Information processing device, information processing method, and program

Abstract

To identify the order in which a user purchased commodities in an area.SOLUTION: An information processing apparatus includes: a flow line acquisition unit which acquires flow line information related to movement of a user in an area; a commodity information acquisition unit which acquires commodity information on purchased commodities which have been purchased by the user in the area; a specifying unit which specifies, based on the flow line information, at least one of stay time and stay order of the user for each of predetermined commodity arrangement areas in the area; and an order determination unit which determines the order in which the user purchased the purchased commodities, on the basis of the commodity information acquired by the commodity information acquisition unit and at least one of the stay time and the stay order for each of the commodity arrangement areas specified by the specifying unit.SELECTED DRAWING: Figure 9

Description

Technical Field

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

Background Art

[0002] Conventionally, in stores such as supermarkets and shopping malls, in order to grasp the consumption trends of users and further increase sales, it has been required to analyze information on how consumers move within the facility.

[0003] For example, in Patent Document 1, by receiving a signal transmitted from a signal transmitter disposed in a portable container that can be carried by a user in a shopping area and in which goods are stored, a position information collection apparatus has been proposed that accumulates the behavior logs (movement histories) of customer users on a shopping floor.

Prior Art Documents

Patent Documents

[0004]

Patent Document…

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, in the conventional technology described above, although the movement history of customers within the store floor can be known, it is not possible to grasp the order in which customers put goods into a shopping basket or cart while moving within the store floor. In recent years, a mechanism for purchasing goods by scanning the code information of goods with a user terminal has been introduced, and in such a mechanism, the order of purchase of goods by customers can be grasped, but a large amount of costs are incurred for the introduction of the system. Therefore, an object of the present invention is to grasp the order of purchase of goods by users within an area.

Means for Solving the Problems

[0006] One aspect of the present invention is an information processing device comprising: a movement path acquisition unit that acquires movement path information relating to the movement of a user within an area; a product information acquisition unit that acquires product information of purchased products purchased by the user within the area; a specification unit that identifies at least one of the user's dwell time and stay order for each predetermined product placement area within the area based on the movement path information; and a sequence determination unit that determines the purchase order of the purchased products by the user based on at least one of the dwell time and stay order for each product placement area identified by the specification unit and the product information acquired by the product information acquisition unit. [Effects of the Invention]

[0007] According to one aspect of the present invention, it is possible to understand the order in which users purchase products within an area. [Brief explanation of the drawing]

[0008] [Figure 1] This is a schematic diagram illustrating the user behavior analysis system of the first embodiment. [Figure 2] This figure shows an example of the data structure of a movement pattern dataset. [Figure 3] This is a floor plan of an exemplary store where the user behavior analysis system of the first embodiment is applied. [Figure 4] This figure shows an example of the data structure in a product placement table. [Figure 5] Figure 3 illustrates the movement of a single wireless tag in a store floor. [Figure 6] This figure shows an example of virtual point placement in an exemplary store floor shown in Figure 3. [Figure 7] This diagram illustrates an example of setting up a virtual point. [Figure 8] This diagram shows an example of a virtual circle set up in an exemplary store floor. [Figure 9]It is a block diagram showing the internal configuration of each device of the user behavior analysis system according to the first embodiment. [Figure 10] It is a diagram showing an example of the data configuration of the purchase dataset. [Figure 11] It is a flowchart showing the product acquisition order determination process. [Figure 12] It is a diagram showing an example of the data configuration of the behavior dataset. [Figure 13] It is a flowchart showing the zone matching process for purchased products. [Figure 14] It is a flowchart showing the scoring process. [Figure 15] It is a diagram for explaining a specific example of the scoring process. [Figure 16] It is a diagram showing an example of the data configuration of the acquisition order dataset. [Figure 17] It is a diagram schematically showing the user behavior analysis system according to the second embodiment. [Figure 18] In the user behavior analysis system according to the second embodiment, it is a diagram showing an example of creating a behavior dataset from the flow line data. [Figure 19] It is a diagram showing an example of the data configuration of the display management database. [Figure 20] It is a diagram exemplifying a plurality of stay order condition data corresponding to different condition IDs. [Figure 21] In the user behavior analysis system according to the second embodiment, it is a diagram showing an example of a screen displayed on a display device in a store. [Figure 22] It is a block diagram showing the internal configuration of each device of the user behavior analysis system according to the second embodiment. [Figure 23] It is a flowchart showing the display control process executed by the server according to the second embodiment. [Figure 24] In the user behavior analysis system according to the second embodiment, it is a diagram showing an example of a screen displayed on a display device in a store. [Figure 25] In the user behavior analysis system according to the second embodiment, it is a diagram showing an example of a screen displayed on a display device in a store.

Embodiment for Carrying out the Invention

[0009] Hereinafter, an embodiment of the information processing apparatus, information processing method, and program of the present invention will be described. Hereinafter, as an example of a system including an information processing apparatus, a user behavior analysis system that identifies line information regarding the movement of a user from the line information of a communication device attached to a commodity carrying tool such as a shopping cart or a shopping basket will be described.

[0010] For example, in a store where a plurality of sales floors are mixed, such as a supermarket, a large number of customers move inside the store to obtain commodities from the sales floors and settle accounts at the cash register. At this time, a shopping basket or a shopping cart is used to pick up the commodities and carry them to the cash register. Therefore, in order to analyze the behavior of the users who use the store, for example, it is conceivable to attach a communication device to the shopping basket or the shopping cart and obtain the movement line of the shopping basket or the shopping cart (that is, the movement line of the communication device) as the movement line regarding the movement of the user.

[0011] Hereinafter, the case where the communication device attached to the shopping cart or the shopping basket is a wireless tag will be described, but the types of communication devices are not limited thereto, and other communication devices such as smartphones, wearable terminals, and tablet terminals may also be used.

[0012] One embodiment of the information processing device includes: a movement path acquisition unit that acquires movement path information regarding the movements of users within an area such as a store; a product information acquisition unit that acquires product information of products purchased by users within the area; an identification unit that identifies at least one of the user's dwell time and stay order for each predetermined product placement area within the area based on the movement path information; and an order determination unit that determines the purchase order of products purchased by users based on at least one of the dwell time and stay order for each product placement area identified by the identification unit and the product information acquired by the product information acquisition unit. With this information processing device, the purchase order of products purchased by users can be grasped based on the user's movement path information, which can be used to improve the design of product sales areas, etc.

[0013] One embodiment of the information processing device includes a reference value calculation unit that calculates a first reference value indicating a statistical amount of the user's dwell time based on movement information. In this case, if purchased items are placed in multiple item placement areas that the user has passed through, the order determination unit increases the likelihood that the user purchased the previously purchased items in the item placement area where the user's dwell time is equal to or greater than the first reference value. This improves the accuracy of determining the purchase order of previously purchased items by the user.

[0014] In one embodiment of the information processing device, if purchased items are placed in multiple item placement areas that the user has passed through, the order determination unit increases the likelihood that the user purchased an item in an item placement area where the user has stayed for a longer period of time, and increases the likelihood that the user purchased a previously purchased item in an item placement area where the user has stayed for a greater number of times. This improves the accuracy of determining the purchase order of previously purchased items by the user.

[0015] In one embodiment of the information processing device, the order determination unit prioritizes the length of the user's stay in each product placement area over the number of times the user has stayed, and identifies product placement areas from among multiple product placement areas where the user is most likely to have purchased previously purchased items. Since a longer stay time can be considered as indicating that the user spent time purchasing the item, prioritizing the stay time improves the accuracy of determining the purchase order of previously purchased items by the user.

[0016] In one embodiment of the information processing device, each product placement area within an area is associated with one of several subdivided product classifications for the products placed there. In this case, the order determination unit determines the purchase order of purchased products by the user based on the product classification of the purchased products in the product placement area where the purchased products are located. For example, if purchased products are placed in multiple product placement areas that the user has passed through, the order determination unit considers the probability of the user purchasing the purchased product in a product placement area to be higher the finer the product classification associated with the product in that area. This improves the accuracy of determining the purchase order of purchased products by the user.

[0017] One embodiment of the information processing device includes a measurement unit that defines virtual points arranged in a grid pattern within an area and measures the time during which each virtual point is continuously included within a predetermined range centered on the position indicated by the movement information, and a dwell time calculation unit that aggregates the measured times of each virtual point measured by the measurement unit for each product placement area and uses this as the dwell time for each product placement area. By calculating the dwell time in this way, it is possible to avoid mismeasuring the dwell time at the boundaries of the product placement areas.

[0018] In one embodiment of the information processing device, the order determination unit determines the order in which the user places items into their shopping cart or shopping basket as the purchase order of the purchased items.

[0019] In one embodiment of the information processing device, the sequence determination unit determines that if the user's dwell time in the product placement area is less than a second criterion value, the purchased product has not been placed in the product transport equipment in that product placement area. For example, if the dwell time in the product placement area is less than the second criterion value, it is considered that the user has merely passed through that product placement area. Therefore, by performing the above determination, the product placement area in which the purchased product has been placed can be identified more accurately.

[0020] In one embodiment of the information processing device, the order determination unit determines the purchase order of products purchased by the user based on at least one of weighting according to the length of time spent in the product placement area and weighting according to the number of times the product placement area is occupied. This allows for flexible setting of the method used to determine the purchase order of products purchased by the user.

[0021] One embodiment of the information processing device is capable of communicating with an output device placed within an area and includes: a movement path acquisition unit that acquires movement path information regarding the movements of users within the area; a stay order acquisition unit that acquires stay order information indicating the order in which users stayed in predetermined product placement areas within the area based on the movement path information; a purchase product prediction unit that predicts products that users will purchase in the future based on the stay order information acquired by the stay order acquisition unit and purchase order information of products sold in the product placement area; and an output control unit that causes an output device placed in the product placement area to output product-related content when a user is present in a product placement area where the products predicted by the purchase product prediction unit are located. This information processing device makes it possible to effectively appeal to users about products they will purchase in the future.

