Information processing device, information processing method and storage medium

Abstract

Provide useful information when it is estimated that the user did not achieve the purpose in the store they visited. Therefore, the information processing device includes: a proximity detection unit that detects the situation that the information terminal carried by the user approaches the store; a purchase behavior detection unit that detects the situation that the user performs a purchase behavior in the store; an accumulation unit that accumulates history information related to the user's visit to the store and the purchase behavior; and a recommendation unit that, when the store visit history information of a first store is detected from the history information of a user, the purchase behavior history information of the first store is not detected, and the store visit history information and purchase behavior history information of a second store are detected, sends recommendation information related to the second store to the information terminal of the user who is detected to be approaching the first store, and the store visit history information and purchase behavior history information of the second store are accumulated after the history information related to the first store.

Description

Technical Field

[0001] The present invention relates to an information processing device, an information processing method and a storage medium, and in particular to a technology for sending information to a user. Background Art

[0002] The following technology is known: providing information that is considered to be useful to users based on their online behavior, for example, performing the following actions: prompting a product page that arouses the user's interest based on the user's browsing history, or selecting and prompting an advertisement corresponding to the user's purchase history from various advertisements.

[0003] Furthermore, by applying these technologies, a technology has been disclosed for providing information that is considered to be useful not only on the Internet but also in actual stores (actual stores) based on user behavior.

[0004] For example, Patent Document 1 discloses a technique for providing information that recommends purchase of products in an actual store based on user profile information and purchase history information.

[0005] Prior art literature

[0006] Patent Literature

[0007] Patent Document 1: Japanese Patent Application Publication No. 2015-14887 Summary of the invention

[0008] Problems to be solved by the invention

[0009] However, the technology disclosed in Patent Document 1 is a technology for sending recommendation information related to a store visited by a user. If the store is a store where the user cannot purchase the target product, appropriate recommendation information cannot be sent.

[0010] The present invention has been made in view of such circumstances, and an object of the present invention is to provide useful information when it is estimated that the user has not achieved the purpose in the store he / she visited.

[0011] Means for solving problems

[0012] The information processing device of the present invention comprises: a proximity detection unit, which detects the situation that an information terminal carried by a user approaches a store; a purchase behavior detection unit, which detects the situation that a user performs purchase behavior in a store; an accumulation unit, which accumulates history information related to the user's visit to the store and the purchase behavior; and a recommendation unit, which, when store visit history information of a first store is detected from the history information of a user and the purchase behavior history information of the first store is not detected, and the store visit history information and purchase behavior history information of a second store are detected, sends recommendation information related to the second store to the information terminal of the user detected to be approaching the first store, wherein the store visit history information and purchase behavior history information of the second store are accumulated after the history information related to the first store.

[0013] Thus, based on the store visit history information and purchase behavior history information of other users related to the actual store, information of a store that is a substitute for the store visited by the user is provided to the user.

[0014] In the recommendation server, the information processing device may include a similarity determination unit that determines similarity between users.

[0015] The recommendation unit performs the transmission based on history information of a similar user determined to be similar to a user to which the transmission is to be directed.

[0016] Thus, the degree of similarity between the user who is detected to be approaching the first store and the user who is the extraction source of the recommended information is determined.

[0017] The proximity detection unit of the information processing device may detect approach to a store using short-range wireless communication.

[0018] This eliminates the need to activate, for example, a GPS function on the information terminal carried by the user in order to detect the user's approach to a store.

[0019] The recommendation unit of the information processing device may transmit the recommendation information after detecting that the user to whom the recommendation information is to be transmitted has checked out of the first store.

[0020] Thus, when the user simply visits the first store, no alternative store is presented as recommended information.

[0021] In the above-mentioned information processing device, the detection of the purchase behavior may be performed based on the use of electronic money, the use of a credit card, or the use of a point card.

[0022] In the use of electronic money, credit cards, and point cards, electronic data is often processed, and thus purchase behavior can be easily detected.

[0023] The recommendation unit of the information processing device may not transmit the recommendation information related to the second store when the second store is not available.

[0024] Recommending a store that cannot be used may be disadvantageous to the user who received the notification.

[0025] Alternatively, when a plurality of the second stores are extracted from the history information of a plurality of users, the recommendation unit of the information processing device may send, as the recommendation information, a store having more purchase behavior history information of the plurality of users among the extracted second stores.

[0026] The next store where users who have not been detected to have made purchases at the first store have visited the most and made purchases is recommended to the user as the second store.

[0027] The accumulation unit of the information processing device may accumulate the history information associated with time information at which the user performed an action, and the recommendation unit may send the recommendation information using the time information.

[0028] That is, the recommendation information is sent in consideration of the time when the user was detected to have visited the first store and the time when another user made a purchase at the second store visited after the first store.

[0029] The recommendation unit of the information processing device may transmit the recommendation information using weather information related to a location of a user to whom the recommendation information is to be transmitted.

[0030] For example, when the second store to be transmitted as the recommendation information is crowded and there is a possibility of waiting time, the second store can be selected by considering whether the waiting time can be spent in an appropriate state.

[0031] The information processing method of the present invention is executed by an information processing device, and comprises: a proximity detection step, which detects the situation that an information terminal carried by a user is close to a store; a purchase behavior detection step, which detects the situation that the user performs purchase behavior in the store; an accumulation step, which accumulates history information related to the user's visit to the store and the purchase behavior; and a recommendation step, which, when the store visit history information of a first store is detected from the history information of a user, the purchase behavior history information of the first store is not detected, and the store visit history information and purchase behavior history information of a second store are detected, sends recommendation information related to the second store to the information terminal of the user detected to be close to the first store, and the store visit history information and purchase behavior history information of the second store are accumulated after the history information related to the first store.

[0032] This information processing method makes it possible to provide useful information when it can be estimated that the user has not achieved the purpose in the store he / she visited.

[0033] The program of the present invention is a program that causes an information processing device to execute steps corresponding to the above steps. The storage medium of the present invention is a medium storing the above program. Through these, the processing of the above information processing device is realized.

[0034] Effects of the Invention

[0035] According to the present invention, when it can be estimated that the user has not achieved the purpose in the store he / she visited, useful information can be provided. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 This is an explanatory diagram of a network including a recommendation server according to an embodiment of the present invention.

[0037] Figure 2 It is an explanatory diagram of the functional structure of the recommendation server according to the embodiment.

[0038] Figure 3 This is a diagram showing an example of a history DB.

[0039] Figure 4 is a block diagram of a computer device that can be used in the embodiments.

[0040] Figure 5 This is a flowchart related to the first embodiment.

[0041] Figure 6 This is a flowchart of the recommended information extraction process according to the first embodiment.

[0042] Figure 7 This is a flowchart of the recommended information selection process according to the first embodiment.

[0043] Figure 8 This is a flowchart related to the second embodiment.

[0044] Fig. 9 This is a flowchart of the recommended information extraction process according to the second embodiment.

[0045] Fig.10 This is a flowchart of the recommended information selection process according to the second embodiment.

[0046] Fig.11 This is a flowchart of another example of the recommended information extraction process.

[0047] Fig.12 This is a flowchart of another example of the recommended information selection process.

[0048] Fig.13This is a flowchart of another example 2 of the recommended information extraction process. DETAILED DESCRIPTION

[0049] Hereinafter, the embodiments will be described in the following order.

[0050] <1. System structure>

[0051] <2. Examples of each treatment>

[0052] <2-1. First embodiment>

[0053] <2-2. Second embodiment>

[0054] <2-3. Another example of recommended information extraction processing>

[0055] <2-4. Another example of recommended information selection processing>

[0056] <2-5. Another example 2 of recommended information extraction processing>

[0057] <3. Modifications>

[0058] <4. Conclusion>

[0059] <5. Program and storage medium>

[0060] <1. System structure>

[0061] Figure 1 A configuration example of a network system including the recommendation server 1 according to the embodiment is shown.

