Multiple meter detection and processing using motion data

• US 10,712,361 B2
• Filed: 06/30/2017
• Issued: 07/14/2020
• Est. Priority Date: 11/30/2011
• Status: Active Grant

• Current Assignee: Nielsen Company (US) LLC (RR Donnelley & Sons Company)
• Original Assignee: Nielsen Company (US) LLC (RR Donnelley & Sons Company)
• Inventors: Jain, Anand, Stavropoulos, John, Neuhauser, Alan, Lynch, Wendell, Kuznetsov, Vladimir, Crystal, Jack
• Primary Examiner(s): Le, Toan M
• Application Number:
  US15/638,799
• Publication Number:
  US 20170299624A1
• Time in Patent Office: 1,110 Days
• CPC Class Codes: G01P 15/00 Measuring acceleration; Measuring deceleration; Measuring shock, i.e. sudden change of acceleration
• CPC Class Codes: G01P 15/18 in two or more dimensions
• CPC Class Codes: G06F 17/00 Digital computing or data processing equipment or methods, specially adapted for specific functions information retrieval, database structures or file system structures therefor G06F16/00
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data

Abstract

Systems and methods are disclosed for identifying portable devices carried by the same person. An example method includes receiving media exposure data from one of a plurality of portable computing devices. Receiving segmented accelerometer data from the plurality of portable computing devices. Forming accelerometer classification data for the plurality of portable computing devices based on features extracted from the segmented accelerometer data. Comparing the accelerometer classification data for the first portable computing device to accelerometer classification data for the plurality of portable computing devices, to determine if the accelerometer classification data for the first portable computing device is sufficiently similar to a second portable computing devices. The example method further includes identifying the first and second portable computing device as being physically carried by a same person when the comparison determines that the accelerometer classification data for the first portable computing device is sufficiently similar to the second computing devices.

Claims

1. A system for detecting multiple portable computing devices being physically carried by one person, the system comprising:

an input to receive media exposure data from a first one of a plurality of portable computing devices;

the input to receive segmented accelerometer data from the plurality of portable computing devices, the segmented accelerometer data segmented based on accelerometer data associated with at least one of walking, stopping, running, or sitting;

a processor, operatively coupled to the input, to extract features from the segmented accelerometer data, the processor forming accelerometer classification data for respective ones of the plurality of portable computing devices based on the extracted features;

the processor to compare the accelerometer classification data for the first portable computing device to at least one of (i) the accelerometer classification data for respective ones of the plurality of portable computing devices and (ii) stored accelerometer classification data for respective ones of the plurality of portable computing devices, to determine if accelerometer classification data for the first portable computing device is sufficiently similar to a second one of the plurality of portable computing devices;

the processor to identify the first portable computing device and the second portable computing device as being physically carried by a same person when the comparison determines that the accelerometer classification data for the first portable computing device is sufficiently similar to the accelerometer classification data for the second portable computing device; and

in response to identifying the first and second portable computing devices as being carried by the same person, the processor to credit the media exposure data for one of the first portable computing device or the second portable computing device, thereby reducing inaccurate media measurement results.

2. The system of claim 1, wherein the media exposure data comprises at least one of (i) ancillary codes detected from audio, (ii) one or more signatures extracted from audio, (iii) a web page, (iv) application data, and (v) metadata.

3. The system of claim 1, wherein the plurality of portable computing devices are associated with a group.

4. The system of claim 1, wherein the accelerometer classification data and stored accelerometer classification data comprise raw accelerometer data processed in one of a time domain and a frequency domain.

5. The system of claim 1, wherein the comparison of the accelerometer classification data for the first portable computing device to the accelerometer classification data for respective ones of the plurality of computing devices is based on one of (1) cross-correlation, (2) absolute Manhattan distance, (3) Euclidean distance, and (4) dynamic time warping.

6. The system of claim 1, wherein the accelerometer classification data for the first portable computing device is sufficiently similar when the similarity is above a predetermined threshold.

