Purchase data analysis apparatus, method, and program

Abstract

To provide a purchase data analysis apparatus, method and program for properly clustering stores having similar characteristics.SOLUTION: A purchase data analysis apparatus includes a customer information acquisition unit, a customer feature quantity generation unit, a store feature quantity generation unit, and a store clustering unit. The customer information acquisition unit acquires customer information including an action time related to a purchase, for each customer. The customer feature quantity generation unit generates customer feature quantities representing action patterns of customers, for each customer, on the basis of the action time. The store feature quantity generation unit generates store feature quantities representing features of customers who visit the stores, for each store, on the basis of the customer feature quantities. The store clustering unit clusters the stores using the store feature quantities.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] Embodiments of the present invention relate to a purchase data analysis apparatus, method, and program.

Background Art

[0002] A purchase data analysis apparatus is used for analyzing purchase history data with customer IDs added, such as ID-POS data. These days, it has become possible to obtain purchase history data spanning multiple stores or multiple companies through settlement systems, point systems, and the like. Therefore, using a clustering method that classifies multiple data into multiple sets with similar characteristics, the stores are clustered (classified) into multiple sets with similar characteristics (hereinafter referred to as store clusters), and it is required to analyze the stores using the clustering results.

[0003] For store clustering, information such as sales time, sales amount, and product sales is used. However, since information such as product sales depends on the product lineup of the store, out-of-stock items in the store, and the sales method of the store, the store clustering result reflects the differences in the product lineup, out-of-stock items, and sales method of the store, and it is difficult to appropriately reflect the characteristics of the store. For example, when the product lineups between stores are different and there are few common products, it is difficult to obtain an appropriate clustering result in store clustering using product sales. Thus, in store clustering using only sales time and sales amount, the amount of information for performing store clustering is small, and it is difficult to appropriately cluster the stores.

[0004] Furthermore, store analysis using store clustering results employs statistical and machine learning techniques to predict product sales. For example, sales forecasts are made using past sales history of products. When forecasting sales of a product not currently carried at a particular store, past sales history from other stores belonging to the same store cluster is used. However, if stores with similar characteristics are not properly classified, and sales history from stores with different characteristics is used, accurate sales forecasts cannot be expected. Therefore, it is important to obtain store clusters in which stores or companies with similar characteristics are properly classified. The store analysis results can also be used to analyze companies that have one or more stores. [Prior art documents] [Patent Documents]

[0005] [Patent Document 1] Japanese Patent Publication No. 2003-44653 [Overview of the project] [Problems that the invention aims to solve]

[0006] The problem that this invention aims to solve is to provide a purchase data analysis device, method, and program that can appropriately cluster stores with similar characteristics. [Means for solving the problem]

[0007] To solve these problems, the purchase data analysis device of this embodiment comprises a customer information acquisition unit, a customer feature generation unit, a store feature generation unit, and a store clustering unit. The customer information acquisition unit acquires customer information for each customer, including the time of actions related to purchases. The customer feature generation unit generates customer features representing the customer's behavior pattern for each customer based on the time of actions. The store feature generation unit generates store features representing the characteristics of visiting customers for each store based on the customer features. The store clustering unit clusters stores using the store features. [Brief explanation of the drawing]

[0008] [Figure 1] Figure 1 shows an example of the configuration of a purchasing data analysis device according to an embodiment. [Figure 2] Figure 2 is a flowchart illustrating the processing procedure of the analysis process performed by the purchasing data analysis device according to the embodiment. [Figure 3] Figure 3 shows an example of customer features. [Figure 4] Figure 4 shows an example of customer features. [Figure 5] Figure 5 shows an example of customer features. [Figure 6] Figure 6 shows an example of clustering results for customers. [Figure 7] Figure 7 shows an example of store features. [Figure 8] Figure 8 shows an example of clustering results for stores. [Figure 9] Figure 9 is a schematic diagram illustrating the processing flow in the analysis process. [Figure 10] Figure 10 shows an example of the display screen that appears when a store cluster is specified. [Figure 11] Figure 11 shows an example of the display screen that appears when a store is specified. [Figure 12] Figure 12 shows an example of a display screen that appears when a product is selected. [Figure 13] Figure 13 shows an example of a display screen that accepts input for the name and information of a store cluster. [Figure 14] Figure 14 shows an example of a display screen that accepts input for the name and information of a store cluster label. [Figure 15] Figure 15 shows an example of data stored in the store cluster label storage unit. [Figure 16] Figure 16 shows an example of data stored in the store cluster label storage unit. [Figure 17]FIG. 17 is a diagram showing an example of data stored in the store cluster label storage unit. [Figure 18] FIG. 18 is a diagram for specifically explaining the effect of the purchase data analysis apparatus according to the embodiment. [Figure 19] FIG. 19 is a schematic diagram schematically showing the flow of processing in the analysis processing according to the modification. [Figure 20] FIG. 20 is a block diagram illustrating the hardware configuration of the purchase data analysis apparatus according to the application example.

Embodiment for Carrying Out the Invention

[0009] Hereinafter, embodiments of a purchase data analysis apparatus, method, and program will be described in detail with reference to the drawings. In the following description, components having substantially the same functions and configurations are denoted by the same reference numerals, and duplicate description will be made only when necessary.

