System for detecting discard of user equipment

Abstract

The server device may determine a state of the user device using a model, the state of the user device based on the event-based data, where the model is configured based on historical event data associated with historical events relating to a plurality of other user devices associated with the user device. The server device may determine a location associated with the user device based on a state indicating that the user device may have been abandoned in the physical environment. The server device may perform actions associated with the user device and the location.

Description

Technical Field

[0001] The present disclosure generally relates to detecting the status of a user device and, for example, to detecting when a user device may be abandoned or otherwise lost. Background Art

[0002] User devices (such as wireless communication devices, mobile phones, environmental sensor devices, wearable devices, handheld scanning devices, etc.) are increasingly being used for a variety of purposes. Many user devices are easily lost, stolen, misplaced, or otherwise abandoned, especially if the user devices are, for example, mobile, relatively small in size, and / or owned by forgetful users. Lost user devices can negatively impact users and / or organizations in a variety of ways, including resources lost to recover and / or replace the lost user device, productivity lost because the user device cannot be used for productive purposes when lost, etc. To ensure that devices are not abandoned, there is a need to detect when a user device is at risk of being lost or has been lost, and there is a need to facilitate the recovery of lost user devices. Summary of the invention

[0003] Certain embodiments described herein relate to methods. The method may include receiving, by a server device, event-based data associated with a user device, wherein the user device is communicatively coupled to the server device via an access point in a physical environment. The method may include determining, by the server device and using a model, a state of the user device, the state of the user device being based on the event-based data, wherein the model is configured based on historical event data associated with historical events, the historical events involving multiple other user devices associated with the user device. The method may include determining, by the server device and based on a state indicating that the user device may have been abandoned in the physical environment, a location associated with the user device. The method may include performing, by the server device, an action associated with the user device and the location.

[0004] Certain embodiments described herein relate to a system. The system may include one or more memories and one or more processors coupled to the one or more memories. The system may be configured to receive event-based data associated with a user device via an access point. The system may be configured to determine a state of the user device using a model and based on one or more event parameters associated with the event-based data. The system may be configured to determine that the user device may have been abandoned in a physical environment based on the state. The system may be configured to perform an action associated with locating the user device.

[0005] Certain embodiments described herein relate to a tangible machine-readable medium storing a system instruction set. When executed by one or more processors of the system, the instruction set may cause the system to receive event-based data associated with a user device in a physical environment. When executed by one or more processors of the system, the instruction set may cause the system to monitor a state of the user device using a model, the state being based on one or more events identified in the event-based data. When executed by one or more processors of the system, the instruction set may cause the system to provide an indication of the state to a management device to allow the management device to monitor abandonment of the user device. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] Figure 1 is an illustration of an example implementation 100 associated with detecting abandonment of a user device as described herein.

[0007] Figure 2 is an illustration of another example implementation 100 associated with detecting abandonment of a user device as described herein.

[0008] Figure 3 is an illustration of an example of training and using a machine learning model as described herein in connection with detecting abandonment of a user device.

[0009] Figure 4 is an illustration of an example environment in which the systems and / or methods described herein may be implemented.

[0010] Figure 5 yes Figure 4 An illustration of example components of one or more devices.

[0011] Figure 6 is a flow diagram of an example process involving detecting abandonment of a user device as described herein. DETAILED DESCRIPTION

[0012] The following detailed description of example embodiments refers to the accompanying drawings.The same reference numbers in different drawings may identify the same or similar elements.

[0013] Different people and different organizations can use various user devices in various ways. Whether for personal use or business use, user devices are easily lost, stolen, misplaced or otherwise abandoned. As used herein, "abandonment" can refer to all types of losses, intentional or otherwise. Many user devices are mobile and can be operated at any location, which can make it difficult to find abandoned user devices. In addition, many user devices are relatively small in size, which can also make it difficult to recover abandoned user devices. Lost user devices can have a negative impact on users and / or organizations in various ways, including recovering and / or replacing the resources lost by the lost user device, the productivity lost because the user device cannot be used for productive purposes when it is lost, etc. To ensure that the device is not abandoned, it is necessary to detect when the user device is at risk of being lost or has been lost, and it is necessary to facilitate the recovery of the lost user device.

[0014] Certain embodiments described herein are conducive to preventing user devices from being abandoned, and also assist in recovering abandoned user devices or user devices that may be abandoned. For example, event-based data (including activity-based data) from user devices can be collected to determine the use and / or non-use patterns in various activities and various periods. Using the collected data, a model can be generated to track the monitored user devices to determine when the user devices may or have been abandoned. In the case where the user device may be abandoned, various actions can be taken to facilitate the recovery of the user device, including sending an alarm directly to the user device, and sending other instructions that can facilitate the recovery of the user device to other devices. As a result, the user device may be less likely to be abandoned, more likely to be recovered and more likely to be recovered faster than traditional recovery methods. By reducing the abandonment of user devices, the negative impact on users and organizations that utilize user devices can be reduced, which can include reducing the resources that may be lost in other ways to recover and / or replace abandoned user devices, because the user device can be used instead of being abandoned to improve the productivity of both users and user devices, etc.

[0015] Figure 1 is an illustration of an example implementation 100 associated with detecting abandonment of a user device. Figure 1 As shown, the exemplary implementation 100 includes a user device, an external device, and a modeling device. Figure 4 and Figure 5 These devices are described in more detail.

[0016] As shown in the reference numeral 110, the modeling device can receive various information about the use of various user devices (e.g., from user device 1 to user device N). The information can be received from the user device and / or from an external device (e.g., a device separate from the user device and / or the modeling device). The information (collectively referred to herein as "event-based data") includes event-based and activity-based data associated with the user device. As further described herein, a variety of event-based data can be provided to the modeling device to facilitate the generation of a model for use in determining whether a particular user device is likely to be abandoned and / or whether it is beneficial to recover a particular user device.

