includes a customer information receiving unit 481, a customer profile identification unit 482, a classification unit 483, and an analysis information presentation unit 484. The control unit 480 is a control device that controls the operation of the customer classifier 403, and is, for example, a processor such as a CPU. The control unit 480 loads the program 471 from the storage unit 470 into the memory 450 and executes it. In this way, the control unit 480 realizes the functions of the customer information receiving unit 481, the customer profile identification unit 482, the classification unit 483, and the analysis information presentation unit 484.

[0054] When requesting an estimation of the effectiveness of a policy, the service request device 600 first transmits customer information for the target business for which the policy is to be implemented to the customer classification device 403. Upon receiving the customer information, the customer information receiving unit 481 accepts input of the customer information. Here, the customer information accepted by the customer information receiving unit 481 includes information about multiple customers.

[0055] The customer profile identification unit 482 identifies second attribute information of a customer by attribute-expanding the customer information, i.e., the customer's first attribute information, and associates the second attribute information with the customer. The customer profile identification unit 482 expands the customer information, i.e., identifies the customer profile, for each of the multiple customers whose customer information has been accepted.

[0056] The classification unit 483 classifies the plurality of customer images into predetermined clusters. There are no particular limitations on the method by which the classification unit 483 classifies each customer image into a cluster. For example, the classification unit 483 may summarize the second attribute information included in each customer image into a plurality of words using an LLM, and classify customer images in which a plurality of words are similar to each other into the same cluster.

[0057] The analysis information presenting unit 484 transmits the results of the analysis of the clusters classified by the classifying unit 483 as analysis information to the service request device 600. The method of analyzing the analysis information is not particularly limited, and may be, for example, an RFM analysis using various data related to each cluster.

[0058] <Fourth Embodiment> In the second embodiment, the case where customer information is transmitted from the service request device 600 has been described. However, the customer information may be transmitted from another device, i.e., a customer information providing device. Hereinafter, with reference to FIG. 10 , an example configuration of a policy effect estimation system 700 that communicates with a customer information providing device 801 will be described. The policy effect estimation system 700 is an information processing system for estimating policy effects. As shown in FIG. 10 , the policy effect estimation system 700 includes a policy effect estimation device 400 and a customer information database 710.

[0059] The customer information database 710 stores customer identification information 711 in association with first attribute information 712 and second attribute information 713. The customer identification information 711 is information for identifying each customer and may be, for example, a unique numerical value assigned to each customer. The first attribute information 712 is at least one word extracted from the customer information and is information that may relate to the customer's consumption behavior. Specifically, the first attribute information 712 is, for example, information that can be extracted from customer consumption behavior information included in the customer information. The second attribute information 713 is character information that expands on the first attribute information 712 and is a word that indicates a specific attribute related to the customer's consumption behavior. For example, the second attribute information 713 is a psychographic attribute expanded based on the first attribute information 712.

[0060] FIG. 12 shows an example configuration of a service request device 800 and a customer information providing device 801 that can communicate with the policy effect estimation system 700. As shown in FIG. 12, the policy effect estimation system 700 is connected to the service request device 800 and the customer information providing device 801. The customer information providing device 801 is an information processing device that provides customer information to the policy effect estimation system 700, and is owned, for example, by a company that has a database of customer information for stores that plan to implement a policy. The customer information providing device 801 transmits customer information for a target business in which a policy is to be implemented to the policy effect estimation system 700. The service request device 800 is an information processing device that requests the policy effect estimation system 700 to estimate the policy effect, and is owned, for example, by a user who requests the policy effect estimation, i.e., a business operator that operates the target business. The service request device 800 receives a customer profile for the target business in which a policy is to be implemented from the policy effect estimation device 400. Then, the service request device 800 transmits policy information of a policy to be implemented to the policy effect estimation system 700 and receives interest level information related to the policy from the policy effect estimation system 700 .

[0061] The company that owns the customer information providing device 801 is not particularly limited as long as it is a company that can acquire customer information from the store where the policy is planned to be implemented. For example, it may be a company that operates a predetermined payment system or a company that operates a receipt collection business. The method by which the customer information providing device 801 acquires customer information is not particularly limited. For example, it may acquire customer information by communicating with a predetermined information processing device owned by a business operator that operates the target business. The customer information providing device 801 includes a customer information database 811 and a customer information transmission unit 821. The customer information database 811 stores customer information related to customers of the target business. The customer information transmission unit 821 is a communication means that transmits the customer information stored in the customer information database 811 to the policy effect estimation system 700.

