JP7433807B2 - Product recommendation device, information processing program and product recommendation method - Google Patents

Description

Embodiments of the present invention relate to a product recommendation device, an information processing program, and a product recommendation method.

When recommending products to customers for sales promotion, it is common to consider the customer's attributes, purchase history, etc., and decide on products that the customer is interested in as recommended products.
However, the recommended products determined as described above tend to be biased from customer to customer, making it difficult to determine products that newly arouse the customer's interest as recommended products.
Note that recommended products are sometimes determined at random; in this case, there is an increased possibility that a product that is of no interest to the customer will be selected as a recommended product.
Under these circumstances, it has been desired to be able to recommend products in a way that can more efficiently stimulate customer interest.

Japanese Patent Application Publication No. 2016-197424

The problem to be solved by the present invention is to provide a product recommendation device, an information processing program, and a product recommendation method that can recommend products so as to more efficiently stimulate customer interest.

The product recommendation device of the embodiment includes a first calculation means, a second calculation means, a determination means, and a notification means. The first calculation means calculates the degree of dispersion of purchase amounts for all purchased products for each unit period of one customer. The second calculation means calculates the variation amount by subtracting the average purchase amount from the maximum purchase amount of all purchase items for each unit period of the customer. The determining means determines that when the degree of dispersion calculated by the first calculating means is larger than a predetermined threshold value, the higher the dispersion amount calculated by the second calculating means, the higher the price belongs to the high price range. The candidate products are narrowed down to candidate products sold at , and recommended products are determined as at least some of the narrowed-down candidate products . The notification means notifies the customer of the recommended product determined by the determination means.

1 is an overall configuration diagram of a product sales system according to an embodiment. FIG. 2 is a schematic diagram showing an example of the layout of a store in which the product sales system shown in FIG. 1 has been introduced. FIG. 2 is a block diagram showing the main circuit configuration of the server in FIG. 1. FIG. 4 is a diagram schematically showing the structure of data records included in the member database in FIG. 3. FIG. 4 is a diagram schematically showing the structure of data records included in the recommendation database in FIG. 3. FIG. FIG. 2 is a block diagram showing a main circuit configuration of the information processing device in FIG. 1. FIG. 4 is a flowchart showing the procedure of sales processing by the processor in FIG. 3. 4 is a flowchart showing the procedure of sales processing by the processor in FIG. 3. 4 is a flowchart showing the procedure of analysis processing by the processor in FIG. 3.

Hereinafter, a product sales system having a function of recommending products to shoppers will be explained using drawings.
FIG. 1 is an overall configuration diagram of a product sales system 100 according to this embodiment.
The product sales system 100 includes a server 10, a POS (point-of-sale) terminal 20, a checkout machine 30, an information processing device 40, a monitoring terminal 50, a communication network 60, and an access point 70. Although FIG. 1 shows one each of the POS terminal 20, accounting machine 30, and monitoring terminal 50, and two each of the information processing device 40 and the access point 70, the numbers thereof are arbitrary. In many cases, a plurality of POS terminals 20 and accounting machines 30 are included in the product sales system 100. However, it is not essential to include the POS terminal 20 and the monitoring terminal 50. In many cases, a large number of information processing devices 40 are included in the product sales system 100.

The product sales system 100 is configured by enabling a server 10, a POS terminal 20, an accounting machine 30, an information processing device 40, and a monitoring terminal 50 to communicate with each other via a communication network 60. However, the information processing device 40 and the monitoring terminal 50 are connected to the communication network 60 via the access point 70 by wireless communication with the access point 70. It is desirable that the access points 70 be arranged so that any information processing device 40 located anywhere on the sales floor in the store can communicate with any one of the access points 70 .
Note that the server 10, POS terminal 20, accounting machine 30, or monitoring terminal 50 may be connected to the communication network 60 via the access point 70 by wireless communication with the access point 70.
The communication network 60 transmits various data exchanged between the server 10, the POS terminal 20, the accounting machine 30, and the access point 70. As the communication network 60, the Internet, a VPN (virtual private network), a LAN (local area network), a public communication network, a mobile communication network, etc. can be used alone or in an appropriate combination.

FIG. 2 is a schematic diagram showing an example of the layout of a store that has introduced the product sales system 100.
The information processing device 40 is operated by a shopper M1 who shops in the store. The information processing device 40 is attached to, for example, a shopping cart (hereinafter simply referred to as a cart) C. The information processing device 40 may be carried by the shopper M1. As the information processing device 40, for example, an existing portable information processing device such as a tablet computer, a smartphone, or a smart watch can be used. The information processing device 40 may be installed in a store and used temporarily by the shopper M1, or may be owned by the shopper M1. Information processing device 40 accepts an operation by shopper M1 and notifies server 10 of the content of this operation.

The server 10 generates data representing a list of purchased products (hereinafter referred to as a product list) based on the notification from the information processing device 40. The product list may be in any data format as long as it is possible to identify each registered purchased product. However, the product list includes at least a product code. In a situation where multiple identical products are registered as purchased products, the product list includes the number of purchased items associated with the product code. However, the product list may not include the number of purchased items, but may include multiple identical product codes. The product list may include any other information regarding the product, such as product name or unit price. The server 10 also has a function of processing information for recommending products to the shopper M1. In other words, the server 10 has a function as a product recommendation device.

After the shopper M1 finishes selecting all the products he wants to purchase, he goes to the checkout corner G1. The accounting machine 30 is installed in the accounting corner G1. The accounting machine 30 acquires the product list from the information processing device 40 and calculates the price for the purchased products shown in the product list based on the unit price of the purchased products shown in the product master. The product master is a database for managing products that describes product codes, names, unit prices, etc. for each product sold at a store. The product master is typically stored on the server 10. However, the accounting machine 30 may store the product master. Further, a server different from the server 10 may store the product master. Then, the accounting machine 30 settles the price according to the operation by the shopper M1. For payment, cash, credit card, electronic money, points, barcodes, virtual currency, etc. can be used. Thereafter, the shopper M1 bags the products at the soccer table SU, if necessary.