[0062] In the network system of this embodiment, a recommendation server 1 for providing recommendation information to a user (member) is connected to a user terminal 2 held by the user via a communication network 3. The recommendation information provided by the recommendation server 1 to the user is information related to a store 200. Specifically, it is information about the products handled by the store 200 or information about the store 200 itself.

[0063] Figure 1 A store 200A is shown as an example of the store 200. The store 200A is provided with a credit card terminal 4, an electronic money terminal 5, and a point card terminal 6, and each terminal is connected to a communication network 3.

[0064] In addition, a credit card system 7 , an electronic money system 8 , and a point card system 9 are connected to the communication network 3 in correspondence with the credit card terminal 4 , the electronic money terminal 5 , and the point card terminal 6 , respectively.

[0065] The credit card terminal 4 is a terminal for confirming the validity of the credit card used by the user when purchasing goods. The credit card information read by the credit card terminal 4 is sent to the credit card system 7 via the communication network 3. In the credit card system 7, the validity of the credit card is confirmed and the result is sent to the credit card terminal 4.

[0066] In the credit card system 7, in addition to this, various processes are performed, such as management of credit card user information, payment of product prices to stores, and charging of usage fees to users.

[0067] The electronic money terminal 5 is a terminal for using electronic money to purchase goods, and reads information from a card charged with electronic money to make a settlement. The information read by the electronic money terminal 5 is sent to the electronic money system 8. The electronic money system 8 performs a subtraction process of the charged amount according to the use of the electronic money, thereby performing a payment process to the store.

[0068] In addition to this, the electronic money system 8 manages information related to users and electronic money cards, performs addition processing of a recharge amount based on recharge of electronic money, and the like.

[0069] The point card terminal 6 is a terminal for reading a point card for giving points by purchasing goods, etc., using points by purchasing goods and enjoying discounts, etc. The information of the point card read by the point card terminal 6 is transmitted to the point card system 9 via the communication network 3. The point card system 9 performs a subtraction process of the number of points corresponding to the use of the points, and performs a payment process for paying the store an amount corresponding to the number of points used.

[0070] In addition, the point card system 9 manages the user information of the point card, the number of points, the information of stores where the point card can be used, and the like.

[0071] A short-distance wireless transmitter is installed in the store 200A. As an example of the short-distance wireless communication transmitter, a beacon transmitter 10 is used in the following description.

[0072] The beacon transmitter 10 is provided to detect information processing terminals located nearby, and transmits information at regular intervals (such as several hundred milliseconds intervals) using a battery installed inside. The radio wave reaching distance of the beacon transmitter 10 is, for example, several meters to several tens of meters. Therefore, at least the user's arrival at the store 200A can be detected.

[0073] In addition, when the floor area of ​​the store 200A is large, a beacon transmitter 10 having a long radio wave reaching distance may be installed, or a plurality of beacon transmitters 10, 10, ... may be installed.

[0074] In addition, Figure 1, a state where one credit card terminal 4, one electronic money terminal 5, and one point card terminal 6 are provided in the store 200A is shown, but this is an example, and a plurality of credit card terminals 4, 4, ..., a plurality of electronic money terminals 5, 5, ..., and a plurality of point card terminals 6, 6, ... may be provided. In addition, it is not necessary to provide all kinds of terminals, for example, the credit card terminal 4 may not be provided, the electronic money terminal 5 may not be provided, and the point card terminal 6 may not be provided.

[0075] The communication network 3 may be constructed in various ways, such as the Internet, an intranet, an extranet, a LAN (Local Area Network), a CATV (Community Antenna TeleVision) communication network, a virtual private network, a telephone line network, a mobile communication network, and a satellite communication network.

[0076] In addition, various examples can be conceived as the transmission medium constituting all or part of the communication network 3. For example, wired such as IEEE (Institute of Electrical and Electronics Engineers) 1394, USB (Universal Serial Bus), power line transmission, and telephone line can be used, and wireless such as infrared such as IrDA (Infrared Data Association), Bluetooth (registered trademark), 802.11 wireless, cellular phone network, satellite line, and terrestrial digital network can be used.

[0077] The user terminal 2 can receive information (beacon signal) transmitted from the beacon transmitter 10. In addition, the user terminal 2 has software installed therein, and has a function of transmitting the beacon signal received from the beacon transmitter 10 to the recommendation server 1 via the communication network 3.

[0078] The user terminal 2 is a relatively small portable information processing terminal, and may be, for example, a small PC (Personal Computer) with communication functions, a feature phone, a PDA (Personal Digital Assistant), or a smart device such as a smartphone or tablet terminal.

[0079] The user terminal 2 receives recommendation information from the recommendation server 1 .

[0080] The recommendation server 1 is an information processing terminal such as a PC that receives location information from a user terminal 2 and transmits recommendation information to the user terminal 2 as needed.

[0081] Therefore, the recommendation server 1 includes a detection unit 1a, an accumulation unit 1b, a recommendation unit 1c, and a similarity determination unit 1d (see Figure 2 ).

[0082] The detection unit 1 a detects that the user approaches (comes to) a store such as the store 200A. The detection of the state of approaching a store is performed, for example, by transmitting a beacon signal received by the user terminal 2 (or receiving a beacon signal) to the recommendation server 1 .

[0083] In addition, by continuously detecting the approach state, it is possible to detect the user entering and exiting the store. For example, by changing from a state where the approach state to the store 200A is not detected to a state where the approach state is detected, it is detected that the user enters the store 200A. In addition, by changing from a state where the approach state to the store 200A is detected to a state where the approach state is not detected, it is detected that the user exits the store 200A. That is, exit detection is performed by detecting that the user changes from a state where the user is in a specified area to a state where the user is not in the specified area.

[0084] The beacon transmitter 10 may not be used to detect the user entering or leaving the store 200A. For example, this can be achieved by the user terminal 2 carried by the user periodically sending location information based on the GPS (Global Positioning System) to the recommendation server 1. The detection unit 1a can grasp the location relationship between the store 200A and the user by comparing the received location information with the location information of the store 200A, thereby determining whether the user has entered the store 200A. In addition, by capturing changes in location information, it is possible to grasp whether the user has entered or left the store 200A.

[0085] The detection unit 1a detects the user's behavior (purchase behavior) of purchasing goods in the store 200A. The purchase behavior is detected by acquiring purchase information from the credit card system 7, the electronic money system 8, and the point card system 9.

[0086] In addition, the purchase behavior may be detected by the user inputting information. For example, the purchase behavior may be detected by obtaining the product review posted by the user.

[0087] Alternatively, a plurality of beacon transmitters 10, 10, ... may be provided in the store 200A, and the position of the user terminal 2 in the store 200A may be detected by detecting the signal strength of the beacon signal received by the user terminal 2 from each beacon transmitter 10, 10, ... In this case, if there is a sign that the user terminal 2 is located near the cash register, it can be estimated that there is a purchase behavior. If this method is used, it is not necessary to obtain information from the credit card system 7, the electronic money system 8, and the points card system 9 in the detection of the purchase behavior, so it can be achieved by simple processing. Furthermore, the user who exits the store without moving near the cash register can be regarded as a user who has not reliably made a purchase, and the recommendation information described later can be sent. That is, useful information can be provided to users who have not made a purchase behavior through a simple system structure and processing steps.

[0088] The accumulation unit 1b accumulates the user's behavior history information. Specifically, it detects when the user comes to the store 200A and stores the visit history information. In addition, it detects when the user leaves the store 200A and stores the exit history information. Furthermore, it detects when the user makes a purchase and stores the purchase behavior history information.