7. A method for detecting multiple portable computing devices being physically carried by one person, the method comprising:

receiving, by executing an instruction with a processor, media exposure data from a first one of a plurality of portable computing devices;

receiving, by executing an instruction with the processor, segmented accelerometer data from the plurality of portable computing devices, the segmented accelerometer data segmented based on accelerometer data associated with at least one of walking, stopping, running, or sitting;

extracting, by executing an instruction with the processor, features from the segmented accelerometer data and forming accelerometer classification data for respective ones of the plurality of portable computing devices based on the extracted features;

comparing, by executing an instruction with the processor, the accelerometer classification data for the first portable computing device to at least one of (i) the accelerometer classification data for respective ones of the plurality of portable computing devices and (ii) stored accelerometer classification data for respective ones of the plurality of portable computing devices, to determine if accelerometer classification data for the first portable computing device is sufficiently similar to a second one of the plurality of portable computing devices;

identifying, by executing an instruction with the processor, the first portable computing device and the second portable computing device as being physically carried by a same person when the comparison determines that the accelerometer classification data for the first portable computing device is sufficiently similar to the accelerometer classification data for the second portable computing device; and

in response to identifying the first and second portable computing devices as being carried by the same person, crediting, by executing an instruction with the processor, the media exposure data for one of the first portable computing device or the second portable computing device, thereby reducing inaccurate media measurement results.

8. The method of claim 7, wherein the media exposure data comprises at least one of (i) ancillary codes detected from audio, (ii) one or more signatures extracted from audio, (iii) a web page, (iv) application data, and (v) metadata.

9. The method of claim 7, wherein the plurality of portable computing devices are associated with a group.

10. The method of claim 7, wherein the accelerometer classification data and stored accelerometer classification data comprise raw accelerometer data processed in one of a time domain and a frequency domain.

11. The method of claim 7, wherein the comparison of the accelerometer classification data for the first portable computing device to the accelerometer classification data for respective ones of the plurality of computing devices is based on one of (1) cross-correlation, (2) absolute Manhattan distance, (3) Euclidean distance, and (4) dynamic time warping.

12. The method of claim 7, wherein the accelerometer classification data for the first portable computing device is sufficiently similar when the similarity is above a predetermined threshold.

13. A computer readable storage device comprising computer readable instructions which, when executed, cause a processor to at least:

receive media exposure data from a first one of a plurality of portable computing devices;

receive segmented accelerometer data from the plurality of portable computing devices, the segmented accelerometer data segmented based on accelerometer data associated with at least one of walking, stopping, running, or sitting;

extract features from the segmented accelerometer data and form accelerometer classification data for respective ones of the plurality of portable computing devices based on the extracted features;

compare the accelerometer classification data for the first portable computing device to at least one of (i) the accelerometer classification data for respective ones of the plurality of portable computing devices and (ii) stored accelerometer classification data for respective ones of the plurality of portable computing devices, to determine if accelerometer classification data for the first portable computing device is sufficiently similar to a second one of the plurality of portable computing devices;

identify the first portable computing device and the second portable computing device as being physically carried by a same person when the comparison determines that the accelerometer classification data for the first portable computing device is sufficiently similar to the accelerometer classification data for the second portable computing device; and

in response to identifying the first and second portable computing devices as being carried by the same person, credit the media exposure data for one of the first portable computing device or the second portable computing device, thereby reducing inaccurate media measurement results.

14. The storage device of claim 13, wherein the media exposure data comprises at least one of (i) ancillary codes detected from audio, (ii) one or more signatures extracted from audio, (iii) a web page, (iv) application data, and (v) metadata.

15. The storage device of claim 13, wherein the accelerometer classification data and stored accelerometer classification data comprise raw accelerometer data processed in one of a time domain and a frequency domain.

16. The storage device of claim 13, wherein the comparison of the accelerometer classification data for the first portable computing device to the accelerometer classification data for respective ones of the plurality of computing devices is based on one of (1) cross-correlation, (2) absolute Manhattan distance, (3) Euclidean distance, and (4) dynamic time warping.

17. The storage device of claim 13, wherein the accelerometer classification data for the first portable computing device is sufficiently similar when the similarity is above a predetermined threshold.