An attendant counter AT is installed near the checkout corner G1. As described above, in a store that has introduced the product sales system 100, the shopper M1 operates the information processing device 40 and the accounting machine 30, thereby allowing the shopper M1 to complete the process from registering purchased products to payment without the intervention of a store clerk. It can be done by Note that, hereinafter, this mechanism will be referred to as a cart POS service. However, when providing this cart POS service, a situation may arise where the intervention of a store clerk is unavoidable. For example, a specialized salesperson may have to provide information to the shopper M1 who is purchasing medicines. In addition, there are cases where the discount sticker information is registered at the regular price because the information is forgotten to be input into the information processing device 40. Even in the accounting machine 30, errors such as a coin jam or receipt paper exhaustion may occur.
In order to quickly deal with such an event, a clerk in charge of customer service, a so-called attendant M3, is permanently stationed at the attendant counter AT. Further, the attendant counter AT is equipped with a monitoring terminal 50. The attendant M3 can monitor the status of each information processing device 40 and accounting machine 30 using the monitoring terminal 50.

A POS terminal 20 is installed in the accounting corner G2. Furthermore, a stationary scanner SC is installed in the checkout corner G2. Note that a hand-held scanner may be provided instead of or in addition to the scanner SC. When the cashier M2 causes the scanner SC to scan the barcodes of purchased products one by one, the POS terminal 20 registers the product identified by the product code shown in the barcode as a purchased product. When the cashier M2 has finished registering all the purchased products that are the subject of one transaction, the cashier M2 instructs the POS terminal 20 to complete the transaction. In response to this instruction, the POS terminal 20 calculates the price for the registered purchased item, and settles the payment in accordance with the operator's operation through interaction between the cashier M2 and the shopper M1 regarding payment of the price. Payment can be made using cash, credit cards, electronic money, points, barcodes, virtual currency, or gift certificates. Note that the store can also be configured without providing the checkout corner G2. Like the accounting machine 30, the POS terminal 20 may also have a function of accounting based on the product list obtained from the server 10. The cashier M2 may operate the POS terminal 20 for the transaction, while allowing the cashier to perform the transaction face-to-face.

Next, the configurations of the server 10 and the information processing device 40 will be explained. Note that as the POS terminal 20, it is possible to use an existing POS terminal called a face-to-face type as is. As the accounting machine 30, an accounting machine in a semi-self-type POS system can be used as is. Further, the monitoring terminal 50 is similar to an existing monitoring terminal used in an existing self- or semi-self-type product sales system. Therefore, a description of the configurations of the POS terminal 20, accounting machine 30, and monitoring terminal 50 will be omitted.

FIG. 3 is a block diagram showing the main circuit configuration of the server 10. As shown in FIG.
The server 10 includes a processor 11, a main memory 12, an auxiliary storage unit 13, a communication interface 14, and a transmission path 15. The processor 11, main memory 12, and auxiliary storage unit 13 are connected by a transmission line 15, thereby configuring a computer that executes information processing to realize various functions of the server 10.

The processor 11 corresponds to the central part of the computer. The processor 11 executes information processing to realize various functions of the server 10 according to information processing programs such as an operating system, middleware, and application programs. The processor 11 is, for example, a CPU (central processing unit).

The main memory 12 corresponds to the main memory portion of the computer. Main memory 12 includes a nonvolatile memory area and a volatile memory area. The main memory 12 stores at least part of the above information processing program in a nonvolatile memory area. Further, the main memory 12 may store data necessary for the processor 11 to execute processing for controlling each unit in a non-volatile or volatile memory area. The main memory 12 uses a volatile memory area as a work area in which data is appropriately rewritten by the processor 11. For example, the nonvolatile memory area is ROM (read only memory). The volatile memory area is RAM (random access memory).

The auxiliary storage unit 13 corresponds to the auxiliary storage portion of the computer. For example, an EEPROM (electric erasable programmable read-only memory), an HDD (hard disc drive), or an SSD (solid state drive) is used as the auxiliary storage unit 13. The auxiliary storage unit 13 stores data used by the processor 11 to perform various processes and data created by processing by the processor 11. The auxiliary storage unit 13 may also store part of the above information processing program. A part of the storage area of the auxiliary storage unit 13 is used as an area for storing the member database DB1 and the recommendation database DB2. The member database DB1 is a database for managing shoppers M1 who are registered as members of the cart POS service. The recommendation database DB2 is a database for managing products that are candidates for recommendation by a recommendation function (hereinafter referred to as candidate products) to be described later.

The communication interface 14 performs data communication with each unit connected to the communication network 60 according to a predetermined communication protocol.
The transmission line 15 includes an address bus, a data bus, a control signal line, etc., and transmits data and control signals exchanged between the processor 11, the main memory 12, the auxiliary storage unit 13, and the communication interface 14.

FIG. 4 is a diagram schematically showing the structure of the data record DR1 included in the member database DB1.
The member database DB1 is a collection of a plurality of data records DR1 respectively associated with a plurality of shoppers M1 who have registered as members. Data record DR1 includes fields F11, F12, F13, and F14. A membership code for identifying the shopper M1 associated with the data record DR1 from other shoppers is set in the field F11. Attribute data representing the attributes of the shopper M1 with which the data record DR1 is associated is set in the field F12. What kind of attribute the attribute data represents is arbitrary. The attributes represented by the attribute data may include, for example, name, age, gender, preference, family structure, and the like. Field F13 is set with a dispersion degree calculated as described below for shopper M1 with which the data record DR1 is associated. The field F13 is set with a variation amount calculated as described below regarding the shopper M1 with which the data record DR1 is associated. Note that the data record DR1 may include a field in which arbitrary data other than fields F11 to F14 is set.

