Abstract: Systems and methods of using sensor data for coordinate prediction are disclosed herein. In some example embodiments, for a place, a computer system accesses corresponding service data comprising pick-up data and drop-off data for requests, and accesses corresponding sensor data indicating at least one path of mobile devices of the requesters of the requests, with the at least one path comprising at least one of a pick-up path ending at the pick-up location indicated by the pick-up data and a drop-off path beginning at the drop-off location indicated by the drop-off data. In some example embodiments, the computer system generates at least one predicted geographic location using the paths indicated by the sensor data, and stores the at least one predicted geographic location in a database in association with an identification of the place.

Abstract: Systems and methods of using sensor data for coordinate prediction are disclosed herein. In some example embodiments, for a place, a computer system accesses corresponding service data comprising pick-up data and drop-off data for requests, and accesses corresponding sensor data indicating at least one path of mobile devices of the requesters of the requests, with the at least one path comprising at least one of a pick-up path ending at the pick-up location indicated by the pick-up data and a drop-off path beginning at the drop-off location indicated by the drop-off data. In some example embodiments, the computer system generates at least one predicted geographic location using the paths indicated by the sensor data, and stores the at least one predicted geographic location in a database in association with an identification of the place.

Abstract: Systems and methods for detecting and verifying closed places (e.g., claims no longer in business) from trip data are provided. A networked system accesses trip data associated with the POI. The networked system processes the trip data to generate at least two time buckets based on timestamps from the trip data associated with the POI, and calculates trip counts associated with the POI for each of the time buckets. Using a machine learning algorithm and based on the at least two time buckets, the networked system determines that the trip counts show a decline over time that indicates that the POI is likely closed. In response to the determining, the networked system updates a database to indicate the POI is closed.

Abstract: An authoritative candidate is selected for determining a location of a point of interest (POI). Source data including name, address, and location for POIs is received from multiple data sources. The received data is normalized for ease of comparison, and coordinates for each candidate are compared to coordinates of other candidates to determine which candidate if any is an authoritative location for the POI. The candidate locations are compared using two models a metric-based scoring system and a machine learning model that may utilize a gradient boosted decision tree. The authoritative candidate can be used to render digital maps that include the POI. In addition, the authoritative candidate's location can be used to provide vehicle route guidance to the POI.

Abstract: Systems and methods for improving attribute data for a point of interest (POI) are provided. A networked system accesses trip data associated with the POI. The networked system generates, using a processor-implemented clustering algorithm, a first spatial cluster and a second spatial cluster using coordinates corresponding to the POI indicated in the trip data. A centroid for the first spatial cluster and a centroid for the second spatial cluster are identified by the networked system. The networked system determines that a difference in distance between the centroid for the first spatial cluster and the centroid for the second spatial cluster meets or transgresses a centroid distance threshold. In response to the determining, a database is updated to indicate a new attribute for the POI, the new attribute corresponds to an attribute associated with either the first spatial cluster or the second spatial cluster.

Abstract: Systems and methods of identifying incorrect coordinate prediction using route information are disclosed herein. In some example embodiments, a computer system receives route information and geographic image data. The route information corresponds to a servicing of a request associated with a place and indicates a travelled route that has been traveled by a user in traveling from an origin location to the place, and the geographic image data represents a geographic area corresponding to the travelled route. The computer system determines that an initial geographic location stored in a database in association with the place is incorrect based on the route information and the geographic image data using a first deep learning model, and then performs a verification operation based on the determining that the stored initial geographic location of the place is incorrect.

Abstract: Systems and methods of using sensor data for coordinate prediction are disclosed herein. In some example embodiments, for a place, a computer system accesses corresponding service data comprising pick-up data and drop-off data for requests, and accesses corresponding sensor data indicating at least one path of mobile devices of the requesters of the requests, with the at least one path comprising at least one of a pick-up path ending at the pick-up location indicated by the pick-up data and a drop-off path beginning at the drop-off location indicated by the drop-off data. In some example embodiments, the computer system generates at least one predicted geographic location using the paths indicated by the sensor data, and stores the at least one predicted geographic location in a database in association with an identification of the place.

Abstract: Systems and methods for improving accuracy of geographic position data are provided. A networked system mines ticket data from content of the ticket. Based on the ticket data, a determination is made that the ticket indicates an issue with a trip involving a point of interest (POI). The networked system extracts trip data from a trip log corresponding to the trip involving the POI, and identifies, from a data storage, stored attributes of the POI. The networked system analyzes the ticket data, trip data, and attributes to determine a workflow to improve accuracy of the POI, whereby the analyzing comprises determining a priority level to verify accuracy of the POI. The workflow is triggered based on the priority level to verify accuracy of the POI.

Abstract: Systems and methods of using sensor data for coordinate prediction are disclosed herein. In some example embodiments, for a place, a computer system accesses corresponding service data comprising pick-up data and drop-off data for requests, and accesses corresponding sensor data indicating at least one path of mobile devices of the requesters of the requests, with the at least one path comprising at least one of a pick-up path ending at the pick-up location indicated by the pick-up data and a drop-off path beginning at the drop-off location indicated by the drop-off data. In some example embodiments, the computer system generates at least one predicted geographic location using the paths indicated by the sensor data, and stores the at least one predicted geographic location in a database in association with an identification of the place.

