JP2023159525A - Push notification system and method - Google Patents

Abstract

To solve the problem in which an appropriate notification is unable to be sent to a user at an appropriate timing.SOLUTION: A system identifies, regarding each of a plurality of activities each associated with a smart device that detects actions of a user, a time pattern for the activities of the user based on one or more time periods when the detection is performed by the smart device associated with the activities. The system determines a notification and a timing to send the notification based on the identified time pattern regarding the activities, and a current time. The system sends the determined notification to the user at the determined timing to send.SELECTED DRAWING: Figure 1

Description

TECHNICAL FIELD The present invention relates generally to push notifications.

As a technique related to push notifications, a technique disclosed in Patent Document 1 is known. Patent Document 1 discloses a smart notification technology using a calendar, and for example, in paragraph 0003, it states that "Notifications are delivered to a user based in part on a user's current context. The notifications may be based on personal information ( e.g. calendar events, tasks, alerts, and the like) as well as a context of a user. For example, instead of delivering a notification just using a predetermined time, a reminder notification that is based on a task to buy milk may be delivered to a user in response to the user being located near a grocery store. Notifications may also be sent to other users (e.g. that are part of a trusted group) in response to a task being completed by another group member. Notifications for a sequence of tasks are adjusted based on the current user context. Instead of the user viewing all of the tasks at one time, the task notifications are delivered to the user when the task may be performed. For example, after one task is performed, the user may receive a notification that there is another task to perform near their current location. A group member may also create a messageg to another member requesting an action. When a response is made, the requesting group member receives a notification and the other group members may also receive notifications.”

US2014/0372525

According to Patent Document 1, context sensitivity is limited. For example, the user's location information can be known, but the actual status of the user's activities cannot be known. For this reason, it is not possible to send appropriate notifications to users at appropriate timings. As a result, opportunities for engagement with users may be lost.

For each of a plurality of activities each associated with a smart device that detects the user's behavior, the system determines the user's behavior based on one or more times when the smart device associated with the activity was detected. Identify the temporal pattern of the activity. The system determines the notification and the timing for sending the notification based on the time pattern identified for the activity and the current time. The system transmits the determined notification to the user at the determined transmission timing.

According to the present invention, notification engagement can be optimized.

1 shows an overview of a system according to an embodiment of the present invention. 1 shows an example of the configuration of a system according to an embodiment of the present invention. An example of the configuration of a device table is shown below. An example of the structure of an activity table is shown. An example of the structure of a time table is shown. An example of the configuration of a schedule table is shown below. An example of the structure of a message table is shown below. An example of the configuration of a notification history table is shown. An example of the structure of a response history table is shown below. An example of the configuration of an evaluation table is shown. 3 shows a part of the flow of processing performed in an embodiment of the present invention. 3 shows the rest of the flow of processing performed in one embodiment of the present invention.

In the following description, an "interface device" may be one or more interface devices. The one or more interface devices may be at least one of the following:
- One or more I/O (Input/Output) interface devices. The I/O (Input/Output) interface device is an interface device for at least one of an I/O device and a remote display computer. The I/O interface device for the display computer may be a communication interface device. The at least one I/O device may be a user interface device, eg, an input device such as a keyboard and pointing device, or an output device such as a display device.
- One or more communication interface devices. The one or more communication interface devices may be one or more of the same type of communication interface device (for example, one or more NICs (Network Interface Cards)) or two or more different types of communication interface devices (for example, one or more NICs (Network Interface Cards)). It may also be an HBA (Host Bus Adapter).

Also, in the following description, "memory" refers to one or more memory devices, which may typically be a main storage device. At least one memory device in the memory may be a volatile memory device or a non-volatile memory device.

Also, in the following description, "persistent storage" refers to one or more persistent storage devices. The persistent storage device is typically a nonvolatile storage device (eg, auxiliary storage device), and specifically, for example, an HDD (Hard Disk Drive) or an SSD (Solid State Drive).

Furthermore, in the following description, a "storage device" may be at least a memory and a persistent storage device.