FIG. 5 is a diagram schematically showing the structure of a data record DR2 included in the recommendation database DB2.
The recommendation database DB2 is a collection of a plurality of data records DR2 each associated with a plurality of candidate products. Data record DR2 includes fields F21, F22, F23, and F24. In the field F21, a product code for distinguishing the candidate product associated with the data record DR2 from other products is set. One or more area codes for identifying each of several areas (hereinafter referred to as sales area) defined in the sales floor of the store are set in the field F22. A price range code for distinguishing the price range to which the price of the candidate product associated with the data record DR2 belongs from other price ranges is set in field F23. The field F24 is set with a file path for identifying the data file representing the recommendation screen for recommending the candidate product associated with the data record DR2 from other data files. Note that the data file representing the recommended screen is stored in the auxiliary storage unit 13, for example. However, the data file representing the recommended screen may be stored in a storage device provided in another device capable of communicating with the server 10 via the communication network 60.

FIG. 6 is a block diagram showing the main circuit configuration of the information processing device 40. As shown in FIG.
The information processing device 40 includes a processor 41 , a main memory 42 , an auxiliary storage unit 43 , a wireless unit 44 , a touch panel 45 , a scanner 46 , a reader 47 , a position detection unit 48 , and a transmission path 49 . The processor 41, main memory 42, and auxiliary storage unit 43 are connected by a transmission path 49, thereby configuring a computer that executes information processing to realize various functions of the information processing device 40. Note that the functions of the processor 41, main memory 42, auxiliary storage unit 43, and transmission path 49 are generally the same as those of the processor 11, main memory 12, auxiliary storage unit 13, and transmission path 15, and therefore the description thereof will be omitted.

The wireless unit 44 performs data communication with each unit connected to the communication network 60 via wireless communication with the access point 70 according to a predetermined wireless communication protocol. Note that as the wireless unit 44, for example, an existing wireless unit compliant with the IEEE 802.11 standard can be used.

Touch panel 45 includes a display device and a touch sensor. The display device displays an arbitrary screen such as a GUI (graphical user interface) screen under the control of the processor 41. As the display device, a well-known device such as a color LCD (liquid crystal display) can be used. The touch sensor is arranged to overlap the display surface of the display device. The touch sensor detects the touch position of the operator on the display surface of the display device and sends the position information to the processor 41 . A well-known device can be used as the touch sensor.

The scanner 46 has a camera as an imaging unit, and reads code symbols such as barcodes or two-dimensional data codes from images taken by the camera.
The reader 47 reads data such as a member ID recorded on the recording medium. Therefore, the reader 47 is a magnetic card reader when the recording medium is a magnetic card, and is an IC card reader when the recording medium is a contact type IC card. Further, in the case of a recording medium using RFID (radio frequency identification) such as a non-contact IC card or a smartphone, an RFID reader is used as the reader 47.

The position detection unit 48 detects the location of the information processing device 40 . As the position detection unit 48, a well-known device that detects the location based on, for example, a GPS (global positioning system) signal, a beacon signal, or a transmission signal from the access point 70 can be used.

Now, as the hardware of the server 10, for example, a general-purpose network server can be used. The transfer of the server 10 is generally carried out in a state where an information processing program describing information processing for realizing the operation of the server 10 is stored in the main memory 12 or the auxiliary storage unit 13. However, the above information processing program may not be stored in the main memory 12 or the auxiliary storage unit 13, or a different version of the information processing program of the same type may be stored in the main memory 12 or the auxiliary storage unit 13; The information processing program may be transferred separately. Then, the server 10 may be configured by writing the above-mentioned information processing program into the main memory 12 or the auxiliary storage unit 13 according to an operation by an arbitrary worker. The information processing program can be transferred by recording it on a removable recording medium such as a magnetic disk, magneto-optical disk, optical disk, or semiconductor memory, or by communicating via a network.

Next, the operation of the product sales system 100 configured as described above will be explained, focusing on the operation of the server 10. Note that the various information processes described below are merely examples, and the order of some processes may be changed, some processes may be omitted, or other processes may be added as appropriate.

In a situation where the shopper M1 makes a purchase, the processor 11 performs a process described below to accommodate the shopper M1's purchase according to an information processing program stored in the main memory 12 or the auxiliary storage unit 13. information processing (hereinafter referred to as sales processing). Note that the processor 11 individually executes this sales process for each of the information processing devices 40 that are allowed to be used by the shopper M1. In the following explanation regarding sales processing, when simply referring to "information processing device 40" and "shopper M1", it refers to the information processing device 40 that is the target of the sales processing being explained and the shopper M1 who is using it. represents.

7 and 8 are flowcharts showing the procedure of sales processing by the processor 11.
As ACT11, the processor 11 performs initialization. In this initialization, the processor 11 clears at least the product list. Note that the product list is provided in association with each of the plurality of information processing devices 40. Therefore, the processor 11 clears the product list associated with the information processing device 40 that is the target of the sales process.

In ACT 12, the processor 11 waits for an operation to start use. At this time, the processor 11 is in a state of waiting for the start of shopping using the cart C to which the information processing device 40 is attached.

The shopper M1 takes out one of the carts C placed in the cart storage area and performs a predetermined operation on the information processing device 40 attached thereto to start using the cart. For example, the processor 41 causes a screen showing a start button to be displayed on the touch panel 45 in the above-described standby state. When the touch panel 45 detects that the start button has been touched, the processor 41 determines that an operation for starting use has been performed. The processor 41 then notifies the server 10 that the operation for starting use has been performed. Upon receiving this notification, the processor 11 determines YES in ACT12 and proceeds to ACT13.