[0022] The output device may be, for example, a digital signage display, but is not limited to that; any output device capable of displaying images (including videos) within the store is acceptable. Alternatively, the output device may be a user terminal (e.g., a smartphone, tablet, or a terminal attached to a shopping cart). In this case, content corresponding to the product may be sent via push notification depending on the user's location. For example, if an application related to store use is installed on the user terminal, the application may automatically send a notification (push notification) of content corresponding to the product.

[0023] In one embodiment of the information processing device, the output control unit outputs product-related content to the output device when it is estimated, based on movement information, that the user is approaching a position where they can see the image displayed on the display device. This makes it possible to more directly appeal to the user about products they may purchase in the future.

[0024] In one embodiment of the information processing device, when the purchase prediction unit predicts that the user will purchase multiple items in the future, the display control unit simultaneously outputs multiple pieces of content related to each of the multiple items to the output device. This makes it possible to appeal to the user with multiple items using a single output device.

[0025] In one embodiment of the information processing device, when the purchase prediction unit predicts multiple products that the user will purchase in the future, the output control unit outputs multiple pieces of content related to each of the multiple products to the output device, switching them at predetermined intervals. This makes it possible to appeal to the user with multiple products using a single output device.

[0026] In one embodiment of the information processing device, the stay order acquisition unit includes in the user's stay order the product placement areas within the area where the user stayed for a predetermined time or longer, from among the predetermined product placement areas within the area. This eliminates the need to consider product placement areas that the user merely passed through when determining the user's stay order.

[0027] In one embodiment of the information processing device, the output control unit determines the output time and / or output content of product-related content according to the user's movement speed and / or dwell time in the product placement area where the product is placed. This enables content output control that corresponds to the user's behavioral characteristics.

[0028] In one embodiment of the information processing device, the output control unit causes the output device to output information about the location where the product predicted by the purchase product prediction unit is located, or information about the route to that location. This allows the user to be guided to the product.

[0029] One embodiment of the information processing device includes a user count acquisition unit that acquires user count information regarding the number of users present around an output device. In this case, the output control unit determines at least one of the output time, output content, and output timing of product-related content according to the number of users present around the output device located in the product placement area of ​​the product predicted by the purchase product prediction unit. This enables content output control according to the congestion status around the output device.

[0030] In one embodiment of the information processing device, the output control unit shortens the output time of product-related content when other users are within a predetermined distance of the output device compared to when other users are not within a predetermined distance of the output device. This allows for content output time to be adjusted according to circumstances such as whether or not other users are present around the target user.

[0031] In one embodiment of the information processing device, the purchase product prediction unit predicts the products that each of multiple users within an area will purchase in the future. In this case, if the products predicted by the purchase product prediction unit for multiple users include products that have commonalities, and if multiple users are located in the product placement area where the common products are placed, the output control unit causes the output device located in the product placement area to output content related to the common products. This makes it possible to utilize the output device to the greatest extent possible.

[0032] In one embodiment of the information processing device, the purchase prediction unit predicts multiple products that a user will purchase in the future. In this case, the output control unit is located in a product placement area that has commonalities among the multiple products, and when a user is present in the product placement area with commonalities, it causes the output device located in the product placement area with commonalities to output content related to products that meet predetermined conditions among the multiple products. As a result, when multiple products that a user will purchase in the future are predicted, the desired products (for example, products that align with the store's intentions) can be promoted.

[0033] (1) First Embodiment In one embodiment of the user behavior analysis system, the server (an example of an information processing device) acquires movement data showing the movement of users within the store floor, as well as product information of items purchased by those users within the store floor. In this disclosure, "product information" refers to information that individually identifies a product. Product information includes, for example, the product name, product code, and product model. The product code is an identifier that individually identifies a product. Product information for purchased products (e.g., POS data) can be obtained from, for example, a POS system connected to a store's cash register.

[0034] In one embodiment of the user behavior analysis system, the server identifies at least one of the following based on movement information: the user's dwell time and the order in which they stayed in each predetermined zone (an example of a product placement area) within the store floor. Each zone on the store floor is filled with products of the same category sold in the store. "Dwell time" is not limited to the time a user spends in a zone. To ensure that the time a user simply passes through a zone is not included in the dwell time, the dwell time may also be the time a user stays in a zone for a specified period of time or longer. In one embodiment of the user behavior analysis system, the server determines the purchase order of products purchased by the user based on at least one of the dwell time and stay order for each identified zone, and the product information of the purchased products obtained. "The order in which products are purchased" refers to the order in which customers placed products in their shopping carts or baskets with the intention of purchasing them in a store, and is the same as the order in which customers acquired the products in the store (i.e., the order in which products are acquired). Below, a user behavior analysis system according to one embodiment will be described in more detail with reference to the drawings.

[0035] Figure 1 is a schematic diagram showing the user behavior analysis system 1 of this embodiment. As shown in Figure 1, the user behavior analysis system 1 of this embodiment includes, for example, wireless tags 2 attached to carts CT or baskets used by each user in a store, a receiver 3, a store terminal 4, and a server 5. Figure 1 illustrates the case where the wireless tags 2 are attached to the cart CT.

[0036] Wireless tag 2 is an example of a communication device, such as a relatively small wireless communication device. Wireless tag 2 emits radio waves (beacon signals) at predetermined intervals. The beacon signals are emitted in such a way that they can be received within a predetermined range, depending on the surrounding radio wave environment. The beacon signals include, for example, identification information (such as a tag ID) that identifies wireless tag 2. Receiver 3 and server 5 are connected via a network NW to form a location tracking system that determines the location of customers within the store. The network NW can be, for example, a cellular network, Wi-Fi network, the internet, a LAN (Local Area Network), a WAN (Wide Area Network), a public network, a dedicated network, or a wireless base station.

[0037] In one embodiment, the positioning of the wireless tag 2 utilizes the AOA (Angle of Arrival) method, in which a receiver 3 is installed on the ceiling of the store, the receiver 3 receives radio waves (beacon signals) emitted from the wireless tag 2, and calculates the incident angle of the received beacon signals. The receiver 3 measures the incident angle (direction of arrival) of the beacon signals received from the wireless tag 2 and sends the measured incident angle information to the server 5. The server 5 determines the position (XY coordinates) of the wireless tag 2 from the position of the transmitting receiver 3 within the store (position in XYZ coordinates) and the incident angle based on that position. The communication protocol between the wireless tag 2 and the receiver 3 is not limited, but examples include Wi-Fi®, Bluetooth® Low Energy (BLE), etc.

[0038] While the location of the wireless tag 2 can be estimated using a single receiver 3 (locator), it is preferable to install more receivers 3 depending on the strength of the received signal strength (RSSI) of the beacon signal, the store area, and the radio wave environment of the store. For example, it is preferable to place receivers 3 at equal intervals on the ceiling of the store, and to place receivers 3 at shorter intervals in places where positioning accuracy is particularly required, such as in densely populated sales areas. Note that the positioning method for the wireless tag 2 is not limited to the AOA method, and other methods such as the TOA (Time of Arrival) method may also be used. The positioning interval for wireless tag 2 can be set arbitrarily, but it should be set to the time necessary to accurately understand the user's actions (for example, 100ms to 2 seconds).

[0039] Server 5 calculates (acquires) the position of wireless tag 2 and records the movement data set of wireless tag 2. As shown in Figure 2, in the movement data set, the position data (XY coordinate position) of wireless tag 2 and the time data (time information) are associated for each movement ID. In the following explanation, the data in which the position information and time information of wireless tag 2 are associated for a single movement ID will be referred to as "movement data" (an example of movement information). The movement path ID corresponds one-to-one with the tag ID, which is the identification information of the wireless tag 2. In other words, the movement path dataset consists of multiple movement path data, each corresponding to a different tag ID. The movement path data corresponding to one tag ID corresponds to the movement of one user within the store. In the following explanation, the movement data for a given movement ID and the movement data corresponding to the tag ID associated with that movement ID are considered to have the same meaning.

[0040] Server 5 has a store map that includes two-dimensional information on the planar surface of the store floor (an example of an area) that users move around in, and two-dimensional information regarding the boundary positions of each of the multiple zones that divide the store floor. Figure 3 is an exemplary floor plan of a retail store floor (SF). Figure 3 shows an example where the retail store floor (SF) is divided into 16 zones in a 4x4 grid, consisting of zones Z11 to Z44. While Figure 3 shows each zone as a rectangle or square, this is not limited to these shapes; each zone can be any shape depending on the overall floor shape and the products to be displayed.

[0041] Server 5 has a product placement table in which the categories of products to be placed in each zone of the store floor are defined. Figure 4 shows an example of the data structure of the product placement table. In the example shown in Figure 4, each of the zones Z11 to Z44 is assigned at least one of categories C1 to C6 (an example of multiple subdivided product classifications). Here, categories C1, C2, C3, C4, C5, and C6 are subdivided in this order, with category C1 being the broadest category and category C6 being the smallest category, representing the product itself. For example, in zone Z32, categories C1: processed foods, C2: curry, and C3: curry roux are assigned, with categories subdivided in this order. Note that, for example, as shown in zones Z42 and Z43, the setting of category C1 may be the same. As shown in Figure 4, it is not necessary for all categories C1 to C6 to be set for each zone. Figure 4 shows an example where each zone is subdivided into six categories, but this is not the only option, and the number of categories when subdividing can be set arbitrarily.

