Leveraging location data from mobile devices for user classification

• US 9,930,494 B2
• Filed: 10/13/2015
• Issued: 03/27/2018
• Est. Priority Date: 10/13/2015
• Status: Active Grant

• Current Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Original Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Inventors: Liu, Hao, Tang, David, Shen, Zhitao, Zhang, Hua, Kumaran, Vikram
• Primary Examiner(s): Batista, Marcos
• Application Number:
  US14/881,538
• Publication Number:
  US 20170105099A1
• Time in Patent Office: 896 Days
• Field of Search: 4554561-4566, 370338
• CPC Class Codes: H04W 4/021 Services related to particular areas, e.g. point of interest [POI] services, venue services or geofences
• CPC Class Codes: H04W 4/029 Location-based management or tracking services
• CPC Class Codes: H04W 4/30 Services specially adapted for particular environments, situations or purposes
• CPC Class Codes: H04W 4/33 for indoor environments, e.g. buildings

Abstract

Location data is obtained from signals transmitted by a first plurality of mobile wireless devices in a wireless network, wherein the first plurality of mobile wireless devices are moving within a predefined space, and wherein the location data comprises a plurality of location data time points, each location data time point including a timestamp, a unique mobile wireless device identifier, and location information indicating where in the predefined space an associated mobile wireless device is located. For each mobile wireless device, location data time points are aggregated to generate a set of aggregated location data for each mobile wireless device, and the set of aggregated location data is analyzed to determine characteristics corresponding to time-dependent behavior and location-specific behavior of the corresponding mobile wireless device. A user of each corresponding mobile wireless device is classified into a category of a plurality of categories based on the determined characteristics.

Claims

1. A computer-implemented method comprising:

obtaining location data from signals transmitted by a first plurality of mobile wireless devices in a wireless network provided in a predefined physical space, wherein the first plurality of mobile wireless devices are moving within the predefined physical space, and wherein the location data comprises a plurality of location data time points, each location data time point including a timestamp, a unique mobile wireless device identifier, and location information indicating where within the predefined physical space an associated mobile wireless device is located;

aggregating, for each mobile wireless device, location data time points to generate a set of aggregated location data for each mobile wireless device;

analyzing, for each mobile wireless device, the set of aggregated location data to determine characteristics corresponding to time-dependent behavior and location-specific behavior of the corresponding mobile wireless device, including;

a dwell time indicating how long the corresponding mobile wireless device has resided at a particular location inside the predefined physical space;

a measure of floor dispersion indicating a first spatial distribution of the corresponding mobile wireless device within the predefined physical space; and

a measure of location dispersion indicating a second spatial distribution, different from the first spatial distribution, of the corresponding mobile wireless device within the predefined physical space; and

classifying a user of each corresponding mobile wireless device into a category of a plurality of categories based on the determined characteristics.

2. The method of claim 1, further comprising:

performing classification of select sets of aggregated location data to classify the user of each corresponding mobile wireless device into one of a plurality of categories, wherein each set of aggregated location data is classified by a plurality of distinct classifiers;

generating a training data set based on the classified select sets of aggregated location data; and

wherein the classifying further comprises classifying the user of each corresponding mobile wireless device based on the determined characteristics and the training data set.

3. The method of claim 2, comprising:

selecting a subset of unlabeled aggregated data for new mobile wireless device users, and adding the selected subset to the training data set;

labeling each set of aggregated data based on proximity to a nearest classified mobile wireless device user;

generating continuously new aggregated location data pertaining to movement of each new mobile wireless device user; and

determining features based on the continuously generated data.

4. The method of claim 1 further comprising:

obtaining location data from signals transmitted by a second plurality of mobile wireless devices in the wireless network;

aggregating, for each of the second plurality of mobile wireless devices, location data time points to generate a set of aggregated location data;

generating a vector having at least a spatial component and a temporal component for each of the first plurality of mobile wireless devices and the second plurality of mobile wireless devices; and

grouping similar vectors together to form a cluster, and assigning a same category label to vectors within the same cluster.