As ACT13, the processor 11 performs processing for member login. For example, the processor 11 instructs the information processing device 40 to display a predetermined login screen. In response to this instruction, in the information processing device 40, a login screen is displayed on the touch panel 45 under the control of the processor 41. Shopper M1 follows the instructions on the login screen and performs regular operations for member login. A normal operation for member login is, for example, an operation for causing the reader 47 to read data such as a membership code recorded on a recording medium such as a membership card. The processor 41 then notifies the server 10 of the membership code read by the reader 47. In the server 10, the processor 11 searches the member database DB1 for a data record DR1 in which the same member code as the notified member code is set in the field F11. Then, in response to confirmation that the corresponding data record DR1 is included, the processor 11 determines that a normal operation for member login has been performed, and proceeds to ACT14. Note that the processor 11 may perform other processing for member login, such as password verification. Further, when an operation to declare non-membership is performed on the information processing device 40, the processor 11 determines a membership code for the visitor as a login as a visitor, and proceeds to ACT14.

In ACT14, the processor 11 displays a list screen. The list screen is a screen that displays a list of purchased products shown in the product list associated with the information processing device 40. The processor 11 generates a list screen based on the product list associated with the information processing device 40. However, here, the product list is cleared. For this reason, the processor 11 generates a list screen here that does not show information on purchased products, but only shows a predetermined basic form such as a frame line. The processor 11 then instructs the information processing device 40 to display the list screen while transmitting screen data representing the list screen. In response to this instruction, the processor 41 of the information processing device 40 causes the touch panel 45 to display a list screen represented by the received screen data. By visually checking the list screen in this initial state, the shopper M1 can know that it is now possible to register the purchased product. Note that instead of displaying the list screen here, the processor 11 may instruct the information processing device 40 to display, for example, a screen displaying a message urging the user to start registering purchased products.

In ACT 15, the processor 11 checks whether an operation to instruct a change in the purchased product has been performed. If the processor 11 cannot confirm that the corresponding operation has been performed, the processor 11 determines NO and proceeds to ACT16.
In ACT 16, the processor 11 checks whether the sales floor area including the location of the information processing device 40 has changed. If the corresponding change cannot be confirmed, the processor 11 determines NO and proceeds to ACT17.
In ACT 17, the processor 11 confirms whether payment has been instructed. If the processor 11 cannot confirm that the relevant instruction has been given, the processor 11 determines NO and returns to ACT15.
Thus, in ACT15 to ACT17, the processor 11 waits for an instruction to change the purchased product, a change in the sales floor area, and an instruction for payment.

The shopper M1 moves around the sales floor while pushing the cart C, looking for a product to purchase. When the user finds a product to purchase, the user takes the product out of the display area and places it on cart C. At this time, the shopper M1 selects the applicable item as a purchased item by causing the scanner 46 to read the code symbol displayed on the applicable item or by operating the touch panel 45 to specify the applicable item. specify. Furthermore, when the shopper M1 wishes to cancel the purchase of a product that has already been registered as a purchased product, the shopper M1 takes out the product from the cart C and returns it to the sales floor. At this time, the shopper M1 specifies the cancellation and then performs an operation such as causing the scanner 46 to read the code symbol displayed on the applicable item or operating the touch panel 45 to specify the applicable item. will instruct the cancellation of the corresponding product. The processor 41 requests the server 10 to add or cancel the product along with notification of the product.

Upon receiving the addition or cancellation request in this way, the processor 11 determines YES in ACT15 and proceeds to ACT18.
As ACT 18, the processor 11 updates the product table in response to the request. For example, if addition is requested, the processor 11 updates the product table to represent the notified product as a purchased product. Further, for example, if cancellation is requested, the processor 11 updates the product table so as to exclude the notified product from the purchased products.

As ACT19, the processor 11 updates the list screen. For example, the processor 11 generates a list screen based on the product list updated as described above. The processor 11 then instructs the information processing device 40 to display the list screen while transmitting screen data representing the list screen. In response to this instruction, the processor 41 of the information processing device 40 causes the touch panel 45 to display a list screen represented by the received screen data. As a result, the list screen displayed on the touch panel 45 is updated. Thereafter, the processor 11 returns to the standby state in ACT15 to ACT17.

Once the shopper M1 has finished registering all of the products to be purchased as purchased products, the shopper M1 performs a predetermined operation on the information processing device 40 to instruct payment. The instruction is, for example, touching a payment button displayed on the touch panel 45. Then, in the information processing device 40, the processor 41 notifies the server 10 if an operation for instructing payment has been performed. Upon receiving this notification, the processor 11 in the server 10 determines YES in ACT17, and proceeds to ACT20.

As ACT20, the processor 11 displays a guide screen. For example, the processor 11 instructs the information processing device 40 to display a guidance screen while transmitting screen data prepared in advance as representing a predetermined guidance screen. In response to this instruction, the processor 41 of the information processing device 40 causes the touch panel 45 to display a guidance screen represented by the received screen data. The guidance screen is a screen that guides the procedure for proceeding to checkout using the checkout machine 30.
As ACT 21, the processor 11 waits for any of the accounting machines 30 to be designated.

Shopper M1 moves with cart C to checkout corner G1. Then, the shopper M1 follows the guidance on the guidance screen and causes the scanner 46 of the information processing device 40 to read the code symbol displayed on the checkout guidance screen displayed on the touch panel provided in the checkout machine 30 in the vacant state. Note that the processor included in the accounting machine 30 in the idle state causes the accounting information screen to be displayed on the touch panel. The accounting guidance screen represents a code symbol representing data including an accounting machine code for identifying the accounting machine 30. Note that in the information processing device 40, a selection button associated with the accounting machine 30 may be displayed on the touch panel 45. Then, the processor 41 may determine the checkout machine code of the checkout machine 30 designated by the shopper M1 by touching this selection button.