Abstract: Systems and methods of deep learning coordinate prediction using satellite and service data are disclosed herein. In some example embodiments, for each one of a plurality of places, a computer system trains a deep learning model based on training data of the plurality of places. The deep leaning model is configured to generate a predicted geographical location of a place based on satellite image data and service data associated with the place. The training data for each place comprises satellite image data of the place, service data, and a ground truth geographical location of the place. The service data comprises at least one of pick-up data indicating a geographical location at which a provider started transporting a requester in servicing a request associated with the place or drop-off data indicating a geographical location at which the provider completed transporting the requester in servicing the request associated with the place.

Abstract: Systems and methods of identifying incorrect coordinate prediction using route information are disclosed herein. In some example embodiments, a computer system receives route information and geographic image data. The route information corresponds to a servicing of a request associated with a place and indicates a travelled route that has been traveled by a user in traveling from an origin location to the place, and the geographic image data represents a geographic area corresponding to the travelled route. The computer system determines that an initial geographic location stored in a database in association with the place is incorrect based on the route information and the geographic image data using a first deep learning model, and then performs a verification operation based on the determining that the stored initial geographic location of the place is incorrect.

Abstract: An authoritative candidate is selected for determining a location of a point of interest (POI). Source data including name, address, and location for POIs is received from multiple data sources. The received data is normalized for ease of comparison, and coordinates for each candidate are compared to coordinates of other candidates to determine which candidate if any is an authoritative location for the POI. The candidate locations are compared using two models a metric-based scoring system and a machine learning model that may utilize a gradient boosted decision tree. The authoritative candidate can be used to render digital maps that include the POI. In addition, the authoritative candidate's location can be used to provide vehicle route guidance to the POI.

Abstract: An authoritative candidate is selected for determining a location of a point of interest (POI). Source data including name, address, and location for POIs is received from multiple data sources. The received data is normalized for ease of comparison, and coordinates for each candidate are compared to coordinates of other candidates to determine which candidate if any is an authoritative location for the POI. The candidate locations are compared using two models a metric-based scoring system and a machine learning model that may utilize a gradient boosted decision tree. The authoritative candidate can be used to render digital maps that include the POI. In addition, the authoritative candidate's location can be used to provide vehicle route guidance to the POI.

Abstract: An authoritative candidate is selected for determining a location of a point of interest (POI). Source data including name and location for POIs is received from multiple data sources. The received data is normalized for ease of comparison, and coordinates for each candidate are compared to coordinates of other candidates to determine which candidates lie outside of a consensus set of candidates. A consensus set of candidates are those that are each located on a same side of a road segment. Candidates that are not part of the consensus set are eliminated from consideration. Further criteria are then applied to the consensus set of candidates. The authoritative candidate can be used to render digital maps that include the POI. In addition, the authoritative candidate's location can be used to provide vehicle route guidance to the POI.

Abstract: Systems and methods of deep learning coordinate prediction using satellite and service data are disclosed herein. In some example embodiments, for each one of a plurality of places, a computer system trains a deep learning model based on training data of the plurality of places. The deep leaning model is configured to generate a predicted geographical location of a place based on satellite image data and service data associated with the place. The training data for each place comprises satellite image data of the place, service data, and a ground truth geographical location of the place. The service data comprises at least one of pick-up data indicating a geographical location at which a provider started transporting a requester in servicing a request associated with the place or drop-off data indicating a geographical location at which the provider completed transporting the requester in servicing the request associated with the place.

Abstract: Systems and methods for improving attribute data for a point of interest (POI) are provided. A networked system accesses trip data associated with the POI. The networked system generates, using a processor-implemented clustering algorithm, a first spatial cluster and a second spatial cluster using coordinates corresponding to the POI indicated in the trip data. A centroid for the first spatial cluster and a centroid for the second spatial cluster are identified by the networked system. The networked system determines that a difference in distance between the centroid for the first spatial cluster and the centroid for the second spatial cluster meets or transgresses a centroid distance threshold. In response to the determining, a database is updated to indicate a new attribute for the POI, the new attribute corresponds to an attribute associated with either the first spatial cluster or the second spatial cluster.

Abstract: Systems and methods for detecting and verifying closed places (e.g., claims no longer in business) from trip data are provided. A networked system accesses trip data associated with the POI. The networked system processes the trip data to generate at least two time buckets based on timestamps from the trip data associated with the POI, and calculates trip counts associated with the POI for each of the time buckets. Using a machine learning algorithm and based on the at least two time buckets, the networked system determines that the trip counts show a decline over time that indicates that the POI is likely closed. In response to the determining, the networked system updates a database to indicate the POI is closed.

Abstract: An authoritative candidate is selected for determining a location of a point of interest (POI). Source data including name, address, and location for POIs is received from multiple data sources. The received data is normalized for ease of comparison, and coordinates for each candidate are compared to coordinates of other candidates to determine which candidate if any is an authoritative location for the POI. The candidate locations are compared using two models a metric-based scoring system and a machine learning model that may utilize a gradient boosted decision tree. The authoritative candidate can be used to render digital maps that include the POI. In addition, the authoritative candidate's location can be used to provide vehicle route guidance to the POI.

Abstract: An authoritative candidate is selected for determining a location of a point of interest (POI). Source data including name and location for POIs is received from multiple data sources. The received data is normalized for ease of comparison, and coordinates for each candidate are compared to coordinates of other candidates to determine which candidates lie outside of a consensus set of candidates. A consensus set of candidates are those that are each located on a same side of a road segment. Candidates that are not part of the consensus set are eliminated from consideration. Further criteria are then applied to the consensus set of candidates. The authoritative candidate can be used to render digital maps that include the POI. In addition, the authoritative candidate's location can be used to provide vehicle route guidance to the POI.