[0042] Figure 5 illustrates a movement path FL corresponding to a single tag ID in the store floor SF shown in Figure 3. The movement path FL corresponds to the movement path data shown in Figure 2. Based on the movement path data and the store map in which each zone is defined, Server 5 can determine the order of zones the user corresponding to the tag ID stayed in on the store floor (order of stay) and the time spent in each zone (described later). Server 5 obtains information (purchase dataset, described later) from the POS system 7, for example, about products purchased by the user corresponding to the tag ID (purchased products). Server 5 executes a user flow analysis program, described later, based on user flow data for tag IDs, a store map, and information on the user's purchased items, to determine the order in which the user's purchased items were retrieved (i.e., the order in which items were placed in the shopping cart, etc.).

[0043] Referring again to Figure 1, the store terminal 4 is, for example, located in the store's office and equipped with a display panel such as a personal computer or tablet. The store terminal 4 can communicate with the server 5 via a network NW. The store terminal 4 is a terminal that, for example, acquires movement data from the server 5 and displays the movement as shown in Figure 3, or acquires and displays the execution results of a movement analysis program from the server 5, but it is not necessarily an essential component of the user behavior analysis system 1.

[0044] Next, the method for measuring the time users spend in each zone within the store will be explained with reference to Figures 6 to 8. Dwell time refers to the time a customer spends standing still within the store while browsing. As shown in Figure 5, the movement path FL of the wireless tag 2 only contains route information about the customer's actions within the store. Therefore, as will be discussed later, considering dwell time information when determining the order in which products are acquired can improve the accuracy of determining the order in which products are acquired.

[0045] As shown in Figure 6, in the user behavior analysis system 1 of this embodiment, virtual points VP are set in the area including zones Z11 to Z44 within the store in order to measure the time a user spends in the store. In other words, as shown in Figure 7, which is an enlarged view of a part of the area in Figure 6, virtual points VP are defined as the intersections of virtual lines Lx and Ly arranged in a grid along the X and Y axes of the zone area within the store. The time a user spends in the store is then measured for each of these virtual points VP. The distance between two adjacent virtual points VP can be set arbitrarily, but is preferably set according to the positioning accuracy of the location determination system, which consists of the receiver 3 and the server 5. For example, if the positioning accuracy of the location determination system is 50 cm, the distance between two adjacent virtual points VP may be set to 50 cm.

[0046] The specific method for measuring dwell time will be explained with reference to Figure 8. In Figure 8, we assume that the position of wireless tag 2 is Cm. In this case, a virtual circle CR with a predetermined diameter D is set centered on position Cm, and virtual points (virtual points VP1 and VP2 in the example of Figure 8) located within the circular area of ​​this virtual circle CR are identified. The time during which the user remains in the same position, for example, while virtual points VP1 and VP2 are continuously located within the circular area of ​​the virtual circle CR, is measured as the dwell time. Note that in Figure 8, a circular area is used as an example of the predetermined range, but the shape of the predetermined range is not limited to a circle; it may be an ellipse (including shapes that are substantially elliptical including straight lines), a square, or a rectangle.

[0047] As a user moves around the store, the virtual circle CR moves along with the wireless tag 2, and the virtual point VP included within the circular area of ​​the virtual circle CR set on the wireless tag 2 changes. For each virtual point VP set on the store floor, the duration included in the virtual circle CR corresponding to the user is defined as the dwell time. In other words, the dwell time for each virtual point VP is measured for each user (i.e., for each wireless tag 2).

[0048] Here, the dwell time at each virtual point VP may be measured as an accumulated value (cumulative time) or as a maximum value (maximum duration). In the following, when the dwell time is calculated as an accumulated value for a single virtual point, it will be referred to as "dwell time (cumulative value)," and when the dwell time is calculated as a maximum value, it will be referred to as "dwell time (maximum value)." When referring to common items between dwell time (cumulative value) and dwell time (maximum value), it will simply be referred to as "dwell time." Dwell time is an example of measured time. The zone dwell time (cumulative value) is calculated by accumulating the time measured at each virtual point VP without resetting from the time the user enters the zone until they exit. The maximum dwell time in a zone is the maximum duration during which each virtual point VP is located within the circular area of ​​the virtual circle CR from the time the user enters the zone until they exit. The duration during which each virtual point VP is located within the circular area of ​​the virtual circle CR is measured each time, and is reset once it leaves the circular area of ​​the virtual circle CR.

[0049] The size of the virtual circle CR, which is centered on the position of the wireless tag 2, is preferably determined based on the reference interval d (see Figure 8), which is the distance between two adjacent virtual points VP. Specifically, the diameter D of the virtual circle CR is set to be larger than the reference interval d, which is the distance between two adjacent virtual points. The reason for this is as follows.

[0050] That is, if there is no virtual point VP in the circular area of the virtual circle CR, the residence time cannot be measured. If only one virtual point VP is included in the circular area of the virtual circle CR, there is a possibility that, despite the user actually staying in a certain zone, it may be measured that the user is in another zone adjacent to that zone due to the positioning error of the wireless tag 2. In contrast, in this embodiment, at least two virtual points VP are always included in the circular area of the virtual circle CR, and the residence time is measured at each virtual point VP, so that incorrect measurement of the residence time at the zone boundary is avoided.

[0051] In addition, if the diameter D of the virtual circle CR is made too large compared to the reference interval d, the residence time will be measured for virtual points VP that are far from the actual position of the user, deviating from the actual situation. Therefore, the error in the residence time of each virtual point VP in the area becomes large. Therefore, it is preferable to set the diameter D of the virtual circle CR so that 2 to 5 virtual points VP are included in the circular area of the virtual circle CR. For example, it is preferable to set the diameter D of the virtual circle CR so that d < D ≤ 2d is satisfied.

[0052] Based on the residence time for each virtual point VP, the sum of the residence times of each virtual point (total residence time) is calculated for each zone. Thereby, when the user is touring the store, information on how long the user stopped in each zone (that is, information on the residence time for each zone) can be obtained. In other words, the total dwell time for a given zone is the sum of the dwell times calculated for each of the multiple virtual points within a specific zone from the time a user enters the store until they leave. In this case, depending on whether the dwell time for each virtual point is calculated as an accumulated value or as a maximum value, the total dwell time for each zone will also be calculated as either an accumulated value or a maximum value, and either can be applied. The total dwell time for a zone is an example of the aggregated measured time. The total dwell time calculated for a zone is the time during which the user continuously stayed in that zone, and is simply referred to as the "dwell time" in that zone.

[0053] Next, the internal configuration of the user behavior analysis system 1 will be explained with reference to the block diagram in Figure 9.

[0054] As shown in Figure 9, the wireless tag 2 includes, for example, a control unit 21 and a communication unit 22. The control unit 21 is mainly composed of a microcontroller and controls the entire wireless tag 2. For example, the control unit 21 processes the received signal and the transmitted signal (processing of the baseband signal). The communication unit 22 is an interface for communicating with the receiver 3, and for example modulates the transmission signal (e.g., a beacon signal) to the receiver 3 and broadcasts it according to BLE. The beacon signal includes the tag ID of the wireless tag 2.

[0055] As shown in Figure 9, the receiver 3 includes, for example, a radio wave receiving unit 31, an incident angle measuring unit 32, and a communication unit 33. The radio wave receiving unit 31 includes an antenna that receives beacon signals (radio waves) transmitted from the wireless tag 2. The incident angle measuring unit 32 measures the incident angle of the radio waves from the wireless tag 2 received by the radio wave receiving unit 31. The communication unit 33 is an interface for communicating with the wireless tag 2 and the server 5. For example, the communication unit 33 demodulates the received signal from the wireless tag 2. The communication unit 33 also associates the incident angle information measured by the incident angle measuring unit 32 with the tag ID included in the received beacon signal and transmits it to the server 5 via the network NW.

[0056] As shown in Figure 9, the store terminal 4 includes, for example, a control unit 41, a display unit 42, and a communication unit 43. The control unit 41 is mainly composed of a microcontroller and controls the entire store terminal 4. The control unit 41, for example, acquires movement data from the server 5 via the communication unit 43 for display and displays the movement on the display unit 42 as shown in Figure 3, and / or acquires the execution results of the movement analysis program from the server 5 and displays them on the display unit 42. The display unit 42 includes, for example, a display panel such as an LCD (Liquid Crystal Display) panel, and a drive circuit that drives the display panel based on display data acquired from the server 5. The communication unit 43 functions as a communication interface for communicating with the server 5 via the network NW.

[0057] As shown in Figure 9, the server 5 includes, for example, a control unit 51, a storage unit 52, and a communication unit 53. The control unit 51 is mainly composed of a microcontroller and controls the entire server 5. For example, when the microcontroller of the control unit 51 executes a movement analysis program, the control unit 51 functions as a movement acquisition unit 511, a product information acquisition unit 512, a identification unit 513, a sequence determination unit 514, a reference value calculation unit 515, a measurement unit 516, and a dwell time calculation unit 517.

[0058] The movement path acquisition unit 511 acquires movement path data related to the movement of users within the store floor. This movement path data, for example, is information about user movements, as it associates the position (XY coordinate position) of the wireless tag 2 with time data (time information). The product information acquisition unit 512 acquires a purchase dataset from the POS system 7, which includes product information of items purchased by the user on the store floor. The purchase dataset consists of data such as the items purchased by the user and the categories associated with each item. A category is information that indicates the classification of a product. Figure 10 shows an example of the data structure of a purchase dataset. As shown in Figure 10, the purchase dataset is data that associates products purchased by users corresponding to a movement ID with categories C1 to C3 corresponding to each product. Here, it is classified into a maximum of three categories, but the number of categories can be set arbitrarily.