[0010] (Embodiment) FIG. 1 is a diagram showing the configuration of a purchase data analysis apparatus 100 according to the embodiment. The purchase data analysis apparatus 100 is an apparatus used for analyzing purchase data. The purchase data analysis apparatus 100 is connected to a purchase-related database 200 that stores purchase-related data and a customer master 300 that stores customer information via a network or the like. The network is, for example, a LAN (Local Area Network). Note that the connection to the network may be a wired connection or a wireless connection. Also, the network is not limited to a LAN, and may be the Internet, a public communication line, or the like.

[0011] The purchase-related database 200 stores purchase-related data. The purchase-related data is customer information including the time of action of purchase-related actions related to purchases. The purchase-related data includes purchase data with a customer ID (hereinafter referred to as customer ID) added. The purchase data with a customer ID added is, for example, ID-POS data. In the purchase-related data, each customer included in the purchase data can be distinguished using the ID. The purchase data includes time information of the purchase-related actions. The time information is the time of action of the purchase-related actions. The purchase-related actions are actions related to purchases. The purchase-related actions are, for example, actions such as settlement, visiting a store, and picking up a product. The approval time can be obtained from, for example, the information on receipts and point cards. The time of visiting a store can be obtained from, for example, in-store cameras installed in the store. The time when a product is picked up can be obtained from, for example, the ID-POS data of self-checkout stores. In addition, when using image information such as in-store cameras, a person can be identified and customers can be distinguished by using face recognition technology for images.

[0012] The customer master 300 stores customer master data. The customer master data is information about customers. The customer master data includes, for example, the age, generation, gender, address, favorite food, favorite topic, etc. for each customer. The customer master data is, for example, information obtained from membership registration information, input information of questionnaires, and various analysis results.

[0013] The purchase data analyzer 100 includes a customer information acquisition unit 101, a customer feature quantity generation unit 102, a customer clustering unit 103, a customer cluster data storage unit 104, a store feature quantity generation unit 105, a store clustering unit 106, a store cluster data storage unit 107, a store cluster aggregation unit 108, a store cluster display unit 109, a store cluster management unit 110, and a store cluster label storage unit 111.

[0014] The customer information acquisition unit 101 acquires purchase-related data for each customer. Purchase-related data is acquired, for example, from the purchase-related database 200. In this case, the customer information acquisition unit 101 may acquire only the information used for generating customer features described later from the information stored in the purchase-related database 200, or it may acquire all the data. Furthermore, if customer master data is required for generating customer features, the customer information acquisition unit 101 acquires the necessary customer master data. Customer master data is acquired, for example, from the customer master 300.

[0015] The customer information acquisition unit 101 may acquire only data that matches the specified conditions. For example, it may acquire only purchase-related data acquired during a specified acquisition period, or it may acquire only purchase-related data that matches the specified type of store. In addition, it may acquire only purchase-related data of customers in the age group specified by the user.

[0016] The customer feature generation unit 102 generates customer features for each customer that represent the customer's behavior patterns and habits, based on the time of the actions included in the acquired purchase-related data. In this process, the customer feature generation unit 102 generates customer features for each customer who has used a particular store. In other words, the customer feature generation unit 102 uses the time information of purchase-related actions acquired by the customer information acquisition unit 101 to feature the customer. The customer's behavior patterns are, for example, the customer's habits regarding purchase-related actions. As customer features, periodic explanatory variables and the elapsed time from entering the store to performing a purchase-related action are used.

[0017] The customer clustering unit 103 clusters each customer based on customer features. In this process, the customer clustering unit 103 performs clustering using customer features for each customer included in the purchase-related data of that store, for each store, and classifies each customer into one of several customer clusters. Existing clustering methods that classify multiple data into multiple sets with similar characteristics can be used for customer clustering. For example, the k-means method or hierarchical clustering methods can be used, but the unit is not limited to these methods. The customer clustering unit 103 stores the clustering results for each customer in the customer cluster data storage unit 104.

[0018] The customer cluster data storage unit 104 stores the clustering results for each customer. The clustering results are stored in a table format consisting of, for example, a combination of columns such as customer ID, store ID, and customer cluster ID.

[0019] The store feature generation unit 105 generates store features for each store that represent trends in the behavioral patterns and habits of customers who visit the store, based on customer features. For example, the store feature generation unit 105 uses the clustering results of each customer using customer features to calculate the number of customers belonging to each customer cluster for each store, and generates the calculation results as store features.

[0020] The store clustering unit 106 clusters each store using store features. In this process, the store clustering unit 106 performs clustering on each store based on the store features and classifies each store into one of several store clusters. Existing clustering methods that classify multiple data into multiple sets with similar characteristics can be used for store clustering. For example, the k-means method or hierarchical clustering methods can be used, but the unit is not limited to these methods. The store clustering unit 106 stores the clustering results for each store in the store cluster data storage unit 107.

[0021] The store cluster data storage unit 107 holds the clustering results for each store. The clustering results are stored in a table format consisting of, for example, a combination of store ID and store cluster ID columns.

