Detecting events from features derived from ingested signals

• US 10,628,601 B2
• Filed: 07/19/2019
• Issued: 04/21/2020
• Est. Priority Date: 02/09/2018
• Status: Active Grant

• Current Assignee: Banjo, Inc.
• Original Assignee: Banjo, Inc.
• Inventors: Patton, Damien, Guereca-Pinuelas, Armando, Leung, K W Justin, Gratton, Christian
• Primary Examiner(s): Aspinwall, Evan
• Application Number:
  US16/516,684
• Publication Number:
  US 20190347436A1
• Time in Patent Office: 277 Days
• Field of Search: 707737
• CPC Class Codes: G06F 21/554 involving event detection and direct action
• CPC Class Codes: G06F 21/604 Tools and structures for managing or administering access control systems
• CPC Class Codes: G06F 21/6218 to a system of files or objects, e.g. local or distributed file system or database
• CPC Class Codes: H04L 63/04 for providing a confidential data exchange among entities communicating through data packet networks
• CPC Class Codes: H04L 63/0428 wherein the data content is protected, e.g. by encrypting or encapsulating the payload

Abstract

The present invention extends to methods, systems, and computer program products for detecting events from features derived from ingested signals. In one aspect, an event detection infrastructure determines that characteristics of multiple (and possibly private and/or non-private) signals, when considered collectively, indicate an event of interest to one or more parties. In another aspect, an evaluation module determines that characteristics of one or more (possibly private and/or non-private) signals indicate a possible event of interest to one or more parties. A validator then determines that characteristics of one or more other (possibly private and/or non-private) signals validate the possible event as an actual event of interest to the one or more parties. Signal features can be used to compute probabilities of events occurring.

Claims

1. A method, implemented at a computing system comprising one or more processors, the method comprising:

accessing, a normalized private data signal that is private to an organization based on data privacy access settings associated with the normalized private data signal for the organization;

deriving, at the one or more processors, first one or more features of the normalized private data signal;

determining that the first one or more features do not satisfy conditions to be identified as an event;

accessing, at the one or more processors, another normalized data signal based on any data privacy access settings associated with the other normalized data signal;

deriving, at the one or more processors, second one or more features of the other normalized data signal;

aggregating, at the one or more processors, the first one or more features with the second one or more features into aggregated features;

configuring access to the aggregated features based on data privacy aggregation settings associated with the normalized private data signal and data privacy aggregation settings associated the other normalized data signal;

detecting an event from the aggregated features.

2. The method of claim 1, wherein aggregating the first one or more features with the second one or more features into aggregated features comprises:

detecting a possible event from the first one or more features; and

validating the possible event as an actual event based on the second one or more features.

3. The method of claim 2, wherein accessing the other normalized data signal comprises accessing the other normalized data signal from among:

another normalized private data signal private to the organization, a normalized non-private data signal, or a normalized public data signal.

4. The method of claim 3, wherein accessing the other normalized data signal comprises accessing a normalized non-private data signal controlled by another organization.

5. The method of claim 1, further comprising including the normalized private data signal in a signal sequence;

determining that the other normalized data signal has sufficient temporal similarity to the normalized private data signal;

determining that the other normalized data signal has sufficient spatial similarity to the normalized private data signal; and

including the other normalized data signal in the signal sequence.

6. The method of claim 5, wherein aggregating the first one or more features with the second one or more features into aggregated features comprises deriving features of the signal sequence from the first one or more features and the second one or more features.

7. The method of claim 6, wherein deriving features of the signal sequence comprises deriving one or more of:

a percentage, a count, a histogram, or a duration.

8. The method of claim 1, further comprising, prior to aggregating the first one or more features with the second one or more features:

accessing the data privacy settings; and

determining that the data privacy settings permit aggregation of the normalized private data signal with the other normalized data signal.

9. The method of claim 1, wherein accessing the other normalized data signal comprises accessing the other normalized signal selected from among:

another normalized private data signal private to the organization, a normalized non-private data signal, or a normalized public data signal.