[0059] The identification unit 513 identifies at least one of the following for each predetermined zone within the store floor, based on the movement data: the user's dwell time and the order in which they stayed. Specifically, based on the movement data in the movement data set (see Figure 2) and the store map, the identification unit 513 identifies the entry and exit times for each zone from the time the user enters the store until they leave, and creates an action data set (described later; see Figure 12) that associates an action order ID with each zone the user stayed in, thereby identifying at least one of the following for each zone: the user's dwell time and the order in which they stayed. The order determination unit 514 determines the purchase order of purchased items by the user based on at least one of the dwell time and stay order for each zone identified by the identification unit 513, and the purchase dataset acquired by the product information acquisition unit 512. The order determination unit 514 determines the purchase order of purchased items as the order in which the items were placed in the cart or basket owned by the user. The order determination unit 514 may determine the purchase order of purchased items by the user based on the category of the purchased items in the zone where the purchased items are located.

[0060] In one embodiment, if purchased items are located in multiple zones that the user has passed through, the sequence determination unit 514 increases the likelihood that the user purchased the purchased items in the zone where the user's dwell time is equal to or greater than a first reference value indicating the statistical amount of the user's dwell time. In one embodiment, if purchased items are placed in multiple zones that the user has passed through, the sequence determination unit 514 increases the likelihood that the user purchased an item in a zone where the user has stayed for a longer period of time, and increases the likelihood that the user purchased a previously purchased item in a zone where the user has stayed for a greater number of times. In one embodiment, the sequence determination unit 514 prioritizes the length of time a user stays in each zone over the number of times the user has stayed in a zone, and identifies a zone among multiple zones where the user is most likely to have purchased the previously purchased item. In one embodiment, each zone within the area is associated with one of several subdivided categories for the products it places, and the order determination unit 514 determines the purchase order of the purchased products by the user based on the product classification of the purchased products in the zone where the purchased products are placed. In one embodiment, if purchased items are located in multiple zones that the user has passed through, the sequence determination unit 514 increases the likelihood that the user purchased the item in a zone where the product classification associated with the item is more detailed. In one embodiment, the order determination unit 514 determines the order in which items are placed in the shopping cart or shopping basket owned by the user as the purchase order of the purchased items. In one embodiment, the sequence determination unit 514 determines that if the user's dwell time in a zone is less than a second criterion value, the user has not placed any purchased items in their shopping basket or cart within that zone. In one embodiment, the order determination unit 514 determines the purchase order of products purchased by the user based on at least one of weighting according to the length of time spent in a zone and weighting according to the number of times the user has stayed in a zone.

[0061] The reference value calculation unit 515 calculates the median dwell time of users (an example of a first reference value indicating a statistical quantity) based on the movement data. Alternatively, the mean dwell time or quartiles may be calculated instead of the median dwell time.

[0062] As explained with reference to Figures 6 to 8, the measurement unit 516 and the dwell time calculation unit 517 are provided to calculate the dwell time of each user in each zone. Specifically, the measurement unit 516 defines virtual points arranged in a grid pattern within the store floor and measures the time during which each virtual point is continuously included within a predetermined range centered on the position indicated by the movement data. The dwell time calculation unit 517 aggregates the measured times of each virtual point measured by the measurement unit for each zone and uses this as the dwell time for each zone.

[0063] Storage 52 is a large-capacity storage device such as an HDD (Hard Disk Drive), and stores, for example, a store map (see Figure 3), a customer flow dataset (customer flow DS; see Figure 2), a purchase dataset (purchase DS; see Figure 10), a product placement table (see Figure 4), an action dataset (action DS; described later), and an acquisition order dataset (acquisition order DS; described later). Each piece of data in storage 52 can be updated, added, or deleted as appropriate in response to access from the control unit 51. The communication unit 53 functions as a communication interface for communicating with the receiver 3, the store terminal 4, and the POS system 7 via the network NW.

[0064] Next, with reference to Figures 11 to 16, the product acquisition order determination process performed by the server 5 will be described. The product acquisition order determination process is performed by the control unit 51 of the server 5 executing a movement analysis program. At the time the product retrieval order determination process is executed, it is assumed that a purchase dataset is associated with each movement ID linked to a tag ID. In other words, it is assumed that the movement paths of individual users, associated with their movement IDs, are linked to the data of their purchased products.

[0065] Figure 11 is a flowchart showing the process for determining the order in which to acquire products. Referring to Figure 11, the control unit 51 sequentially extracts one movement data to be processed (movement data with a movement ID associated with one tag ID) from multiple movement data aggregated within a predetermined period, and executes the processing in steps S4 to S14 (steps S2, S16). First, the control unit 51 obtains the purchase dataset for the traffic flow ID to be processed from the POS system 7 (step S4). Next, the control unit 51 generates an action dataset based on the traffic flow data for the traffic flow ID to be processed (step S6). Generating the action dataset is equivalent to identifying at least one of the following based on the traffic flow data: the dwell time of users in each zone and the order in which they stayed in each zone.

[0066] Figure 12 shows an example of the data structure of a behavioral dataset based on movement data associated with a single movement ID. The behavioral dataset contains data that shows a user's behavior within the store, sequentially showing each zone the user stayed in and the time spent in each zone. Here, the time spent in each zone is calculated as explained with reference to Figures 6 to 8. Note that if the time spent in a zone by a user is short and it is considered that the user merely passed through that zone, the record for that zone does not need to be included in the behavioral dataset. The behavior dataset contains records for each behavior sequence ID, which is assigned in chronological order of movement. Each record includes the zone, the time spent in that zone, and the number of times the zone was spent. Here, the number of stays indicates the number of times the player stayed in the same zone. For example, in "Action Sequence ID: 5", the player first stayed in zone Z42, and in "Action Sequence ID: 7", the player stayed in zone Z42 for the second time, so the number of stays corresponding to "Action Sequence ID: 7" is "2".

[0067] Next, the control unit 51 calculates the median dwell time based on the behavior data set (step S8). In the example shown in Figure 12, the median dwell time for each behavior sequence ID: 1 to 11 is calculated. The median dwell time is calculated in order to assign an appropriate score in the scoring process described later. Specifically, in the scoring process described later, a score is assigned to each purchased item for each zone in order to identify the zone in which the item was purchased. By calculating the median dwell time, it is possible to set the score while taking into account the user's movement speed and behavioral habits within the store.

[0068] Next, the control unit 51 performs zone matching processing for the purchased items (step S10). The zone matching processing will be described later. The control unit 51 assigns an action order ID to each purchased product based on the average score assigned to each zone in the zone matching process (step S12). Furthermore, the control unit 51 determines the product acquisition order based on the action order ID associated with each purchased product (step S14). The process for determining the product acquisition order will be explained in detail later.

[0069] Next, the zone matching process for purchased items performed in step S10 will be explained with reference to Figure 13. Figure 13 is a flowchart showing the zone matching process for purchased items. In this flow, the purchase dataset obtained in step S4 and the behavior dataset generated in step S6 are referenced. In the following explanation, in the purchase dataset, one record associated with a product (purchased product) is referred to as "purchase data". In the behavior dataset, one record associated with each behavior sequence ID is referred to as "behavior data".

[0070] Referring to Figure 13, the control unit 51 sequentially extracts one purchase data to be processed from the purchase dataset and executes the processes in steps S32 to S40 (steps S30, S42). The processes in steps S32 to S40 identify the zones in which the purchased items included in the purchase data to be processed may have been acquired by the user and assign a score to them. First, the control unit 51 extracts the zone value from the extracted purchase data to be processed (step S32). For example, in Figure 10, the purchase data for "Product: Delicious Curry" is associated with "Category C1: Processed Foods", "Category C2: Curry", and "Category C3: Curry Roux", and it can be seen from the product placement table that the zone corresponding to these categories is zone Z32. In step S32, it is not necessary to identify zones based on all categories shown in the purchase data; for example, a zone value of 1 or more may be extracted based on category C1 alone. For example, in Figure 10, category C1 of the purchase data shown for "Product: Delicious Beef Curry" is "Meat," and zones Z42 and Z43 may be extracted by referring to the product placement table based on this category C1.

[0071] Next, the control unit 51 determines whether the zone value extracted in step S32 matches a zone corresponding to any of the action sequence IDs in the action dataset. That is, the control unit 51 sequentially extracts one action data to be processed from the purchase dataset and executes the processes in steps S36 and S38 (steps S34 and S40). In other words, the control unit 51 determines whether the zone value extracted in step S32 matches the zone value extracted from the action data (step S36), and if they match, it executes a scoring process for that zone (step S38). For example, in Figure 10, zone Z32 extracted from the purchase data for "Product: Delicious Curry" matches zone Z32 for "Action Sequence ID: 8" in the action dataset. By performing zone matching on purchased items, zones (zones corresponding to one or more action sequence IDs) where the user may have purchased the item are identified. The scoring process assigns a score to each zone indicating the likelihood that the purchased item was acquired, when a zone corresponding to one or more action sequence IDs is identified for each purchased item.

[0072] Furthermore, if the dwell time indicated by the behavioral data to be processed is less than a predetermined value (an example of the second criterion value), it can be determined that the user has not placed purchased items in the cart or basket in the corresponding zone, and processing of that behavioral data does not need to be performed. This is because if the dwell time in a zone is very short, it can be assumed that the user simply passed through that zone.

[0073] Figure 14 shows a detailed flow of the scoring process in step S38. The control unit 51 extracts a category score (step S50), a dwell time score (step S52), and a dwell time score (step S54) for the target zone (the zone matched in step S36 of Figure 13). The control unit 51 extracts the category score, the number of dwells score, and the dwell time score, respectively, based on the category score criteria, the number of dwells score criteria, and the dwell time score criteria, which will be described later.

[0074] In step S50, the control unit 51 processes the zones that were matched in step S36 of Figure 13, and assigns a category score to the zones to be processed, depending on the category level at which they were identified. If a more specific category can be identified, it is considered more likely that the product was purchased in the zone being processed, and therefore a higher category score is assigned. In other words, if the purchased product is located in multiple zones that the user has passed through, the control unit 51 increases the category score for the zone to which the product is associated, the more specific the zone, thereby increasing the likelihood that the user purchased the product in that zone. An example of category scoring criteria is shown below.