5. The method of claim 1, wherein classifying comprises:

classifying, for location data having a relatively longer dwell time and relatively smaller measures of floor dispersion and location dispersion, the user of the corresponding mobile wireless device as belonging to a permanent category; and

classifying, for location data having a relatively shorter dwell time and relatively larger measures of floor dispersion and location dispersion, the user of the corresponding mobile wireless device as belonging to a temporary category.

6. The method of claim 1, comprising:

receiving feedback from a classification procedure, the feedback indicating whether the classification of the user of the corresponding mobile wireless device based on the determined characteristics is consistent with a classification performed by the classification procedure; and

updating, when classification is not consistent, the determined characteristics based on the feedback.

7. The method of claim 1, wherein the measure of floor dispersion indicates an amount of time that the corresponding mobile wireless device remains on one floor of a plurality of floors in the predefined physical space and the measure of location dispersion indicates how the corresponding mobile wireless is moving across any of the plurality of floors.

8. An apparatus comprising:

a network interface unit configured enable network communications;

a memory; and

a processor coupled to the network interface unit and configured to;

obtain location data from signals transmitted by a first plurality of mobile wireless devices in a wireless network provided in a predefined physical space, wherein the first plurality of mobile wireless devices are moving within the predefined physical space, and wherein the location data comprises a plurality of location data time points, each location data time point including a timestamp, a unique mobile wireless device identifier, and location information indicating where within the predefined physical space an associated mobile wireless device is located;

aggregate, for each mobile wireless device, location data time points to generate a set of aggregated location data for each mobile wireless device, and store the set of aggregated location data in the memory;

analyze, for each mobile wireless device, the set of aggregated location data to determine characteristics corresponding to time-dependent behavior and location-specific behavior of the corresponding mobile wireless device, including;

a dwell time indicating how long the corresponding mobile wireless device has resided at a particular location inside the predefined physical space;

a measure of floor dispersion indicating a first spatial distribution of the corresponding mobile wireless device within the predefined physical space; and

a measure of location dispersion indicating a second spatial distribution, different from the first spatial distribution, of the corresponding mobile wireless device within the predefined physical space; and

classify a user of each corresponding mobile wireless device into a category of a plurality of categories based on the determined characteristics.

9. The apparatus of claim 8, wherein the processor is further configured to:

perform classification of select sets of aggregated location data to classify the user of each corresponding mobile wireless device into one of a plurality of categories, wherein each set of aggregated location data is classified by a plurality of distinct classifiers;

generate a training data set based on the classified select sets of aggregated location data, and store the training data in memory; and

classify the user of each corresponding mobile wireless device based on the determined characteristics and the training data set.

10. The apparatus of claim 9, wherein the processor is further configured to:

select a subset of unlabeled aggregated data for new mobile wireless device users, and adding the selected subset to the training data set;

label each set of aggregated data based on proximity to a nearest classified mobile wireless device user;

generate continuously new aggregated location data pertaining to movement of each new mobile wireless device user; and

determine features based on the continuously generated data.

11. The apparatus of claim 8, wherein the processor is further configured to:

obtain location data from signals transmitted by a second plurality of mobile wireless devices in the wireless network;

aggregate, for each of the second plurality of mobile wireless devices, location data time points to generate a set of aggregated location data, and store the aggregated location data in memory;

generate a vector having at least a spatial component and a temporal component for each of the first plurality of mobile wireless devices and the second plurality of mobile wireless devices; and

group similar vectors together to form a cluster, and assigning a same category label to vectors within the same cluster.

12. The apparatus of claim 8, wherein the processor is further configured to:

classify, for location data having a relatively longer dwell time and relatively smaller measures of floor dispersion and location dispersion, the user of the corresponding mobile wireless device as belonging to a permanent category; and

classify, for location data having a relatively shorter dwell time and relatively larger measures of floor dispersion and location dispersion, the user of the corresponding mobile wireless device as belonging to a temporary category.