[0017] In some embodiments, the event-based data received from the user device may include data identifying the event, a timestamp associated with the event, and / or a location associated with the event, etc. Example events associated with the user device may include: events associated with scanning performed using the user device (e.g., barcode scanning events, quick response (QR) code scanning events, radio frequency identifier (RFID) scanning events, etc.), events associated with using an application of the user device (e.g., launching a predetermined application and / or interacting with a predetermined application, using a predetermined function of an application, etc.), events associated with charging a battery of the user device (e.g., connecting to a power source, disconnecting from a power source, reaching full charge, reaching a predetermined battery charge threshold, etc.), events associated with battery replacement of the user device (e.g., disconnecting the battery, connecting to a power source, etc.), events associated with accelerometer data of the user device (e.g., motion detected, no motion detected, a particular type and / or pattern of motion detected, motion that meets a threshold type and / or pattern, etc.), and / or events associated with a step counter of the user device (e.g., steps detected, no steps detected within a threshold time period, steps within a predetermined time period, etc.), etc.

[0018] In addition to or in lieu of event-based data provided by the user device, the external device may also provide event-based data to the modeling device. The external device further described herein is a device that interacts with the user device in some way. For example, a transaction server device may be a device that the user device communicates with when the user device is used to perform a transaction (e.g., a purchase, an agreement, or other types of transactions). Other external devices may include, for example, an access point (AP) (e.g., a Wi-Fi access point, a cellular access point, a switch device, etc.) that the user device can use to access an internal and / or external network, a camera device (e.g., a security camera, a camera attached to other devices, etc.) that can capture images and / or videos of the user device, a workforce management device that tracks the user device (e.g., via check-in and check-out procedures), a geo-fence device (e.g., a geo-fence server, a geo-fence beacon, etc.) that tracks the location of the user device, and / or a smart cabinet (e.g., via RFID, tracking the user device, a container or shelf that charges the user device, etc.), etc.

[0019] In some embodiments, the event-based data received from the external device may include, for example, events associated with a user device checking in and / or out (e.g., a time associated with a user device checking in and / or out, a threshold time period associated with a user device checking in or checking out, etc.), events associated with an access point to which the user device is connected (e.g., connecting to an access point, disconnecting from an access point, signal strength or other measurement associated with the quality of the connection to the access point, etc.), events associated with geo-fencing (e.g., a user device entering / leaving a predefined location or area, a user device being in a particular area for a predetermined time period, etc.), events associated with a transaction (e.g., a user device being used to perform at least a portion of a transaction), events associated with a smart storage device (e.g., adding and / or removing a user device to and / or from a smart storage device, charging a user device by starting or ending a smart storage device, a predetermined time period associated with charging and / or storing a user device associated with a smart storage device, etc.), and / or events associated with a camera device (e.g., a tracking system detecting, via a camera device, a user device entering and / or leaving a particular location, other activities detected at a location associated with a user device, etc.).

[0020] Although some event-based data are described as being provided by user devices, and other event-based data are described as being provided by external devices, the above are merely examples. In some aspects, any event-based data described herein may be provided by user devices and / or external devices.

[0021] Whether transmitted by an external device or a user device, event-based data can be transmitted in various ways. For example, event-based data can be transmitted via wired communication and / or wireless communication. In some aspects, the external device and / or the user device can transmit event-based data in the following ways: in near real time (e.g., based on communication upon detection of an event), periodically (e.g., based on a schedule), manually based on user input, and / or based on triggering events (e.g., when the user device signs in, when a specific application on the user device is used), etc. In some embodiments, event-based data can be transmitted one event at a time, and / or in batches, and can depend on the type of communication and / or the device transmitting the event-based data.

[0022] As indicated by reference numeral 120, the modeling device may generate a model. For example, the modeling device may use event-based data received from the user device and / or the external device to generate a model that can be used to determine whether the user device is likely to be abandoned. In some embodiments, the model may receive event-based data as input, such as any information associated with the event-based data described herein (e.g., information identifying an event associated with the user device, a timestamp associated with the event, a location associated with the event, etc.). The model may provide as an output an indication of whether the event-based data (provided as input) indicates that the user device is likely to be abandoned.

[0023] As described herein, in some embodiments, event-based data may be associated with historical event parameters that can be used to generate models, including as training data for machine learning models. The historical event parameters may include, for example: user interaction metrics associated with other user devices (e.g., indicating the number of times associated with the event occurred at other user devices), application metrics associated with applications installed on multiple other user devices (e.g., indicating the use of the application by other user devices), location information identifying the location of multiple other user devices during the historical event, power consumption metrics associated with batteries of multiple other user devices (e.g., including the charging status of the other user devices), and / or replacement intervals associated with replacing batteries of multiple other user devices, as well as other examples associated with different types of event data as described herein.

[0024] In some embodiments, the model is manually generated (e.g., via user input and selection of which portions of the event data and which corresponding events may lead to a determination that the user device is likely abandoned). In some aspects, the model may be machine generated. For example, as described herein (e.g., in conjunction with Figure 3 ) As described in more detail, the modeling device may use one or more machine learning methods to generate a model.

[0025] In some embodiments, the modeling device may generate different models for different types of user devices and / or for different locations. For example, some user devices have different functions, and different events may affect the possibility of abandonment differently for different types of user devices (for example, a user device without a scanning feature may not need a model that takes into account the time since the last scan, or a larger user device is less likely to be abandoned than a smaller device with the same event data). In the case where the location is associated with different risk levels, different locations may use different models in the case. For example, a location where user devices are abandoned more often (for example, based on historical data of the location, the size of the location, specific characteristics of the location, etc.) may use a model that is more likely to identify the user device as abandoned than another model associated with a location associated with a lower risk level. Although different models can be used for different situations, in some embodiments, the distinction (for example, the type of user device, the risk of the location, etc.) can be a feature of a single model rather than being used to generate a separate model.

[0026] In some aspects, the model may be generated based on a pattern of activity. For example, certain events may be more likely to indicate abandonment in a particular sequence or pattern if they occur in a particular sequence or pattern. For example, the expected activity pattern of a remote user device used to scan objects in a warehouse or retail location may be as follows: the device checks out from the smart storage area, disconnects from the charger, scans periodically throughout the day, returns to the smart storage area, and connects to the charger (e.g., at the end of the day). In the case where the user device has an expected activity pattern, deviations from the pattern may indicate that the user device has been abandoned. For example, after being disconnected from the charger, the user device may be considered abandoned if the user device does not scan anything within a first threshold time period. As another example, if the user device does not return to the smart storage area within a second threshold time period, the user device may be considered abandoned. As described herein, thresholds may be part of the model, and these thresholds may be manually derived and / or machine generated (e.g., via machine learning).