[0075] [Category Score Criteria] • If category C6 (product) can be identified in the zone, the category score = 2.0 • If category C5 can be identified in the zone, the category score = 1.8 • If category C4 can be identified in the zone, the category score = 1.6 • If category C3 can be identified in the zone, the category score = 1.4 • If category C2 can be identified in the zone, the category score = 1.2 • If category C1 can be identified in the zone, the category score = 1.0

[0076] For example, in Figure 10, the purchase data for "Product: Delicious Curry" ("Category C1: Processed Foods", "Category C2: Curry", "Category C3: Curry Roux") matches zone Z32 in categories C1, C2, and C3 when referring to the product placement table. When multiple categories are identified in a matched zone in this way, the highest category score value is assigned to the zone according to the category score criteria. In this example, the category score is 1.4.

[0077] In step S52, the control unit 51 assigns a dwell time score to the target zone according to the number of times the matched zone was spent in step S36 in Figure 13. If the number of times the user has spent in a zone is higher, it is considered more likely that the user purchased a product in that zone, and therefore a higher dwell time score is assigned. In other words, if the purchased product is located in multiple zones that the user has passed through, the control unit 51 refers to the number of times the user has spent in each zone in the behavioral dataset shown in Figure 12, for example, and increases the dwell time score for the zone in which the user has spent more time than in other zones, thereby increasing the likelihood that the user purchased a product in that zone. An example of the criteria for scoring the number of dwell times is shown below.

[0078] [Dwell time score criteria] • If the number of stays is 5 or more, the stay count score = 1.8 • If the number of stays is 4, the stay count score = 1.6 • If the number of stays is 3, the stay count score = 1.4 • If the number of stays is 2, the stay count score = 1.2 · When the number of stays is 1, the stay count score = 1.0

[0079] In step S54, the control unit 51 assigns a residence time score to the target zone according to the residence time in the zone that matched in step S36 of FIG. 13. When the residence time in the zone is longer, it is considered that the user spent time selecting products in that zone and is likely to have purchased products, so a larger residence time score is assigned. That is, when purchased products are arranged in a plurality of zones passed by the user, for example, referring to the residence time for each zone in the action data set shown in FIG. 12, the longer the residence time of the user in the plurality of zones, the larger the residence time score for that zone, thereby increasing the possibility that the user purchased products in that zone. An example of the residence time score criteria is shown below. When extracting the residence time score, the median STM (an example of the first reference value indicating the statistical amount of the residence time) of the residence time calculated in step S8 (FIG. 11) is referred to.

[0080] [Residence Time Score Criteria] · When STM×2.5 ≤ residence time, the residence time score = 1.8 · When STM×2.0 ≤ residence time < STM×2.5, the residence time score = 1.6 · When STM×1.5 ≤ residence time < STM×2.0, the residence time score = 1.4 · When STM×1.0 ≤ residence time < STM×1.5, the residence time score = 1.2 · When residence time < STM×1.0, the residence time score = 1.0

[0081] In this case, the control unit 51, acting as a reference value calculation unit, calculates the median STM of the user's dwell time based on the movement data. If purchased items are located in multiple zones that the user has passed through, the control unit 51 increases the dwell time score in the zones where the user's dwell time is equal to or greater than the median STM, thereby increasing the likelihood that the user purchased the purchased items in those zones.

[0082] Next, the control unit 51 calculates the average score of the category score, dwell count score, and dwell time score extracted in steps S50, S52, and S54, and assigns the average score to the zone to be processed (step S56). The scoring process described above assigns an average score to each purchased item for each of the one or more zones from which the user may have acquired the item.

[0083] The zone matching process for purchased items shown in Figure 13, described above, will now be explained in detail based on the product placement table in Figure 4, the purchase dataset exemplified in Figure 10, and the behavior dataset exemplified in Figure 12. Note that the median dwell time STM in the behavior dataset in Figure 12 is, for example, 5 seconds. Furthermore, the explanation will be an example using the category score criterion, dwell count score criterion, and dwell time score criterion described above.

[0084] (i) When purchase data corresponding to "Product: Delicious Curry" is extracted First, purchase data corresponding to "Product: Delicious Curry" is extracted from the purchase dataset (Figure 10) (Step S30). This purchase data for "Product: Delicious Curry" is associated with the product categories "Category C1: Processed Foods", "Category C2: Curry", and "Category C3: Curry Roux". "Category C1: Processed Foods", "Category C2: Curry", and "Category C3: Curry Roux" correspond to the categories set in zone Z32 among the product categories set for each zone in the product placement table (Figure 4) (Step S32). Zone Z32 corresponds to "Action Sequence ID: 8" in the action dataset (Figure 12) (Steps S34, S36). The category score for zone Z32 is 1.4 because it matches category C3 ("Curry Roux") among the categories set for zone Z32 in the product placement table (Step S50). The dwell time score for zone Z32 is 1 because the dwell time for zone Z32 in the behavior dataset is "1" (Step S52). The dwell time score for zone Z32 is 1.6 because the dwell time for zone Z32 in the behavior dataset is "10 seconds" (Step S54). Therefore, the average score for zone Z32 for "Behavior Order ID: 8" is 1.333333 (Steps S38, S56).

[0085] (ii) When purchase data corresponding to "Product: Direct-from-the-farm potatoes" is extracted Next, purchase data corresponding to "Product: Direct-from-the-Farm Potatoes" is extracted from the purchase dataset (Figure 10) (Step S30). In this purchase data, "Product: Direct-from-the-Farm Potatoes" is associated with category C1 ("Vegetables"). Category C1 ("Vegetables") corresponds to the category set in zone Z21 among the product categories set for each zone in the product placement table (Figure 4) (Step S32). Zone Z21 corresponds to "Action Sequence ID: 2" in the action dataset (Figure 12) (Steps S34, S36). The category score for zone Z21 is 1 because it matches category C1 ("Vegetables") among the categories set for zone Z21 in the product placement table (step S50). The dwell time score for zone Z21 is 1 because the dwell time for zone Z21 in the behavior dataset is "1" (step S52). The dwell time score for zone Z21 is 1.8 because the dwell time for zone Z21 in the behavior dataset is "20 seconds" (step S54). Therefore, the average score for zone Z21 for "behavior sequence ID: 2" is 1.266667 (steps S38, S56).

[0086] (iii) When purchase data corresponding to "Product: Directly Sourced Carrots" is extracted Next, purchase data corresponding to "Product: Direct-from-the-Farm Carrots" is extracted from the purchase dataset (Figure 10) (Step S30). In this purchase data, "Product: Direct-from-the-Farm Carrots" is associated with category C1 ("Vegetables"). Category C1 ("Vegetables") corresponds to the category set in zone Z21 among the product categories set for each zone in the product placement table (Figure 4) (Step S32). Zone Z21 corresponds to "Action Sequence ID: 2" in the action dataset (Figure 12) (Steps S34, S36). The category score for zone Z21 is 2 because it matches category C6 ("Directly Sourced Carrots") among the categories set for zone Z21 in the product placement table (step S50). The dwell time score for zone Z21 is 1 because the dwell time for zone Z21 in the behavior dataset is "1" (step S52). The dwell time score for zone Z21 is 1.8 because the dwell time for zone Z21 in the behavior dataset is "20 seconds" (step S54). Therefore, the average score for zone Z21 for "Action Sequence ID: 2" is 1.600000 (steps S38, S56).

[0087] (iv) When purchase data corresponding to "Product: Delicious Beef Curry" is extracted Next, purchase data corresponding to "Product: Delicious Curry Beef" is extracted from the purchase dataset (Figure 10) (Step S30). This purchase data for "Product: Delicious Curry Beef" is associated with category C1 ("Meat"). Category C1 ("Meat") corresponds to the category set for zones Z42 and Z43 among the product categories set for each zone in the product placement table (Figure 4) (Step S32). Zones Z42 and Z43 correspond to "Action Sequence ID: 5", "Action Sequence ID: 6", and "Action Sequence ID: 7" in the action dataset (Figure 12) (Steps S34, S36).

[0088] The average scores for "Action Order ID: 5", "Action Order ID: 6", and "Action Order ID: 7" are as follows (Steps S38, S56). The category score for zone Z42 with "Action Order ID: 5" is 1 because category C1 ("Meat") can be identified among the categories set for zone Z42 in the product placement table (Step S50). The dwell time score for zone Z42 with "Action Order ID: 5" is 1 because the dwell time for zone Z42 with "Action Order ID: 5" in the action dataset is 1 (Step S52). The dwell time score for zone Z42 with "Action Order ID: 5" is 1.2 because the dwell time for zone Z42 with "Action Order ID: 5" in the action dataset is 6 seconds (Step S54). Therefore, the average score for zone Z42 with "Action Order ID: 5" is 1.066667 (Steps S38, S56).

[0089] The category score for zone Z43 for "Action Order ID: 6" is 1.2 because category C2 ("Beef") can be identified among the categories set for zone Z43 in the product placement table (Step S50). The dwell time score for zone Z43 for "Action Order ID: 6" is 1 because the dwell time for zone Z43 for "Action Order ID: 6" in the action dataset is 1 (Step S52). The dwell time score for zone Z43 for "Action Order ID: 6" is 1.8 because the dwell time for zone Z43 for "Action Order ID: 6" in the action dataset is 30 seconds (Step S54). Therefore, the average score for zone Z43 for "Action Order ID: 6" is 1.333333 (Steps S38, S56).

[0090] The category score for zone Z42 with "Action Order ID: 7" is 1 because category C1 ("Meat") can be identified among the categories set for zone Z42 in the product placement table (Step S50). The dwell time score for zone Z42 with "Action Order ID: 7" is 1.2 because the dwell time for zone Z42 with "Action Order ID: 7" in the action dataset is 2 (Step S52). The dwell time score for zone Z42 with "Action Order ID: 7" is 1 because the dwell time for zone Z42 with "Action Order ID: 7" in the action dataset is 2 seconds (Step S54). Therefore, the average score for "Action Order ID: 7" is 1.066667 (Steps S38, S56).

