JP2020155002A - Terminal, server device, method and program - Google Patents

Abstract

To detect abnormality with high accuracy according to attribute of a user.SOLUTION: A terminal comprises: detection means for detecting that abnormality occurs in a user to be a target of abnormality detection, using predetermined first information as input, by a detection model created in advance according to attribute of the user; display means for, if the occurrence of the abnormality is detected, displaying a confirmation screen for the user to confirm whether or not the occurrence of the abnormality is correct; and update means for updating the detection model according to a confirmation result of the confirmation screen.SELECTED DRAWING: Figure 2

Description

The present invention relates to terminals, server devices, methods and programs.

In many cases, mobile terminals such as smartphones are always carried or possessed by the user. For this reason, in recent years, safety confirmation systems or abnormality detection systems using mobile terminals such as smartphones have become widespread. In such a system, it is detected that some abnormality has occurred in the user, for example, when the smartphone is not operated for a predetermined time or when the position information of the smartphone does not move.

Japanese Unexamined Patent Publication No. 2008-098744

However, conventionally, attributes such as the user's gender, age, and occupation have not been taken into consideration, and the accuracy of abnormality detection may not be high.

For example, in general, the time and frequency of operating smartphones on a daily basis differ between the elderly and young people such as students. Therefore, if the abnormality detection is performed uniformly on the condition that the smartphone is not operated for a predetermined time, the abnormality detection accuracy of both the elderly and the student or either of them may decrease.

The present invention has been made in view of the above points, and an object of the present invention is to detect an abnormality with high accuracy according to a user attribute.

In order to achieve the above object, in the embodiment of the present invention, an abnormality occurs in the user by a detection model created in advance according to the attribute of the user to be the target of abnormality detection by inputting a predetermined first information. The detection means for detecting that the abnormality has occurred, the display means for displaying a confirmation screen for causing the user to confirm whether or not the occurrence of the abnormality is correct when the occurrence of the abnormality is detected, and the confirmation screen. It has an update means for updating the detection model according to the confirmation result.

Abnormalities can be detected with high accuracy according to user attributes.

It is a figure which shows an example of the whole structure of the abnormality occurrence notification system which concerns on this embodiment. It is a figure which shows an example of the functional structure of the abnormality occurrence notification system which concerns on this embodiment. It is a flowchart which shows an example of the creation process of a detection model. It is a flowchart which shows an example of the detection of abnormality occurrence and the notification processing. It is a figure which shows an example of the confirmation message. It is a flowchart which shows an example of the update process of a detection model. It is a figure which shows an example of the hardware configuration of a computer.

Hereinafter, embodiments of the present invention (hereinafter, also referred to as “the present embodiment”) will be described. In the present embodiment, it is possible to detect with high accuracy that an abnormality has occurred in the user according to the attributes of the user who uses the terminal such as a smartphone, and notify the predetermined notification destination of the detection result. The notification system 1 will be described. In the following, the user who is the target of detecting the occurrence of an abnormality will be referred to as the "detection target user", and the user who is notified that the abnormality has occurred in the detection target user will be referred to as the "notification destination user".

In the present embodiment, a detection model for detecting (or determining) whether or not an abnormality has occurred in the detection target user is created according to the attributes of the detection target user, and then the abnormality detection is performed by this detection model. Do. As a result, in the present embodiment, it is possible to detect an abnormality according to the attribute of the detection target user.

Further, in the present embodiment, the detection model is updated as appropriate according to the life pattern and characteristics of the detection target user. As a result, it is possible to detect anomalies with higher accuracy according to the lifestyle pattern and identification of the user to be detected.

<Overall configuration of error notification system 1>
First, the overall configuration of the abnormality occurrence notification system 1 according to the present embodiment will be described with reference to FIG. FIG. 1 is a diagram showing an example of the overall configuration of the abnormality occurrence notification system 1 according to the present embodiment.

As shown in FIG. 1, the abnormality occurrence notification system 1 according to the present embodiment includes a server device 10 and a plurality of terminals 20. Further, the server device 10 and each terminal 20 are connected via, for example, a communication network N such as the Internet.

The server device 10 is, for example, a computer or a computer system that manages each terminal 20.