[0089] The time information at which the behavior was detected is managed in association with each piece of history information.

[0090] The recommendation unit 1c transmits the recommendation information to the user terminal 2. When transmitting the recommendation information, it is roughly divided into executing extraction processing of the recommendation information, selection processing of the recommendation information, transmission processing, etc.

[0091] In the recommended information extraction process, the recommended information extraction process is performed when the detection unit 1a detects that the user has entered the store 200A or has exited the store 200A. In other words, the recommended information to be sent to the user is extracted.

[0092] The recommended information selection process is a process of selecting information to be actually transmitted from among the recommended information extracted in the extraction process.

[0093] In the transmission process, a process of transmitting the selected recommended information to the target user terminal 2 is performed.

[0094] The specific processing of each process will be described later.

[0095] The similarity determination unit 1d determines the similarity between users or determines whether they are similar. The similarity or similarity is determined based on user information of each user, or based on store information with user's store visit history or store information with purchase behavior history.

[0096] In order to execute the above-mentioned various processes, the recommendation server 1 can access a database. In addition, the "database" will be referred to as "DB (Database)" below. Figure 1 and Figure 2 , a user DB 50 , a store DB 51 , and a history DB 52 are exemplified as DBs that the recommendation server 1 can access.

[0097] The user DB 50 stores information related to users who receive recommendation information provided by the recommendation server 1. For example, personal information such as login password, name, age, gender, annual income, address, email address, and hobbies is stored in association with a user ID (Identification) that can identify a user.

[0098] In addition, in this embodiment, the user's current location information may be stored. The current location information may be received from the user terminal 2 by GPS information of the user terminal 2 or by beacon signal information received from the beacon transmitter 10 installed in the store 200 .

[0099] The store DB 51 stores information of actual stores such as the store 200. For example, the store name, location information such as GPS information or address information, product ID of the processed product or product type information (sales type information) that can identify the processed product group, contact information such as a telephone number or an email address, etc. are stored in association with the store ID that can uniquely identify each store.

[0100] The history DB 52 stores the user's store visit history information, store exit history information, and purchase behavior history information. This information is stored sequentially when the user visits the store, exits the store, and makes a purchase. Specifically, Figure 3 As shown, each record is stored in time series. Each record, for example, adopts a structure that associates date and time information, user ID, behavior category, and target store information with a history ID that can uniquely identify the record. The behavior category indicates whether the record is a record related to the user's visit to the store, a record related to the store's exit, or a record related to the purchase behavior.

[0101] In addition, this structure is just an example, and other information may be included, or some information may not be included.

[0102] in addition, Figure 3This is a diagram that extracts a part of various behavior records of two users (U09921 and U00764) from each record stored in the record DB 52. That is, the part showing the following records is extracted: user A (U09921) came to the store 200A and left the store without buying anything, then came to the store 200B and bought some goods and left the store; and user B (U00764) came to the store 200A and left the store without buying anything.

[0103] Each of the above DBs (user DB50, store DB51, and history DB52) can be implemented in any manner as long as they can be accessed by the recommendation server 1. For example, all of the DBs can be formed in a storage unit in the same system as the recommendation server 1, or part or all of the DBs can be separated and set in a computer system such as a remote location. Of course, it is not necessary for each DB to be formed in one device (such as an HDD, etc.). In addition, it is not necessary for each DB to be constituted as one DB. For example, the information stored in the history DB52 can also be stored and managed by multiple DBs (such as a DB storing history information related to store visits and store exits and a DB storing history information related to purchase behavior, etc.). The above DBs are merely examples of storage units for information associated with the processing of the implementation method in the form of one DB.

[0104] Figure 4 Show composition Figure 1 The hardware structure of the information processing device such as the terminals of the recommendation server 1, user terminal 2, credit card system 7, electronic money system 8, and point card system 9 is shown. The terminals of the recommendation server 1, user terminal 2, credit card system 7, electronic money system 8, and point card system 9 are configured by means of a computer capable of information processing and information communication. Figure 4 It can be implemented by a computer device as shown.

[0105] exist Figure 4 In the computer device, a CPU (Central Processing Unit) 101 executes various processes according to a program stored in a ROM (Read Only Memory) 102 or a program downloaded from a storage unit 108 to a RAM (Random Access Memory) 103. The RAM 103 also stores data and the like required for the CPU 101 to execute various processes.

[0106] The CPU 101 , the ROM 102 , and the RAM 103 are connected to each other via a bus 104 . An input / output interface 105 is also connected to the bus 104 .

[0107] An input unit 106 , an output unit 107 , a storage unit 108 , and a communication unit 109 are connected to the input / output interface 105 .

[0108] The input unit 106 is composed of a keyboard, a mouse, and a touch panel.

[0109] The output unit 107 is composed of a display including an LCD (Liquid Crystal Display), a CRT (Cathode Ray Tube), an organic EL (Electroluminescence) panel, a speaker, and the like.

[0110] The storage unit 108 is composed of a HDD (Hard Disk Drive) or a flash memory device, etc.

[0111] The communication unit 109 performs communication processing or inter-device communication via the communication network 3 .

[0112] A media drive 110 is connected to the input / output interface 105 as needed, and a removable medium 111 such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory is appropriately mounted to write or read information to or from the removable medium 111 .

[0113] In such a computer device, data or programs are uploaded or downloaded through communication by the communication unit 109. In addition, data or programs can be transferred via the removable medium 111.

[0114] The CPU 101 performs processing operations according to various programs, thereby executing necessary information processing and communication with terminals constituting each system such as the recommendation server 1 , the user terminal 2 , the credit card system 7 , the electronic money system 8 , and the point card system 9 .

[0115] In addition, the information processing device is not limited to Figure 4 Such a computer device may be constituted singly or a plurality of computer devices may be constituted as a system, wherein the information processing device constitutes a terminal, and the terminal constitutes each system such as the recommendation server 1, the user terminal 2, the credit card system 7, the electronic money system 8, and the point card system 9. The plurality of computer devices may be systematized through a LAN or the like, or may be remotely configured through a VPN or the like using the Internet or the like. The plurality of information processing devices may also include an information processing device as a server group (cloud) that can be used by a cloud computing service.

[0116] The functions of the recommendation server 1 are functions implemented by processing executed by the CPU 101 in the information processing device according to a program. However, all or part of the processing of each structure described below may be implemented by hardware.

[0117] In addition, when each function is realized by software, it is not necessary to realize each function by a separate program. A plurality of functions may be processed by one program, or a single function may be realized by the cooperation of a plurality of program modules.

[0118] In addition, each function may be distributed to a plurality of information processing devices. Furthermore, one function may be implemented by a plurality of information processing devices.

[0119] <2. Examples of each treatment>

[0120] The processing executed by the recommendation server 1 will be described with reference to the drawings.

[0121] <2-1. First embodiment>

[0122] In the first embodiment, referring to Figure 5 An example of sending recommended information when a user visits a store 200 will be described. In this example, when a user visits a store, an alternative store is extracted based on the behavior of other users and provided as recommended information.

[0123] In step S101, the recommendation server 1 determines whether behavior information has been received. Behavior information refers to information on a user's visit to a predetermined store 200, information on a user's exit from a store 200, and information on a user's purchase behavior in the store 200. Before receiving such behavior information, the recommendation server 1 waits in step S101.

[0124] The recommendation server 1 that has received the behavior information performs processing to store the behavior information in the history DB 52 in the next step S102. Thus, one record is stored in the history DB 52.

[0125] In step S103, the recommendation server 1 determines whether the behavior type of the stored behavior information is related to store visit. If the behavior information is other than information related to store visit, that is, information related to store exit or information related to purchase behavior, the recommendation server 1 moves to the waiting state of step S101.