When the code symbol is read by the scanner 46, the processor 41 in the information processing device 40 notifies the server 10 of the data indicated in the code symbol. Upon receiving this notification, the processor 11 in the server 10 determines YES in ACT21, and proceeds to ACT22.
As the ACT 22, the processor 11 accesses the accounting machine 30 identified by the accounting machine code included in the data notified as described above via the communication network 60, and retrieves the product list associated with the information processing device 40. Send.

Note that the processor provided in the accounting machine 30 may retrieve the product list from the server 10. For example, a code symbol indicating an identification code for identifying the information processing device 40 is displayed on the touch panel 45 of the information processing device 40 . When the code symbol is read by the scanner provided in the checkout machine 30, the processor provided in the checkout machine 30 accesses the server 10 via the communication network 60. Then, the processor provided in the checkout machine 30 requests the server 10 to transmit a product list associated with the information processing device 40 identified by the identification code indicated by the code symbol. In the server 10, the processor 11 sends the product list to the accounting machine 30 in response to the request.

Further, when the processor 11 transmits a product list to any one of the accounting machines 30 and that accounting machine 30 is capable of starting accounting, the processor provided in that accounting machine 30 sends the product list to one of the accounting machines 30. You can also import it. In this case, the processor provided in the checkout machine 30 transfers the product list to another checkout machine 30 if the checkout cannot be started because the checkout process is already in progress.

When the product list is transmitted to the accounting machine 30 via the communication network 60, the processor provided in the accounting machine 30 calculates the payment amount for the purchase of all the products included in the product list, and calculates the payment amount with respect to the purchase of all the products included in the product list. Perform accounting procedures for Note that this accounting process may be similar to that performed in accounting machines of existing semi-self checkout systems, for example. When the processor provided in the accounting machine 30 finishes the accounting process, it notifies the server 10 of the completion of the accounting process, along with transmitting result data representing the result of the accounting process. The result data may include the amount paid, the method used for the payment, data corresponding to the payment method such as the amount of change, or the date and time when the payment was completed.

In ACT 23, the processor 11 waits for the transaction to be completed. If the processor 11 is notified of the completion of the transaction from the accounting machine 30 as described above, the processor 11 determines YES and proceeds to ACT24.
As ACT24, the processor 11 updates the history database. Processor 11 updates the history database to include data records including membership codes, product lists, and results data, for example. The history database may be stored in the auxiliary storage unit 13 or may be stored in a storage device provided in a device other than the server 10 that can communicate via the communication network 60. The processor 11 then returns to ACT11 and repeats the processing from ACT11 onwards in the same manner as described above.

Now, at a predetermined execution timing, the processor 11 executes information processing as described below (hereinafter referred to as analysis processing) according to the information processing program stored in the main memory 12 or the auxiliary storage unit 13. . It is assumed that the execution timing is, for example, timing outside the store's business hours for each period such as one day, one week, or one month. However, the execution timing may be during business hours.

FIG. 9 is a flowchart showing the procedure of analysis processing by the processor 11.
In ACT 41, the processor 11 selects any one of the registered shoppers M1. Note that in the following explanation regarding the analysis process, the shopper M1 selected here will be referred to as a target customer. The processor 11, for example, selects one of the data records contained in the member database DB1. Then, the processor 11 sets the shopper M1 identified by the membership code set in the field F11 of the data record DR1 selected in this way as the target customer.

As ACT 42, processor 11 collects the purchase amount for each transaction regarding the target customer. For example, the processor 11 extracts all data records containing the target customer's membership code from the history database. The processor 11 then collects the payment amount included in the extracted data record as the purchase amount. Note that some kind of discount may be applied to the payment amount, and whether or not such a discount is applied may differ for each transaction. Therefore, the processor 11 may use the product price before applying the discount as the purchase amount. Processor 11 may collect only purchase amounts related to transactions during a predetermined collection period, such as the most recent month.

In ACT 43, the processor 11 calculates the degree of dispersion of the collected purchase amounts. As the degree of dispersion, for example, well-known statistical values such as variance, standard deviation, or coefficient of variation can be used. For example, the processor 11 takes the total amount of shopping for one shopping as one piece of data, and calculates the scatter value for the total amount of shopping for each of the multiple shopping. In this case, the unit period is a period during which one shopping is performed. However, the processor 11 may set the total amount of purchases related to one or more purchases made within a predetermined unit period, such as one day, one week, or one month, as the total shopping amount. As the processor 11 executes information processing based on the information processing program, the computer with the processor 11 as a core functions as a calculation means for calculating the degree of dispersion.

In ACT 44, the processor 11 calculates the amount of variation in the collected purchase amounts. The amount of variation is a value that represents an index of variation in purchase amounts in monetary terms. Processor 11, for example, finds the maximum amount of the collected purchase amounts. Further, the processor 11 calculates, for example, the average amount of the collected purchase amounts. Then, the processor 11 determines the amount obtained by subtracting the average amount from the maximum amount as the variation amount. The processor 11 may calculate the variation amount according to a predetermined algorithm, such as by finding the maximum and minimum amounts of the collected purchase amounts, and calculating the variation amount by subtracting the minimum amount from the maximum amount.

As ACT45, the processor 11 updates the member database DB1. For example, the processor 11 finds a data record DR1 in which the same membership code as the target customer's membership code is set in the field F11 from the membership database DB1. Then, the processor 11 rewrites the fields F13 and F14 of the data record DR1 with the dispersion degree and variation amount calculated in ACT43 and ACT44.