The terminal 20 is, for example, a computer used by a detection target user or a notification destination user. Examples of the terminal 20 include smartphones, tablet terminals, notebook PCs (personal computers), portable game devices, wearable terminals, and the like. Hereinafter, the terminal 20 used by the detection target user is also referred to as a “notification source terminal 20”, and the terminal 20 used by the notification destination user is also referred to as a “notification destination terminal 20”. For example, when a couple notifies each other of the occurrence of an abnormality, one terminal 20 also functions as a notification source terminal 20 and also as a notification destination terminal 20. Similarly, the same user may be the detection target user and the notification destination user.

The configuration of the abnormality occurrence notification system 1 shown in FIG. 1 is an example, and may be another configuration. For example, the abnormality occurrence notification system 1 may not include the server device 10.

<Functional configuration of error notification system 1>
Next, the functional configuration of the abnormality occurrence notification system 1 according to the present embodiment will be described with reference to FIG. FIG. 2 is a diagram showing an example of the functional configuration of the abnormality occurrence notification system 1 according to the present embodiment.

As shown in FIG. 2, the abnormality occurrence notification system 1 according to the present embodiment includes an input unit 101, a model creation unit 102, an information acquisition unit 103, an abnormality detection unit 104, and a confirmation unit 105 as functional units. , A notification unit 106, a specific unit 107, and a model update unit 108. Further, the abnormality occurrence notification system 1 according to the present embodiment has a detection model storage unit 110 as a storage unit.

The input unit 101 inputs, for example, a user attribute for creating a detection model, or inputs a notification destination when an abnormality occurs. Examples of the user attribute include the age, gender, occupation, and the like of the detection target user. In addition to these, user attributes may include, for example, height, weight, health condition, medical history, and the like. Further, examples of the notification destination include an e-mail address and a telephone number of the notification destination user, a user ID that can uniquely identify the user on the abnormality occurrence notification system 1, and the like.

The model creation unit 102 creates a detection model according to the user attribute input by the input unit 101. Then, the model creation unit 102 stores (saves) the created detection model in the detection model storage unit 110. At this time, the model creation unit 102 stores, for example, information that can identify the detection target user (for example, the user ID of the detection target user) and the notification destination in the detection model storage unit 110 in association with the detection model. The abnormality detection of the detection target user is performed by the detection model associated with the user ID of the detection target user.

However, for example, when the notification source terminal 20 has the detection model storage unit 110, the detection model storage unit 110 may store the detection model and the notification destination in association with each other (that is, the detection target). It is not necessary to associate the user ID of the user with the detection model.)

As described above, the detection model is a model for detecting whether or not an abnormality has occurred in the detection target user. In the detection model, for example, at least one of various information such as the position information and operation history of the notification source terminal 20, various sensor information (for example, sensor information such as an acceleration sensor and a motion sensor), and battery information is input. , Detects whether or not an abnormality has occurred in the user to be detected (more accurately, whether or not an abnormality has occurred). The various information described above may include various information such as calendar information, weather information, time information, and information based on voice input.

As such a detection model, any model that can detect whether or not an abnormality has occurred in the detection target user by inputting various information can be adopted. For example, a machine learning model (time series model or statistics) can be adopted. Includes model models), models based on rule-based methods, etc.

The information acquisition unit 103 acquires various information necessary for such detection and update when detecting whether or not an abnormality has occurred in the detection target user or when updating the detection model.

The abnormality detection unit 104 detects (determines) whether or not an abnormality has occurred in the detection target user by using various information acquired by the information acquisition unit 103 and the detection model stored in the detection model storage unit 110. ).

When the abnormality detection unit 104 detects the occurrence of an abnormality, the confirmation unit 105 displays a message (confirmation message) for confirming the authenticity of the abnormality occurrence with the detection target user on the notification source terminal 20. By using the confirmation result of this confirmation message for updating the detection model, it is possible to improve the detection accuracy of the occurrence of an abnormality by the detection model.

When the abnormality detection unit 104 detects the occurrence of an abnormality, the notification unit 106 notifies the notification destination (that is, the notification destination terminal 20) of the occurrence of the abnormality. As a result, the notification destination user can know that some abnormality (or a predetermined abnormality) has occurred in the detection target user.