[0126] On the other hand, if the stored behavior information is information related to store visits, the recommendation server 1 proceeds to step S104. In addition, in the following embodiments and examples, the case where the history information of user B visiting the store 200A is stored is described. Figure 3 The following describes a case where a record of history ID=17052958915 is stored in the history DB 52 shown and the processing of step S104 and step S105 is executed.

[0127] In step S104 , a process of extracting recommended information to be presented to user B is executed. Next, in step S105 , the recommendation server 1 executes a selection process of selecting a part of the extracted recommended information.

[0128] Reference Figure 6 and Figure 7 An example of the extraction process and selection process of recommended information will be described.

[0129] exist Figure 6 In the illustrated recommendation information extraction process, in step S201, the recommendation server 1 extracts users who visited the first store but left the store without making any purchases. Here, the extracted users may also include user B, but are basically users other than user B. The first store is the store visited by user B, who is the target of the recommendation information, that is, store 200A. This is to send information about alternative stores to user B who visited store 200A as recommendation information.

[0130] Specifically, if Figure 3 As shown in the history DB 52, one or more users (hereinafter referred to as "first extracted users") including user A who came to the store 200A but did not make any purchase and then left the store are extracted.

[0131] Next, in step S202, the recommendation server 1 further extracts (screens) users who visit the second store. Specifically, the process of further narrowing down the users who visit the second store from the first extracted users extracted in step S201 is performed. The users further narrowed down in step S202 are recorded as "second extracted users".

[0132] The second store is a store different from the first store (that is, the store 200A), and may be the store 200B or the store 200C.

[0133] Alternatively, the number of users may be narrowed down to those who visited the second store within a specified time period such as 30 minutes from the closing time of the store 200A, which is the first store. This is because, even if the user who visited the second store, for example, 24 hours after the closing time of the first store is extracted, it is difficult to determine that the user visited the second store instead of the first store. That is, the longer the time from the closing time of the first store to the visit to the second store, the lower the possibility of visiting the second store as a substitute store for the first store.

[0134] Therefore, by narrowing down to users who visited the second store within a specified time such as 30 minutes, 1 hour, or 2 hours from the exit of the first store, the possibility of sending appropriate recommendation information to user B can be increased. Of course, the specified time can also be made variable using the location information of the first store and the second store. That is, when the first store and the second store are located at a relatively close distance, the specified time can be shortened, and when the first store and the second store are located at a relatively far distance, the specified time can be extended. In this way, the possibility of being able to suggest an appropriate alternative store to user B in consideration of the location information of each store can be further increased.

[0135] In addition, it is also possible to visit or leave other stores between leaving the first store and visiting the second store. Specifically, if user A visits store 200C after leaving store 200A, but leaves without making any purchases there, and then makes a purchase in store 200B visited next, it is highly likely that the alternative store suggested to user B is not store 200C visited after store 200A but store 200B where the purchase was made. In addition, if store 200B is located near store 200A and the products sold are similar, it is highly likely that user B who left store 200A will then visit store 200B if recommendation information is not sent.

[0136] Even in such a case, according to the present structure, information about the store where the purchase was made is sent to user B as recommendation information instead of information about the store visited before the purchase was made. Therefore, user B can omit unnecessary visits to the store and use his time effectively.

[0137] After extracting the second extraction user, in step S203, the recommendation server 1 extracts one or more second stores. Since there are one or more second extraction users, the second stores extracted here are also one or more. The extracted second stores are recorded as "extracted stores". In addition, for example, for three second extraction users, the second stores may be repeated, so the extracted stores are not necessarily limited to three.

[0138] The recommended server 1 of the store is extracted and ended Figure 6 The recommended information extraction process shown is then executed Figure 7 The recommended information selection process shown ( Figure 5 Step S105).

[0139] In the recommendation information selection process, in step S301, the recommendation server 1 performs selection based on the product category, and selects only the stores whose sales category information at least partially overlaps with the store 200A as the first store from one or more extracted stores.

[0140] This is because the possibility that a store whose sales category information does not overlap with that of the first store is a substitute store for the first store is low.

[0141] Next, in step S302, the recommendation server 1 performs selection using the time information. Specifically, the case where the store 200A is a restaurant will be described as an example.

[0142] For example, assume that user C exits store 200A without buying anything (eating and drinking), and then goes to store 200C to make a purchase (eating and drinking). User C makes a purchase in store 200C around 12 o'clock, which is considered to be lunch time.

[0143] On the other hand, Figure 3 As shown, user B comes to store 200A at 7:00 AM, which is unlikely to be lunch time. Therefore, it may be inappropriate to suggest store 200C to user B as an alternative store based on the behavior history of user C.

[0144] In step S302, the process of further narrowing down the stores selected as the substitute store candidates by the use time information is performed in this way. Therefore, the store 200C is excluded from the substitute store candidates.

[0145] Next, in step S303, the recommendation server 1 selects whether the store is available. Each store 200, 200, ... may not be available. For example, a store on a fixed holiday, a store outside business hours, or a closed store. In addition, a store that is difficult to use due to a congested situation may also be excluded as an unavailable store. Since it is not desirable to present such a store 200 to the user as a substitute store, such a store 200 is excluded through the selection process of step S303.

[0146] Information such as regular holidays and business hours is stored in the store DB 51, for example.

[0147] In addition, only one or all of the selection processes of step S301 to step S303 may be performed. In addition, if the number of alternative stores to be presented to user B is set to 1, the subsequent selection process may not be performed when the number of alternative stores becomes 1. In addition, if the number of alternative stores to be presented to user B is 0, that is, there is no recommended information, all of the selection processes may be performed, and the subsequent process may not be performed and the process ends when the number of alternative stores becomes 0.

[0148] Next, in step S304, the recommendation server 1 determines whether to perform further selection processing. Specifically, it determines whether more information than the recommended information scheduled to be presented to the user B remains as an alternative store.

[0149] If no further selection process is performed, the process ends without performing step S305. Figure 7 On the other hand, when further selection processing is performed, the process proceeds to step S305.

[0150] In step S305, the recommendation server 1 further narrows down the alternative store candidates remaining as a result of the selection processes from step S301 to step S303. Specifically, a specified number of stores are selected in descending order of purchase behavior. The specified number refers to the number of alternative store information items that are recommended information presented to user B. When one alternative store is presented to the user, one alternative store is selected in descending order of purchase behavior, and when three alternative stores are presented, three alternative stores are selected in descending order of purchase behavior.

[0151] In addition, the store with many purchase behaviors refers to a store where many users visit the store instead of the first store (store 200A).

[0152] In step S305, instead of selecting a store with many purchase behaviors, selection may be made based on other information. For example, the store order may be based on the store closest to the current value of user B, or a store without user B's store visit history may be selected.

[0153] return Figure 5 The recommendation server 1 that has performed the recommendation information extraction process of step S104 and the recommendation information selection process of step S105 performs a process of determining whether the transmission timing has arrived in step S106. If the transmission timing has not arrived, the process of step S106 is continued. That is, in step S106, it waits until the transmission timing arrives.

[0154] The sending timing is a time when it is considered appropriate to send the recommendation information to the user. For example, after detecting that user B has exited the store 200A. In addition, in this embodiment, the user's store visit detection, store exit detection, and purchase behavior detection are performed, but even if the store visit detection is not performed, the time considered to be the store exit can be used as the sending timing. For example, if the average stay time of user B in the store 200A visited is 1 hour, the sending timing can also be 1 hour after the store visit detection of user B.

[0155] By performing such processing, even if only the user's store visit detection is performed, recommendation information can be sent to the user at an appropriate timing. This can reduce the processing load of the recommendation server 1 and reduce the consumption of the storage area of ​​the history DB 52, and can also send recommendation information.

