Associating user logs using geo-point density
A method for clustering geolocations using geo-point density includes receiving a user log of geolocation data extracted from user interactions with at least one electronic device. A density is determined relative to other geo-points for each geo-point in a set of geo-points extracted from the user log. Lower density geo-points in the set are merged into higher density geo-points in the set to result in a merged set of geo-points, and clusters of geo-points are identified from the merged set. Merging the geo-points tends to preserve frequently occurring geo-points while reducing those that constitute noise, which improves the reliability of identifying the clusters. Core geo-points of the user log are selected from the clusters. The core geo-points of the user log can be compared to core geo-points of other use logs to identify associations between the user logs.
1. A computer-implemented method, comprising:
extracting, via a processor, a plurality of geo-points from geolocation information in a plurality of logs;
determining, via the processor, two core geo-points from the plurality of geo-points, each core geo-point of the two core geo-points being representative of geolocation information in a corresponding log of the plurality of logs;
associating, via the processor, two logs corresponding to the two core geo-points to a common user based on respective features of the two core geo-points; and
selecting, via the processor, a personalized content for the common user based on the two core geo-points.
2. The computer-implemented method of claim 1 , further comprising:
determining respective densities of the plurality of geo-points; and
determining the two core geo-points based on the respective densities of the plurality of geo-points.
3. The computer-implemented method of claim 1 , further comprising:
determining a first number of geo-points of the plurality of geo-points;
determining a second number of geo-points that are within a designated area surrounding a geo-point of the plurality of geo-points; and
determining a density of the geo-point based on the first number of geo-points and the second number of geo-points.
4. The computer-implemented method of claim 3 , wherein the designated area surrounding the geo-point is determined based on an average distance between each geo-point to its nearest neighbor in the plurality of geo-points.
5. The computer-implemented method of claim 1 , further comprising:
determining respective densities of a first geo-point and a second geo-point of the plurality of geo-points; and
filtering out the first geo-point from the plurality of geo-points based on a density of the first geo-point being less than a density of the second geo-point.
6. The computer-implemented method of claim 1 , further comprising:
determining respective densities of the plurality of geo-points; and
iteratively merging, based on an order of respective densities of the plurality of geo-points, the plurality of geo-points until each geo-point in the plurality of geo-points has either been used as a given geo-point to merge another geo-point or has been merged.
7. The computer-implemented method of claim 1 , further comprising:
computing a cluster of geo-points from the plurality of geo-points using a k-means based clustering method; and
designating a center of the cluster of geo-points as a core geo-point of the two core geo-points.
8. The computer-implemented method of claim 1 , wherein the respective features of the two core geo-points comprise respective geolocation data; and the method further comprising:
computing a distance between the two core geo-points based on the respective geolocation data; and
associating the two logs based on the distance being below a threshold value.
9. The computer-implemented method of claim 1 ,
wherein the selecting comprises detecting the common user being located at a location corresponding to one of the two core geo-points.
10. A non-transitory computer-readable media having a plurality of executable instructions embodied thereon, which, when executed by one or more processors, cause the one or more processors to perform operations, comprising:
obtaining a plurality of geo-points from a plurality of logs;
determining a plurality of clusters of geo-points based on respective densities of the plurality of geo-points;
designating a plurality of core geo-points to represent respective clusters of the plurality of clusters of geo-points;
identifying associations among the plurality of logs based on the plurality of core geo-points; and
personalizing a content for a user based on the associations among the plurality of logs.
11. The computer-readable media of claim 10 , the operations further comprising:
determining a density of a geo-point of the plurality of geo-points based on a ratio between a first number of geo-points that are within a designated area surrounding the geo-point and a second number of geo-points of the plurality of geo-points.
12. The computer-readable media of claim 10 , the operations further comprising:
merging a first geo-point into a second geo-point based on a density of the first geo-point being less than a density of the second geo-point, and a distance between the first geo-point and the second geo-point being less than a threshold value.
13. The computer-readable media of claim 10 , wherein designating the plurality of core geo-points comprises:
computing a cluster of geo-points from the plurality of geo-points using a k-means based clustering method; and
designating a core geo-point for the cluster of geo-points based on a center point of the cluster of geo-points.
14. The computer-readable media of claim 10 , wherein identifying the associations comprises:
calculating distances among the plurality of core geo-points; and
applying the distances to a probabilistic model to determine the associations among the plurality of logs.
15. The computer-readable media of claim 10 , the operations further comprising:
assigning two logs to the user based on an association between the two logs; and
selecting the content for the user based on respective core geo-points of the two logs.
16. A computer-implemented system, comprising:
means for receiving a plurality of geo-points associated with a plurality of logs;
means for determining a plurality of core geo-points based on respective densities of the plurality of geo-points;
means for identifying associations among the plurality of logs based on the plurality of core geo-points; and
means for selecting a content for a user based on the associations among the plurality of logs.
17. The system of claim 16 , further comprising:
means for determining respective densities of the plurality of geo-points; and
means for iteratively merging, based on an order of respective densities of the plurality of geo-points, the plurality of geo-points.
18. The system of claim 16 , wherein the means for determining the plurality of core geo-points comprises:
means for computing a cluster of geo-points from the plurality of geo-points using a clustering process; and
means for designating a core geo-point for the plurality of geo-points based on the cluster of geo-points.
19. The system of claim 16 , wherein the means for identifying associations comprises:
means for calculating distances among the plurality of core geo-points; and
means for identifying the associations among the plurality of logs based on the distances.
20. The system of claim 16 , further comprising:
means for assigning two logs to the user based on an association between the two logs; and
means for personalizing a content for the user based on respective core geo-points of the two logs.