[0091] As explained above, using the product placement table in Figure 4, the purchase dataset exemplified in Figure 10, and the behavior dataset exemplified in Figure 12 as examples, each product (purchased product) in the purchase dataset is associated with a behavior order ID, zone, category score, dwell count score, dwell time score, and average score.

[0092] Figure 15 shows a score table summarizing the product, action sequence ID, zone, and each score as a result of the specific zone matching process described above. Figure 15 shows score table ST1 immediately after executing the zone matching process (step S10 in Figure 11), and score table ST2 after executing step S12 on score table ST1. As shown in Figure 15, for example, in score table ST1, multiple action sequence IDs are associated with "Product: Delicious Beef Curry". The product "Delicious Curry Beef" shown in score table ST1 has the highest average score for zone Z43 with "Action Order ID: 6". When the average score is highest, the product is most likely to have been acquired. Therefore, as shown in score table ST2 in Figure 15, "Action Order ID: 6" is assigned to the product "Delicious Curry Beef".

[0093] Figure 16 shows an example of the data structure of an acquisition order dataset. The acquisition order dataset shown in Figure 16 is obtained by performing step S14 in Figure 11 on the score table ST2 in Figure 15. Specifically, in score table ST2, "Product: Delicious Curry", "Product: Locally Sourced Potatoes", "Product: Locally Sourced Carrots", and "Product: Delicious Curry Beef" are assigned "Action Order ID: 8", "Action Order ID: 2", "Action Order ID: 2", and "Action Order ID: 6", respectively. Therefore, by sorting each product in score table ST2 in ascending order of the corresponding Action Order ID, the product acquisition order is determined to be "Product: Locally Sourced Potatoes" ("Product: Locally Sourced Potatoes", "Product: Locally Sourced Carrots", "Product: Delicious Curry Beef", "Product: Delicious Curry"). In the example shown in Figure 16, the same action sequence ID is associated with the two products ("directly sourced potatoes" and "directly sourced carrots"), and the same product acquisition sequence ("1") is assigned to these two products.

[0094] As described above, in one embodiment, the control unit 51 determines the purchase order of products purchased by the user based on at least one of a dwell time score (an example of weighting) corresponding to the length of time spent in a zone, and a dwell time score (an example of weighting) corresponding to the number of times the user has stayed in a zone. Alternatively, the purchase order of products purchased by the user may be determined by calculating an average score without considering the category score.

[0095] As described above, in the user behavior analysis system 1 of one embodiment, the server 5 acquires movement data related to the movement of users within the store floor and acquires product information of purchased items purchased by users within the store floor from the POS system 7. Furthermore, based on the movement data, the server 5 identifies at least one of the user's dwell time and stay order for each predetermined zone within the store floor, and determines the order in which users acquired purchased items (purchase order) based on at least one of the identified dwell time and stay order for each zone, along with the product information of purchased items. Therefore, it is possible to understand the order in which users purchased items within the store floor based on the user's movement data. Knowing the order in which customers purchased items while moving within the store floor allows for analysis of customer purchasing behavior, which can be reflected, for example, in store layout design that leads to increased sales. In this case, as mentioned above, the purchase order of purchased items by users is determined by calculating an average score based on the dwell time score, dwell time score, and category score, which has the advantage of being computationally intensive.

[0096] The above-mentioned user behavior analysis system 1 allows us to understand the order in which users purchase (acquire) products within a store, which enables us to improve the store's layout. For example, if a user repeatedly lingers in a zone where previously purchased items are located, it suggests that the product placement is inappropriate. In such cases, the store layout can be improved to reflect the user's purchase order. Furthermore, if a product that is frequently purchased first is located far from the store entrance, it can be moved closer to the entrance to boost sales of other products.

[0097] In one embodiment, the control unit 51 may prioritize the length of the user's stay in each zone over the number of times the user has stayed, and identify a zone among multiple zones where the user is most likely to have purchased the item they have already bought. This is because a longer stay in a particular zone suggests that the user spent time choosing an item, and therefore it is more likely that the user purchased the item in a zone with a longer stay than in a zone with a higher number of stays. To prioritize the length of the user's stay in each zone over the number of times the user has stayed, the stay time score should be set to a relatively larger value than the stay count score.

[0098] (2) Second embodiment Next, a user behavior analysis system according to a second embodiment will be described. In the second embodiment of the user behavior analysis system, the system predicts products that users will purchase in the future based on their behavior within the store and outputs content related to those products to a display device placed within the store. The content is not limited to any particular type, but could include advertisements that contribute to the promotion of related products. The display device could be, for example, digital signage, but is not limited to that; any display device that can be used within the store is acceptable. Product-specific content may be sent as a push notification depending on the user's location. For example, if an application related to using the store is installed on the user's terminal, the application may automatically notify (push) product-specific content.

[0099] In the first embodiment of the user behavior analysis system, the system identifies the order in which users acquire purchased items (purchase order) based on the user's actual movement information within the store and the associated information on the items purchased by the user, and accumulates an acquisition order dataset (Figure 16) for a large number of users. This dataset associates the items (product groups) that users actually purchase with their behavioral characteristics within the store. In other words, from the acquisition order dataset, it is possible to pattern the behavior of users who purchase items (product groups), for example, a user who acquires (purchases) items in the order of item D → item E → item G within the store is likely to purchase item J. By displaying advertising content in zones where item G or related items are located for users who acquire items in this patterned order, it is possible to effectively appeal to those users with item G or related items. Examples of effective ways to promote a product include including explanations of the product's advantages in the content, and offering coupons that are contingent on purchasing the product. Below, a user behavior analysis system according to one embodiment will be described in more detail with reference to the drawings.

[0100] Figure 17 is a schematic diagram of the user behavior analysis system 1A of this embodiment. In the following description, the same reference numerals are used for components identical to those in the system shown in Figure 1, and redundant explanations will not be provided. As shown in Figure 17, the difference between the user behavior analysis system 1A and the user behavior analysis system 1 (Figure 1) is that a display device 8 (an example of an output device) is placed within the store. In Figure 17, only one display device 8 is shown, but this is not limited to this; two or more may be placed within the store, or they may be placed in each zone. The display device 8 can communicate with the server 5A via the network NW. The content displayed on the display device 8 is controlled by the server 5A.

[0101] In this embodiment, in order to output timely and appropriate content to users moving within the store in real time, the server 5A sequentially generates behavioral datasets from the user's real-time movement data, as shown in Figure 18. Figure 18 shows an example of creating behavioral dataset 102 from movement data 101 for each movement ID associated with a tag ID. The method for creating the behavioral dataset is the same as when creating Figure 12. That is, the dwell time in each zone along the movement path is calculated from the movement path data, as explained with reference to Figures 6 to 8. The behavioral sequence ID is assigned in the order of the passage of time along the movement path. In this embodiment, the behavioral dataset is created in real time for users moving between stores, which is different from the case in Figure 12. The behavioral dataset may also include the number of dwell times, as in Figure 12.

[0102] In this embodiment, server 5A stores the display management database. Figure 19 shows an example of the data structure of the display management database. The display management database associates a condition ID that identifies the conditions applied to the actions of users in the store, the products that the user will purchase in the future when the conditions are met (referred to as "predicted purchase products"), a content ID that identifies the content to be output when the conditions are met, and an output destination ID for the said content.

[0103] The Condition ID is an identifier that identifies information (condition information) that serves as a condition for the acquisition order (purchase order) of products that a user is likely to purchase, including the target product (purchase prediction product). The acquisition order (purchase order) of products that a user is likely to purchase is created by accumulating data (acquisition order dataset in Figure 16) of product acquisition orders of users who have previously purchased the target product (purchase prediction product). Condition information is an example of purchase order information. The output destination ID is identification information that identifies one or more display devices 8 located within the store. Although not shown in the diagram, the content corresponding to each content ID and the display device 8 (one of the multiple display devices 8 in the store) corresponding to each output destination ID are predefined. Content includes images (including still images and videos) and audio. The following explanation uses the example where the content is an image.

[0104] Figure 20 illustrates multiple stay order condition data corresponding to different condition IDs. It shows examples of condition information corresponding to condition IDs. All of the condition information shown in Figure 20 indicates conditions corresponding to "Purchase prediction product: Fukujinzuke A". In Figure 20, the condition information corresponding to "Condition ID: R01" defines the order in which the user stays in each zone. Based on the user's behavioral dataset, if the user moves in the order of Zone Z21 ("Category C1: Vegetables" zone) → Zone Z43 ("Category C1: Meat" zone) → Z32 ("Category C1: Processed Foods" zone), it is estimated that the user is likely to be a customer purchasing curry. If a user in the store moves in an order that satisfies the conditions indicated by "Condition ID: R01," it is estimated that the user is a customer purchasing curry. For example, by displaying the content of "Fukujinzuke A advertisement" on the display device 8 located in Zone Z32, it is possible to appeal to curry-purchasing customers with Fukujinzuke A and induce impulse purchases.

[0105] In Figure 20, the condition information corresponding to "Condition ID: R02" specifies the order of stay in each zone and the duration of stay. Based on the user's behavioral dataset, if the conditions are met—staying in zone Z21 for 10 seconds or more → staying in zone Z43 for 10 seconds or more → staying in zone Z32 for 5 seconds or more—it is estimated that the user is likely to be a curry purchaser. Including the duration of stay in the conditions can improve the accuracy of the estimation that a user is likely to be a curry purchaser. In the condition information corresponding to "Condition ID: R03," the order of stays in zones and the duration of stay are defined, similar to "Condition ID: R02," but the fact that the customer repeatedly stayed in a specific zone (zone Z21 in this example) is taken into consideration. By including the fact that the customer repeatedly stayed in a specific zone as a condition, the accuracy of estimating the likelihood that the customer is a curry purchaser can be improved.