[0027] In some embodiments, the model may define one or more thresholds, the satisfaction of which provides an indication of whether the user device is likely to be abandoned. For example, the model may provide one or more thresholds, and one or more values ​​(e.g., values ​​associated with the above-mentioned events and / or features described elsewhere herein) are compared with the one or more thresholds. For example, the time since the last accelerometer input provided by the user device may be compared with a threshold time period, and if the threshold is reached, the user device may be given a value associated with the abandonment probability associated with the threshold. In some embodiments, the model may have multiple thresholds associated with a single value (e.g., multiple time thresholds associated with the last accelerometer input), each of which is associated with a different abandonment probability value. In some embodiments, in addition to or as an alternative to a single threshold, a combination of thresholds can be used to indicate the possibility of abandonment. In some embodiments, the thresholds associated with the model may be based on empirical evidence, user input, machine learning, or a combination thereof.

[0028] As mentioned above, Figure 1 Other examples can be found in the Figure 1 Different than described. Figure 1 The number and arrangement of devices shown in are provided as examples. In practice, Figure 1 There may be additional devices, fewer devices, different devices, or differently arranged devices compared to the devices shown in FIG. Figure 1 Two or more of the devices shown in the figure may be implemented in a single device, or Figure 1 The single device shown in may be implemented as multiple distributed devices. Additionally or alternatively, Figure 1 The device set (e.g., one or more devices) shown in FIG. 1 may execute the functions described as being performed by Figure 1 One or more functions performed by another set of devices shown in .

[0029] Figure 2 is an illustration of another example implementation 200 associated with detecting abandonment of a user device. Figure 2 As shown, the exemplary implementation 200 includes a user device, a server device, and a management device. Figure 4 and Figure 5 These devices are described in more detail.

[0030] As shown in example implementation 200, a server device may be used to detect whether a user device may be abandoned and take one or more actions designed to recover the user device, as described herein. In this example, the user device may be a user device used to scan an object (e.g., a barcode), and the user device is depicted as being left on a shelf (e.g., in a warehouse, a retail store location, an office, a residential area, etc.) some time after being used to scan the object.

[0031] As shown in reference numeral 210, the user device may transmit event-based data and the server device may receive event-based data, such as event-based data as described herein. The user device may be communicatively coupled to the server device via an access point within the physical environment. In this example, the event-based data indicates that the scan was performed by the user device at a specific time (e.g., indicated by a timestamp: 10:15:30 25 / 10 / 2022) and is associated with a specific location (e.g., indicated by the user device's connection to a specific access point labeled AP 7). Although the example shows only a single event, in some embodiments, multiple events may be transmitted to the server device.

[0032] In some embodiments, a user device may be configured to report event-based data on a schedule, based on the user device becoming communicatively coupled to an access point, and / or sensor data (e.g., accelerometer data, Global Positioning System (GPS) data, etc.) indicating that a particular type of event associated with the user device has occurred, etc.

[0033] As shown in reference numeral 220, the server device may use a model (such as the model described herein) to determine the state of the user device. The model may be configured based on historical event data associated with historical events involving other user devices. For example, as described herein, the model may be generated based at least in part on event-based data received from other user devices and / or other external devices. In some aspects, the historical event data may include data identifying a first value of an event parameter associated with other user devices that have not been abandoned, and a second value of an event parameter associated with other user devices that have been abandoned. As further described in detail herein, in some embodiments, the historical data may be used as training data for a machine learning model.

[0034] In some embodiments, the server device may obtain a model for determining the likelihood of abandonment from a modeling device described herein. In this example 200, the likelihood of abandonment is identified as 0.85, which may indicate that the likelihood that the user device is abandoned or will be abandoned is relatively high (e.g., in the range of 0 to 1). To use the model, the server device may provide at least a portion of the event-based data as input and receive as output a value indicating the likelihood of abandonment (e.g., a Boolean true / false value, a percentage likelihood value, etc.). The use of the model is described in more detail herein, such as in conjunction with Figure 1 and Figure 3 .

[0035] In some embodiments, the state may indicate whether the user device has been abandoned within a threshold time period while the user was using the user device in a physical environment. For example, the state may indicate that the user device has been abandoned after being outside of a predefined physical space for a threshold time period. In some embodiments, the state indicates a probability that the user device has been abandoned, the probability meeting a probability threshold associated with a user device being abandoned by the user. For example, where the model is associated with one or more thresholds, a threshold may be used to indicate that the user device is abandoned.

[0036] As shown in reference numeral 230, the server device may determine a location associated with the user device. A determination indicating that the user device may have been abandoned may be made based at least in part on a state (e.g., determined by a model). In some embodiments, the location may be based on event-based data. For example, in example embodiment 200, the last known location of the user device may be associated with an access point labeled AP 7. In an area served by multiple access points, the access point with which the user device last communicated may be used to determine the general location of the user device (e.g., near access point AP 7). In some embodiments, more detailed location data may be recorded. For example, an object scanned by a user device may be associated with a specific location that may be used for a location associated with a scanning event. In another example, a wireless signal strength value associated with one or more wireless access points may be used to indicate the location of the user device (e.g., via triangulation using different signal strength values ​​associated with different access points, such as provided by the user device and / or access point).

[0037] As shown in reference numeral 240, the server device may send information identifying the location of the user device and the management device may receive information identifying the location of the user device. For example, the management device may be associated with a user or other device that recovers the user device (e.g., automatically recovered by a drone or other robotic device), and the information identifying the location of the user device may facilitate the management device to recover the user device. As an example, the management device may be another user device that may notify or not notify the user device based on the proximity of the management device to the user device location. In this case, the management device may be associated with a user who may be able to search and recover the user device based on the last known location of the user device. As another example, the management device may communicate with one or more other user devices, and the management device may be able to send an alarm to at least one of the other user devices to indicate that the user associated with (multiple) other user devices recovers the user device (e.g., based on the user device location). In this case, the management device may select other devices based on the proximity to the user device location, and provide other devices with information indicating that the user of the other device should search for the user device.

[0038] As shown in reference numeral 250, the server device may provide a user device alert and the user device may receive the user device alert. The user device alert may cause the user device to take certain actions designed to facilitate recovery of the user device. For example, the user device alert may cause the user device to emit a sound, vibrate, flash a light, etc.