[0156] The recommendation server 1 that determines that the transmission timing has arrived in step S106 performs a transmission process in step S107. In this transmission process, the substitute store information as the recommended information is transmitted to the user B as the transmission target.

[0157] In addition, if the result of various selections in the recommended information selection process of step S105 is that there is no alternative store as the recommended information, the process of step S106 and step S107 is not performed and the process ends. Figure 5 In addition, when the information to be recommended is not extracted in the recommended information extraction process of step S104, the process of step S106 and step S107 is not performed and the process ends.

[0158] In addition, Figure 7 When the congestion status of each store is obtained in step S303, the recommended information can be sent in combination with the congestion status in the sending process of step S107. Thus, the user can determine whether to use the alternative store based on the congestion status. That is, information useful to the user can be provided.

[0159] <2-2. Second embodiment>

[0160] In the second embodiment, recommendation information is sent when the user exits the store 200. Figure 8 Provide explanation.

[0161] Furthermore, a case will be described in which user B comes to store 200A, which is the first store, and then leaves the store without making any purchase.

[0162] The recommendation server 1 determines whether the behavior information has been received in step S121, and stores the received behavior information in the history DB 52 in step S122. The processing of steps S121 and S122 is similar to Figure 5 The processing of step S101 and step S102 is the same processing, so the detailed description is omitted.

[0163] In step S123, the recommendation server 1 determines whether the behavior category of the behavior record stored in the previous process is related to store withdrawal. If a behavior record other than the behavior category related to store withdrawal is received, the recommendation server 1 executes the process of step S121 again.

[0164] On the other hand, when receiving the behavior record whose behavior category is related to store closing, the recommendation server 1 performs the recommended information extraction process in step S124 and executes the recommended information selection process in step S125.

[0165] Reference Fig. 9 and Fig.10 , the recommended information extraction process of step S124 and the recommended information selection process of step S125 are explained.

[0166] In the present embodiment, the recommendation information is sent when the user quits a store. Therefore, in addition to sending information about other stores different from the quitting store, it is also possible to send information related to the store that just quit.

[0167] Here, an example in which information related to a store that has just closed down is transmitted as recommendation information will be described.

[0168] Recommended Server 1 execution Fig. 9 The process of step S221 is regarded as the recommended information extraction process. In step S221, the recommended information related to the store 200A where the user B checked out is extracted.

[0169] Examples of the information related to the store 200A include information on products that are popular in the store 200A, products whose sales are increasing rapidly, and products that the user B is interested in.

[0170] Next, the recommendation server 1 performs Fig.10 In each process after step S321, recommended information is selected from the extracted information.

[0171] First, in step S321, selection is made based on the user's interest or hobby information. The user's interest or hobby information is stored in, for example, the user DB 50. This information may be information directly input by the user, or may be estimated information extracted from the user's purchase behavior history, or may be estimated information extracted from the store information visited by the user.

[0172] Next, in step S322, the recommendation server 1 performs selection processing according to the user's moving path. For example, when a plurality of beacon transmitters 10, 10, ... are provided in the store 200A, the moving path of the user B in the store 200A can be grasped. It is conceivable to use such information to grasp the corners that the user has passed through and the corners that he has not passed through, and select the product information related to the products sold in the corners that he has not passed through. By executing the selection processing of both step S321 and step S322, the products of interest that the user B may not have noticed are selected as the recommended information. Specifically, when the store 200A is a bookstore and the works of the author that the user B likes are temporarily moved from the usual place to the special corner, the user B may not notice it and exit the store 200A. By grasping the moving path of the user B in the store 200A, such a situation is estimated, and the appropriate recommended information is selected.

[0173] The positions of the products in the store may be stored in, for example, the store DB 51. In addition, instead of understanding the positions of the products, information on which attributes of users tend to visit each corner may be used based on the movement paths and preference information of each user.

[0174] The recommendation server 1 determines whether selection processing is required in step S323, and then performs further selection processing in step S324. The processing of steps S323 and S324 is the same as the processing of steps S304 and S305 described above, and thus detailed description thereof is omitted.

[0175] return Figure 8 The recommendation server 1 that has extracted and selected the recommendation information executes a transmission process in step S126.

[0176] <2-3. Another example of recommended information extraction processing>

[0177] Reference Fig.11 right Figure 8 Another example of the recommended information extraction process of step S124 will be described.

[0178] In another example of the recommended information extraction process, the following example is described: instead of simply extracting users who visited the first store but left without making any purchases and then visited the second store and made any purchases, users who are similar to the user to whom the recommended information is to be sent are also extracted from these users.

[0179] First, in step S241, the recommendation server 1 extracts similar users. Similar users include users with similar store visit histories, users with similar store purchase behaviors, users with similar family structures, and users with similar address information (users with close addresses).

[0180] Next, in step S242 and step S243, the recommendation server 1 extracts the first extracted user and the second extracted user from these similar users. The process of step S242 is similar to the process of step S243 described above. Figure 6 The processing of step S201 is the same as that of step S243. Figure 6 The process is the same as step S202, so the detailed description is omitted.

[0181] Finally, in step S244, the recommendation server 1 extracts one or more second stores. Figure 6 The same processing as the processing of step S203.

[0182] In addition, this process is Figure 8 The processing of step S124 is described as an example, but it can also be used as Figure 5 In other words, it is also possible to perform the recommended information extraction process of step S104. Fig.11 The processing is regarded as the recommended information extraction processing performed based on the user B coming to the store 200A.

[0183] <2-4. Another example of recommended information selection processing>

[0184] Reference Fig.12 right Figure 8 Another example of the recommended information selection process in step S125 will be described.

[0185] In another example of the recommended information selection process, the recommended information is selected in consideration of weather information.

[0186] In this example, the user B exits the store 200A as an opportunity to execute Figure 8 Step S124 and other recommended information extraction processing, and then execute Fig.12 A series of processing shown.

[0187] In the recommended information selection process using weather information, in step S341 , the recommendation server 1 acquires address information of the store 200A which is the first store from the store DB 51 .

[0188] Next, in step S342, the recommendation server 1 obtains the current weather information of the store 200A. The weather information is obtained from a website that publishes weather information via the communication network 3. The weather information refers to, for example, temperature information, weather information such as sunny or rainy, and wind speed information.

[0189] In step S343, the recommendation server 1 selects a store according to the weather information from one or more stores that have been extracted. For example, if the weather information is rain, a store that can be reached from the current location of user B, that is, store 200A, without going outside the house but only moving in an underground passage or a roofed arcade is preferred. In addition, stores located in the same building are also preferred. At this time, the crowdedness of the store can also be further considered. For example, if it is known that the store is crowded and waiting, a store with a roofed waiting space can also be selected.

[0190] If it is known that the outside temperature is low and the shops are crowded, a shop where you can wait indoors may be selected. In addition, a shop may be selected based on pollen information or ultraviolet information. Moreover, it is preferred that whether such information is used is based on user information of user B. For example, whether the user is suffering from hay fever is determined based on whether the user is suffering from hay fever.

[0191] When the outside temperature is low and the shops are crowded but there is no wind, you can also choose a shop where you can wait in the sun.

[0192] In addition, when the temperature is mild and it is sunny or cloudy, the selection based on weather information may not be performed. In this case, for example, Figure 7 or Fig.10 In addition, Fig.10 In the above example, the selection based on interest / hobby information in step S321 selects product information, but here the process is to select store information. In the selection process based on the moving path in step S322, an example of selecting because it is estimated that the user missed the information of the corners that he did not pass through was described, but here the information of the corners that the user passed through is obtained, and the product information that the user is looking for is estimated based on the corners that the user passed through, so that the store selection is performed based on the estimated information.