Also, in the following description, a "processor" refers to one or more processor devices. The at least one processor device is typically a microprocessor device such as a CPU (Central Processing Unit), but may be another type of processor device such as a GPU (Graphics Processing Unit). At least one processor device may be single-core or multi-core. The at least one processor device may be a processor core. The at least one processor device may be a broadly defined processor device such as a hardware circuit (for example, an FPGA (Field-Programmable Gate Array) or an ASIC (Application Specific Integrated Circuit)) that performs part or all of the processing.

In addition, in the following explanation, information such as "xxx table" may be used to explain information that can be obtained from an input in response to an input, but the information may be data of any structure, and the information can be expressed as It may also be a learning model such as a neural network that generates. Therefore, the "xxx table" can be called "xxx information." In addition, in the following explanation, the configuration of each table is an example, and one table may be divided into two or more tables, or all or part of two or more tables may be one table. It's okay.

In addition, in the following explanation, functions may be explained using the expression "kkk part", but functions may be realized by one or more computer programs being executed by a processor, or one or more computer programs may be executed by a processor. It may be realized by the above hardware circuit (for example, FPGA or ASIC). When a function is realized by a program being executed by a processor, the specified processing is performed using a storage device and/or an interface device as appropriate, so the function may be implemented as at least a part of the processor. good. A process described using a function as a subject may be a process performed by a processor or a device having the processor. Programs may be installed from program source. The program source may be, for example, a program distribution computer or a computer-readable recording medium (for example, a non-temporary recording medium). The description of each function is an example, and a plurality of functions may be combined into one function, or one function may be divided into a plurality of functions.

In addition, in the following description, common parts of reference signs may be used when the same type of elements are described without distinguishing them, and reference signs may be used when the same type of elements are described separately. .

An embodiment of the present invention will be described below.

FIG. 1 shows an overview of a system according to an embodiment of the present invention.

One purpose of this embodiment is to guide target consumers to the ES (entrance showcase) 20. A "target consumer" is an example of a user, and is one target among a plurality of consumers. ES20 is a smart device installed in a shared space of a building. The ES 20 is an example of a smart device, and displays information representing a plurality of products, and receives orders for products desired by the user or feedback from consumers. A "product" may be a product or a service.

There is a smart appliance 5 in the target consumer's residence 11. The smart appliance 5 is an example of a smart device. Examples of the smart appliances 5 include an alarm clock 5A, a hair dryer 5B, a bedroom light 5C, a washing machine 5D, and a refrigerator 5E.

When the target consumer performs an action such as approaching or operating the smart device (S1), the smart device detects the action and sends detection information including information representing the time of detection to the notification server 40. (S2).

The notification server 40 is an example of a push notification system. The notification server 40 performs activity classification (S3) and time recording (S4) every time it receives target consumer detection information from any of the smart devices such as the smart appliance 5 and the ES 20. In the activity classification, the notification server 40 classifies the activity corresponding to the source smart device of the detection information based on the device table 201 having information regarding smart devices and the activity table 202 representing the relationship between the smart appliance 5 and the activity. Identify. In time recording, the notification server 40 records the time represented by the detection information in the time table 203 for the identified activity.

After a certain period of time (for example, one week or more) has passed, information representing time for each activity is accumulated in the time table 203.

The notification server 40 constructs a routine based on the time table 203 (S5). In the routine construction, the notification server 40 specifies, for each activity, the time pattern of the routine of the activity (the time pattern of the routine in a fixed period). The routine time pattern may be a statistic of the time of the activity, such as an average time or an average time period, and may also include days of the week and dates. The fixed period may be any period such as daily, weekly, or monthly. The notification server 40 records the routine construction results (information representing the time pattern for each activity) in the schedule table 204. As a result, schedule table 204 represents a time pattern for each activity for the target consumer.

The notification server 40 determines the notification (message) and the transmission timing of the notification based on the schedule table 204, the current time, and the message table 205 representing messages for each activity (S6). The notification server 40 transmits the determined notification to the target consumer at the determined transmission timing (S7). This notification is sent as a push notification. The push notification may be sent by email or via an app installed on the consumer terminal 10.

The notification is displayed on the target consumer's consumer terminal 10. The consumer terminal 10 is a consumer's information processing terminal (typically a mobile terminal), and may be a smartphone, for example.