When updating the detection model, the identification unit 107 identifies the life pattern and characteristics (hereinafter, also referred to as “user characteristics”) of the detection target user by using various information acquired by the information acquisition unit 103. Here, the life pattern is a habitual life pattern, for example, "Weekday 9:00 to 17:00 is in a specific place (for example, school)", "Weekday wake-up time". 7:00 ”,“ Every Friday from 20:00 to 21:30 is in a specific place (for example, a specific English conversation school) ”,“ Every day from 22:00 to 7:00 is sleeping ”, etc. Can be mentioned. Further, the user characteristic is some characteristic that accompanies the user's behavior, and examples thereof include "inflection and rhythm when voice input is performed".

The model update unit 108 uses the life pattern and user characteristics specified by the specific unit 107 to use the detection model associated with the corresponding detection target user among the detection models stored in the detection model storage unit 110. To update. As a result, the detection model is updated according to the life pattern and user characteristics of the detection target user, and the abnormality can be detected with higher accuracy.

It should be noted that various configurations can be considered as to which functional unit and storage unit each of the server device 10 and the terminal 20 has. For example, the server device 10 has a model creation unit 102, a model update unit 108, and a detection model storage unit 110, and an input unit 101, an information acquisition unit 103, an abnormality detection unit 104, a confirmation unit, and a notification. The terminal 20 may have a unit 106 and a specific unit 107. Alternatively, for example, the server device 10 has a confirmation unit 105 and a notification unit 106, and the input unit 101, the model creation unit 102, the information acquisition unit 103, the abnormality detection unit 104, the specific unit 107, and the model. The terminal 20 may have an update unit 108 and a detection model storage unit 110. Alternatively, for example, the terminal 20 may have all the functional units and all the storage units (in this case, the abnormality occurrence notification system 1 may not include the server device 10).

<Detection model creation process>
Next, a process of creating a detection model (processing of creating a detection model) for detecting whether or not an abnormality has occurred in the detection target user will be described with reference to FIG. FIG. 3 is a flowchart showing an example of the detection model creation process. The detection model creation process is started using the notification source terminal 20 as an initial setting when, for example, the detection target user starts using the abnormality occurrence notification system 1.

First, the input unit 101 inputs the user attribute and the notification destination (step S101). These user attributes and notification destinations are input by, for example, a detection target user or the like using the notification source terminal 20. However, for example, a user other than the detection target user may input the user attribute and user ID of the detection target user and the notification destination.

Next, the model creation unit 102 creates a detection model according to the user attribute input in step S101 above. Then, the model creation unit 102 stores the created detection model in the detection model storage unit 110 in association with the notification destination input in step S101 above (step S102). At this time, the user ID or the like of the detection target user may be associated with the detection model and stored in the detection model storage unit 110. The detection model storage unit 110 may be possessed by the notification source terminal 20 or may be possessed by the server device 10.

Here, in step S102, the model creation unit 102 detects, for example, one of a plurality of detection models predetermined as a template (template) according to the user attribute input in step S101. All you have to do is select and create a model.

Further, in the present embodiment, only the user attribute and the notification destination are input in the above step S101, but detailed conditions (detection conditions) related to abnormality detection may be input. For example, it may be possible to input the location information of the time zone, stay area, facility, etc. to be detected for abnormality, the type of abnormality to be detected, and the like as detection conditions. As a result, in step S102 described above, it is possible to create a detection model according to the input detection conditions.

The time zone targeted for abnormality detection is the time zone in which processing for detecting whether or not an abnormality has occurred in the detection target user (“abnormality occurrence detection and notification processing” described later) is executed. Is. By inputting this time zone, for example, the detection target user can prevent the abnormality detection from being performed in a time zone other than the input time zone.

The type of abnormality to be detected is the type of action, action, or state detected as an abnormality. Examples of such actions, actions, or states include "no movement in the position information of the notification source terminal 20", "falling of the detection target user", and the like. By inputting the type of this abnormality, for example, the detection target user can only detect an abnormality related to the input type of abnormality.

<Detection of abnormal occurrence and notification processing>
Next, the process of detecting whether or not an abnormality has occurred in the detection target user and notifying a predetermined notification destination when the abnormality is detected (abnormality occurrence detection and notification processing) is shown in the figure. This will be described with reference to 4. FIG. 4 is a flowchart showing an example of abnormality occurrence detection and notification processing.