[0039] In this way, in the event that a user device may be abandoned, various actions may be taken to facilitate recovery of the user device, including sending an alert directly to the user device and sending other instructions to other devices that may facilitate recovery of the user device object. As a result, the user device may be less likely to be abandoned, more likely to be recovered, and more likely to be recovered faster than traditional recovery methods. By reducing abandonment of user devices, negative impacts on users and organizations that utilize user devices may be reduced, which may include reducing resources that may otherwise be lost to recover and / or replace abandoned user devices, improving productivity of both the user and the user device because the user device can be used rather than abandoned, etc.

[0040] As mentioned above, Figure 2 Other examples can be found in the Figure 2 Different than described. Figure 2 The number and arrangement of devices shown in are provided as examples. In practice, Figure 2 There may be additional devices, fewer devices, different devices, or differently arranged devices compared to the devices shown in FIG. Figure 2 Two or more of the devices shown in the figure may be implemented in a single device, or Figure 2The single device shown in may be implemented as multiple distributed devices. Additionally or alternatively, Figure 2 The device set (e.g., one or more devices) shown in FIG. 1 may perform the operations described by Figure 2 One or more functions performed by the set of devices shown in .

[0041] Figure 3 3 is a diagram illustrating an example 300 of training and using a machine learning model in connection with detecting abandonment of a user device. The machine learning model training and use described herein may be performed using a machine learning system. The machine learning system may include, or may be included in, a computing device, a server, a cloud computing environment, etc. (such as a modeling device, a server device, and / or a management device described in more detail elsewhere herein).

[0042] As shown at reference numeral 305, a machine learning model can be trained using an observation set. The observation set can be obtained from training data (e.g., historical data), such as data collected during one or more processes described herein. In some embodiments, the machine learning system can receive the observation set (e.g., as input) from one or more user devices and / or external devices, as described elsewhere herein.

[0043] As shown in the reference numeral 310, the observation set includes a feature set. The feature set may include a variable set, and the variables may be referred to as features. A specific observation may include a set of variable values ​​(or feature values) corresponding to the variable set. In some embodiments, the machine learning system may determine the variables of the observation set and / or the variable values ​​of the specific observation based on input received from one or more user devices and / or external devices. For example, the machine learning system may identify a feature set (e.g., one or more features and / or feature values) by extracting the feature set from structured data, by performing natural language processing to extract the feature set from unstructured data, and / or receiving input from an operator.

[0044] For example, the feature set of the observation set may include a first feature of location type (e.g., based on existing observations, each location type may be associated with a different abandonment risk), a second feature of time since last activity (e.g., the last time the user device was used to scan an object or perform some other activity), a third feature of time since last accelerometer input (e.g., the last time the user device moved in a manner that triggered an accelerometer output), etc. As shown, for the first observation, the value of the first feature may be 5 (e.g., indicating a relatively high risk based on a range of 1 to 5, where 1 is a low risk and 5 is a high risk), the value of the second feature may be 480 (e.g., indicating 480 minutes since the user device was last used for an activity (such as for scanning an object)), the value of the third feature may be 60 (e.g., indicating 60 minutes since the user device last reported accelerometer input), etc. These features and feature values ​​are provided as examples and may be different in other examples. For example, a feature set may include features associated with event-based data described herein, such as one or more of the following: time since last activity (e.g., a specific activity or any activity), time since last scan associated with a user device, time since last accelerometer input was detected by the user device, time since an application associated with the user device was accessed and / or used, time since the user device was last charged, time since last battery associated with the user device was replaced, time since last transaction associated with the user device, time since last communication with an access point, time since the user device was last detected by another device (e.g., a camera, RFID scanner, smart storage device, geo-fencing device, etc.), and / or time when the user device last checked in or out, etc. While the above features are listed as time since last event occurred, other features may include measurements other than time associated with the same event, such as a location feature associated with any of the above events (e.g., the location where the event last occurred may have a value corresponding to a risk value); an absolute time feature associated with any of the above events (e.g., one day, one week, one month, one year, etc.); a count feature associated with the number of occurrences of any of the above events, etc. Other characteristics may include characteristics associated with the type of user device, a location associated with the user device (e.g., separate from the location of any particular event), a user of the user device (e.g., the user of the device and / or the last event associated with the device may have a value corresponding to a risk value), and the like.

[0045] As indicated by reference numeral 315, the set of observations may be associated with a target variable. The target variable may represent a variable having a numerical value, may represent a variable having a value that falls within a range of values ​​or having certain discrete possible values, may represent a variable selected from one of a plurality of options (e.g., one of a plurality of categories, classifications, or labels), and / or may represent a variable having a Boolean value. The target variable may be associated with a target variable value, and the target variable value may be specific to an observation. In example 300, the target variable is the likelihood of abandonment, which has a value of 0.3 for the first observation (e.g., indicating that the confidence measure, between 0 and 1, that the observation indicates that the corresponding user device was abandoned).

[0046] The feature sets and target variables described above are provided as examples only, and other examples may be different from those described above. For example, for a target variable indicating the possibility of loss (relative to intentional deprivation), the feature set may include the time since the user equipment last communicated with the access point, etc.

[0047] The target variable may represent the value that the machine learning model is being trained to predict, and the feature set may represent the variables that are input to the trained machine learning model to predict the value of the target variable. The observation set may include the target variable values, so that the machine learning model can be trained to recognize patterns in the feature set that lead to the target variable values. A machine learning model trained to predict the value of the target variable may be referred to as a supervised learning model.

[0048] In some embodiments, a machine learning model may be trained on an observation set that does not include a target variable. This may be referred to as an unsupervised learning model. In this case, the machine learning model may learn patterns from the observation set without labeling or supervision, such as by using clustering and / or association to identify groups of related items within the observation set, and may provide outputs indicative of these patterns.

[0049] As shown in reference numeral 320, the machine learning system can train a machine learning model using the observation set and using one or more machine learning algorithms (such as a regression algorithm, a decision tree algorithm, a neural network algorithm, a k-nearest neighbor algorithm, a support vector machine algorithm, etc.). After training, the machine learning system can store the machine learning model as a trained machine learning model 325 for use in analyzing new observations.