[0193] Next, the recommendation server 1 determines in step S344 whether further selection processing is required, and in step S345 selects stores in descending order of purchase behavior. Figure 7 The processing of step S304 and step S305 is the same, so the detailed description is omitted.

[0194] In order to perform the selection process based on the weather information described in this example, information related to the weather information may also be stored in the store DB 51. For example, information such as whether the waiting area is suitable in rainy days is stored for each store. In addition, information related to the weather information may also be stored in the history DB 52. For example, the weather at that time may also be stored in the store visit record. In the case where there is less store visit history information on rainy days and more store visit history information on sunny days compared to other stores, there is a high possibility that the store suggested as recommended information on rainy days is not suitable, so it may be considered not to select it in the recommended information selection process.

[0195] <2-5. Another example 2 of recommended information extraction processing>

[0196] In the recommended information extraction process, as described above, for example, when user B comes to store 200A or leaves store 200A, recommended information about stores other than store 200A is presented to user B based on the behavior history other than user B. At this time, for example, if user A comes to store 200B after store 200A and makes a purchase, store 200B becomes a candidate for a substitute store presented to user B.

[0197] In another example 2 of the recommended information extraction process, it is considered whether user A selects store 200B and visits the store. Specifically, it is considered whether user A selects store 200B according to a search operation.

[0198] If user A visits store 200A but leaves without making any purchases and then performs a search operation to obtain information related to store 200B, it can be considered that there is a high possibility that a search operation was performed to find an alternative store other than store 200A. For user A, there is a high possibility that store 200B, the information obtained in this way, is an alternative store to store 200A. That is, for user B, there is a high possibility that store 200B is also an alternative store to store 200A. For details, refer to Fig.13 Provide explanation.

[0199] In step S261 , the recommendation server 1 extracts users who visited the store 200A, which is the first store, but left the store without making any purchase.

[0200] Next, in step S262, the recommendation server 1 narrows down the users who have visited the second store as a result of the search operation. Information on the presence or absence of the search operation is transmitted from the user terminal 2 to the recommendation server 1 and stored in the DB. For example, for each record stored in the history DB 52 related to the store visit, information on whether or not there is a search operation is stored.

[0201] In step S263, the recommendation server 1 extracts the second store. In this way, the information of the second store that the user has visited and then searched for after visiting the first store is extracted.

[0202] In addition, the recommendation information selection process performed by the recommendation server 1 may be: Figure 7 The processing shown can also be Fig.10 The processing shown can also be Fig.12 At this time, Fig.10 The processing of step S321 or step S322 shown is the processing content described in another example of the recommended information selection processing.

[0203] <3. Modifications>

[0204] In the above Figure 7 , Fig.10 as well as Fig.12 In the recommended information selection process shown, a predetermined number of information may be randomly selected from the information of each store or the information related to the product extracted by the recommended information extraction process previously executed. This can prevent the recommended information from being biased, and thus provide versatile recommended information.

[0205] It is also possible to cooperate with a POS (Point Of Sale) system to grasp the purchase behavior of the user. In this way, the purchase behavior can be grasped reliably.

[0206] In addition, the purchase behavior may also include a request to pick up the goods, a purchase reservation, and a delivery request, etc. That is, it is not limited to the behavior of paying for the goods in the store 200 and receiving the purchased goods, but also includes the situation where the goods are received in the store but the payment is made later, and the situation where the goods are delivered later but the payment is made in advance, etc.

[0207] In the above examples, various processes are performed while logging in to the recommendation information providing service provided by the recommendation server 1 in order to store the user ID information in the history DB 52, but logging in is not necessarily required. For example, at the stage where the visit behavior, exit behavior, and purchase behavior to the store 200 are understood, the recommendation server 1 may issue a temporary ID valid for several hours or one day to the user.

[0208] The recommendation server 1 transmits the provided recommendation information, which is the recommendation information related to the store 200 that has just closed down or the recommendation information of the substitute store of the closed down store 200. Therefore, although the user A who closed down from the store 200A made a purchase in the store 200B one week later, it is highly possible that it is inappropriate to estimate the store 200B as the substitute store of the store 200A. That is, it is not necessary to know whether each behavior performed in the store 200A and the store 200B in a whole week was performed by the same person.

[0209] Therefore, as long as it is possible to determine that store visits, store exits, and purchases performed within a certain short period of time such as a few hours are performed by the same person, sufficient results can be achieved even if the behavior of each user is monitored using a temporary ID that is valid for a short period of time.

[0210] The combination of the above-mentioned processes is only one method. Therefore, no matter how the processes are combined, various effects can be obtained. For example, as a recommended information selection process, Figure 7 After each selection process of steps S301, S302, and S303, it can also be executed Fig.10 In addition, the recommended information extraction process and the recommended information selection process of different examples may be combined.

[0211] In addition, in the above examples, an example of sending actual store information as recommendation information is described, but a virtual shopping mall or virtual store launched on the network may be suggested as a substitute store. A visit to a virtual shopping mall or virtual store can be processed as a visit to a store by starting to browse the page of a store launched on the network. In addition, it can be processed as a withdrawal from a store by stopping browsing the page. Furthermore, the substitute store suggested to the user in response to the withdrawal from a store may be another virtual store belonging to the virtual shopping mall, or may be an actual store not on the network.

[0212] Specifically, when user A comes to the actual store 200A but does not make a purchase, and then comes to the store 200B and makes a purchase, the recommendation server 1 accumulates this series of history information. Next, when user B, who is a similar user to user A, comes to the virtual store 200A' of store 200A opened in the virtual shopping mall (starts browsing the web page) and then exits the store (ends browsing the page), recommendation information about store 200B is sent to user B. At this time, the recommendation information sent to user B may be information related to the actual store 200B or information related to the virtual store 200B' of store 200B. Of course, information about both store 200B and store 200B' may be sent.

[0213] <4. Conclusion>

[0214] As described in the above examples, the recommendation server 1 has: a proximity detection unit (detection unit 1a), which detects the situation where the information terminal (user terminal 2) carried by the user approaches the store 200; a purchase behavior detection unit (detection unit 1a), which detects the situation where the user performs a purchase behavior in the store 200; an accumulation unit 1b, which accumulates history information related to the user's visit to the store 200 and the purchase behavior; and a recommendation unit 1c, which sends recommendation information related to the second store to the information terminal of the user who is detected to be approaching the first store when the store visit history information of the first store is detected from the history information of a user, the purchase behavior history information of the first store is not detected, and the store visit history information and purchase behavior history information of the second store accumulated after the history information related to the first store are detected.

[0215] Thus, based on the store visit history information and purchase behavior history information of other users related to the actual store, information of a store that is a substitute for the store visited by the user is provided to the user.

[0216] Therefore, for example, in the case of a restaurant, if a user who cannot dine at a first restaurant has a second restaurant where another user has dined instead of the first restaurant, the second restaurant is recommended as an alternative restaurant.

[0217] Of course, the first store or the second store is not limited to a restaurant, and may be a store that handles stationery, etc. For example, if the first store does not have the stationery you want to buy, a stationery store that can serve as a second store as an alternative store is recommended.

[0218] In addition, if the second store recommended to the user detected to be close to the first store is located too far from the user, it is likely to be inappropriate. According to this structure, a store that other users who actually came to the first store later visited and made purchases are recommended to the user as an alternative store, thereby increasing the possibility of making appropriate recommendations to the user who is near the first store.

[0219] Furthermore, by suppressing the transmission and reception of unnecessary information, it is possible to reduce the processing burden of the recommendation server 1 and the processing burden of the user terminal 2 held by the user. In addition, by providing appropriate information, the burden on the user who performs a new search, etc. is reduced, and the processing burden on the user terminal 2 during the search is suppressed.