In ACT 46, the processor 11 checks whether there is a shopper M1 who has not yet been selected among the shopper M1 who has registered as a member. If there is a shopper M1 who has not yet been selected, the processor 11 determines YES, and repeats the processing from ACT41 onwards in the same manner as described above. However, when the processor 11 executes ACT41 again, it selects a different shopper from the shopper selected in ACT41 in the current analysis process. That is, the processor 11 selects a data record DR1 different from the data record DR1 selected so far in ACT41.

As a result, the processor 11 successively selects each of the registered shoppers M1 as target customers, and repeats ACT42 to ACT45 regarding the target customers. When the processor 11 finishes executing ACT42 to ACT45 with all registered shoppers M1 as target customers, it determines NO in ACT46 and ends the analysis process.

By the way, in the information processing device 40, the processor 41 notifies the server 10 of the position detected by the position detection unit 48 at every predetermined timing. When the processor 11 in the server 10 is notified of the location, it determines the sales floor area that includes the location. If the currently determined sales area is different from the previously determined sales area, the processor 11 determines YES in ACT16 in FIG. 7, and proceeds to ACT25 in FIG.

In ACT 25, the processor 11 checks whether there is a candidate product associated with the sales floor area determined as described above. For example, the processor 11 checks whether the recommendation database DB2 includes at least one data record DR2 in which the field F22 has the same area code as the area code of the sales floor area determined as described above. Then, if the corresponding data record DR2 is not included in the recommendation database DB2, the processor 11 determines NO, and returns to the standby state of ACT15 to ACT17 in FIG. If there is no product to be recommended in the sales floor area to which the shopper M1 moves, the processor 11 returns to the standby state in ACT15 to ACT17. However, if the corresponding data record DR2 is included in the recommendation database DB2, the processor 11 determines YES in ACT25 and proceeds to ACT26.

In ACT 26, the processor 11 checks whether the shopper M1 using the information processing device 40 is a member. If the user is a member, the processor 11 determines YES and proceeds to ACT27.
In ACT 27, the processor 11 checks whether the dispersion degree for the shopper M1 is large. Processor 11, for example, finds data record DR1 associated with shopper M1 from membership database DB1. Then, the processor 11 determines YES if the dispersion degree set in the field F13 of the corresponding data record DR1 is equal to or greater than a predetermined threshold value. If the processor 11 determines YES, the processor 11 proceeds to ACT28. Note that the processor 11 may determine YES when the degree of dispersion is larger than a threshold value.

In ACT 28, the processor 11 determines recommended products based on the first condition from among the candidate products associated with the determined sales floor area. The first condition is determined to narrow down candidate products to be recommended products without being based on the attributes and purchase history of the shopper M1. For example, if the variation amount is less than a predetermined first amount, the processor 11 narrows down the candidate products to those whose price range is in the first price range among the corresponding candidate products. For example, if the variation amount is greater than or equal to a predetermined first amount and less than a second amount, the processor 11 narrows down the candidate products to those whose price range is in the second price range among the corresponding candidate products. . For example, if the variation amount is greater than or equal to a predetermined second amount, the processor 11 narrows down the candidate products to those whose price range is in the third price range among the corresponding candidate products. Note that here, the first amount < the second amount < the third amount, and the first price range < the second price range < the third price range. In other words, in this example, the processor 11 narrows down the search to candidate products with higher prices as the amount of variation increases. If there are no candidate products that belong to the price range according to the amount of variation, the processor 11 narrows down the candidate products to the candidate products that belong to the price range adjacent to the price range. If the number of candidate products remaining after the above narrowing down is less than or equal to the specified number, the processor 11 selects the remaining candidate products as recommended products. If the number of candidate products remaining after the above narrowing down is greater than the specified number, the processor 11 randomly selects a specified number of candidate products from among the remaining candidate products and sets them as recommended products.

Note that the processor 11 may select recommended products from among the candidate products remaining after the above narrowing down, according to a predetermined rule. For example, the processor 11 ensures that a plurality of candidate products belonging to the same price range are selected as recommended products evenly. Alternatively, the processor 11 may, for example, set the frequency at which a plurality of candidate products belonging to the same price range are selected as recommended products to be individually determined for each candidate product. Note that the above specified number is typically 1, but may be 2 or more. The first condition may be arbitrarily determined by the creator of the information processing program, the administrator of the server 10, or the like.

On the other hand, if the degree of dispersion is less than the threshold, the processor 11 determines NO in ACT 27 and proceeds to ACT 29. The processor 11 may determine NO in ACT 27 when the degree of dispersion is less than or equal to the threshold value.
In ACT 29, the processor 11 determines recommended products based on the second condition from among the candidate products associated with the determined sales floor area. The second condition is determined to narrow down the candidate products to be recommended products to a specified number based on the attributes and purchase history of the shopper M1. For this narrowing down, well-known processing can be used, for example, to narrow down candidate products to candidate products that are likely to be of interest to shopper M1, taking attribute data of shopper M1 into consideration. Alternatively, it is possible to use a well-known process such as narrowing down to candidate products that the shopper M1 is likely to purchase by considering the purchase history of the shopper M1. The second condition may be arbitrarily determined by the creator of the information processing program, the administrator of the server 10, or the like.
Through ACT27 to ACT29 described above, the processor 11 determines recommended products based on the dispersion degree. As the processor 11 executes information processing based on the information processing program, the computer having the processor 11 as a central part functions as a determining means.