[0050] As an example, the machine learning system may obtain training data for an observation set based on event-based data received from user devices and external devices. For example, event-based data from various devices may be received and stored over time for use as training data. In some embodiments, the training data may be further based on user input. For example, the user input may indicate which observations (e.g., event-based data and / or corresponding user devices) are associated with actual abandonment instances. As shown in reference numeral 315, the target variable may be updated based on actual instances of abandonment or non-abandonment (e.g., 1 may indicate that the observation is associated with the user device being abandoned, while 0 indicates that the observation is associated with the user device not being abandoned).

[0051] As indicated by reference numeral 330, the machine learning system may apply the trained machine learning model 325 to the new observation, such as by receiving the new observation and inputting the new observation into the trained computer learning model 325. As shown, as an example, the new observation may include a first feature of location type, a second feature of time since the last activity, a third feature of time since the last accelerometer input, and the like. The machine learning system may apply the trained machine learning model 325 to the new observation to generate an output (e.g., a result). The type of output may depend on the type of machine learning model and / or the type of machine learning task being performed. For example, such as when supervised learning is employed, the output may include a predicted value of a target variable. Additionally or alternatively, such as when unsupervised learning is employed, the output may include information identifying the cluster to which the new observation belongs and / or information indicating the similarity between the new observation and one or more other observations.

[0052] As an example, the trained machine learning model 325 may predict that the target variable value for the abandonment probability of the new observation is 0.75, as shown in reference numeral 335. Based on the prediction, the machine learning system may provide a first recommendation, may provide an output for determining the first recommendation, may perform a first automatic action, and / or may cause the first automatic action to be performed (e.g., by instructing another device to perform the automatic action), etc. The first recommendation may include, for example, a recommendation to restore the user device associated with the new observation, a recommendation to send an alert to the user device or other device, etc. The first automatic action may include, for example, transmitting a notification to a management device or other device to enable the restoration of the user device associated with the new observation; transmitting an alert to the user device so that the user device takes an action that facilitates discovery, such as playing a sound, vibrating, activating a light, etc.

[0053] As another example, if the machine learning system predicts a value of 0.3 for the target variable of abandonment likelihood, then the machine learning system may provide a second (e.g., different) recommendation (e.g., marking the corresponding user device as potentially abandoned, but not yet in need of recovery) and / or may perform or cause performance of a second (e.g., different) automatic action (e.g., alerting another device, such as another user device, in a location near the user device that corresponds to the new observation).

[0054] In some embodiments, the trained machine learning model 325 may classify (e.g., cluster) new observations in clusters, as shown in reference numeral 340. Observations within a cluster may have a threshold similarity. As an example, if the machine learning system classifies the new observation in a first cluster (e.g., observations associated with user devices that may be abandoned), the machine learning system may provide a first recommendation, such as the first recommendation described above. Additionally or alternatively, the machine learning system may perform a first automatic action based on classifying the new observation into the first cluster and / or may cause the first automatic action to be performed (e.g., by instructing another device to perform the automatic action), such as the first automatic action described above.

[0055] As another example, if the machine learning system classifies the new observation into a second cluster (e.g., observations associated with user devices at risk of being abandoned), then the machine learning system may provide a second (e.g., different) recommendation (e.g., a recommendation to notify nearby devices of the user device associated with the new observation) and / or may perform or cause performance of a second (e.g., different) automatic action, such as sending an alert to the user device.

[0056] In some embodiments, recommendations and / or automatic actions associated with a new observation may be based on a target variable value having a particular label (e.g., a classification or categorization), may be based on whether the target variable value satisfies one or more thresholds (e.g., the target variable value is greater than a threshold, less than a threshold, equal to a threshold, falls within a threshold range, etc.), and / or may be based on a cluster into which the new observation is classified.

[0057] In some embodiments, feedback information can be used to retrain the trained machine learning model 325. For example, feedback can be provided to the machine learning model. The feedback can be associated with actions performed based on recommendations provided by the trained machine learning model 325 and / or automatic actions performed or caused by the trained machine learning model 325. In other words, the recommendations and / or actions output by the trained machine learning model 325 can be used as input to retrain the machine learning model (e.g., a feedback loop can be used to train and / or update the machine learning model). For example, the feedback information can include a manually provided value to indicate whether a user device associated with an observation is actually abandoned.

[0058] In this way, the machine learning system can apply a rigorous and automated process to detect whether a user device is likely to be abandoned. The machine learning system is able to recognize and / or identify tens, hundreds, thousands, or millions of features and / or feature values ​​for tens, hundreds, thousands, or millions of observations, thereby improving accuracy and consistency and reducing latency associated with detecting whether a user device is likely to be abandoned, relative to the need to allocate computing resources for tens, hundreds, or thousands of operators to manually detect whether a user device is likely to be abandoned using features or feature values.

[0059] As mentioned above, Figure 3 Provided as an example. Other examples can be combined with Figure 3 Different than described.

[0060] Figure 4 4 is an illustration of an example environment 400 in which the systems and / or methods described herein may be implemented. Figure 4 As shown, environment 400 may include user device 410, external device 420, modeling device 430, server device 440, management device 450, and network 460. The devices of environment 400 may be interconnected via wired connections, wireless connections, or a combination of wired and wireless connections.

[0061] User device 410 includes one or more devices that are capable of receiving, generating, storing, processing, and / or providing information related to detecting a state of a user device, e.g., detecting when a user device may be abandoned or otherwise lost, as described elsewhere herein. User device 410 may include a communication device and / or a computing device. For example, user device 410 may include a wireless communication device, a mobile phone, user equipment, a laptop computer, a tablet computer, a desktop computer, a game console, a set-top box, a wearable communication device (e.g., a smart watch, a pair of smart glasses, a head-mounted display, or a virtual reality headset), a handheld scanning device, or a similar type of device.

[0062] The external device 420 includes one or more devices that are capable of receiving, generating, storing, processing, providing and / or routing information associated with the state of a user device, such as providing event-based data to facilitate determining the abandonment probability associated with the user device, as described elsewhere herein. The external device 420 may include a communication device and / or a computing device. For example, the external device 420 may include a server, such as an application server, a client server, a network server, a database server, a host server, a proxy server, a virtual server (e.g., executed on computing hardware), or a server in a cloud computing system. In some embodiments, the external device 420 includes computing hardware used in a cloud computing environment.