[0062] The service request device 800 includes a customer image information receiving unit 830, a policy information transmitting unit 840, and an interest level information receiving unit 850. The customer image information receiving unit 830 is a communication means that receives customer image information from the policy effect estimation device 400. The policy information transmitting unit 840 is a communication means that transmits policy information input by a user to the policy effect estimation device 400. The interest level information receiving unit 850 is a communication means that receives interest level information from the policy effect estimation device 400.

[0063] Next, an example of the operation of each device during customer image identification will be described with reference to FIG. 13 . First, when the customer information providing device 801 transmits customer information (step S501), the customer information receiving unit 441 receives the customer information and registers the customer information in the customer information database 710. Next, the customer image identifying unit 442 identifies a customer image including second attribute information that is an extension of the customer information received in step S201 (step S502) and registers the second attribute information in the customer information database 710 (step S503). Next, the classification unit 443 classifies each customer image into a predetermined cluster based on the second attribute information identified in step S502 (step S504). Next, the output unit 446 transmits the clusters classified in step S504 to the service request device 800 as customer image information (step S505).

[0064] <Fifth Embodiment> In the fourth embodiment and the like, the user devised and generated the measures. However, the measures may be generated in a predetermined device, i.e., a measure generation device 920. The measure generation device 920 shown in FIG. 14 is an information processing device that generates measures to be implemented in stores of a target business. Hereinafter, with reference to FIG. 15 , an example configuration of a measure effect estimation system 900 that communicates with the measure generation device 920 will be described. The measure effect estimation system 900 is an information processing system for estimating the effect of a measure. The measure effect estimation system 900 supplies a customer image identified based on customer information to the measure generation device 920 and receives measure information generated by the measure generation device 920. As shown in FIG. 14 , the measure effect estimation system 900 includes a measure effect estimation device 400, a customer information database 710, and an interest level information database 910.

[0065] The interest level information database 910 stores the degree of interest in each measure linked to customer identification information 911. In the example of Fig. 15, as the degree of interest in each of two measures, a degree of interest in a first measure 912 and a degree of interest in a second measure 913 are linked to the customer identification information 911. The customer identification information 911 is information for identifying each customer, and may be, for example, a unique numerical value assigned to each customer.

[0066] Next, with reference to FIG. 16 , an example of the operation of each device during policy generation will be described. First, when the customer information providing device 801 transmits customer information (step S601), the customer information receiving unit 441 receives the customer information. Next, the customer image identification unit 442 identifies a customer image including second attribute information that is an extension of the customer information received in step S601 (step S602). Next, the customer image identification unit 442 transmits the customer image information including the second attribute information identified in step S602 to the policy generation device 920 (step S603). It is assumed that the policy generation device 920 is configured to generate a policy upon receiving the customer image information. Next, the policy generation device 920 transmits the generated policy information to the policy effect estimation device 400 (step S604). Next, the output unit 446 transmits the policy information received in step S604 to the service request device 600 (step S605).

[0067] The user checks the received policy information and determines whether or not to estimate the effects of the policy information. If the effects are to be estimated, the user inputs this into the service request device 800. The service request device 800 instructs the policy effect estimation device 400 to estimate the policy effects in response to the input. Multiple policies may be sent in step S605. The user may check multiple policies and select only those policies for which the effects are to be estimated from the multiple policies. Figure 17 shows an example of the operation of each device when estimating policy effects when only those policies for which the effects are to be estimated are selected from the multiple policies.

[0068] First, the service request device 800 transmits the selected policy information to the policy effect estimation device 400 (step S701). The user may select one or more policy information items. Next, the estimation unit 445 estimates the customer's level of interest in the policy indicated by the policy information based on the customer profile identified in step S602 and the policy information received in step S701 (step S702). If multiple policy information items are received in step S701, the estimation unit 445 estimates the customer's level of interest for each policy information item. Next, the output unit 446 transmits the interest level information generated in step S702 to the service request device 600 (step S703). The service request device 600 presents the received interest level information to the user.

[0069] In the above example, the degree of interest in each measure is estimated after the user selects the measure. However, the timing of estimating the degree of interest in each measure is not particularly limited, and may be, for example, after receiving the measure information in step S604. In this case, the estimation unit 445 estimates the degree of interest after step S604 and registers the estimation result in the interest level information database 910. Then, upon receiving the measure information selected by the user, the estimation unit 445 extracts the customer's degree of interest in the selected measure information from the interest level information database 910.