[0220] As another example of recommended information extraction processing ( Fig.11 ), the recommendation server 1 has a similarity determination unit 1d for determining similarities between users, and the recommendation unit 1c may also send based on history information of similar users determined to be similar to the user to be sent.

[0221] Thus, the degree of similarity between the user who is detected to be approaching the first store and the user who is the extraction source of the recommended information is determined.

[0222] Therefore, it is possible to increase the possibility that an appropriate alternative store will be recommended as the second store to the user who is the target of the recommendation information (the destination of the recommendation information).

[0223] As described in the description of the detection unit 1 a , the approach detection unit (detection unit 1 a ) may detect approach to the store 200 using short-range wireless communication (for example, communication using a beacon signal).

[0224] This eliminates the need to activate, for example, a GPS function on the user terminal 2 carried by the user in order to detect the approach to the store 200 .

[0225] Most of the short-range wireless communications can communicate with low power consumption. In particular, it is low power consumption compared to always starting the GPS function. Therefore, it is possible to suppress the battery consumption of the user terminal 2 carried by the user, thereby improving the convenience of the user.

[0226] As described in the second embodiment and the like, the recommendation unit 1c may transmit the recommendation information after detecting that the user to whom the information is to be transmitted has checked out of the first store.

[0227] Thus, simply because the user comes to the first store, no alternative store is presented as recommended information.

[0228] Specifically, when a user visits the first store but leaves without making a purchase, the second store as an alternative store is provided as recommendation information. This prevents recommendation information from being sent to users who plan to make a purchase at the first store, i.e., users who do not need a recommendation for an alternative store.

[0229] As described in the description of the detection unit 1a, the detection of the purchase behavior may be performed based on the use of electronic money, the use of a credit card, or the use of a point card.

[0230] Electronic data is often processed in the use of electronic money, credit cards, and point cards, and thus, purchase behavior can be easily detected.

[0231] Therefore, while detecting store visit or store exit using GPS or short-distance wireless communication, purchasing behavior can be reliably detected, thereby preventing the transmission of useless recommendation information.

[0232] As described in the first embodiment, the recommendation unit 1c may not transmit the recommendation information related to the second store when the second store is not available.

[0233] Recommending a store that cannot be used may be disadvantageous to the user who received the notification.

[0234] Therefore, by not recommending such a second store, it is possible to improve the convenience of the user.

[0235] As in the first embodiment ( Figure 7 ), when the recommendation unit 1c extracts multiple second stores from the history information of multiple users, the store with more purchase behavior history information of the multiple users among the extracted second stores can be sent as recommendation information.

[0236] The next store where users who have not been detected to have made purchases at the first store have visited the most and made purchases is recommended to the user as the second store.

[0237] Therefore, by recommending a store that suits the purpose of the user as the second store, the possibility of providing information useful to the user can be increased.

[0238] As in the first embodiment ( Figure 7 ) as described in the above, the accumulation unit 1b may accumulate history information associated with time information of user's behavior, and the recommendation unit 1c may send recommendation information using the time information.

[0239] That is, the recommendation information is sent in consideration of the time when the user was detected to have visited the first store and the time when another user made a purchase at the second store visited after the first store.

[0240] For example, the second store that should be suggested as an alternative store at lunch time is different from the second store that should be suggested as an alternative store at dinner time. According to this structure, the second store is extracted and sent as recommendation information in consideration of time information, thereby further improving the possibility of making appropriate recommendations.

[0241] Another example of recommended information selection processing is as follows ( Fig.12 ) as described in the foregoing, the recommendation unit 1c may send the recommendation information using weather information related to the location of the user to whom the recommendation information is sent.

[0242] For example, when the second store to be transmitted as the recommendation information is crowded and there is a possibility of waiting time, the second store can be selected by considering whether the waiting time can be spent in an appropriate state.

[0243] Therefore, specifically, stores that require waiting outside on rainy days or days with low temperatures can be excluded from the second store candidates, and information with high user satisfaction can be provided. In addition, stores that can ensure a moving path that can be moved without being exposed to wind and rain on rainy days can be preferentially extracted as second store candidates.

[0244] As described in the second embodiment, the recommendation server 1 includes the detection unit 1a that detects that a user has checked out of the store 200 and the recommendation unit 1c that sends recommendation information based on information about the store 200 to the user whose checkout was detected.

[0245] In this way, after the users to whom the recommended information is to be sent are narrowed down to the users whose checkout is detected, the recommended information is sent.

[0246] Therefore, compared with the case where recommended information is transmitted only by visiting a store, it is possible to suppress unnecessary transmission of information, thereby improving the convenience and satisfaction of the user.

[0247] Furthermore, by suppressing the transmission and reception of unnecessary information, it is possible to reduce the processing burden on the server and the processing burden on the information terminal held by the user.

[0248] As described in the description of the detection unit 1 a , the detection unit 1 a may detect that the user has performed a purchase in the store 200 .

[0249] The need for recommended information is determined and filtered based on whether the user has made a purchase.

[0250] Therefore, the possibility of transmitting information useful to the user as recommended information can be increased.

[0251] As described in the description of the detection unit 1 a , the detection unit 1 a may detect the exit of the store by changing from a state where the user is detected to be in the predetermined area to a state where the user is not detected.

[0252] Therefore, the method of detecting store visits is used to detect store exits.

[0253] That is, there is no need to install a dedicated device for store exit detection in the store, which can help reduce costs.

[0254] As described in the description of the detection unit 1 a , the detection unit 1 a may detect the exit from the store 200 using short-distance wireless communication.

[0255] This eliminates the need to activate, for example, a GPS function on the user terminal 2 carried by the user in order to detect the user's exit from the store 200 .

[0256] Most of the short-range wireless communications can communicate with low power consumption. In particular, it is low power consumption compared to always starting the GPS function. Therefore, it is possible to suppress the battery consumption of the user terminal 2 carried by the user, thereby improving the convenience of the user.

[0257] As in the second embodiment ( Fig. 9 or Fig.10 ) as described in the foregoing, the recommendation unit 1c may send information related to the store 200 in which delisting is detected as recommendation information when no purchase behavior is detected in the store 200 in which delisting is detected.

[0258] The store that the user visits is likely to be a store that matches the user's interests and hobbies. Recommendation information about such a store is sent to the user.

[0259] For example, if a user visits a bookstore but leaves without making any purchases, useful information about a corner that the user has not visited can be sent as recommended information. In this case, the possibility that information known to the user is sent as recommended information is low, which can increase the possibility of providing useful information.

[0260] As described in the first embodiment and the like, when no purchase behavior is detected in the store 200 where closure is detected, the recommendation unit 1c may transmit information on an alternative store different from the store 200 where closure is detected as the recommendation information.

[0261] In a store 200 where no purchase behavior is detected despite the user's visit, there is a high possibility that the user has not achieved his or her purpose. An alternative store to such a store 200 is presented to the user.

[0262] By recommending an alternative store based on the store 200 visited by the user, the user is saved the trouble of searching for a store to be visited next, and thus highly convenient information can be provided.

[0263] As described in the first embodiment, the recommendation unit 1c may not transmit the recommendation information related to the substitute store when the substitute store is not available.

[0264] Recommending an unavailable store 200 may be disadvantageous to the user who received the notification.

[0265] Therefore, by not recommending such alternative stores, it is possible to improve user convenience.

[0266] As in the first embodiment ( Figure 7 ) as described in the above, the recommendation unit 1c may send the store 200 where more purchase behaviors are detected as recommendation information when a plurality of alternative stores are extracted.

[0267] The possibility that the store 200 in which many purchase behaviors are detected is suitable for many users is high. Such a store 200 is recommended as an alternative store to the user in whom no purchase behaviors are detected.

[0268] Therefore, by recommending a store that suits the user's purpose as an alternative store, the possibility of providing information that is useful to the user can be increased.

