Machine learning model for combining device presence data with physical venue transaction data
AssigneeAdentro, Inc.
InventorKai Umezawa
Life of this patent
Estimating the term…
Abstract
A device detection system is configured to determine whether a device is located within a venue. The device detection system detects pings from the device to determine whether the device is located within the venue. The device detection system receives point-of-sale data for a transaction from an entity associated with the venue. The point-of-sale data may be compared to data describing the detected device to determine whether the transaction was conducted by a customer associated with the detected device. The device detection system may correlate the point-of-sale data with content that was viewed by the customer to determine whether the content resulted in the customer visiting a venue and making purchases at the venue.
Claims20
A non-transitory computer readable storage medium comprising stored instructions, the instructions when executed cause at least one processor to:
- access a set of training data comprising a first set of wireless signals received from devices physically located within a geographic boundary of a venue and a second set of wireless signals received from devices physically located outside of the geographic boundary of the venue, wherein at least one of the second set of wireless signals are received during a time when the venue is closed;
- train a machine-learned model using the set of training data and configured to determine whether devices are located inside or outside the geographic boundary of the venue based on characteristics of wireless signals received from the devices;
- detect, by a wireless access point of a venue, a ping from a device;
- determine that the device is located within the venue by applying the machine-learned model to characteristics of the detected ping from the device;
- validate that the device is located within the venue based on an action performed by the device within the venue;
- and
- in response to validating that the device is located within the venue, re-train the machine-learned model based at least in part on the detected ping from the device.
A method comprising:
- accessing a set of training data comprising a first set of wireless signals received from devices physically located within a geographic boundary of a venue and a second set of wireless signals received from devices physically located outside of the geographic boundary of the venue, wherein at least one of the second set of wireless signals are received during a time when the venue is closed;
- training a machine-learned model using the set of training data and configured to determine whether devices are located inside or outside the geographic boundary of the venue based on characteristics of wireless signals received from the devices;
- detecting, by a wireless access point of a venue, a ping from a device;
- determining that the device is located within the venue by applying the machine-learned model to characteristics of the detected ping from the device;
- validating that the device is located within the venue based on an action performed by the device within the venue;
- and
- in response to validating that the device is located within the venue, re-training the machine-learned model based at least in part on the detected ping from the device.
A system comprising a hardware processor and a non-transitory computer readable storage medium storing executable instructions that, when executed by the hardware processor, cause the system to perform steps comprising:
- accessing a set of training data comprising a first set of wireless signals received from devices physically located within a geographic boundary of a venue and a second set of wireless signals received from devices physically located outside of the geographic boundary of the venue, wherein at least one of the second set of wireless signals are received during a time when the venue is closed;
- training a machine-learned model using the set of training data and configured to determine whether devices are located inside or outside the geographic boundary of the venue based on characteristics of wireless signals received from the devices;
- detecting, by a wireless access point of a venue, a ping from a device;
- determining that the device is located within the venue by applying the machine-learned model to characteristics of the detected ping from the device;
- validating that the device is located within the venue based on an action performed by the device within the venue;
- and
- in response to validating that the device is located within the venue, re-training the machine-learned model based at least in part on the detected ping from the device.
Dependencies are inferred from the claim text.