By the way, if the shopper M1 using the information processing device 40 is not a member, the processor 11 determines NO in ACT26 and proceeds to ACT30.
In ACT 30, the processor 11 determines recommended products based on the third condition from among the candidate products associated with the determined sales floor area. The third condition is defined so that recommended products are determined without being based on the attributes and purchase history of shopper M1. For example, the processor 11 randomly selects a specified number of candidate products from among the candidate products associated with the determined sales floor area, and sets them as recommended products. Note that instead of randomly selecting candidate products, the processor 11 may select candidate products such that predetermined candidate products are selected as recommended products in a predetermined order. Furthermore, the processor 11 may select candidate products such that the frequency with which each candidate product is selected as a recommended product is a predetermined frequency for each candidate product. The third condition may be arbitrarily determined by the creator of the information processing program, the administrator of the server 10, or the like.

After the processor 11 finishes determining recommended products in ACT27, ACT29, or ACT30, the process proceeds to ACT31 in any case.
In ACT31, the processor 11 displays a recommendation screen. The recommendation screen is a screen for recommending recommended products to the shopper M1. The processor 11 obtains, for example, screen data identified by a screen file path set in field F24 of a data record associated with a candidate product determined as a recommended product. The processor 11 then instructs the information processing device 40 to display the recommended screen while transmitting the acquired screen data. In response to this instruction, the processor 41 of the information processing device 40 causes the touch panel 45 to display a recommended screen represented by the received screen data. Thereafter, the processor 11 returns to the standby state at ACT15 to ACT17 in FIG. Note that the processor 11 may instruct the information processing device 40 to return the display on the touch panel 45 to the list screen in response to an instruction from the shopper M1 on the information processing device 40. Alternatively, the processor 41 in the information processing device 40 may return the display on the touch panel 45 to the list screen in response to an instruction from the shopper M1 in the information processing device 40. Further, if the processor 11 proceeds to ACT19 after executing ACT18 with the recommended screen displayed on the touch panel 45, the processor 11 changes the display on the touch panel 45 to a list screen.

The shopper M1 can recognize recommended products by visually viewing the recommendation screen displayed on the touch panel 45. In other words, the processor 11 notifies the shopper M1 of the recommended products. As the processor 11 executes information processing based on the information processing program, the computer having the processor 11 as a core functions as a notification means.

As described above, according to the server 10, even for the shopper M1 who is a member, if the dispersion of the purchase amount for the shopper M1 is large, the attributes and purchase history of the shopper M1 are Recommend recommended products determined without any basis.

The degree of dispersion of purchase amounts is likely to be small if, for example, a housewife or the like repeatedly purchases everyday consumables and the like within a certain budget. On the other hand, the degree of dispersion of purchase amounts is likely to be large when a single person or the like freely purchases novel products that they would not normally purchase. In other words, in the case of a shopper M1 whose purchase amount dispersion is large, even if a product with a low correlation with the attributes or purchase history of the shopper M1 is recommended, it will stimulate the interest of the shopper M1 and the recommended product will be It may lead to a purchase. As a result, it is possible to induce impulse purchases by the shopper M1 and to temporarily increase sales. Furthermore, if shopper M1 likes the recommended product, repeat purchases can be expected in the future, and long-term sales can be expected to improve. In this embodiment, the recommendation screen is displayed using the shopper M1 entering a new sales area as a trigger. Therefore, if products sold in the sales floor area where shopper M1 entered are recommended products, the effect of inducing impulse purchases can be further enhanced.

Furthermore, the shopper M1 who has a large variation in amount may tend to allow the amount of one purchase to be higher than usual. Therefore, by recommending products in a higher price range to the shopper M1 who has a larger amount of variation, it is possible to promote the purchase of products in such a higher price range, and an increase in sales can be expected.

Note that in the case of shopper M1 whose purchase amount dispersion is small, even if a new product is recommended, there is a risk that it will be ignored. Further, such recommendation of a product that is ignored by the shopper M1 may cause the shopper M1 to feel bothered. However, if the dispersion of purchase amounts for the shopper M1 who is a member is small, the server 10 recommends recommended products determined based on the shopper M1's attributes and purchase history. That is, for such a shopper M1, products that are highly correlated with the shopper M1 are recommended based on the purchase history and preferences. This allows products to be recommended efficiently.

This embodiment can be modified in various ways as follows.
The processor 11 does not need to perform narrowing down based on the variation amount when determining recommended products in ACT28.

The processor 11 may consider at least one of the attributes and purchase history of the shopper M1 when selecting a recommended product from among the candidate products remaining after the narrowing down in ACT28. For example, if there is a candidate product that matches the preferences of the shopper M1 among the candidate products remaining after narrowing down, the processor 11 may determine the candidate product as the recommended product.

The processor 11 can change the algorithm for determining recommended products based on the dispersion degree as appropriate, such as determining a candidate product with a higher sales price as the recommended product as the dispersion degree increases.

When a data record is added to the history database, the processor 11 may perform analysis processing only on the shopper M1 identified by the membership code included in the added data record.

Sales processing for each of the plurality of shoppers M1 may be distributed by a plurality of processors provided in the server 10 or by processors provided in each of the plurality of servers.

The processor 11 determines the timing and timing at which the processor 11 determines recommended products, such as when a predetermined time arrives, when the shopper M1 requests product recommendation, or when a store staff member instructs product recommendation. The timing of notifying shoppers of recommended products can be changed arbitrarily. Further, the timing at which the processor 11 determines the recommended product and the timing at which the shopper M1 is notified of the recommended product may be different.

The notification of the recommended product to the shopper M1 may be carried out through any device, such as a signage device fixedly installed in the store or an information terminal such as a smartphone owned by the shopper M1. good. Further, the notification of the recommended product to the shopper M1 is not limited to screen display, and may be performed by any means such as outputting a voice message or printing a printed material.

All or part of the sales processing may be executed by the processor 41 in the information processing device 40.

The recommended product may be a product provided for eating and drinking at a store. The purchase amount may be the price for eating and drinking the product.

Each function realized by the processor 11 through information processing can be partially or entirely realized by hardware that executes information processing not based on a program, such as a logic circuit. Further, each of the above functions can also be realized by combining software control with hardware such as the above logic circuit.