[0022] The store cluster aggregation unit 108 aggregates purchase-related data for each store cluster based on the clustering results of the stores. For example, the store cluster aggregation unit 108 performs various aggregations required by the store cluster display unit 109 based on the store clusters stored in the store cluster data storage unit 107 and the user's specifications. Alternatively, it may perform aggregation of all items without receiving a user's specifications.

[0023] For example, if a user specifies a store cluster, the system will perform aggregation processes such as obtaining a list of stores belonging to the specified store cluster, aggregating the age-specific ratios of demographic attributes such as age and gender of customers who visited the stores belonging to the specified store cluster, calculating the average value of customer features for customers in the stores belonging to the specified store cluster, and aggregating the quantity and amount of product sales at the stores belonging to the specified store cluster. In addition, as part of the aggregation process, sales forecasts for all products and for each product may be made for the specified store cluster based on the sales performance of the specified store cluster.

[0024] Furthermore, if a user specifies a store, aggregation processing such as summarizing the sales performance of the specified store will be performed. For example, sales forecasts for all products and for each product will be made for the specified store based on the sales performance of stores belonging to the store cluster to which the specified store belongs.

[0025] Furthermore, when a user specifies a product, aggregation processing is performed, such as aggregating the sales performance of the specified product by store cluster and aggregating the sales performance of the specified product by store. In addition, as part of the aggregation processing, sales forecasts for the specified product may be performed for each store cluster. If the number of stores to be aggregated is large, the aggregation processing may be limited to stores with high sales performance of the specified product.

[0026] In addition, customer master data may be used to perform aggregation for each store cluster in addition to purchase-related data. In this case, the store cluster aggregation unit 108 performs aggregation for each store cluster using purchase-related data and customer master data.

[0027] Furthermore, the store cluster aggregation unit 108 may aggregate purchase-related data for each store cluster, as well as for each store cluster label. A store cluster label is a cluster group composed of one or more store clusters, and can be freely set by the user. Typically, a store cluster label consists of multiple store clusters. Aggregation at the store cluster label level may be performed using the same method as the aggregation for each store cluster described above, or it may be performed using a different method.

[0028] The store cluster display unit 109 displays the aggregated results of purchase-related data. For example, the store cluster display unit 109 displays the aggregated results obtained by the store cluster aggregation unit on the display according to the user's operation. In addition to the aggregated results, the store cluster name, store cluster information, store cluster label name, store cluster label information, etc., may also be displayed on the display.

[0029] The store cluster management unit 110 manages store clusters and store cluster labels. For example, the store cluster management unit 110 accepts input such as the name of the store cluster, store cluster information, the name of the store cluster label, and store cluster label information, and stores the accepted name and information in the store cluster label storage unit 111. The data stored in the store cluster label storage unit 111 is read and used by the store cluster aggregation unit 108 and the store cluster management unit 110.

[0030] The store cluster label storage unit 111 holds the names and information entered in the store cluster management unit 110. For example, if the store cluster management unit 110 receives input for the name of a store cluster or the name of a store cluster label, the store cluster label storage unit 111 updates the stored name to the entered name.

[0031] Next, the operation of the clustering process performed by the purchase data analysis device 100 will be described. Figure 2 is a flowchart showing an example of the steps of the clustering process performed by the purchase data analysis device 100. The clustering process is a process that uses purchase-related data for each store to classify multiple stores into multiple sets with similar characteristics. Note that the processing steps in each process described below are merely examples, and each process can be modified as appropriate as possible. In addition, steps in the processing steps described below can be omitted, replaced, and added as appropriate, depending on the embodiment.

[0032] (Clustering process) (Step S201) First, the customer information acquisition unit 101 acquires purchase-related data from the purchase-related database 200. The purchase-related data includes purchase data from multiple stores. The customer information acquisition unit 101 then identifies the customers included in the purchase-related data based on the customer IDs included in the acquired purchase-related data, and acquires customer master data related to the identified customers from the customer master 300.

[0033] (Step S202) Next, the customer feature generation unit 102 generates customer features for each store based on the acquired purchase-related data. Figures 3-5 are diagrams showing examples of customer features. Figures 3-5 show the customer features of a specific customer who visited a specific store. Also, in Figures 3-5, customer features are generated using purchase-related data over one month. Figures 3 and 4 show an example of using periodic explanatory variables as customer features. The customer features shown in Figures 3 and 4 are also called customer visit patterns. Figure 5 shows an example of using the elapsed time from entering the store to purchase-related behavior as customer features.

[0034] In Figure 3, the payment time over a one-month period is used as the purchase-related time, and the number of payments per day of the week and per one-hour time slot is used as customer characteristics. The payment time is obtained, for example, from information on receipts or loyalty cards.

[0035] In Figure 4, the entry and exit times for one month are used as purchase-related times, and the number of entries and exits for each day of the week and for each one-hour time slot are used as customer features. Entry and exit times are obtained, for example, using data acquired from surveillance cameras that capture customers entering and leaving the store.

[0036] Note that while the examples in Figures 3 and 4 use the number of times each purchasing action occurred over a month, the average number of times per day may also be used. For example, the average number of times per day can be calculated by dividing the number of times in Figures 3 and 4 by the number of days in that month and used.

[0037] In Figure 5, the time of entry into the store and the time the product was picked up are used as purchase-related times to calculate the elapsed time from entering the store to picking up the product, and the number of times each 10-minute interval elapsed is used as a customer feature. The time the product was picked up is obtained, for example, from ID-POS data at stores with automated payment systems.