[0063] Modeling device 430 includes one or more devices that can receive, generate, store, process, provide and / or route information associated with a model for determining a user device state, as described elsewhere herein. Modeling device 430 may include a communication device and / or a computing device. For example, modeling device 430 may include a server, such as an application server, a client server, a network server, a database server, a host server, a proxy server, a virtual server (e.g., executed on computing hardware), or a server in a cloud computing system. In some embodiments, modeling device 430 includes computing hardware used in a cloud computing environment.

[0064] The server device 440 includes one or more devices that can receive, generate, store, process, provide and / or route information associated with detecting the state of a user device and facilitating recovery of the user device, as described elsewhere herein. The server device 440 may include a communication device and / or a computing device. For example, the server device 440 may include a server, such as an application server, a client server, a network server, a database server, a host server, a proxy server, a virtual server (e.g., executed on computing hardware), or a server in a cloud computing system. In some embodiments, the server device 440 includes computing hardware used in a cloud computing environment.

[0065] The management device 450 includes one or more devices that are capable of receiving, generating, storing, processing, providing and / or routing information associated with facilitating the recovery of a user device that may be abandoned, as described elsewhere herein. The management device 450 may include a communication device and / or a computing device. For example, the management device 450 may include a wireless communication device, a mobile phone, a user equipment, a laptop computer, a tablet computer, a desktop computer, a game console, a set-top box, a wearable communication device (e.g., a smart watch, a pair of smart glasses, a head-mounted display or a virtual reality headset), a handheld scanning device or a similar type of device; a server, such as an application server, a smart watch, a pair of smart glasses, a head-mounted display or a virtual reality headset), a handheld scanning device or a similar type of device; a server, such as an application server, a client server, a network server, a database server, a host server, a proxy server, a virtual server (e.g., executed on computing hardware), or a server in a cloud computing system. In some embodiments, the management device 450 includes computing hardware used in a cloud computing environment.

[0066] Network 460 includes one or more wired networks and / or wireless networks. For example, network 460 may include a wireless wide area network (e.g., a cellular network or a public land mobile network), a local area network (e.g., a wired local area network or a wireless local area network (WLAN), such as a Wi-Fi network), a personal area network (e.g., a Bluetooth network), a near field communication network, a telephone network, a private network, the Internet, and / or a combination of these or other types of networks. Network 460 enables communication between devices in environment 400.

[0067] Figure 4 The number and arrangement of devices and networks shown in are provided as examples. Figure 4 There may be additional devices and / or networks, fewer devices and / or networks, different devices and / or networks, or differently arranged devices and / or networks compared to the devices and / or networks shown in FIG. Figure 4 Two or more of the devices shown in the figure may be implemented in a single device, or Figure 4 A single device shown in the environment 400 may be implemented as multiple distributed devices. Additionally or alternatively, a device set (eg, one or more devices) of the environment 400 may perform one or more functions that are described as being performed by another device set of the environment 400.

[0068] Figure 5 4 is a diagram of example components of device 500, which may correspond to user device 410, external device 420, modeling device 430, server device 440, and / or management device 450. In some embodiments, user device 410, external device 420, modeling device 430, server device 440, and / or management device 450 include one or more devices 500 and / or one or more components of device 500. Figure 5 As shown in , device 500 may include a bus 510 , a processor 520 , a memory 530 , an input component 540 , an output component 550 , and a communication component 560 .

[0069] The bus 510 includes one or more components that enable wired and / or wireless communication between components of the device 500. The bus 510 may enable the components of the device 500 to communicate with each other, such as by operational coupling, communication coupling, electronic coupling, and / or electrical coupling. Figure 5Two or more components in are coupled together. Processor 520 includes a central processing unit, a graphics processing unit, a microprocessor, a controller, a microcontroller, a digital signal processor, a field programmable gate array, an application specific integrated circuit, and / or another type of processing component. Processor 520 is implemented in hardware, firmware, or a combination of hardware and software. In some embodiments, processor 520 includes one or more processors that can be programmed to perform one or more operations or processes as described elsewhere herein.

[0070] The memory 530 includes volatile and / or non-volatile memory. For example, the memory 530 may include random access memory (RAM), read-only memory (ROM), a hard drive, and / or other types of memory (e.g., flash memory, magnetic memory, and / or optical memory). The memory 530 may include internal memory (such as RAM, ROM, or a hard drive) and / or removable memory (e.g., a removable memory connected via a universal serial bus). The memory system 530 may be a non-transitory computer-readable medium. The memory 530 stores information, instructions, and / or software (e.g., one or more software applications) related to the operation of the device 500. In some embodiments, the memory 530 includes one or more memories, which are coupled to one or more processors (e.g., processor 520), such as via a bus 510.

[0071] Input component 540 enables device 500 to receive input, such as user input and / or sensed input. For example, input component 540 may include a touch screen, keyboard, keypad, mouse, button, microphone, switch, sensor, global positioning system sensor, accelerometer, gyroscope and / or actuator. Output component 550 enables device 500 to provide output such as via a display, speaker and / or light emitting diode. Communication component 560 enables device 500 to communicate with other devices such as via wired connection and / or wireless connection. For example, communication component 560 may include a receiver, a transmitter, a transceiver, a modem, a network interface card and / or an antenna.

[0072] Device 500 may perform one or more operations or processes described herein. For example, a non-transient computer-readable medium (e.g., memory 530 and / or storage component 640) may store an instruction set (e.g., one or more instructions or codes) to be executed by processor 520. Processor 520 may execute an instruction set to perform one or more operations or processes described herein. In some embodiments, execution of an instruction set by one or more processors 520 causes one or more processors 520 and / or device 500 to perform one or more operations or processes described herein. In some embodiments, hardwired circuitry may be used instead of instructions or in combination with instructions to perform one or more operations or processes described herein. Additionally or alternatively, processor 520 may be configured to perform one or more operations or processes as described herein. Therefore, the embodiments described herein are not limited to any particular combination of hardware circuitry and software.

[0073] Figure 5 The number and arrangement of components shown in FIG. 5 are provided as examples. Figure 5 The arrangement comparison shown in may include additional components, fewer components, different components, or differently arranged components. Additionally or alternatively, a component set (e.g., one or more components) of device 500 may perform one or more functions that are described as being performed by another component set of device 500.