Although several embodiments of the invention have been described, these embodiments are presented by way of example and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, substitutions, and changes can be made without departing from the gist of the invention. These embodiments and their modifications are included within the scope and gist of the invention, as well as within the scope of the invention described in the claims and its equivalents.
Below, the invention described in the original claims of the present application will be added.
[Additional note 1] Calculation means for calculating the degree of dispersion of the customer's purchase amount for each unit period,
determining means for determining recommended products based on the dispersion degree calculated by the calculating means;
Notifying means for notifying the customer of the recommended product determined by the determining means;
A product recommendation device equipped with
[Additional note 2] If the dispersion degree is larger than a predetermined threshold, the determining means selects a plurality of candidate products as candidates for the recommended product, taking into account the attributes of the customer and the purchase history of the customer. narrowing down the candidate products, and further determining at least some of the narrowed-down candidate products as the recommended products;
Product recommendation device described in Appendix 1.
[Additional note 3] When the dispersion degree is larger than a predetermined threshold value, the determining means determines that the higher the dispersion of the plurality of purchase amounts for each unit period of the customer, the higher the price range. Narrowing down to the candidate products sold at this price,
Product recommendation device described in Appendix 2.
[Additional note 4] When the dispersion degree is smaller than a predetermined threshold, the determining means narrows down the plurality of candidate products by considering at least one of the customer's attributes and the customer's purchase history. , further determining at least some of the candidate products narrowed down as the recommended products;
The product recommendation device described in Appendix 2 or 3.
[Appendix 5] The computer installed in the product recommendation device,
a calculation means for calculating the degree of dispersion of the customer's purchase amount for each unit period;
determining means for determining recommended products based on the dispersion degree calculated by the calculating means;
Notifying means for notifying the customer of the recommended product determined by the determining means;
An information processing program that functions as
[Appendix 6] In the product recommendation device,
Calculate the dispersion of the customer's purchase amount for each unit period,
Determine recommended products based on the calculated dispersion degree,
Notifying the customer of the determined recommended product;
Product recommendation method.

100... Product sales system, 10... Server, 20... POS terminal, 30... Accounting machine, 40... Information processing device, 50... Monitoring terminal, 60... Communication network, 70... Access point, 11, 41... Processor, 12, 42 ...Main memory, 13, 43...Auxiliary storage unit, 14...Communication interface, 15, 49...Transmission path, 41...Processor, 42...Main memory, 43...Auxiliary storage unit, 44...Wireless unit, 45...Touch panel, 46... Scanner, 47...reader, 48...position detection unit.

Claims (6)

a first calculation means for calculating the degree of dispersion of purchase amounts for all purchased products for each unit period of one customer;
a second calculation means that calculates a variation amount by subtracting the average purchase amount from the maximum purchase amount of all purchased products for each unit period of the customer;
If the degree of dispersion calculated by the first calculation means is larger than a predetermined threshold, the higher the dispersion amount calculated by the second calculation means, the more the product will be sold at a price that belongs to a higher price range. determining means for narrowing down the candidate products to be recommended as at least some of the narrowed-down candidate products ;
Notifying means for notifying the customer of the recommended product determined by the determining means;
A product recommendation device equipped with a first calculation means for calculating the degree of dispersion of purchase amounts for all purchased products for each unit period of one customer;
a second calculation means that calculates a variation amount by subtracting the minimum amount of the purchase amount from the maximum amount of the customer's purchase amount for each unit period;
If the degree of dispersion calculated by the first calculation means is larger than a predetermined threshold, the higher the dispersion amount calculated by the second calculation means, the more the product will be sold at a price that belongs to a higher price range. determining means for narrowing down the candidate products to be recommended as at least some of the narrowed-down candidate products;
Notifying means for notifying the customer of the recommended product determined by the determining means;
A product recommendation device equipped with When the dispersion degree calculated by the first calculating means is smaller than a predetermined threshold, the determining means selects the plurality of candidate products based on at least one of the customer's attributes and the customer's purchase history. narrowing down the candidate products that the customer is likely to be interested in, taking into consideration the above, and further determining at least some of the narrowed down candidate products as the recommended products;
The product recommendation device according to claim 1. When the dispersion degree calculated by the first calculating means is smaller than a predetermined threshold, the determining means selects the plurality of candidate products based on at least one of the customer's attributes and the customer's purchase history. narrowing down the candidate products that the customer is likely to purchase by taking into consideration the above, and further determining at least some of the candidate products among the narrowed down candidate products as the recommended products;
The product recommendation device according to claim 1 . The computer equipped with the product recommendation device,
a first calculation means for calculating the degree of dispersion of purchase amounts for all purchased products for each unit period of one customer;
a second calculation means that calculates a variation amount by subtracting the average purchase amount from the maximum purchase amount of all purchased products for each unit period of the customer;
If the degree of dispersion calculated by the first calculation means is larger than a predetermined threshold, the higher the dispersion amount calculated by the second calculation means, the more the product will be sold at a price that belongs to a higher price range. determining means for narrowing down the candidate products to be recommended as at least some of the narrowed-down candidate products ;
Notifying means for notifying the customer of the recommended product determined by the determining means;
An information processing program that functions as In the product recommendation device,
Calculate the dispersion of purchase amounts for all purchased products for each unit period for one customer,
Calculating the variation amount by subtracting the average purchase amount from the maximum purchase amount of all purchased products for each unit period of the customer,
If the calculated degree of dispersion is larger than a predetermined threshold, the higher the calculated dispersion amount, the more narrowed down the candidate products will be narrowed down to those sold at prices that belong to a higher price range. Determine recommended products as at least some candidate products in
Notifying the customer of the determined recommended product;
Product recommendation method.