[0070] Furthermore, the policy effect estimation device 400 may estimate the degree of interest in a policy not only for customers of the target business but also for other people. For example, the policy effect estimation device 400 may estimate the degree of interest in a policy not only for customers of the store where the policy is planned to be implemented but also for customers of other stores. For example, consider a case where a user who manages four stores, Store A, Store B, Store C, and Store D, estimates the effect of a policy planned to be implemented at Store A. First, the customer information providing device 801 transmits customer information for each of Store A, Store B, Store C, and Store D to the policy effect estimation device 400. The policy effect estimation device 400 expands the attributes of the customer information for each of Store A, Store B, Store C, and Store D and registers it in the customer information database 710. The policy effect estimation device 400 estimates the degree of interest in a policy for each of customers of Store A, Store B, Store C, and Store D.

[0071] FIG. 18 shows an example of interest level information transmitted to the service request device 800 in such a case. In the example shown in FIG. 18 , the percentage of customers interested in each campaign is output in the form of a pie chart. That is, each pie chart shown in FIG. 18 is composed of the percentage of customers interested in each campaign and the percentage of customers not interested. By viewing the graph shown in FIG. 18 , the user can understand that at Store A, more customers are expected to be interested in the "Tuna Festival" campaign than in the "Singapore Fair" campaign. Furthermore, the user can understand that more customers at Store A are expected to be interested in the "Tuna Festival" campaign than at Stores B and C. In this way, by presenting the interest level of customers of the target business in a campaign and the interest level of others in a campaign, the user can confirm whether the campaign is particularly useful for the customers of the target business. While the interest level information is output in the form of a pie chart in the example shown in FIG. 18 , the output format of the interest level information is not limited to this and may be output in various graphical formats, such as a bar graph or a line graph. The interest level information may also be output using text. The interest level information may be output as a combination of graphs and text.

[0072] <Other Embodiments, etc.> When performing output in each embodiment, the display content may be changed based on information about the display to which the output is made. Examples of display information include the size of the display and the ratio of the vertical length to the horizontal length of the display. Based on the display information, the display content may be changed, for example, so that the larger the display size, the larger the size of characters, graphs, and other figures. In this case, an upper limit may be set so that the display content is not displayed larger than a predetermined size on the display. Similarly, the display content may be changed so that the smaller the display size, the smaller the size of characters, graphs, and other figures. Furthermore, a lower limit may be set so that the display content is not displayed smaller than a predetermined size on the display. In addition, the display position may be changed, or certain items may not be displayed on the same screen.

[0073] In another aspect, the display content may be changed depending on the processing power of the information processing device that performs the processing for displaying on the display. For example, when the processing power of the information processing device is low, the content to be displayed or the amount of information to be displayed may be reduced compared to when the processing power is high. Regarding the processing power, predetermined specifications such as memory size may be referenced, or the operating status of the processor or the execution status of tasks may be referenced.

[0074] <Example of Hardware Configuration> Hereinafter, with reference to FIG. 19 , a case where each functional configuration of a policy effect estimation device according to the present disclosure is realized by a combination of hardware and software will be described.

[0075] The policy effect estimation device according to the present disclosure can realize the above-described functions using a computer 11 including the hardware configuration shown in the figure. The computer 11 may be a portable computer such as a smartphone or tablet terminal, or a stationary computer such as a PC. The computer 11 may be a dedicated computer designed to realize each device, or may be a general-purpose computer. The computer 11 can realize the desired functions by installing a predetermined program.

[0076] The computer 11 has a bus 21, a processor 30, a memory 40, a storage device 50, an input / output interface 60 (an interface is also called an I / F (Interface)), and a network interface 70. The bus 21 is a data transmission path through which the processor 30, the memory 40, the storage device 50, the input / output interface 60, and the network interface 70 transmit and receive data to and from each other. However, the method of connecting the processor 30 and the like to each other is not limited to bus connection.

[0077] The processor 30 is a processor such as a CPU, a GPU, an FPGA, etc. The memory 40 is a main storage device realized using a RAM (Random Access Memory) or the like.

[0078] The storage device 50 is an auxiliary storage device realized using a hard disk, SSD, memory card, ROM (Read Only Memory), etc. The storage device 50 stores programs for realizing desired functions. The processor 30 reads these programs into the memory 40 and executes them to realize the various functional components of each device.

[0079] The input / output interface 60 is an interface for connecting the computer 11 with input / output devices. For example, the input / output interface 60 is connected to an input device such as a keyboard and an output device such as a display device.

[0080] The network interface 70 is an interface for connecting the computer 11 to a network.

[0081] Although an example of a hardware configuration for the present disclosure has been described above, the above-described embodiment is not limited to this. Any processing in the present disclosure can also be realized by causing a processor to execute a computer program.