Assume that the target consumer responds to the notification. The response includes a predetermined operation on the consumer terminal 10 (for example, a button operation indicating approval or rejection of a notification) and a predetermined action on the ES 20 (for example, when the target consumer is detected by the sensor of the ES 20). may approach ES20).

When such a response is made, response information that is information related to the response is input to the notification server 40 (S8). The response information includes the ID of the notification (message) to which the response was made, information regarding the content of the response (for example, information representing the operated button or information representing the target consumer's behavior toward the ES20), and the response to the notification (message). The information may include information representing the time when the The notification server 40 records information regarding the presence or absence of a response to the notification sent in S7 (for example, information regarding whether response information was input within a certain period of time after the notification was sent) in the response history table 207. (S9).

The notification server 40 performs an evaluation process that includes evaluating the engagement of the sent notification for each sent notification based on the response history table 207 (S10). The notification server 40 records the results of the evaluation process in the evaluation table 208.

The evaluation process may include machine learning described below. Information 209 (eg, a machine learning model such as a neural network, or parameters) according to the results of machine learning may be used for the decision at S6 (eg, deciding when to send a notification).

According to S1 to S7 described above, for example, the following can be expected. That is, the target consumer uses the smart appliance 5 in the residence 11 in his/her daily life. The frequency of use varies depending on the type of smart appliance 5. For example, the alarm clock 5A may be used every day, and the washing machine 5D may not be used every day, but every week. For each smart appliance 5, an operation (for example, power on or power off) on the smart appliance 5 is detected, and a history of the detection (operation) is accumulated in the time table 203. Similarly, the target consumer's behavior toward the ES 20 is also detected, and a history of detection (behavior) is accumulated in the time table 203. Based on the time table 203, a time pattern of activities associated with the smart device, ie, activities in the target consumer's life, is identified. Based on the time pattern identified for the activity and the current time, a notification and its sending timing are determined, and the determined notification is sent to the target consumer at the determined timing. This optimizes notifications and their timing, which in turn optimizes notification engagement.

Further, according to S8 to S10 described above, the following can be expected, for example. In other words, notification engagement is evaluated based on the presence or absence of a response to the notification. Based on this evaluation result, it is possible to understand the engagement effect of the notification, and thereby, for example, it is possible to improve the notification server 40 or feed back the engagement effect to a supplier (for example, a retail store).

This embodiment will be described in detail below.

FIG. 2 shows an example of the configuration of a system according to this embodiment.

Smart appliance 5 has one or more sensors 51 and a controller 52 connected thereto. The controller 52 transmits detection information including information representing the operation or action (operation or action by the consumer) detected using the sensor 51 or the like to the notification server 40 via the communication network 41A.

The ES 20 includes one or more cameras 61, one or more separate sensors 62, and a controller 63 connected to them. The controller 63 transmits detection information including information representing the operation or action (operation or action by the consumer) detected using the camera 61, another sensor 62, etc. to the notification server 40 via the communication network 41B. do.

Consumer terminal 10 includes a communication interface device 71, an I/O interface device 72, a storage device 74, and a processor 75 connected thereto. Communication interface device 71 and I/O interface device 72 are examples of interface devices. A UI (User Interface) device 73 is connected to the I/O interface device 72 . The UI device 73 may be an input device and a display device, or may be a touch panel in which they are integrated. A notification is received from the notification server 40 via the communication network 41C, and the notification is stored in the storage device 74. Processor 75 displays the notification on UI device 73. The UI device 73 receives a click (tap) as an example of a response to the notification. The processor 75 transmits response information representing the response to the notification server 40.

Notification server 40 includes communication interface devices 81A-81C, a storage device 82, and a processor 83 connected thereto.

Communication interface devices 81A-81C communicate via communication networks 41A-41C, respectively. Two or more communication networks 41 among the communication networks 41A to 41C may be the same communication network, and similarly, two or more communication interface devices 81 among the communication interface devices 81A to 81C may be the same communication interface device 81. But that's fine. The communication network 41 may be a communication network such as the Internet or a WAN (Wide Area Network).

Storage device 82 stores information and computer programs executed by processor 83. The information stored in the storage device 82 is, for example, a device table 201, an activity table 202, a time table 203, a schedule table 204, a message table 205, a notification history table 206, a response history table 207, and an evaluation table 208. Tables 201-208 may exist for each consumer, and the illustrated tables 201-208 may correspond to target consumers.