[0269] As in the first embodiment ( Figure 7 ) as described in the specification, the recommendation unit 1c may send the recommendation information by using the time information when the user performed the behavior.

[0270] That is, the alternative store is sent as the recommendation information in consideration of the time when the user's visit to the store 200 is detected and the time when the purchase behavior of other users in the alternative store is detected.

[0271] For example, it is considered that the alternative store to be presented at lunch time is different from the alternative store to be presented at dinner time. According to this configuration, alternative stores are extracted in consideration of time information and sent as recommendation information, thereby further increasing the possibility of making appropriate recommendations.

[0272] Another example of recommended information selection processing is as follows ( Fig.12 ) as described in the foregoing, the recommendation unit 1c may send the recommendation information using weather information related to the location of the user to whom the recommendation information is sent.

[0273] For example, when there are many alternative stores sent as recommendation information and there is a possibility of waiting time, it is possible to select and present an alternative store in consideration of whether the waiting time can be spent in an appropriate state.

[0274] Therefore, specifically, stores that require waiting outside on rainy days or days with low temperatures can be excluded from candidates for substitute stores, and information with high user satisfaction can be provided. In addition, stores that can ensure a travel path that can be moved without being exposed to wind and rain on rainy days can be preferentially extracted as candidates for substitute stores.

[0275] <5. Program and storage medium>

[0276] The program according to the embodiment is a program that causes the calculation processing device (CPU, etc.) of the recommendation server 1 to execute various processes.

[0277] The program of the embodiment causes the processing device to execute a proximity detection function for detecting that an information terminal carried by a user approaches a store.

[0278] Furthermore, the program of the embodiment causes the processing device to execute a purchase behavior detection function that detects that a user has performed a purchase behavior in a store.

[0279] Furthermore, the program of the embodiment causes the processing device to execute a storage function for storing history information related to the user's visit to a store and purchase behavior.

[0280] In addition, when store visit history information of a first store is detected from the history information of a user and the purchase behavior history information of the first store is not detected, and store visit history information and purchase behavior history information of a second store accumulated after the history information related to the first store are detected, the program of the implementation method causes the computing processing device to execute a recommendation function, which sends recommendation information related to the second store to the information terminal of the user who is detected to be close to the first store.

[0281] That is, the program causes the arithmetic processing device of the information processing device to execute Figures 5 to 13 The procedures for each process are shown.

[0282] By using such a program, one or more information processing devices serving as the recommendation server 1 can be realized.

[0283] Such a program can be stored in advance in a HDD as a storage medium built into a computer device or the like, or in a ROM in a microcomputer having a CPU. Alternatively, it can be temporarily or permanently stored (stored) in a removable storage medium such as a semiconductor memory, a memory card, an optical disk, a magneto-optical disk, or a magnetic disk. In addition, such a removable storage medium can be provided as so-called packaged software.

[0284] Such a program can be installed from a removable storage medium into a personal computer or the like, or can be downloaded from a download site via a network such as a LAN or the Internet.

[0285] Description of symbols

[0286] 1: recommendation server; 1a: detection unit; 1b: accumulation unit; 1c: recommendation unit; 1d: similarity determination unit; 2: user terminal; 3: communication network; 4: credit card terminal; 5: electronic money terminal; 6: point card terminal; 7: credit card system; 8: electronic money system; 9: point card system; 10: beacon transmitter; 50: user DB; 51: store DB; 52: history DB; 200, 200A, 200B, 200C: stores.

Claims

1. An information processing device, comprising: a proximity detection unit for detecting the entry and exit of a store by detecting the state of an information terminal carried by a user approaching the store; The purchase behavior detection department detects the purchase behavior of users in the store; an accumulation unit for accumulating history information related to the user's visit to the store and the purchase behavior in association with information on the time when the user performed the behavior; as well as A recommendation unit, which, when store visit history information of a first store is detected from a user's history information, and purchase behavior history information of the first store is not detected, and store visit history information and purchase behavior history information of a second store are detected, and it is detected that the visit to the second store where the purchase behavior of the one user was performed was within a specified time from a withdrawal from the first store where the purchase behavior of the one user was not detected, sends recommendation information related to the second store to an information terminal of a user different from the one user who was detected to have visited the first store, wherein the second store is a substitute store and the store visit history information and purchase behavior history information of the second store are accumulated after the withdrawal history information related to the first store, When a plurality of the second stores are extracted from the history information of a plurality of users, the recommendation unit transmits, as the recommendation information, a store having more purchase behavior history information of the plurality of users among the extracted second stores.

2. The information processing device according to claim 1, wherein: The information processing device includes a similarity determination unit that determines similarity between users. The recommendation unit performs the transmission based on history information of a similar user determined to be similar to a user to which the transmission is to be directed.

3. The information processing device according to claim 1, wherein: The proximity detection unit detects approach to a store using short-range wireless communication.

4. The information processing device according to claim 1, wherein: The recommendation unit transmits the recommendation information after detecting that the user to whom the recommendation information is to be transmitted has checked out of the first store.

5. The information processing device according to claim 1, wherein: The purchase behavior is detected based on the use of electronic money, the use of a credit card, or the use of a point card.

6. The information processing device according to claim 1, wherein: When the second store is not available, the recommendation unit does not transmit the recommendation information related to the second store.

7. The information processing device according to claim 1, wherein: The recommendation unit transmits the recommendation information at a predetermined timing using the time information.

8. The information processing device according to claim 1, wherein: The recommendation unit transmits the recommendation information using weather information related to a location of a user to which the recommendation information is to be transmitted.

9. An information processing method, comprising: performing the following steps by an information processing device: a proximity detection step, detecting the entry and exit of the store by detecting the state of the information terminal carried by the user approaching the store; The purchasing behavior detection step detects the user's purchasing behavior in the store; An accumulation step of accumulating history information related to the user's visit to the store and the purchase behavior in association with the time information of the user's behavior; as well as A recommendation step, in which, when store visit history information of a first store is detected from the history information of a user, and purchase behavior history information of the first store is not detected, and store visit history information and purchase behavior history information of a second store are detected, and it is detected that the visit to the second store where the purchase behavior of the one user was performed was within a specified time from the exit from the first store where the purchase behavior of the one user was not detected, recommendation information related to the second store is sent to the information terminal of a user different from the one user who was detected to have visited the first store, wherein the second store is used as a substitute store, and the store visit history information and purchase behavior history information of the second store are accumulated after the exit history information related to the first store, In the recommendation step, when a plurality of the second stores are extracted from the history information of a plurality of users, a store having more purchase behavior history information of the plurality of users among the extracted second stores is transmitted as the recommendation information.

10. A computer-readable storage medium storing a program that enables a computer to perform the following functions: Proximity detection function, which detects the entry and exit of a store by detecting the state of the information terminal carried by the user approaching the store; Purchase behavior detection function, which detects the user's purchase behavior in the store; An accumulation function for accumulating history information related to the user's visit to the store and the purchase behavior in association with information on the time when the user's behavior occurred; and A recommendation function, in which, when store visit history information of a first store is detected from a user's store history information, and purchase behavior history information of the first store is not detected, and store visit history information and purchase behavior history information of a second store are detected and it is detected that the visit of the one user to the second store where the purchase behavior was performed was within a specified time from the exit of the first store where the purchase behavior of the one user was not detected, recommendation information related to the second store is sent to the information terminal of a user different from the one user who is detected to have visited the first store, wherein: The second store is a substitute store, and the store visit history information and purchase behavior history information of the second store are accumulated after the store exit history information related to the first store. In the recommendation function, when a plurality of the second stores are extracted from the history information of a plurality of users, a store having more purchase behavior history information of the plurality of users among the extracted second stores is transmitted as the recommendation information.

Images