[0038] Furthermore, the customer feature generation unit 102 may generate customer features based on purchase-related data and customer master data. For example, customer master data such as the customer's age, gender, and preferences may be obtained, and these data may be added to the numbers in Figures 3-5 to obtain the customer features.

[0039] (Step S203) Next, the customer clustering unit 103 performs clustering for each customer using customer features and obtains a customer cluster ID for each customer as a clustering result. The resulting clustering result is in a table format consisting of a combination of customer ID, store ID, and customer cluster ID columns. The clustering result is stored in the customer cluster data storage unit 104. Figure 6 shows an example of the clustering result for each customer.

[0040] (Step S204) Next, the store feature generation unit 105 generates store features for each store using the clustering results for each customer. At this time, the store feature generation unit 105 generates the distribution of customers belonging to each customer cluster ID as store visit features.

[0041] Figure 7 shows an example of store features generated using customer clustering results. Figure 7 shows the store features of a specific store. In the example in Figure 7, the number of customers for each customer cluster ID is used as a store feature. The vertical axis of Figure 7 shows the customer cluster IDs, and the horizontal axis shows the number of customers belonging to each customer cluster ID. Alternatively, instead of the number of customers belonging to each customer cluster ID, the proportion of each customer cluster to the total number of customers in that store may be used.

[0042] (Step S205) Next, the store clustering unit 106 performs store clustering for each store using the store visit features and obtains the store cluster ID for each store as a result of the clustering. The clustering results are stored in the store cluster data storage unit 107.

[0043] Figure 8 shows an example of clustering results for each store. In the example in Figure 8, the clustering results for the stores are stored in a table format consisting of a combination of store ID and store cluster ID columns.

[0044] Figure 9 is a schematic diagram illustrating the process from step S201 to step S205. As shown in Figure 9, in the process from step S201 to step S205, the purchase data analysis device 100 acquires customer characteristics for each customer visiting each store, performs a first clustering on each customer using the customer characteristics, and classifies each customer into a customer cluster. Subsequently, the customer clusters obtained from the first clustering result are counted by store, and the count result is used as store characteristics. Then, a second clustering is performed using the store characteristics, and each store is classified into a store cluster.

[0045] (Step S206) Next, the store cluster aggregation unit 108 performs various aggregations on customer-related data and customer master data, either by store cluster or by store cluster label. The store cluster aggregation unit 108 obtains the data necessary for aggregation from the purchase-related database 200 and the customer master 300, and performs aggregation using the obtained data. When performing aggregation, aggregation may be performed on all pre-configured items, or only on items specified by the user.

[0046] (Step S207) Next, the store cluster display unit 109 accepts the designation of a store, store cluster, store cluster label, or product. For example, a store is designated when the user enters the store name or store ID, a store cluster is designated when the user enters the store cluster name or store cluster ID, a product is designated when the user enters the product name or product ID, and a store cluster is designated when the user enters the store cluster label name or store cluster label ID.

[0047] (Step S208) Next, the store cluster display unit 109 displays the aggregated results for the specified store, store cluster, or product on the display.

[0048] If a specific store cluster is specified in step S207, the aggregated results for the specified store cluster will be displayed. Figure 10 shows an example of the display screen when a store cluster named "Cluster A" is specified. In the example in Figure 10, five items are displayed as aggregated results: "List of relevant stores," "Demographic ratio," "Representative customer visit pattern," "Sales ranking," and "Sales performance and sales forecast by cluster." If a specific store cluster label is specified in step S207, the same items as in Figure 10 should be displayed.

[0049] The "List of Relevant Stores" displays the names of the stores belonging to the "Cluster A" store cluster, along with the names of the companies that operate those stores.

[0050] The "Demographic Ratio" section displays the distribution of demographic attributes of customers who visited stores belonging to "Cluster A." Here, customer gender and age are used as demographic attribute information, and the ratio of visits by age group is displayed separately for men and women.

[0051] The "Representative Customer Visit Pattern" section displays typical customer visit patterns for customers who visited stores belonging to "Cluster A." For example, the "Representative Customer Visit Pattern" section aggregates the number of customers who visited stores belonging to "Cluster A" for each customer cluster, and displays the typical visit pattern for the customer cluster with the largest number of customers.

[0052] The "Sales Ranking" displays the total sales quantity and total sales amount for each product at stores belonging to "Cluster A," sorted by sales amount from highest to lowest.

[0053] The "Sales Performance and Sales Forecast by Cluster" section displays graphs showing the time-series changes for "Overall Sales Performance," "Sales Performance within a Cluster," "Overall Sales Forecast," and "Sales Forecast within a Cluster." The horizontal axis for each item represents the date or month, and the vertical axis represents the amount. "Overall Sales Performance" is the average of past sales amounts for each store, encompassing all stores. "Sales Performance within a Cluster" is the average of past sales amounts for each store, encompassing stores belonging to "Cluster A." "Overall Sales Forecast" is the average of future sales forecast amounts for each store, encompassing all stores. "Sales Forecast within a Cluster" is the average of future sales forecast amounts for each store, encompassing stores belonging to "Cluster A."