A processor 83 that executes a computer program realizes functions such as an activity classification section 211, a pattern identification section 212, a notification determination section 213, a notification transmission section 214, and a notification evaluation section 215.

FIG. 3 shows a configuration example of the device table 201.

The device table 201 is a table in which smart device information regarding target consumers is registered. The device table 201 has records for each smart device. Each record has information such as a device ID 301 and a device type 302.

Device ID 301 represents the ID of a smart device. If there are multiple smart devices of the same type, each smart device has a device ID. Device type 302 represents the type of smart device.

FIG. 4 shows an example of the structure of the activity table 202.

The activity table 202 has a record for each activity. Each record has information such as an activity ID 401, a device ID 402, and an activity name 403.

Activity ID 401 represents an ID of an activity. Device ID 402 represents the ID of a smart device. Activity name 403 represents the name of the activity. According to FIG. 4, activities are associated with smart devices. The correspondence between activities and smart devices may be 1:1 or n:1 (n is an integer of 2 or more).

FIG. 5 shows an example of the configuration of the time table 203.

Time table 203 represents the times at which the target consumer performed the activity. The time table 203 has records for each activity. Each record has information such as an activity ID 401 and a time set 502.

Activity ID 501 represents an ID of an activity. The time set 502 represents the time (time when the activity was performed) for each day of the week. The "time" here may be a time or a period (for example, a set of a start time and an end time). Furthermore, the unit of "time" may include other units (for example, seconds) in addition to hours and minutes.

FIG. 6 shows a configuration example of the schedule table 204.

Schedule table 204 represents routine times for the target consumer's activities. The schedule table 204 has a record for each activity. Each record has information such as an activity ID 601 and an average time 602.

Activity ID 601 represents an ID of an activity. Average time 602 represents the average time the activity took place. The "time" here may also be a time or a period (for example, a set of a start time and an end time). Moreover, "time" may include other units (for example, seconds or days of the week) in addition to hours and minutes. The number of average times associated with one activity may be one or more.

FIG. 7 shows an example of the structure of the message table 205.

Message table 205 represents messages (notifications) that may be sent. Message table 205 has a record for each message. Each record has information such as a message ID 701, an activity ID 702, a keyword 703, and a message 704.

Message ID 701 represents the ID of the message. Activity ID 702 represents the ID of the activity. Keyword 703 represents one or more keywords. Message 704 represents a text sentence as a message.

According to FIG. 7, a message is associated with at least one of an activity and a keyword. Additionally, keywords are associated with the activities.

The keywords and messages may be keywords and messages prepared by the supplier (eg, retail store). For example, for an activity corresponding to hair dryer 5B (activity ID "2"), a retailer or manufacturer of hair dryer "AAA" registers "hair style" etc. as a keyword and sends a message "Looking to try new". hairstyles? Try out new AAA hair dryer! Find it in ES!” Preferably, the message includes a guide to the ES 20.

The number of keywords associated with one message may be one or more.

The number of messages associated with one activity may be one or more. For the identified activity, the messages determined to be sent may be all or some of the messages associated with the activity.

FIG. 8 shows a configuration example of the notification history table 206.

The notification history table 206 represents the history of notifications. The notification history table 206 has a record for each notification. Each record has information such as a notification ID 801, date and time 802, and keyword 803.

The notification ID 801 represents the ID of the notification. The date and time 802 represents the date and time when the notification (message) was sent. Keyword 803 represents a keyword corresponding to the sent notification.

In place of or in addition to the keyword, the ID of the activity may be included in the record. Furthermore, the message ID of the transmitted message may be included in the record in order to associate information indicating which message has been transmitted.

FIG. 9 shows a configuration example of the response history table 207.

The response history table 207 represents the response history. The response history table 207 has a record for each response. Each record has information such as a response ID 901, date and time 902, keyword 903, click 904, and visit ES 905.

Response ID 901 represents the ID of the response. The date and time 902 represents the date and time when a response was made (for example, if both a click and a visit ES were made, the earlier, later, or both dates and times). A keyword 903 represents a keyword corresponding to a response.