[0074] Figure 6 is a flow chart of an example process 600 associated with a system for detecting abandonment of a user device. In some embodiments, Figure 6 One or more process blocks of may be performed by a device server (e.g., server device 440). In some embodiments, Figure 6 One or more process blocks of may be performed by another device or device group independent of or including the server device, such as a server device (e.g., server device 440), an external device (e.g., external device 420), a modeling device (e.g., modeling device 430), and / or a management device (e.g., management device 450). Additionally or alternatively, Figure 6 One or more process blocks of may be performed by one or more components of device 500, such as processor 520, memory 530, input component 540, output component 550, and / or communication component 560.

[0075] like Figure 6As shown, process 600 may include receiving event-based data associated with a user device, wherein the user device is communicatively coupled to a server device via an access point in a physical environment (block 610). For example, a server device may receive event-based data associated with a user device, wherein the user device is communicatively coupled to the server device via an access point in the physical environment, as described above. In some embodiments, the user device may be communicatively coupled to the server device via an access point within the physical environment. Event-based data may be received from a user device based on at least one of the user device being configured to report event-based data according to a plan, the user device becoming communicatively coupled to an access point, or sensor data associated with a sensor of the user device indicating that a particular type of event associated with the user device has occurred.

[0076] like Figure 6 As further shown, the processor 600 may include using a model to determine the state of a user device, the state of the user device is based on event-based data, wherein the model is configured based on historical event data associated with historical events, and these historical events involve multiple other user devices associated with the user device (box 620). For example, a server device uses a model to determine the state of a user device, the state of the user device is based on event-based data, wherein the model is configured based on historical event data associated with historical events, and these historical events involve multiple other user devices associated with the user device, as described above. In some embodiments, the model is configured based on historical event data associated with historical events, and the historical events involve multiple other user devices associated with the user device. The state may indicate whether the user device has been abandoned within a threshold time period when the user uses the user device in a physical environment. The state may also indicate the probability that the user device has been abandoned, wherein the probability meets the probability threshold associated with the user device being abandoned by the user. In some embodiments, the process 600 may include generating a model. For example, a server device or another device may generate, train and / or retrain a model, as described herein.

[0077] In some embodiments, the historical event data includes first historical event data and second historical event data, the first historical event data identifying a first value of an event parameter associated with a user device that has not been abandoned, and the second historical event data identifying a second value of an event parameter associated with a user device that has been abandoned. The event parameters may include: user interaction metrics associated with other user devices, application metrics associated with applications installed on other user devices, location information identifying the location of other user devices during the historical event, power consumption metrics associated with batteries of other user devices, and / or replacement intervals associated with replacing batteries of other user devices.

[0078] like Figure 6As further shown, process 600 may include determining a location associated with the user device based on a state indicating that the user device may have been abandoned in the physical environment (block 630). For example, the server device may determine a location associated with the user device based on a state indicating that the user device may have been abandoned in the physical environment, as described above.

[0079] like Figure 6 As further shown, process 600 may include performing an action associated with the user device and the location (block 640). For example, the server device may perform an action associated with the user device and the location, as described above. In some embodiments, performing the action includes causing the user device to activate an alarm via a speaker of the user device, or providing location information identifying the location of the user device to the management device.

[0080] although Figure 6 An example block diagram of process 600 is shown. In some implementations, process 600 may be performed in conjunction with Figure 6 Compared to the blocks depicted in the process 600, additional blocks, fewer blocks, different blocks, or differently arranged blocks may be included. Additionally or alternatively, two or more blocks of the blocks of process 600 may be executed in parallel.

[0081] The above disclosure provides illustration and description, but is not intended to be exhaustive or to limit the embodiments to the precise form disclosed. Modifications and variations may be made in light of the above disclosure or may be acquired from practice of the embodiments.

[0082] As used herein, the term "component" is intended to be interpreted broadly as hardware, firmware, and / or a combination of hardware and software. As used herein, each of the terms "tangible machine-readable medium", "non-transitory machine-readable medium", and "machine-readable storage device" is explicitly defined as a storage medium (e.g., a disk of a hard disk drive, a digital versatile disk, an optical disk, a flash memory, a read-only memory, a random access memory, etc.), on which machine-readable instructions (e.g., codes in the form of software and / or firmware) can be stored. Instructions may be stored to any suitable duration, such as permanently extending the time period storage (e.g., when executing a program associated with the instruction) or a short time period storage (e.g., when the instruction is cached, during buffering, etc.). Further, as used herein, each of the terms "tangible machine-readable medium", "non-transitory machine-readable medium", and "machine-readable storage device" is explicitly defined to exclude propagation signals. That is, as used in any claim herein, "tangible machine-readable medium", "non-transitory machine-readable medium", and "machine-readable storage device", etc. should not be interpreted as being implemented as a propagation signal.

[0083] As used herein, satisfying a threshold may refer to a value that is greater than a threshold, greater than or equal to a threshold, less than a threshold, less than or equal to a threshold, equal to a threshold, not equal to a threshold, etc., depending on the context.

[0084] It will be apparent that the systems and / or methods described herein can be implemented in various forms of hardware, firmware, or a combination of hardware and software. The actual dedicated control hardware or software code used to implement these systems and / or methods is not a limitation of the implementation. Therefore, the operation and behavior of the systems and / or methods are described herein without reference to specific software code - it should be understood that software and hardware can be designed to implement the systems and / or methods based on the description herein.

[0085] Even if a particular combination of features is described in the claims and / or disclosed in the specification, these combinations are not intended to limit the disclosure of the various embodiments. In fact, many of these features can be combined in a manner not specifically set forth in the claims and / or not disclosed in the specification. Although each dependent claim listed below may directly depend on only one claim, the disclosure of the various embodiments includes the combination of each dependent claim with each other claim in the claims. As used herein, a phrase citing "at least one" of a list of items refers to any combination of these items, including a single member. As an example, "at least one of a, b, or c" is intended to cover: a, b, c, ab, ac, bc, and abc, as well as any combination with multiple identical items.