The information acquisition unit 103 acquires various information (or various information for comparison with the detection conditions defined or set in the detection model) stored in the detection model storage unit 110 and input to the corresponding detection model. (Step S201). As described above, such various information includes position information and operation history of the notification source terminal 20, various sensor information (for example, sensor information such as an acceleration sensor and a motion sensor), battery information, calendar information, weather information, and the like. Examples include time information and information based on voice input (for example, information on the presence / absence of voice input and information on intonation and rhythm when voice input is performed). The information acquisition unit 103 may acquire one or more of these various pieces of information used for inputting the detection model.

Next, the abnormality detection unit 104 includes various information acquired in step S201 and the corresponding detection model stored in the detection model storage unit 110 (that is, the detection model used for abnormality detection of the detection target user). Is used to detect whether or not an abnormality has occurred in the detection target user (step S202). Here, for example, when the detection model is a machine learning model, the abnormality detection unit 104 may detect whether or not an abnormality has occurred by inputting various acquired information into the detection model. Further, for example, when the detection model is a model based on the rule-based method, the abnormality detection unit 104 performs various acquired information and a detection condition (rule) defined or set in the detection model (that is, a rule table). By comparing with, it is sufficient to detect whether or not an abnormality has occurred.

Then, if the occurrence of an abnormality is not detected, the process returns to the process of step S201. As a result, for example, the processes of steps S201 and S202 are repeatedly executed at predetermined time intervals.

On the other hand, when the occurrence of an abnormality is detected, the confirmation unit 105 displays a confirmation message on the notification source terminal 20 for confirming the authenticity of the abnormality occurrence with the detection target user (step S203). Here, an example of the confirmation message 1000 displayed on the notification source terminal 20 is shown in FIG. In the confirmation message 1000 shown in FIG. 5, the message "Have an abnormality occurred? Notify the registration notification destination of the occurrence of the abnormality" is displayed. Further, the confirmation message 1000 shown in FIG. 5 includes a cancel button 1100. At this time, if the abnormality occurrence is an error (that is, the abnormality detection is an erroneous detection), the detection target user presses the cancel button 1100. As a result, it is possible to improve the detection accuracy of the detection model by updating the detection model described later (step S205 described later). The confirmation message 1000 shown in FIG. 5 may include other buttons such as a "notify" button and a "re-notify after n minutes" button, for example.

The confirmation message 1000 displayed by the confirmation unit 105 may be hidden after a predetermined time (for example, 1 minute, 3 minutes, etc.) has elapsed.

Next, the notification unit 106 notifies the notification destination terminal 20 that an abnormality has occurred in the detection target user (step S204). As a result, the notification destination user can know that an abnormality has occurred in the detection target user.

Here, the notification unit 106 may immediately notify the notification destination when the occurrence of an abnormality is detected in the above step S202, or after the confirmation message 1000 is displayed in the above step S203, the notification unit 106 may notify the notification destination. When a predetermined time (for example, 1 minute, 3 minutes, etc.) elapses without pressing the cancel button 1100, the notification destination may be notified.

Next, the model update unit 108 updates the detection model using the confirmation result by the confirmation message 1000 displayed in step S203 (that is, for example, whether or not the cancel button 1100 is pressed) (step). S205). That is, when the cancel button 1100 is pressed, it means that the abnormality detection was incorrect, so that the model update unit 108 can perform, for example, the same abnormality detection (or the same abnormality detection in the same time zone). Update the detection model so that it does not happen. As a result, the detection model is updated according to the confirmation result of the detection target user in the confirmation message 1000, and the detection accuracy of the detection model can be improved (that is, the detection target user feeds back the truth of the abnormality detection result. Therefore, the detection accuracy of the detection model can be improved.) For example, the detection model may be updated by prompting the detection target user to change the detection condition and then using the changed detection condition.

The process of step S203 may not be executed (that is, the process of step S203 may be omitted). In this case, the process of step S205 may not be executed either.

<Detection model update process>
Next, a process of updating the detection model according to the life pattern and user characteristics of the detection target user will be described with reference to FIG. FIG. 6 is a flowchart showing an example of the detection model update process. The detection model update process is repeatedly executed, for example, every preset period.

The information acquisition unit 103 acquires various information used for updating the detection model (step S301). As described above, examples of the various information include position information and operation history of the notification source terminal 20, various sensor information, battery information, calendar information, weather information, time information, information based on voice input, and the like. The information acquisition unit 103 may acquire one or more of these various information used for updating the detection model. The type of information acquired in this step, the acquisition period, and the like may be different from those in step S201 described above.