[0106] As shown in Figure 19, in the display management database, the content IDs output when the condition corresponding to the condition ID is met may be different. For example, in the example shown in Figure 20, "Condition ID: R02" and "Condition ID: R03" share the same dwelling zone, but they differ in whether or not they include a zone that has been repeatedly stayed in. In this way, different content IDs may be assigned depending on whether or not a zone that has been repeatedly stayed in is included, and different content may be displayed. Furthermore, if multiple conditions corresponding to different condition IDs overlap, a priority may be set for these multiple conditions. In other words, if multiple conditions are met simultaneously, the content corresponding to the condition with the relatively higher priority among those conditions will be displayed.

[0107] Server 5A, based on the movement data of each user moving within the store, determines the order in which users stay in zones, and if the zone matches the conditions for the target product (predicted purchase product) placed in that zone, and if a user is present in that zone, it causes the display device 8 located in that zone to output content related to the predicted purchase product. An example of the content to be output to the display device 8 is shown in Figure 21. Figure 21 shows an example of a screen displayed on a display device 8 in a store in the user behavior analysis system 1A of this embodiment. The example in Figure 21 is an example of a screen displayed on a display device 8 installed in a zone where "Fukujinzuke A," a predicted purchase item, is sold.

[0108] Screen G1 is an example of what is displayed when there are no users who meet the criteria for the target product (predicted purchase product) located in the zone, and it includes general information unrelated to a specific product. Screens G2 and G3 are examples of what is displayed when there are users who meet the criteria for the predicted purchase product located in the zone. Screens G2 and G3 show example screens displayed on the display device 8 when the predicted purchase product is "Fukujinzuke A" and the content related to that product is "Advertisement for Fukujinzuke A". Screen G2 is, for example, the content of content ID: CT02 corresponding to "Condition ID: R02" in Figure 20. Screen G3 is, for example, the content of content ID: CT03 corresponding to "Condition ID: R03" in Figure 20. As shown in Screen G3, in addition to the information shown in Screen G2, content displayed as a condition for repeatedly staying in a specific zone (zone Z21 in this example) may include information that encourages the purchase of the product (e.g., "Coupon Gift" in the example of Screen G3). This can encourage users who are hesitant to purchase the product to make a purchase.

[0109] Figure 22 is a block diagram showing the internal configuration of each device in the user behavior analysis system 1A of this embodiment. The differences between the block diagram in Figure 22 and Figure 9 will be explained below. In the user behavior analysis system 1A, the storage 52 of server 5A stores the display management database (Figure 19) described above. The control unit 51 of server 5A functions as a movement path acquisition unit 511, a stay order acquisition unit 521, a purchase item prediction unit 522, and an output control unit 523 when the microcontroller executes a display management program. The stay order acquisition unit 521 acquires a behavioral dataset (an example of stay order information) that shows the order in which users stayed in predetermined zones within the store floor, based on the movement data. The purchase product prediction unit 522 predicts the products that the user will purchase in the future (predicted purchase products) based on the dwell time sequence indicated by the behavioral dataset acquired by the dwell time sequence acquisition unit 521 and the condition information of products sold in the zone (an example of purchase sequence information). The condition information of products sold in the zone corresponds to the condition ID shown in Figure 19 and is information that serves as a condition for the acquisition sequence (purchase sequence) of products that the user is likely to purchase as predicted purchase products. A specific example of condition information is illustrated in Figure 20. The output control unit 523 causes the display device 8 located in a zone to output content related to the predicted purchase item when a user is present in a zone where the predicted purchase item (predicted purchase item) is located.

[0110] Next, the operation of the user behavior analysis system 1A will be explained with reference to Figure 23. Figure 23 is a flowchart showing the display control process that is sequentially executed by the control unit 51 of the server 5A. In the flowchart in Figure 23, the tag ID of the wireless tag 2 attached to carts and baskets moving within the store is selected as the target for processing, and advertisements and other content are displayed to the user carrying the cart or basket at the appropriate time.

[0111] In the flowchart of Figure 23, the tag ID to be processed is sequentially determined from among multiple tag IDs, and the processes in steps S62 to S76 are executed (steps S60, S78). First, the control unit 51 acquires movement data corresponding to the tag ID to be processed (step S62). This movement data is, for example, the movement data from the time the user started moving their cart or basket upon entering the store until the present time. Next, as explained with reference to Figure 18, the control unit 51 creates an action dataset from the acquired movement data (step S64).

[0112] The control unit 51 determines whether the behavior dataset created in step S64 satisfies the condition corresponding to the condition ID included in the display management database (step S66). If the condition corresponding to the condition ID is satisfied (step S66: YES), it determines the condition ID to be processed and performs the processing in steps S70 to S74 in order (steps S68, S76). First, the control unit 51 refers to the display management database to determine the predicted purchase product corresponding to the condition ID to be processed (step S70). Next, the control unit 51 determines whether the tag ID to be processed satisfies a predetermined output condition based on the behavioral dataset created in step S64 (step S72). An example of an output condition is that "the wireless tag 2 corresponding to the tag ID to be processed is located in the zone where the predicted purchase product determined in step S70 is located." For example, if the predicted purchase product corresponding to the condition satisfied in step S66 is "Fukujinzuke A," the user corresponding to the tag ID to be processed is estimated to be a curry purchaser, and the output condition is satisfied when that user is in the zone where "Fukujinzuke A" is located.

[0113] If the control unit 51 determines that the output conditions are met (step S72: YES), the control unit 51 performs content output control (step S74). The content output control refers to the display management database and outputs the content with the content ID corresponding to the condition ID to be processed to the display device 8 corresponding to the output destination ID (for example, the display device 8 in the zone where "Fukujinzuke A" is located). For example, if a user is moving in a sequence of stay that satisfies the conditions shown in "Condition ID: R01" in Figure 20, based on the behavioral dataset created from the movement data of users in the store, the content of "Fukujinzuke A advertisement" corresponding to "Condition ID: R01" is output to the display device 8 installed in the zone where Fukujinzuke A is located when the user is in that zone. This makes it possible to directly appeal to curry-buying customers with Fukujinzuke A.

[0114] Another example of an output condition in step S72 is that "it is estimated that the user corresponding to the wireless tag 2 corresponding to the tag ID to be processed is approaching a position where the image displayed on the display device 8 installed in the zone where the predicted purchase product determined in step S70 is located can be seen." For example, the range in which the image displayed on the display device 8 can be seen can be set in advance, and the output condition can be set when the user corresponding to the wireless tag 2 corresponding to the tag ID to be processed enters that range. By setting such an output condition, the predicted purchase product can be appealed to the user more directly. Server 5A continuously recognizes the location data of the wireless tag 2 to be processed, and since the location information and the direction in which each display device 8 in the store faces are known, it can estimate whether the user corresponding to the wireless tag 2 is approaching a position where they can recognize the image displayed on the display device 8. For example, if the user is approaching the display device 8 from the direction in which the display device 8 faces, and the distance between the user and the display device 8 is within a predetermined value, it may be determined that the user is "approaching a position where they can recognize the image."

[0115] As described above, the user behavior analysis system 1A of this embodiment acquires behavioral data showing the order in which users stay in different zones within the store floor, based on movement data. Based on this order of stay and conditional information associated with predicted purchase products (an example of purchase order information), it predicts which products a user will purchase. Furthermore, if a target user is present in a zone where predicted purchase products are located, the user behavior analysis system 1A outputs content related to those products to the display device 8 located in that zone. This makes it possible to effectively appeal to individual users who are likely to purchase the products in the future. Furthermore, if a single display device is placed in a specific zone within a store, it's possible to display content that is highly effective for each user. For example, one user might see content related to products they are likely to purchase in the future, while another user sees content related to products they are likely to purchase in the future. In short, by displaying content tailored to each individual target user, it's possible to achieve effective sales promotion.

[0116] It should be noted that multiple predicted purchase items may be identified for a single user. For example, this may occur when multiple condition IDs satisfy the conditions indicated in step S66 of the flowchart in Figure 23, and these condition IDs correspond to different predicted purchase items located in the same zone. In one embodiment, when multiple products that a user will purchase in the future are predicted, the control unit 51 may output multiple contents related to each of the multiple products to the display device 8 simultaneously, or switch between them at predetermined intervals and output them to the display device 8. This makes it possible to appeal to the user with multiple predicted purchase items.

[0117] Figure 24 shows an example of the display when the conditions corresponding to two predicted purchase items, "curry" and "stew," are met for the user. When the conditions corresponding to "curry" are met, the content displayed is content to promote "Fukujinzuke A." When the conditions corresponding to "stew" are met, the content displayed is content to promote "French bread C." The display device 8 for displaying the content is placed near the sales area where both processed foods and bread are sold.

[0118] Figure 24 shows screen G11 as an example of simultaneously outputting content related to "Fukujinzuke A" and "French Bread C" to the display device 8. Screen G11 is an example where one screen is divided into two, and content related to "Fukujinzuke A" and "French Bread C" are displayed simultaneously. Figure 24 shows screens G12 and G13 as an example of outputting content related to "Fukujinzuke A" and "French Bread C" to the display device 8, switching between them at predetermined intervals. In this example, content related to "Fukujinzuke A" (screen G12) and content related to "French Bread C" (screen G13) are output, switching between them at predetermined intervals.

[0119] In one embodiment, zones within a store floor where a user has stayed for a predetermined amount of time or longer may be included in the order of stay. This eliminates the need to consider zones where the user has merely passed through. As a result, the accuracy of determining whether a user meets the conditions corresponding to the predicted purchase items is improved.