[0086] Unless explicitly stated, any element, action or instruction used in this article should not be interpreted as key or necessary. In addition, as used in this article, the article "a" and "an" are intended to include one or more items, and can be used interchangeably with "one or more". In addition, as used in this article, the article "the" is intended to include one or more items quoted in conjunction with the article "the", and can be used interchangeably with "the one or more". In addition, as used in this article, the term "set (set)" is intended to include one or more items (for example, related items, unrelated items or related items and unrelated items The combination of items), and can be used interchangeably with "one or more". If it is intended to represent only one item, the phrase "only one" or similar language is used. In addition, as used in this article, the term "has (has)", "have (have)", "have (having)" or the like are intended to be open terms. In addition, unless explicitly stated otherwise, the phrase "based on" is intended to represent "at least partially based on". Furthermore, as used herein, the term "or" when used in series is intended to be inclusive and may be used interchangeably with "and / or" unless expressly stated otherwise (e.g., if used in combination with "either" or "only one").

Claims

1. A method for detecting the status of a user device, the method include: receiving, by a server device, event-based data associated with the user device, wherein the user device is communicatively coupled to the server device via an access point within the physical environment; determining, by the user device and using a model, a state of the user device based on the event-based data, wherein the model is configured based on historical event data associated with historical events involving a plurality of other user devices associated with the user device; determining, by the server device and based on the status indicating that the user device may have been abandoned in the physical environment, a location associated with the user device; as well as An action associated with the user device and the location is performed by the server device.

2. The method of claim 1 , wherein the event-based data is received from the user device based on the user device being configured to report the event-based data according to at least one of: plan, The user equipment becomes communicatively coupled to the access point, or Sensor data associated with the user equipment sensor indicates that a particular type of event associated with the user equipment has occurred. 3 . The method of claim 1 , wherein the status indicates whether the user device has been abandoned within a threshold time period while the user is using the user device in a physical environment.

4. The method of claim 1, wherein the status indicates a probability that the user device has been abandoned, the probability satisfying a probability threshold associated with the user device being abandoned by a user.

5. The method of claim 1, wherein the historical event data include: first historical event data identifying a first value of an event parameter associated with a first set of a plurality of other user devices that were not abandoned, and Second historical event data identifying a second value of the event parameter associated with a second set of a plurality of other user devices that were abandoned.

6. The method of claim 5, wherein the event parameters include at least one of the following: user interaction metrics associated with the plurality of other user devices, application metrics associated with applications installed on the plurality of other user devices, location information identifying locations of the plurality of other user devices during the historical event, power consumption metrics associated with the plurality of other user device batteries, or A replacement interval associated with replacing batteries of the plurality of other user devices.

7. The method of claim 1, wherein performing the action comprises at least one of: causing the user device to activate an alarm via a speaker of the user device; or Location information identifying the location of the user equipment is provided to a management device.

8. A system for detecting the status of a user device, the system include: one or more memories; as well as one or more processors, the one or more processors coupled to the one or more memories, the one or more processors configured to: receiving, via an access point, event-based data associated with the user equipment, wherein the user equipment is communicatively coupled to the access point in a physical environment associated with an entity; determining a state of the user device using a model and based on one or more event parameters associated with the event-based data, wherein the model is configured based on historical event data associated with historical events involving historical usage of a plurality of other user devices associated with the entity, and wherein the one or more event parameters are associated with one or more events involving the user device; determining, based on the state, that the user equipment is likely to be abandoned in the physical environment; as well as An action associated with locating the user equipment is performed.

9. The system of claim 8, wherein the event-based data is received from the user device based on the user device being configured to report the event-based data while communicatively coupled with the access point.

10. The system of claim 8, wherein the status indicates at least one of the following: whether the user device may be currently being used by the user or may have been abandoned by the user, or The location of the user equipment.

11. A system as claimed in claim 8, wherein the model includes a machine learning model, which is trained based on the historical event data, and the historical event data indicates that users of the multiple other user devices involved in the historical event abandoned the multiple other user devices after their respective historical uses of the multiple other user devices.

12. The system of claim 8, wherein the one or more events are associated with at least one of: interacting with a user of the user device, an application installed on said other user device that performs an application operation, the type of movement of the user equipment, the user equipment has not moved within a threshold period of time, an operation involving charging of a battery of said user equipment, or Replace the battery in the user device.

13. The system of claim 8, wherein the one or more processors are configured to perform the actions by: causing the user device to activate an audible alarm, or An indication is provided to a management device that the user equipment may have been abandoned.

14. The system of claim 8, wherein the one or more processors are configured to perform the actions by: determining a location of the user equipment in the physical environment; and Location information identifying the location of the user equipment is provided to a management device.

15. A tangible machine-readable medium storing a set of instructions for detecting a state of a user device, the set of instructions include: One or more instructions that, when executed by one or more processors of a system, cause the system to: receiving event-based data associated with the user device in a physical environment, wherein the user equipment is communicatively coupled to an access point in the physical environment, and wherein the event-based data is received from the user equipment via the access point; monitoring a state of the user device using a model, the state being based on one or more events identified in the event-based data, wherein the model is configured based on historical event data associated with historical events involving a plurality of other user devices associated with the user device, and wherein the status indicates whether the user equipment has been abandoned; as well as An indication of the status is provided to a management device to allow the management device to monitor abandonment of the user equipment.

16. A tangible machine-readable medium as described in claim 15, wherein the event-based data is received from the user device based on the user device being configured to report the event-based data based on a sensor of the user device indicating that a particular type of event associated with abandonment of the user device may have occurred.

17. The tangible machine-readable medium of claim 15, wherein the event-based data indicates at least one of: Whether the user equipment has been used by the user within the threshold event period, whether the battery of the user device is charged, or The location of the user equipment.

18. The tangible machine-readable medium of claim 15, wherein the historical event data include: first historical event data, the first historical event data identifying a first value of an event parameter associated with a first set of the plurality of other user devices, the first set of the other user devices not being abandoned in association with a first corresponding historical event of the historical event, and Second historical event data, wherein the second historical event data identifies a second value of an event parameter associated with a second set of the plurality of user devices, the second set of the other user devices being abandoned in association with the second corresponding historical event of the historical events.

19. The tangible machine-readable medium of claim 15, wherein the user device and the plurality of other user devices are associated with the same entity.

20. The transitory machine-readable storage medium of claim 15, wherein the one or more instructions, when executed by the one or more processors, further cause the one or more processors to: An indication of the status is provided to the user equipment to allow the user equipment to indicate the status via a user interface of the user equipment.

Images