Click 904 indicates whether or not a click has been made. “YES” means that the notification was clicked (for example, within a certain period of time) after the notification was issued. “NO” means that the notification was not clicked (for example, within a certain period of time) after the notification was issued.

Visit ES 905 indicates whether the ES 20 has been visited. “YES” means that the ES 20 has been visited (for example, within a certain period of time) after being notified. “NO” means that the ES 20 has not been visited (for example, within a certain period of time) after the notification was made.

A response record is associated with a notification record (a record in the notification history table 206) corresponding to the response. For example, the response ID 901, date and time 902, and keyword 903 may be the same as the notification ID 801, date and time 802, and keyword 803 in the notification record corresponding to the response. Alternatively, instead of or in addition to the response ID, the ID of the notification corresponding to the response may be included in the record. Furthermore, the ID of the message as the notification may be included in the record.

FIG. 10 shows a configuration example of the evaluation table 208.

Rating table 208 represents ratings of notification engagement. The evaluation table 208 has a record for each response. Each record has information such as response ID 1001, date and time 1002, keyword 1003, click 1004, visit ES 1005, and evaluation 1006.

The response ID 1001, date and time 1002, keyword 1003, click 1004, and visit ES 1005 are the same as the information 901 to 905 described above. Evaluation 1006 represents the evaluation result (eg, "GOOD" or "BAD"). In this embodiment, the evaluation 1006 is "GOOD" when both the click 1004 and the visit ES 1005 are "YES", and the evaluation 1006 is "BAD" when the click 1004 or the visit ES 1005 is "NO".

11 and 12 show the flow of processing performed in this embodiment.

The processes performed in this embodiment are broadly classified into processes 1 to 5.

<Processing 1>

Process 1 includes S111 to S117, and occurs each time the target consumer acts on a smart device.

For example, when the target consumer operates the smart appliance 5 in the residence 11 (for example, turns on the power), the smart appliance 5 detects the operation (S111). The smart appliance 5 transmits detection information representing the detected behavior to the notification server 40. Also, for example, when the target consumer walks in the lobby of a building and approaches the ES 20, the ES 20 detects the proximity (S112). The ES 20 transmits detection information representing the detected proximity to the notification server 40. When the notification server 40 receives detection information from the smart appliance 5 or ES 20, the activity classification unit 211 stores the received detection information (S113). The detection information may include information representing the type of smart device. The activity classification unit 211 specifies the device ID corresponding to the type of the smart device that sent the detection information from the device table 201, and specifies the activity corresponding to the specified device ID from the activity table 202 (S114).

For example, when the target consumer switches off the smart appliance 5 operated in S111, the smart appliance 5 detects the switch off (S115). The smart appliance 5 transmits detection information representing the detected switch-off to the notification server 40. Further, for example, when the target consumer leaves the ES 20, the ES 20 detects the loss of the person (S116). The ES 20 transmits detection information representing the detected loss to the notification server 40. The detection information in this paragraph is information that means that the target consumer will stop the activity. When the notification server 40 receives such detection information from the smart appliance 5 or the ES 20, the activity classification unit 211 uses the time indicated by the received detection information and the time indicated by the detection information stored in S113 to The time for the activity specified in S114 is specified, and the specified time is recorded in the time table 203 (S117).

<Processing 2>

Process 2 includes S121 and S122. Process 2 is started periodically (eg, weekly).

The pattern identifying unit 212 starts processing 2. The pattern identifying unit 212 identifies a time pattern (average time in this embodiment) for each activity from the time table 203 (S121). The pattern specifying unit 212 records the specified time pattern in the schedule table 204 for each activity (S122). The pattern identifying unit 212 ends processing 2.

<Processing 3>

Process 3 includes S131 to S136. For example, process 3 is started periodically (eg, every minute).

The notification determining unit 213 refers to the schedule table 204 and determines whether or not the time obtained by subtracting x minutes (x is an arbitrary value) from the average time 602 (for example, start time) corresponds to the current time T for any activity. is determined (S131). If the determination result in S131 is false (S131: No), Process 3 ends (S131 is started in the next cycle).