13. The apparatus of claim 8, wherein the processor is further configured to:

receive feedback from a classification procedure, the feedback indicating whether the classification of the user of the corresponding mobile device based on the determined characteristics is consistent with a classification performed by the classification procedure; and

update, when classification is not consistent, the determined characteristics based on the feedback.

14. The apparatus of claim 8, wherein the measure of floor dispersion indicates an amount of time that the corresponding mobile wireless device remains on one floor of a plurality of floors in the predefined physical space and the measure of location dispersion indicates how the corresponding mobile wireless is moving across any of the plurality of floors.

15. A non-transitory computer readable storage media encoded with instructions that, when executed by a processor of a computing device, cause the processor to:

obtain location data from signals transmitted by a first plurality of mobile wireless devices in a wireless network provided in a predefined physical space, wherein the first plurality of mobile wireless devices are moving within the predefined physical space, and wherein the location data comprises a plurality of location data time points, each location data time point including a timestamp, a unique mobile wireless device identifier, and location information indicating where within the predefined physical space an associated mobile wireless device is located;

aggregate, for each mobile wireless device, location data time points to generate a set of aggregated location data for each mobile wireless device, and store the set of aggregated location data in memory;

analyze, for each mobile wireless device, the set of aggregated location data to determine characteristics corresponding to time-dependent behavior and location-specific behavior of the corresponding mobile wireless device, including;

a dwell time indicating how long the corresponding mobile wireless device has resided at a particular location inside the predefined physical space;

a measure of floor dispersion indicating a first spatial distribution of the corresponding mobile wireless device within the predefined physical space; and

a measure of location dispersion indicating a second spatial distribution, different from the first spatial distribution, of the corresponding mobile wireless device within the predefined physical space; and

classify a user of each corresponding mobile wireless device into a category of a plurality of categories based on the determined characteristics.

16. The non-transitory computer readable storage media of claim 15, wherein the processor is configured to:

perform classification of select sets of aggregated location data to classify the user of each corresponding mobile wireless device into one of a plurality of categories, wherein each set of aggregated location data is classified by a plurality of distinct classifiers;

generate a training data set based on the classified select sets of aggregated location data; and

classify the user of each corresponding mobile wireless device based on the determined characteristics and the training data set.

17. The non-transitory computer readable storage media of claim 15, wherein the processor is configured to:

obtain location data from signals transmitted by a second plurality of mobile wireless devices in the wireless network;

aggregate, for each of the second plurality of mobile wireless devices, location data time points to generate a set of aggregated location data, and store the aggregated data in memory;

generate a vector having at least a spatial component and a temporal component for each of the first plurality of mobile wireless devices and the second plurality of mobile wireless devices; and

group similar vectors together to form a cluster, and assigning a same category label to vectors within the same cluster.

18. The non-transitory computer readable storage media of claim 15, wherein the processor is configured to:

classify, for location data having a relatively longer dwell time and relatively smaller measures of floor dispersion and location dispersion, the user of the corresponding mobile wireless device as belonging to a permanent category; and

classify, for location data having a relatively shorter dwell time and relatively larger measures of floor dispersion and location dispersion, the user of the corresponding mobile wireless device as belonging to a temporary category.

19. The non-transitory computer readable storage media of claim 15, wherein the processor is configured to:

receive feedback from a classification procedure, the feedback indicating whether the classification of the user of the corresponding mobile wireless device based on the determined characteristics is consistent with a classification performed by the classification procedure; and

update, when classification is not consistent, the determined characteristics based on the feedback.

20. The non-transitory computer readable storage media of claim 15, wherein the measure of floor dispersion indicates an amount of time that the corresponding mobile wireless device remains on one floor of a plurality of floors in the predefined physical space and the measure of location dispersion indicates how the corresponding mobile wireless is moving across any of the plurality of floors.