[0082] In the above examples, the program includes instructions (or software code) that, when loaded into a computer, cause the computer to perform one or more functions described in the embodiments. The program may be stored on a non-transitory computer-readable medium or a tangible storage medium. By way of example and not limitation, computer-readable medium or tangible storage medium includes random-access memory (RAM), read-only memory (ROM), flash memory, solid-state drive (SSD) or other memory technology, CD-ROM, digital versatile disc (DVD), Blu-ray disc or other optical disk storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage device. The program may also be transmitted on a transitory computer-readable medium or communication medium. By way of example and not limitation, transitory computer-readable medium or communication medium includes electrical, optical, acoustic, or other forms of propagated signals.

[0083] Although the present disclosure has been described above with reference to the embodiments, the present disclosure is not limited to the above-described embodiments. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure. Furthermore, each embodiment can be combined with other embodiments as appropriate.

[0084] Each drawing is merely an example for describing one or more embodiments. Each drawing may not relate to only one particular embodiment, but may also relate to one or more other embodiments. As will be understood by those skilled in the art, various features or steps described with reference to any one drawing can be combined with features or steps shown in one or more other drawings to create, for example, an embodiment not explicitly shown or described. Not all features or steps shown in any one drawing are necessary to describe an exemplary embodiment, and some features or steps may be omitted. The order of steps described in any drawing may be changed as appropriate.

[0085] A part or all of the above-described embodiments can be described as, but not limited to, the following supplementary notes.

[0086] (Appendix A1) A policy effectiveness estimation device comprising: a customer profile identification means for identifying a customer profile based on predetermined information about the customer; a policy reception unit for receiving predetermined policy information; and an output means for outputting the customer's level of interest in a policy estimated based on the customer profile and the predetermined policy information.

[0087] (Supplementary Note A2) The policy effect estimation device according to Supplementary Note A1, further comprising: an estimation unit that estimates a degree of interest of the customer in the policy based on the customer profile and the predetermined policy information.

[0088] (Appendix A3) The policy effect estimation device described in Appendix A2, wherein the policy reception means receives the policy information as text information, and the estimation means estimates the interest level by expanding words extracted from the policy information to identify second attribute information.

[0089] (Appendix A4) The policy effect estimation device according to Appendix A2 or A3, wherein the estimation means estimates the interest level for each customer by calculating a degree of association between second attribute information included in the customer image and second attribute information obtained by extending the policy information.

[0090] (Appendix A5) The policy effect estimation device according to any one of Appendices A2 to A4, further comprising a classification means for classifying the plurality of customer images into predetermined clusters, wherein the estimation means generates statistical information by performing statistical processing for each of the clusters after estimating the interest level for each of the customers.

[0091] (Supplementary Note A6) The policy effect estimation device according to any one of Supplementary Notes A1 to A5, wherein the policy reception means receives the policy information by presenting a query display for prompting a user to input the policy information.

[0092] (Supplementary Note A7) The policy effect estimation device according to any one of Supplementary Notes A1 to A6, wherein the policy reception means receives the policy information generated by a predetermined policy generation device by supplying the customer image to the policy generation device.

[0093] (Appendix A8) The policy effect estimation device described in any of Appendices A1 to A7, wherein the policy reception means receives the policy information including information regarding the timing of implementing the policy, and the output means outputs the interest level estimated taking into account the timing of implementation.

[0094] (Supplementary Note A9) The policy effect estimation device according to any one of Supplementary Notes A1 to A8, wherein the output means presents the degree of interest in the policy of the customer and the degree of interest in the policy of other people than the customer in a comparable manner.

[0095] (Appendix A10) The policy effect estimation device according to any one of Appendices A1 to A9, wherein the customer image identification means identifies the customer image in a manner that allows it to be presented to the user, and the output means outputs the customer image that has been identified in a manner that allows it to be presented together with the interest level.

[0096] (Appendix B1) A method for estimating the effectiveness of a policy, in which a computer identifies a customer profile based on specified information about the customer, accepts specified policy information, and outputs the customer's level of interest in a policy estimated based on the customer profile and the specified policy information.

[0097] (Appendix C1) A program for estimating the effectiveness of a policy, which causes a computer to execute the following processes: a process for identifying a customer profile based on predetermined information about the customer; a process for accepting predetermined policy information; and a process for outputting the customer's level of interest in a policy estimated based on the customer profile and the predetermined policy information.

[0098] (Appendix D1) A policy effectiveness estimation device comprising: a classification means for classifying a plurality of customer profiles identified based on predetermined information about a plurality of customers into predetermined clusters; a policy reception unit for receiving predetermined policy information; and an analysis information presentation means for outputting statistical information generated by performing statistical processing for each cluster regarding the customer's level of interest in a policy estimated based on the customer profile and the predetermined policy information.