[0054] Furthermore, if a specific store is specified in step S207, the aggregated results for the specified store will be displayed. Figure 11 shows an example of the display screen that appears when a store named "○○ Store" is specified. In the example in Figure 11, the display screen shows the "Store Name," the "Company Name of the Company Operating ○○ Store," the "Affiliated Cluster," and the "Aggregated Results for ○○ Store."

[0055] The "Affiliated Cluster" column displays the name of the store cluster to which "Store XX" belongs. The summary results display two items: a "Table of Actual and Forecasted Product Sales" and a "Graph of Actual and Forecasted Product Sales." The "Table of Actual and Forecasted Product Sales" displays the actual quantity, actual amount, quantity forecast, and amount forecast for each product. Here, not only products handled by "Store XX" but also products handled by other stores in the store cluster to which "Store XX" belongs are displayed. The "Graph of Actual and Forecasted Product Sales" displays the time-series changes of the total past sales amount and the forecast amount of future sales for "Store XX." In addition, the forecast amount is displayed as both the forecast amount based on the sales performance of all stores and the forecast amount based on the sales performance of stores belonging to the store cluster "Cluster B," to which "Store XX" belongs.

[0056] Furthermore, if a specific product is specified in step S207, the aggregated results for the specified product will be displayed. Figure 12 shows an example of the display screen that appears when a product with the product name "〇△□× Cookies (12 pieces)" is specified. In the example in Figure 12, four items are displayed as aggregated results: "Table of actual and projected product sales by cluster," "Graph of actual and projected product sales by cluster," "Table of actual and projected product sales by store," and "Graph of actual and projected product sales by store."

[0057] The "Table of Actual and Forecasted Product Sales by Cluster" displays the actual quantity, actual value, quantity forecast, and value forecast for "〇△□× Cookies (12 pieces)" for each store cluster. The "Graph of Actual and Forecasted Product Sales by Cluster" displays the time-series changes of the total past sales and the forecasted future sales amount for "〇△□× Cookies (12 pieces)". In addition, the forecast amount is displayed as both the forecast amount based on the sales performance of all stores and the forecast amount based on the sales performance of stores belonging to a specific store cluster.

[0058] The "Table of Actual and Forecasted Product Sales by Store" displays the actual quantity, actual value, quantity forecast, and value forecast for "〇△□× Cookies (12 pieces)" for each store. The "Graph of Actual and Forecasted Product Sales by Cluster" displays the time-series changes of the total past sales amount and the forecast amount for future sales of "〇△□× Cookies (12 pieces)". In addition, the forecast amount is displayed as both the forecast amount based on the sales performance of all stores and the forecast amount based on the sales performance of a specific store.

[0059] Thus, in the process of step S208, aggregate items corresponding to the store, store cluster, store cluster label, or product specified in step S207 are displayed. The aggregate items displayed at this time are not limited to the examples in Figures 10-12. For example, for ease of viewing, only some of the aggregate items displayed in Figures 10-12 may be displayed, different aggregate items from those displayed in Figures 10-12 may be displayed, or aggregate items according to user specifications may be displayed. Furthermore, the information displayed on the graph is not limited to the examples in Figures 10-12. For example, for ease of viewing, only some of the information displayed on the graph in Figures 10-12 may be displayed, different information from that displayed in Figures 10-12 may be displayed, or information according to user specifications may be displayed. For example, in the "Graph of Actual and Forecasted Product Sales," the change in sales quantity may be displayed instead of the change in sales amount.

[0060] (Step S209) If a specific store cluster or store cluster label is specified, the user can enter the name and information of the specified store cluster or store cluster label. In this case, a new store cluster label can also be created by entering the name and information of the store cluster label. The store cluster management unit 110 determines whether or not the name and information of the store cluster or store cluster label has been entered. If the name and information of the store cluster or store cluster label has been entered, the process proceeds to step S210; if the name and information of the store cluster or store cluster label has not been entered, the process proceeds to step S211.

[0061] Figure 13 shows an example of a display screen that accepts input for the name and information of a store cluster. In the example in Figure 13, the user can enter a new name for the specified store cluster in the "New Cluster Name" input field. In addition, in the example in Figure 13, the user can enter information about the specified store cluster in free text format in the "Information" input field. Alternatively, a set of typical input examples may be provided in advance, allowing the user to input information about the store cluster by selecting one of the examples.

[0062] Figure 14 shows an example of a display screen that accepts input for the name and information of a new store cluster label. In the example in Figure 14, the user can freely set the store clusters belonging to the store cluster label by entering the name of the new store cluster label in the "New Label Name" input field and the ID or name of the store cluster in the "Cluster Selection 1" - "Cluster Selection 3" input fields. There can be any number of store clusters that can be entered. The user can also freely enter information about the store cluster label in the "Information" input field. Alternatively, a selection of typical input examples may be provided in advance, allowing the user to enter information about the store cluster label by selecting one of the examples.

[0063] (Step S210) The store cluster management unit 110 stores the information entered in step S209 in the store cluster label storage unit 111. For example, when it receives input of a new name for a store cluster or store cluster label, the store cluster management unit 110 changes the current name to the newly entered name and stores the changed name in the store cluster label storage unit 111.