10. The method of claim 9, wherein accessing the other normalized data signal comprises accessing a normalized non-private data signal controlled by another organization.

11. The method of claim 1, wherein the normalized private data signal is one of:

an image from a traffic camera feed, a 911 call, weather data, IoT device data, satellite data, satellite imagery, a sound clip from a listening device, data from air quality sensors, a sound clip from radio communication, crowd sourced traffic information, or crowd sourced road information.

12. The method of claim 1, wherein deriving the first one or more features comprises deriving the first one or more features from a first single source probability assigned to the normalized private data signal;

wherein deriving the second one or more features comprises deriving the second one or more features from a second single source probability assigned to the other normalized data signal;

wherein aggregating the first one or more features with the second one or more features comprises aggregating the first single source probability and the second single source probability into a multisource probability; and

wherein detecting an event from the aggregated features comprises detecting an event from the multisource probability.

13. A method, implemented at a computing system comprising one or more processors, comprising:

accessing a normalized non-private data signal, that is controlled by an organization, based on data privacy access settings associated with the normalized non-private data signal for the organization;

deriving first one or more features of the normalized non-private data signal;

determining that the first one or more features do not satisfy conditions to be identified as an event;

accessing another normalized data signal based on any data privacy access settings associated with the other normalized data signal;

deriving second one or more features of the other normalized data signal;

aggregating the first one or more features with the second one or more features into aggregated features;

configuring access to the aggregated features based on the data privacy aggregation settings associated with the normalized private data signal and the data privacy aggregation settings associated with the other normalized data signal; and

detecting an event from the aggregated features.

14. The method of claim 13, wherein aggregating the first one or more features with the second one or more features into aggregated features comprises:

detecting a possible event from the first one or more features; and

validating the possible event as an actual event based on the second one or more features.

15. The method of claim 14, wherein accessing the other normalized data signal comprises accessing the other normalized data signal selected from among:

a normalized private data signal private to the organization, another normalized non-private data signal, or a normalized public data signal.

16. The method of claim 15, wherein accessing the other normalized data signal comprises accessing a normalized non-private data signal controlled by another organization.

17. The method of claim 13, further comprising including the normalized non-private data signal in a signal sequence;

determining that the other normalized data signal has sufficient temporal similarity to the normalized non-private data signal;

determining that the other normalized data signal has sufficient spatial similarity to the normalized non-private data signal; and

including the other normalized data signal in the signal sequence.

18. The method of claim 17, wherein aggregating the first one or more features with the second one or more features into aggregated features comprises deriving features of the signal sequence from the first one or more features and the second one or more features.

19. The method of claim 17, wherein deriving features of the signal sequence comprises deriving one or more of:

a percentage, a count, a histogram, or a duration.

20. The method of claim 13, further comprising, prior to aggregating the first one or more features with the second one or more features:

accessing the data privacy settings; and

determining that the data privacy settings permit aggregation of the normalized non-private data signal with the other normalized data signal.

21. The method of claim 13, wherein accessing the other normalized data signal comprises accessing the other normalized data signal, the other normalized data signal selected from among:

a normalized private data signal, another normalized non-private data signal, or a normalized public data signal.

22. The method of claim 21, wherein accessing the other normalized data signal comprises accessing a normalized non-private data signal controlled by another organization.

23. The method of claim 13, wherein the normalized non-private data signal is one of:

an image from a traffic camera feed, a 911 call, weather data, IoT device data, satellite data, satellite imagery, a sound clip from a listening device, data from air quality sensors, a sound clip from radio communication, crowd sourced traffic information, or crowd sourced road information.

24. The method of claim 13, wherein deriving first one or more features comprises deriving the first one or more features from a first single source probability assigned to the normalized non-private data signal;

wherein deriving second one or more features comprises deriving the second one or more features from a second single source probability assigned to the other normalized data signal;

wherein aggregating the first one or more features with the second one or more features comprises aggregating the first single source probability and the second single source probability into a multisource probability; and

wherein detecting an event from the aggregated features comprises detecting an event from the multisource probability.