If the determination result in S131 is true (S131: Yes), the notification determining unit 213 determines a message to be notified from the message table 205 (S132). The determined message may be at least one of (a) and (b) below.
(a) S131: Message associated with the corresponding activity. Thereby, it is possible to send a message (notification) suitable for the corresponding activity in S131: Yes. Note that if there are multiple messages corresponding to the activity, all or some of the messages may be the messages to be notified.
(b) S131: Yes, a message containing a keyword corresponding to the corresponding activity. Thereby, in addition to the message suitable for the activity corresponding to S131: Yes, it is also possible to send a message suitable for the activity related to the activity. Note that if there are multiple messages including the keyword corresponding to the activity in S131: Yes, all or some of the messages may be the messages to be notified.

The notification determining unit 213 determines whether or not the time obtained by subtracting y minutes (y is a natural value smaller than x) from the average time 602 (for example, the start time) corresponds to the current time T for the activity corresponding to S131: Yes. (S133).

If the determination result in S133 is true (S133: Yes), the notification transmitter 214 transmits the message determined in S132 to the target consumer (the target consumer's consumer terminal 10) (S134), and The transmission date and time, keywords, etc. are recorded in the notification history table 206 (S135).

The message may be a message regarding guidance to the ES 20. The transmission timing of the message is a time earlier than the average time 602 of the activity corresponding to the message. Therefore, an appropriate message regarding the activity can be given to the target consumer before the target consumer starts the activity, thereby motivating the target consumer to go to the ES20.

Note that in the example shown in FIG. 12, the message transmission timing for all activities is y minutes before the average time 602, but the transmission timing may differ depending on the activity (for example, the value of y is y minutes before the average time 602). may be different depending on the Furthermore, the transmission time may be determined in advance for at least one activity based on the average time 602.

<Processing 4>

Process 4 includes S141 to S143. Process 4 is performed when the target consumer responds to the transmitted message (notification).

Assume that the consumer terminal 10 receives a click (tap) on a message (S141). In this case, the consumer terminal 10 transmits response information representing the click (tap) to the notification server 40. Further, it is assumed that the ES 20 detects the proximity of the target consumer based on information from the camera 61 of the ES 20 or another sensor 62 (S142). In this case, the ES 20 transmits response information representing the proximity to the notification server 40. The response information includes the ID of the notification (message) to which the response was made. The notification evaluation unit 215 receives the response information, uses the response information to identify the notification from the notification history table 206, and records information regarding the response for the specified notification in the response history table 207 (S143).

<Processing 5>

Process 5 includes S151. For example, process 5 is started periodically (eg, weekly).

The notification evaluation unit 215 performs an evaluation process that includes evaluating the engagement of the sent notification for each sent notification based on the response history table 207 (S151).

For each notification sent, the presence or absence of a response is determined by click 1004 (whether the target consumer clicked (tapped) on the consumer terminal 10 on which the notification was displayed) and visit ES 1005 (target consumption by ES 20). whether the proximity (visit) of a person is detected). The notification evaluation unit 215 sets the evaluation 1006 to "GOOD" for notifications where both the click 1004 and the visit ES 1005 are "YES", and sets the evaluation 1006 to "BAD" for notifications where the click 1004 or the visit ES 1005 is "NO". . The click 1004 "YES" and the visit ES 1005 "YES" mean an estimation that the visit to the ES 20 was made due to the motivation given to the target consumer by the message sent. If such an estimate is obtained, it can be interpreted (evaluated) that the notification has an engagement effect. Note that the target consumer's residence 11 is a room in a building (typically an apartment), and the ES 20 may be installed in a shared space of the building.

Furthermore, the evaluation process in S151 may include machine learning. Machine learning uses the timing of sending notifications corresponding to each activity (for example, notifications that include keywords corresponding to the activity) and the evaluation results of the notifications corresponding to the activity. It is sufficient to learn the optimal transmission timing based on the time pattern (for example, the value of y in S133). For example, a machine learning model may be a model that inputs a value representing an activity or keyword and outputs a value representing transmission timing. The model may be a neural network or another type of model. In the machine learning of the model, a value representing an activity or keyword corresponding to a notification (message) whose evaluation 1006 is "GOOD" and the transmission timing of the notification may be used as training data. The information 209 (see FIG. 1) according to the results of such machine learning may include a learned model, or may include the value of y obtained using the model. Since the results of machine learning are reflected in determining the sending timing, the notification (message) can be sent at a more appropriate timing (timing that increases the possibility of guiding the target consumer to the ES20).