[0099] (Appendix E1) A policy effectiveness estimation device comprising: a customer information receiving means for receiving first attribute information extracted from predetermined information about a customer; a customer image identification means for identifying a customer image including second attribute information expanded based on the first attribute information; a policy receiving unit for receiving predetermined policy information; and an output means for outputting the customer's degree of interest in a policy estimated based on the customer image and the predetermined policy information.

[0100] Some or all of the elements (e.g., configurations and functions) described in Appendixes A2 to A10 that are dependent on Appendix A1 may also be dependent on Appendix B1, Appendix C1, and Appendix E1 in the same dependency relationship as Appendixes A2 to A10. Some or all of the elements described in any appendix may be applied to various hardware, software, recording means for recording software, systems, and methods.

[0101] This application claims priority based on Japanese Patent Application No. 2024-32786, filed March 5, 2024, the disclosure of which is incorporated herein by reference in its entirety.

[0102] REFERENCE SIGNS LIST 100 Policy effect estimation device 110 Customer information reception unit 120 Customer image identification unit 130 Policy reception unit 140 Output unit 400 Policy effect estimation device 403 Customer classification device 410, 450 Memory 420, 460 Communication unit 430, 470 Storage unit 431, 471 Program 440, 480 Control unit 441, 481 Customer information reception unit 442, 482 Customer image identification unit 443, 483 Classification unit 444 Policy reception unit 445 Estimation unit 446 Output unit 484 Analysis information presentation unit 600, 800 Service request device 801 Customer information provision device 610, 811 Customer information database 620, 821 Customer information transmission unit 630, 830 Customer image information reception unit 640, 840 Measure information transmitting unit 650, 850 Interest level information receiving unit 700, 900 Measure effect estimation system 710 Customer information database 711 Customer identification information 712 First attribute information 713 Second attribute information 910 Interest level information database 911 Customer identification information 912 Interest level in first measure 913 Interest level in second measure 920 Measure generation device 11 Computer 21 Bus 30 Processor 40 Memory 50 Storage device 60 Input / output interface 70 Network interface

Claims

1. A policy effectiveness estimation device comprising: a customer profile identification means for identifying a customer profile based on predetermined information about the customer; a policy reception means for receiving predetermined policy information; and an output means for outputting the customer's level of interest in a policy estimated based on the customer profile and the predetermined policy information.

2. The policy effect estimation device according to claim 1, further comprising an estimation means for estimating the customer's level of interest in the policy based on the customer profile and the predetermined policy information.

3. The policy effectiveness estimation device described in claim 2, wherein the policy reception means receives the policy information as text information, and the estimation means estimates the interest level by expanding words extracted from the policy information to identify second attribute information.

4. A policy effect estimation device as described in claim 2 or 3, wherein the estimation means estimates the interest level for each customer by calculating the degree of association between second attribute information included in the customer profile and second attribute information that is an extension of the policy information.

5. A policy effectiveness estimation device as described in claim 2 or 3, further comprising a classification means for classifying the plurality of customer profiles into predetermined clusters, wherein the estimation means generates statistical information by statistical processing for each cluster after estimating the interest level for each customer.

6. A policy effect estimation device according to any one of claims 1 to 3, wherein the policy reception means receives the policy information by presenting a query display for inputting the policy information to a user.

7. A policy effect estimation device according to any one of claims 1 to 3, wherein the policy reception means receives the policy information generated by a predetermined policy generation device by supplying the customer image to the policy generation device.

8. A policy effectiveness estimation device according to any one of claims 1 to 3, wherein the policy reception means receives the policy information including information regarding the timing of the implementation of the policy, and the output means outputs the degree of interest estimated taking into account the timing of the implementation.

9. A policy effectiveness estimation device according to any one of claims 1 to 3, wherein the output means presents the degree of interest in the policy of the customer and the degree of interest in the policy of people other than the customer in a manner that allows them to compare them.

10. A policy effectiveness estimation device according to any one of claims 1 to 3, wherein the customer image identification means identifies the customer image in a manner that allows it to be presented to a user, and the output means outputs the customer image identified in a manner that allows it to be presented together with the interest level.

11. A method for estimating the effectiveness of a policy, in which a computer identifies a customer profile based on specified information about the customer, accepts specified policy information, and outputs the customer's level of interest in a policy estimated based on the customer profile and the specified policy information.

12. A program for estimating the effectiveness of a policy that causes a computer to perform the following processes: identifying a customer profile based on specified information about the customer; receiving specified policy information; and outputting the customer's level of interest in a policy estimated based on the customer profile and the specified policy information.

Images