[0064] Figure 15 shows an example of a store cluster stored in the store cluster label storage unit 111. In the example in Figure 15, the store cluster is stored in a table format with columns for the store cluster ID, the store cluster name, and information about the store cluster.

[0065] Figures 16 and 17 show examples of store cluster labels stored in the store cluster label storage unit 111. In the example in Figure 16, the columns for the store cluster label ID (label ID) and information about the store cluster label are stored in a table format. In the example in Figure 17, the columns for the store cluster label ID (label ID) and the ID of the store cluster belonging to the store cluster label (cluster ID) are stored in a table format.

[0066] (Step S211) Users can change the designation of stores, store clusters, store cluster labels, or products. The store cluster display unit 109 accepts new designations from the user for stores, store clusters, store cluster labels, or products.

[0067] (Step S212) If a new specification is accepted, the process returns to step S208, and the store cluster display unit 109 changes the aggregated results displayed on the display according to the change in specification.

[0068] If no new specifications are entered, the purchase data analysis device 100 will terminate the clustering process. Alternatively, the clustering process may be terminated if the user requests that the clustering process be terminated.

[0069] (Effects of the embodiment) The effects of the purchase data analysis device 100 according to this embodiment will be described below.

[0070] Analyzing purchase history data requires classifying multiple stores included in the data into multiple store clusters composed of stores with similar characteristics. Clustering methods using information such as sales time, sales amount, and product sales tend to have results that depend on factors such as product assortment, stockouts, and sales methods. As a result, stores with similar product assortments and sales methods may be classified into the same store cluster, failing to reflect customer characteristics. In such cases, the accuracy of analyses such as sales forecasting using store clusters decreases.

[0071] To address these challenges, the purchase data analysis device 100 according to this embodiment can acquire customer information for each customer, including the time of actions related to purchases; generate customer features representing the customer's behavior pattern based on the time of actions; generate store features representing the characteristics of customers who visit the store based on the customer features; and cluster stores using the store features. Customer information includes, for example, purchase history data for each customer. Customer information is, for example, ID-POS data, and is obtained, for example, from a purchase-related database 200. Customer features represent, for example, the customer's lifestyle habits and behavior patterns regarding purchases. Customer features include, for example, periodic explanatory variables or the elapsed time from entering the store to the purchase-related action. Discretized values ​​of the above explanatory variables or elapsed time may also be used as customer features. Periodic explanatory variables are, for example, customer visit patterns as shown in Figures 3 and 4. Furthermore, customer master data may be acquired from a customer master 300, and customer features may be generated based on customer information and customer master data.

[0072] With the above configuration, the purchase data analysis device 100 according to this embodiment can appropriately cluster stores with similar characteristics. For example, customer visit patterns representing the habitual nature of purchase timing can be generated for each customer from purchase history data, a set of customer visit patterns for each customer can be obtained for each store, and store visit features can be generated by aggregating these sets. Since the store visit features generated in this way reflect the customer's behavior patterns and habits, they reflect the characteristics of the customers who use that store. Customer characteristics are thought to change depending on the store's location and how the store is used. Therefore, the store visit features reflect the customer characteristics that result from the store's location and how the store is used. For this reason, by clustering each store using the store features generated using customer visit patterns, it is possible to obtain store clusters that reflect the characteristics of customer behavior patterns and habits. Furthermore, even if stores have different product assortments, stock shortages, and sales methods, stores with similar customer behavior patterns and habits can be classified into the same cluster, so it is possible to obtain store clusters that reduce or eliminate the influence of store product assortments, stock shortages, and sales methods. This makes it possible to obtain store clusters that consider not only the store's address information but also the surrounding environment of the store.

[0073] Furthermore, the purchase data analysis device 100 according to this embodiment can cluster each customer based on customer characteristics and cluster stores based on the customer clustering results. By using the customer clustering results, it is possible to generate store characteristics that better reflect the characteristics of visiting customers.

[0074] Furthermore, the purchase data analysis device 100 according to this embodiment can aggregate customer information for each store cluster based on the clustering results of the stores and display the aggregated customer information results. For example, it can make sales forecasts for a specified store cluster or a specified store based on the sales performance of a specified store cluster or the store cluster to which the specified store belongs. This makes it possible to make sales forecasts for products that have not been handled before by using the sales performance of other stores belonging to the same store cluster. Because it is possible to use the sales performance of stores with similar characteristics, it is possible to make highly accurate sales forecasts.

[0075] Figure 18 is a diagram illustrating the effects of the purchase data analysis device 100 according to this embodiment. Figure 18 shows maps of the surrounding areas of stores A, B, and C. Stores A, B, and C are assumed to be stores of the same company. Stations A and B are adjacent stations. Stores A and B are adjacent to the same station A and have similar addresses and coordinates. However, because stores A and B are located on opposite sides of station A, many of the customers who visit store A are business district users, while many of the customers who visit store B are shopping district users, resulting in different customer characteristics. Consequently, the product assortment and sales trends differ between stores A and B. Therefore, even if sales forecasts for products not currently handled by store A are made using sales data from store B, which is located nearby, the accuracy of the sales forecast will be low.