[0120] In one embodiment, the control unit 51 may determine the output time and content of content related to the predicted purchase items according to the user's movement speed and dwell time in the zone where the predicted purchase items are located. For example, since the user's movement speed and behavioral habits within the store can be determined from the median dwell time in the user behavior dataset, if the target user moves quickly, it is advisable to shorten the output time of the content output from the display device 8 or simplify the content output. This enables content output control that corresponds to the user's behavioral characteristics. Furthermore, if the target user's dwell time in the zone is short, it is advisable to change the content output from the display device 8 and switch to content intended for a different user.

[0121] In one embodiment, since it can be determined from the user's behavior dataset, such as the dwell time and number of dwells, that the user is looking for a predicted purchase item, information regarding the location where the predicted purchase item is located or the route to that location may be output to the display device 8. This allows the user to be guided to the product. Figure 25 shows screen G14, which is an example screen containing information regarding the location where the predicted purchase item is located or the route to that location. In screen G14, the predicted purchase item is an example of "Fukujinzuke A". Window 80 of screen G14 shows the route 83 from the location 81 where the display device 8 is installed to the location 82 where the predicted purchase item (Fukujinzuke A) is located.

[0122] In one embodiment, the control unit 51 may function as a user count acquisition unit that acquires user count information regarding the number of users present around the display device 8. In this case, the control unit 51 determines at least one of the output time, output content, and output timing of product-related content according to the number of users present around the display device 8, which is located in the product placement area of ​​the product predicted by the purchase product prediction unit. For example, if there are many other users around the display device 8 in addition to the target user, outputting content for a long time to the target user may not effectively appeal to many other users. In such cases, it is advisable to shorten the content output time or simplify the content. Additionally, starting the output of content when the target user approaches the display device 8 (when the distance to the display device 8 is shorter) makes it easier to appeal to that user with the product.

[0123] In one embodiment, if another user is within a predetermined distance of the display device 8 located in the product placement area of ​​the product predicted by the purchase product prediction unit, the output time of the product-related content is shortened compared to when there are no other users. For example, when displaying content related to the predicted purchase product to a user, 15 seconds of content is displayed when another user is nearby, and 30 seconds of content is displayed when no other user is nearby. In other words, by shortening the content display time relatively when other users are nearby and cannot view the content slowly, the content output time can be adjusted according to the situation, such as whether or not other users are present around the target user.

[0124] In one embodiment, if the predicted purchase items for multiple users include items that are common to all users, and multiple users are located near the display device 8, the display device 8 in that zone will output content related to the common items. This allows for the most effective use of the display device 8.

[0125] For example, consider a scenario where, among multiple users, there are users who meet the conditions for predicting the purchase of "curry" (curry customers), and other users who meet the conditions for predicting the purchase of "curry" (stew customers). In this case, if both curry and stew tend to be purchased together with dessert products, then dessert products can be identified as products that share commonalities with the predicted purchase items, curry and stew. In this case, for example, content related to dessert products can be output to the display device 8, which is placed between processed foods and desserts. Furthermore, if both curry and stew tend to be purchased together with meat products (for example, chicken or beef), then meat products can be identified as products that share commonalities with the predicted purchase items of curry and stew. In this case, for example, content related to meat products can be output to the display device 8, which is placed between processed foods and meat products.

[0126] As mentioned above, multiple predicted purchase items may be identified for a single user. In such cases, in one embodiment, content related to products that meet predetermined conditions among the multiple predicted purchase items may be output to a display device 8 located in a common zone. The "predetermined conditions" are not limited, but for example, they may be the store's inventory status or the store's desire to actively sell the product. This allows for the promotion of products that align with the store's intentions among the multiple predicted purchase items. Examples of a "common zone" include the central position of multiple zones where the multiple predicted purchase items are located, or the boundary position of multiple zones.

[0127] For example, the above-described embodiment explained a case where data is exchanged between the store terminal 4 and the server 5 via a network NW, but this is not limited to that. Data can also be exchanged between the store terminal 4 and the server 5 via storage media such as a USB (Universal Serial Bus) memory, an SD (Secure Digital) memory card, an HDD device, or an SSD (Solid State Drive). The receiver 3 and server 5 are not limited to communicating via a network NW; they may also communicate one-to-one via wired or wireless connections. The functions of server 5 and store terminal 4 may be distributed across multiple devices, or the storage 52 of servers 5 and 5A may be distributed across multiple devices.

[0128] Although embodiments of the information processing device, information processing method, and program have been described above, the present invention is not limited to the embodiments described above. Furthermore, the above embodiments can be improved or modified in various ways without departing from the spirit of the present invention. [Explanation of Symbols]

[0129] 1,1A...User behavior analysis system 2… Wireless tags 21... Control Unit 22... Communications Department 3…Receiver 31...Radio wave receiving unit 32...Incidence angle measurement section 33... Communications Department 4…Store terminals 41... Control Unit 42...Display section 43… Communications Department 5.5A…Server 51... Control Unit 511…Flow line acquisition part 512…Product information acquisition department 513…Specific part 514...Order determining section 515...Reference value calculation unit 516...Measurement unit 517... Dwell time calculation unit 521... Stay Order Acquisition Unit 522... Purchase Product Prediction Department 523...Output control unit 52...Storage 53... Communications Department 7…POS system 8...Display device 101...Movement flow data 102…Behavioral dataset NW...Network CT... Cart SF...Store floor Z11~Z44... Zone

Claims

1. A movement path acquisition unit that acquires movement path information regarding the movement of users within the area, A product information acquisition unit that acquires product information of products purchased by the user within the area, Based on the aforementioned movement information, an identification unit identifies at least one of the user's dwell time and order of stay for each predetermined product placement area within the area, An order determination unit determines the purchase order of the purchased products by the user based on at least one of the dwell time and stay order for each product placement area identified by the identification unit, and product information acquired by the product information acquisition unit. The system includes a reference value calculation unit that calculates a first reference value indicating a statistical amount of the user's dwell time based on the aforementioned movement information, The sequence determination unit, when the purchased product is placed in multiple product placement areas that the user has passed through, increases the likelihood that the user purchased the purchased product in the product placement area where the user's dwell time is equal to or greater than a first criterion value among the multiple product placement areas. Information processing device.

2. The sequence determination unit, when the purchased product is placed in multiple product placement areas that the user has passed through, increases the likelihood that the user purchased the product in the product placement area where the user has stayed for a longer period of time, and increases the likelihood that the user purchased the purchased product in the product placement area where the user has stayed for a greater number of times. The information processing apparatus described in claim 1.

3. The sequence determination unit, for each product placement area, prioritizes the length of the user's stay over the number of times the user has stayed, and identifies the product placement area from among the multiple product placement areas where the user is most likely to have purchased the previously purchased product. The information processing apparatus according to claim 2.

4. Each product placement area within the aforementioned area is associated with one of several subdivided product classifications. The order determination unit determines the purchase order of the purchased products by the user based on the product classification of the purchased products in the product placement area where the purchased products are located. An information processing apparatus according to any one of claims 1 to 3.

5. The order determination unit, when the purchased product is located in multiple product placement areas that the user has passed through, determines that the more detailed the product classification associated with the product is in the product placement area, the higher the probability that the user purchased the purchased product in that product placement area. The information processing apparatus described in claim 4.

6. A measurement unit defines virtual points arranged in a grid pattern within the area, and measures the time during which each virtual point is continuously included within a predetermined range set centered on the position indicated by the movement information. The system includes a dwell time calculation unit that aggregates the measurement times of each virtual point measured by the measurement unit for each product placement area, and uses this as the dwell time for each product placement area. An information processing apparatus according to any one of claims 1 to 5.

7. The order determination unit determines the order in which the products are placed in the product transport equipment owned by the user as the purchase order of the purchased products. An information processing apparatus according to any one of claims 1 to 6.

8. The order determination unit determines that if the user's dwell time in the product placement area is less than the second reference value, the purchased product has not been placed in the product transport device in the product placement area. The information processing apparatus according to claim 7.

9. The order determination unit determines the purchase order of the purchased products by the user based on at least one of the following: weighting according to the length of time spent in the product placement area, and weighting according to the number of times the product placement area is occupied. An information processing apparatus according to any one of claims 1 to 8.

10. An information processing method performed by an information processing device, Steps include obtaining movement information regarding the movement of users within the area, The steps include obtaining product information of purchased items purchased by the user within the area, Based on the aforementioned movement information, the step of identifying at least one of the user's dwell time and order of stay for each predetermined product placement area within the area, A step to determine the purchase order of the purchased products by the user, based on at least one of the dwell time and stay order for each product placement area identified in the aforementioned identification step, and the product information acquired in the step to acquire the product information. The system includes the step of calculating a first reference value that indicates a statistical amount of the user's dwell time based on the aforementioned movement information, An information processing method that, in the step of determining the purchase order, if the purchased product is located in multiple product placement areas that the user has passed through, increases the likelihood that the user purchased the purchased product in the product placement area where the user's dwell time is equal to or greater than a first criterion value among the multiple product placement areas.

11. A program that causes a computer to perform a predetermined method, The aforementioned method, Steps include obtaining movement information regarding the movement of users within the area, The steps include obtaining product information of purchased items purchased by the user within the area, Based on the aforementioned movement information, the step of identifying at least one of the user's dwell time and order of stay for each predetermined product placement area within the area, A step to determine the purchase order of the purchased products by the user, based on at least one of the dwell time and stay order for each product placement area identified in the aforementioned identification step, and the product information acquired in the step to acquire the product information. The system includes the step of calculating a first reference value that indicates a statistical amount of the user's dwell time based on the aforementioned movement information, In the step of determining the purchase order, if the purchased product is located in multiple product placement areas that the user has passed through, the likelihood of the user purchasing the purchased product is increased in the product placement area where the user's dwell time is equal to or greater than a first criterion value among the multiple product placement areas. program.

Images