Although one embodiment has been described above, this is an illustration for explaining the present invention, and is not intended to limit the scope of the present invention only to this embodiment. The invention can also be implemented in various other forms. For example, the specified time pattern may be a time pattern other than the routine time pattern.

40...Notification server

Claims (14)

For each of a plurality of activities to which a smart device that detects a user's action is associated, the activity of the user is determined based on one or more times when the smart device associated with the activity detects the activity. a pattern identification unit that identifies a time pattern;
a notification determination unit that determines a notification and a transmission timing of the notification for the activity based on the time pattern specified for the activity and the current time;
A push notification system comprising: a notification transmitter that transmits a determined notification to the user at a determined transmission timing. a notification evaluation unit that performs evaluation processing including evaluating engagement of the notification based on the presence or absence of the user's response to the notification;
The push notification system according to claim 1, further comprising: One of the smart devices is an entrance showcase installed in a building's common space.
The entrance showcase is a smart device that displays information representing a plurality of products and receives orders or feedback for products desired by the user from the user;
For each notification sent, the presence or absence of a response from the user is determined by whether or not the user performed a predetermined operation on the information processing terminal on which the notification was displayed, and whether the action by the user was detected by the entrance showcase. including whether or not
The push notification system according to claim 2. The evaluation process includes machine learning,
The machine learning is to learn, for each activity, the optimal transmission timing based on the time pattern of the activity, based on the notification transmission timing and the notification evaluation result.
The push notification system according to claim 2. the one or more smart devices are one or more appliances installed in the user's residence;
The push notification system according to claim 1. one of the smart devices other than the one or more appliances is an entrance showcase installed in a common space of the building;
The time pattern for each activity is a routine time pattern,
The entrance showcase is a smart device that displays information representing a plurality of products and receives orders or feedback for products desired by the user from the user;
for at least one activity,
The notification includes a message regarding guidance to the entrance showcase,
The transmission timing is a time earlier than the routine time pattern of the activity;
The push notification system according to claim 5. For each activity, one or more keywords are associated with the activity,
The notification determined for the activity is a message containing a keyword associated with the activity;
The push notification system according to claim 1. The computer detects the user's behavior based on one or more times when the smart device associated with the activity is detected for each of the plurality of activities, each of which is associated with a smart device that detects the user's behavior. identify the temporal pattern of such activity;
determining, by the computer, a notification and the transmission timing of the notification for the activity based on the time pattern identified for the activity and the current time;
Sending the determined notification to the user at the determined transmission timing by the computer;
Push notification method. using a computer to perform an evaluation process including evaluating engagement of the notification based on the presence or absence of the user's response to the notification;
The push notification method according to claim 8. One of the smart devices is an entrance showcase installed in a building's common space.
The entrance showcase is a smart device that displays information representing a plurality of products and receives orders or feedback for products desired by the user from the user;
For each notification sent, the presence or absence of a response from the user is determined by whether or not the user performed a predetermined operation on the information processing terminal on which the notification was displayed, and whether the action by the user was detected by the entrance showcase. including whether or not
The push notification method according to claim 9. The evaluation process includes machine learning,
The machine learning is to learn, for each activity, the optimal transmission timing based on the time pattern of the activity, based on the notification transmission timing and the notification evaluation result.
The push notification method according to claim 9. the one or more smart devices are one or more appliances installed in the user's residence;
The push notification method according to claim 8. one of the smart devices other than the one or more appliances is an entrance showcase installed in a common space of the building;
The time pattern for each activity is a routine time pattern,
The entrance showcase is a smart device that displays information representing a plurality of products and receives orders or feedback for products desired by the user from the user;
for at least one activity,
The notification includes a message regarding guidance to the entrance showcase,
The transmission timing is a time earlier than the time pattern of the activity;
The push notification method according to claim 12. For each activity, one or more keywords are associated with the activity,
The notification determined for the activity is a message containing a keyword associated with the activity;
The push notification method according to claim 8.

Images