Next, the identification unit 107 identifies the life pattern and user characteristics of the detection target user by using various information acquired in step S301 (step S302). Here, the specific unit 107 detects by analyzing position information, operation history, various sensor information, battery information, calendar information, weather information, time information, etc., using, for example, a known time series analysis method or the like. The life pattern of the target user may be specified. Further, the specifying unit 107 may specify the user characteristics by analyzing information or the like based on the voice input by using, for example, a known voice pattern recognition method or the like.

In step S301 described above, the specific unit 107 may specify only one of the life pattern and the user characteristics. In this case, in step S303, which will be described later, the detection model is updated using only one of the life pattern and the user characteristics.

Next, the model update unit 108 updates the detection model of the detection target user by using the life pattern and the user characteristics identified in step S302 above (step S303).

For example, when it is specified that "the notification source terminal 20 is not operated during the daytime 12: 00-14: 00 on weekdays" as the life pattern specified in step S302 above, the model update unit 108 Updates the detection model so that even if the notification source terminal 20 is not operated during the "weekday daytime 12: 00-14: 00 time zone", it is not detected as an abnormality.

Similarly, for example, when a certain Lu Ritsu feature is obtained as the user feature specified in step S302 above, the model update unit 108 performs voice input with a Lu Ritsu feature that deviates from the Lu Ritsu feature. Update the detection model so that it is sometimes detected as abnormal.

In this way, by updating the detection model using the life pattern and user characteristics of the user to be detected, it is possible to detect an abnormality when an action, action, state, etc. that deviates from the daily life pattern occurs, or the user. It is possible to detect an abnormality when voice input or the like is performed with a feature that is out of the feature. Therefore, such an update makes it possible to detect an abnormality of the detection target user with higher accuracy.

Moreover, by repeatedly updating the detection model at predetermined periods (that is, by dynamically updating the detection model at predetermined periods), for example, the detection target user goes on to school, finds a job, or is transferred. It is possible to detect anomalies that follow changes in life patterns due to changes in life patterns, changes in user characteristics due to changes in physical characteristics, and the like.

<Hardware configuration of computer 600>
Finally, the hardware configuration of the computer 600 (for example, the computer that realizes the server device 10 and the computer that realizes the terminal 20) constituting the abnormality occurrence notification system 1 according to the present embodiment will be described with reference to FIG. 7. .. FIG. 7 is a diagram showing an example of the hardware configuration of the computer 600.

The computer 600 shown in FIG. 7 includes an input device 601, a display device 602, an external I / F 603, a RAM (Random Access Memory) 604, a ROM (Read Only Memory) 605, a processor 606, and a communication device 607. , With an auxiliary storage device 608. Each of these hardware is connected so as to be able to communicate with each other via the bus B.

The input device 601 is, for example, a touch panel, various buttons, or the like. The display device 602 is, for example, a display or the like. The computer 600 does not have to have at least one of the input device 601 and the display device 602.

The external I / F 603 is an interface with an external device. The external device includes a recording medium 603a and the like. The computer 600 can read or write the recording medium 603a via the external I / F 603. Examples of the recording medium 603a include a CD (Compact Disc), a DVD (Digital Versatile Disk), an SD memory card (Secure Digital memory card), and a USB (Universal Serial Bus) memory card. In addition, each functional unit of the abnormality occurrence notification system 1 (for example, input unit 101, model creation unit 102, information acquisition unit 103, abnormality detection unit 104, confirmation unit 105, notification unit 106, specific unit 107, model update unit 108, etc. ) Is realized in one or more programs (or modules) may be stored in the recording medium 603a.

The RAM 604 is a volatile semiconductor memory that temporarily holds programs and data. The ROM 605 is a non-volatile semiconductor memory capable of holding programs and data even when the power is turned off. The ROM 605 stores, for example, setting information related to an OS (Operating System), setting information related to a communication network, and the like.

The processor 606 is, for example, an arithmetic unit such as a CPU (Central Processing Unit). Each functional unit of the abnormality occurrence notification system 1 is realized by reading one or more programs stored in the ROM 605 or the auxiliary storage device 608 onto the RAM 604 and executing the process by the processor 606.

The communication device 607 is an interface for connecting the computer 600 to the communication network. One or more programs that realize each functional unit of the abnormality occurrence notification system 1 may be acquired (downloaded) from a predetermined server via the communication device 607.