[0076] Furthermore, the characteristics of customers visiting a store located within a train station differ from those visiting a store located within a station building, which can lead to differences in sales trends. Similarly, when adjacent stations have different scales and surrounding environments, such as between store A and store C, even stores located close together in terms of address and coordinates may have different sales trends due to differences in customer characteristics. In such cases, using sales data from nearby stores to forecast sales for products not currently carried will result in low accuracy.

[0077] In contrast, this embodiment clusters stores using customer visit patterns and the time it takes for customers to pick up products. It is believed that customer behavior patterns and habits are reflected in visit patterns and the time it takes for customers to pick up products. Therefore, by clustering stores using customer features that reflect differences in customer behavior patterns and habits, it is possible to obtain appropriate store clusters in which stores with similar customer characteristics are classified into the same cluster.

[0078] Furthermore, the store analysis results can also be used to analyze companies that have one or more stores. In addition, although each process was performed for each store in this embodiment, the processing can also be performed on a company-by-company basis instead of on a store basis.

[0079] Furthermore, the purchase data analysis device 100 according to this embodiment can manage store cluster labels containing one or more store clusters, retain store cluster labels, and further aggregate customer information for each store cluster label. With this configuration, store cluster labels, which are sets of store clusters with similar characteristics, can be set, and aggregation and analysis can be performed for each store cluster label.

[0080] Furthermore, the purchase data analysis device 100 according to this embodiment can accept input of a store cluster or store cluster label name and update the store cluster or store cluster label name to the input name. The purchase data analysis device 100 according to this embodiment can also further accept input of store cluster or store cluster label information. Users can freely change the name and content of the store cluster label.

[0081] (modified version) In the embodiment described above, the purchase data analysis device 100 performed two clustering operations: one to classify each customer into a customer cluster, and another to classify each store into a store cluster. However, clustering for customers may be omitted. In this case, the customer clustering unit 103 and the customer cluster data storage unit 104 can be omitted. Instead of generating store features using the customer clustering results based on customer features, the store feature generation unit 105 calculates statistics for customer features and uses them as store features. For example, the mean or variance can be used as statistics. Figure 19 is a schematic diagram illustrating the process of generating store clusters in the clustering process according to the modified example. Note that the results of calculations using multiple statistics may also be used as store features.

[0082] (Examples of application) Figure 20 is a block diagram illustrating the hardware configuration of the purchasing data analysis device 2000 according to the application example. The application example is a specific example of the embodiment and each modification, and is a form in which the purchasing data analysis device 2000 is implemented using a computer.

[0083] The purchasing data analysis device 2000 comprises a CPU (Central Processing Unit) 2001, RAM (Random Access Memory) 2002, program memory 2003, auxiliary storage device 2004, and input / output interface 2005 as hardware. The CPU 2001 communicates with the RAM 2002, program memory 2003, auxiliary storage device 2004, and input / output interface 2005 via a bus. In other words, the purchasing data analysis device 2000 of this embodiment is implemented by a computer with this hardware configuration.

[0084] CPU2001 is an example of a general-purpose processor. RAM2002 is used as working memory for CPU2001. RAM2002 includes volatile memory such as SDRAM (Synchronous Dynamic Random Access Memory). Program memory2003 stores data analysis programs for realizing each part according to each embodiment. This data analysis program may be, for example, a program for the computer to realize the functions of the customer information acquisition unit 101, customer feature generation unit 102, customer clustering unit 103, customer cluster data storage unit 104, store feature generation unit 105, store clustering unit 106, store cluster data storage unit 107, store cluster aggregation unit 108, store cluster display unit 109, store cluster management unit 110, and store cluster label storage unit 111. In addition, as program memory2003, for example, ROM (Read-Only Memory), part of auxiliary storage device 2004, or a combination thereof may be used. Auxiliary storage device 2004 stores data non-temporarily. Auxiliary storage device 2004 includes non-volatile memory such as HDD (hard disk drive) or SSD (solID state drive).

[0085] The Input / Output Interface 2005 is an interface for connecting to other devices. For example, the Input / Output Interface 2005 is used to connect to keyboards, mice, databases, and displays.

[0086] The data analysis program stored in program memory 2003 includes computer-executable instructions. When the data analysis program (computer-executable instructions) is executed by the processing circuit, the CPU 2001, it causes the CPU 2001 to perform predetermined processing. For example, when the data analysis program is executed by the CPU 2001, it causes the CPU 2001 to perform a series of processes described with respect to each part of Figure 1. For example, when the computer-executable instructions included in the data analysis program are executed by the CPU 2001, it causes the CPU 2001 to perform a data analysis method. The data analysis method may include steps corresponding to each of the functions of the customer information acquisition unit 101, customer feature generation unit 102, customer clustering unit 103, customer cluster data storage unit 104, store feature generation unit 105, store clustering unit 106, store cluster data storage unit 107, store cluster aggregation unit 108, store cluster display unit 109, store cluster management unit 110, and store cluster label storage unit 111. The data analysis method may also include the steps shown in Figure 2 as appropriate.

[0087] The data analysis program may be provided to the purchasing data analysis device 2000, which is a computer, in a state where it is stored on a computer-readable storage medium. In this case, for example, the purchasing data analysis device 2000 further includes a drive (not shown) for reading data from the storage medium and retrieving the data analysis program from the storage medium. As the storage medium, for example, magnetic disks, optical disks (CD-ROM, CD-R, DVD-ROM, DVD-R, etc.), magneto-optical disks (MO, etc.), semiconductor memory, etc. can be used as appropriate. The storage medium may also be called a non-transitory computer-readable storage medium. Alternatively, the data analysis program may be stored on a server on a communication network, and the purchasing data analysis device 2000 may download the data analysis program from the server using the input / output interface 2005.

[0088] The processing circuit that executes the data analysis program is not limited to a general-purpose hardware processor such as CPU2001, but may also use a dedicated hardware processor such as an ASIC (Application Specific Integrated Circuit). The term "processing circuit" (or "processing unit") includes at least one general-purpose hardware processor, at least one dedicated hardware processor, or a combination of at least one general-purpose hardware processor and at least one dedicated hardware processor. In the example shown in Figure 20, CPU2001, RAM2002, and program memory2003 correspond to the processing circuit.

[0089] Thus, according to any of the embodiments described above, it is possible to provide a purchase data analysis device, method, and program that can appropriately cluster stores with similar characteristics.

[0090] While several embodiments of the present invention have been described, these embodiments are presented as examples only and are not intended to limit the scope of the invention. These embodiments can be carried out in a variety of other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims and their equivalents. [Explanation of Symbols]

[0091] 100, 2000...Purchase data analysis device, 101...Customer information acquisition unit, 102...Customer feature generation unit, 103...Customer clustering unit, 104...Customer cluster data storage unit, 105...Store feature generation unit, 106...Store clustering unit, 107...Store cluster data storage unit, 108...Store cluster aggregation unit, 109...Store cluster display unit, 110...Store cluster management unit, 111...Store cluster label storage unit, 200...Purchase-related database, 300...Customer master, 2001...CPU, 2002...RAM, 2003...Program memory, 2004...Auxiliary storage device, 2005...Input / output interface.

Claims

1. A customer information acquisition unit that acquires customer information for each customer, including the time of purchase-related actions, A customer feature generation unit generates customer features for each customer based on the aforementioned action time, including a periodic explanatory variable that includes the number of purchase-related actions per day of the week, the elapsed time from entering the store to the purchase-related action, the discretized values ​​of the explanatory variable, or the discretized values ​​of the elapsed time. A store feature generation unit generates store feature quantities for each store that represent the characteristics of the customer, based on the statistical data of the customer feature quantities of the customer or the clustering results of the customer based on the customer feature quantities. A store clustering unit that clusters stores using the aforementioned store features, A purchasing data analysis device equipped with the following features.

2. The aforementioned customer information acquisition unit further acquires customer master data relating to customer attributes, The customer feature generation unit generates customer features by adding the customer master data to the customer information. The purchase data analysis apparatus according to claim 1.

3. The system further comprises a customer clustering unit that clusters each customer based on the aforementioned customer characteristics, The store feature generation unit generates the store features based on the customer clustering results. The purchase data analysis apparatus according to claim 1.

4. A store cluster aggregation unit performs aggregation of the customer information for each store cluster based on the clustering results of the aforementioned stores, A store cluster display unit that displays the aggregated results of the customer information, The purchasing data analysis apparatus according to claim 1, further comprising:

5. The store cluster aggregation unit performs a sales forecast for a designated store cluster or a designated store based on the sales performance of the designated store cluster or the store cluster to which the designated store belongs. The purchase data analysis apparatus according to claim 4.

6. A store cluster label management unit manages store cluster labels that include one or more of the aforementioned store clusters, A store cluster label storage unit that holds the store cluster labels, Furthermore, it is equipped with, The store cluster aggregation unit further aggregates the customer information for each store cluster label. The purchase data analysis apparatus according to claim 4.

7. The store cluster label management unit receives input of the name of the store cluster or the store cluster label, The store cluster label storage unit updates the name of the store cluster or the store cluster label to the input name. The purchase data analysis apparatus according to claim 6.

8. The store cluster label management unit further accepts input of information about the store cluster or the store cluster label. The purchase data analysis device according to claim 7.

9. A computer-based information processing method, This involves obtaining customer information for each customer, including the time of purchase-related actions, Based on the aforementioned action times, customer features are generated for each customer, including a periodic explanatory variable that includes the number of purchase-related actions per day of the week, the elapsed time from entering the store to the purchase-related action, the discretized values ​​of the explanatory variable, or the discretized values ​​of the elapsed time. Based on the statistical data of the customer characteristics of the customers who visit the store, or the clustering results of the customers who visit the store based on the customer characteristics, store characteristics representing the characteristics of the customers who visit the store are generated for each store, The stores are clustered using the aforementioned store features, A method for providing this.

10. On the computer, A function to acquire customer information for each customer, including the time of purchase-related actions, Based on the aforementioned action times, a function is provided to generate customer features for each customer that include a periodic explanatory variable containing the number of purchase-related actions per day of the week, the elapsed time from entering the store to the purchase-related action, the discretized values ​​of the explanatory variable, or the discretized values ​​of the elapsed time. A function to generate store features representing the characteristics of the visiting customers for each store, based on the statistical data of the customer features of the visiting customers or the clustering results of the visiting customers based on the customer features, A function to cluster stores using the aforementioned store features, A program to achieve this.

Images