The auxiliary storage device 608 is a non-volatile storage device such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive). The programs and data stored in the auxiliary storage device 308 include, for example, an OS, an application program that realizes various functions on the OS, and one or more programs that realize each functional unit of the abnormality occurrence notification system 1. .. Each storage unit (for example, detection model storage unit 110, etc.) of the abnormality occurrence notification system 1 can be realized by using, for example, an auxiliary storage device 608.

The abnormality occurrence notification system 1 according to the present embodiment is composed of one or more computers 600 having the hardware configuration shown in FIG. 7. The computer 600 shown in FIG. 7 may include a plurality of processors 606 and a plurality of memories (RAM 604, ROM 605, auxiliary storage device 608, etc.). Further, the computer 600 that realizes the terminal 20 may include, for example, various sensors such as an acceleration sensor and a motion sensor, a GPS receiver that receives signals from GPS satellites and measures the current position, and the like. ..

The present invention is not limited to the above-described embodiment disclosed specifically, and various modifications and modifications can be made without departing from the scope of claims. For example, the following modification can be considered.

(Modification example 1)
The detection model may be created without inputting the user attribute in step S101 above. In this case, for example, the model creation unit 102 may create one detection model selected by the user from a plurality of detection models predetermined as templates (templates).

(Modification 2)
The detection model may be updated in step S205 above without displaying the confirmation message in step S203 above. In this case, for example, the model update unit 108 may update the detection model without using the confirmation result.

1 Abnormality occurrence notification system 10 Server device 20 Terminal 101 Input unit 102 Model creation unit 103 Information acquisition unit 104 Abnormality detection unit 105 Confirmation unit 106 Notification unit 107 Specific unit 108 Model update unit 110 Detection model storage unit

Claims (8)

A detection means for detecting that an abnormality has occurred in the user by a detection model created in advance according to the attributes of the user to be detected by using the predetermined first information as an input.
When the occurrence of the abnormality is detected, a display means for displaying a confirmation screen for causing the user to confirm whether or not the occurrence of the abnormality is correct, and
An update means for updating the detection model according to the confirmation result on the confirmation screen, and
Terminal with. Each predetermined period has a specific means for inputting a predetermined second information and identifying at least one of the user's life pattern and characteristics based on the second information.
The update means
The terminal according to claim 1, wherein the detection model is updated by using at least one of the identified life pattern and the feature. Each of the first information and the second information
Information including at least one of the position information of the terminal, the sensor information of the sensor included in the terminal, the battery information of the terminal, the calendar information, the weather information, the time information, and the information based on the voice input of the terminal. , The terminal according to claim 2. The update means
The terminal according to any one of claims 1 to 3, wherein when an operation indicating that the occurrence of the abnormality is performed is performed on the confirmation screen, the detection model is updated so that the abnormality is not detected. The terminal according to any one of claims 1 to 4, which has a notification means for notifying a predetermined notification destination of the occurrence of the abnormality when the occurrence of the abnormality is detected. A server device that is connected to a terminal used by a user who is the target of anomaly detection via a communication network.
A detection means for detecting that an abnormality has occurred in the user by a detection model created in advance according to the attribute of the user by inputting a predetermined first information.
When the occurrence of the abnormality is detected, a display means for displaying a confirmation screen on the terminal for the user to confirm whether or not the occurrence of the abnormality is correct, and
An update means for updating the detection model according to the confirmation result on the confirmation screen, and
Server device with. A detection procedure for detecting that an abnormality has occurred in the user by a detection model created in advance according to the attributes of the user to be detected by using the predetermined first information as an input, and a detection procedure.
When the occurrence of the abnormality is detected, a display procedure for displaying a confirmation screen for causing the user to confirm whether or not the occurrence of the abnormality is correct, and a display procedure.
An update procedure for updating the detection model according to the confirmation result on the confirmation screen, and
How the computer does. A detection procedure for detecting that an abnormality has occurred in the user by a detection model created in advance according to the attributes of the user to be detected by using the predetermined first information as an input, and a detection procedure.
When the occurrence of the abnormality is detected, a display procedure for displaying a confirmation screen for causing the user to confirm whether or not the occurrence of the abnormality is correct, and a display procedure.
An update procedure for updating the detection model according to the confirmation result on the confirmation screen, and
A program that causes a computer to run.

Images