Dynamic event detection system and method

• US 9,043,329 B1
• Filed: 12/18/2014
• Issued: 05/26/2015
• Est. Priority Date: 12/19/2013
• Status: Expired due to Fees

• Current Assignee: Banjo, Inc.
• Original Assignee: Banjo, Inc.
• Inventors: Patton, Damien, Mehta, Rish, Alves, Pedro
• Primary Examiner(s): Jalil, Neveen Abel
• Assistant Examiner(s): BISKEBORN, KRISTOFER M
• Application Number:
  US14/574,966
• Time in Patent Office: 159 Days
• US Class Current: 707/740
• CPC Class Codes: G06F 16/285 Clustering or classification
• CPC Class Codes: G06F 16/29 Geographical information databases
• CPC Class Codes: G06F 16/48 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: G06F 16/487 using geographical or spatial information, e.g. location
• CPC Class Codes: G06F 16/489 using time information
• CPC Class Codes: G06F 16/5866 using information manually generated, e.g. tags, keywords, comments, manually generated location and time information
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9537 Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
• CPC Class Codes: G06F 16/9538 Presentation of query results
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: G06Q 50/01 Social networking

Abstract

A method for dynamic event detection based on content from a set of social networking systems including receiving content from the set of social networking systems, identifying a plurality of content associated with a geofence, the content that was generated within a predetermined time period, determining feature values from the plurality of content for each of a set of features, determining an event probability for the geofence based on the feature values, and detecting an event within the geofence in response to the event probability exceeding a threshold event probability.

Claims

1. A method for dynamic event detection based on content from a set of social networking systems, comprising, at a computing system including hardware:

receiving content from the set of social networking systems;

for each of a plurality of geofences, each geofence representing a geographic region;

identifying a plurality of content, generated within a predetermined time period, that is associated with the geofence;

determining feature values from the plurality of content for each of a set of features;

determining an event probability for the geofence based on the feature values;

detecting an event within the geofence in response to the event probability exceeding a threshold event probability;

for the event, determining a category probability for an event category, based on the feature values;

categorizing the detected event with the event category in response to the category probability exceeding the category probability threshold;

after detecting an event within the geofence;

determining a second event probability for the geofence in response to a trigger event;

determining that the second event probability is below the threshold event probability; and

maintaining the event probability for the geofence at a predetermined event probability for a predetermined period of time after determination that the second event probability falls below the threshold event probability.

2. The method of claim 1, wherein:

determining an event probability for the geofence based on the feature values comprises;

determining the event probability for the geofence based on feature values for a first reduced feature set from the set of features; and

determining a category probability for the event based on the feature values comprises;

determining the category probability for the event based on feature values for a second reduced feature set from the set of features, the second reduced feature set differing from the first reduced feature set.

3. The method of claim 2, wherein determining a category probability for the event comprises:

determining a first category probability for a first event category for the event based on the second reduced feature set;

determining a second category probability for a second event category for the event based on a third reduced feature set; and

categorizing the event with the first event category in response to the first category probability exceeding a first category probability threshold;

categorizing the event with the second event category in response to the second category probability exceeding a second category probability threshold;

sending the event associated with the first event category to a first endpoint in response to the first category probability exceeding the first category probability threshold; and

sending the event associated with the second event category to the second endpoint, different from the first endpoint, in response to the second category probability exceeding the second category probability threshold.

4. The method of claim 2, wherein determining the category probability based on the second set of feature values further comprises determining the category probability based on the event probability.

5. The method of claim 2, wherein the second reduced feature set is associated with the category by a set of machine learning algorithms.

6. The method of claim 5, wherein determining the category probability based on the feature values for the second reduced feature set comprises calculating the category probability based on a formula determined by a second set of machine learning algorithms.

7. The method of claim 1, wherein the trigger event comprises receipt of new content associated with the geofence from a social networking system.

8. A method for dynamic event detection based on content from a set of social networking systems, comprising, at a computing system including hardware:

receiving content from the set of social networking systems;

for each of a set of geofences, each geofence representing a geographic region;

identifying a plurality of content associated with the geofence that was generated within a predetermined time period;

determining feature values from the plurality of content for each of a set of features;

determining an event probability for the geofence based on the feature values; and

detecting an event within the geofence in response to the event probability exceeding a threshold event probability;

after detecting an event within the geofence;

determining a second event probability for the geofence in response to a trigger event;

determining that the second event probability is below the threshold event probability; and

maintaining the event probability for the geofence at a predetermined event probability for a predetermined period of time after determination that the second event probability falls below the threshold event probability.

9. The method of claim 8, further comprising:

determining an identifier for the event based on the plurality of content;

storing the identifier and associated feature values in association with the geofence.

10. The method of claim 9, wherein the identifier is determined based on keywords from the plurality of content.

11. The method of claim 8, further comprising:

concurrently detecting a first and second event within a first and second geofence, respectively, wherein the first and second geofences represent geographically adjacent geographic regions;

determining that the first event and the second event are a common event;

determining a third geofence enclosing the first and second geofences in response to determination that the first event and the second event are a common event; and

associating a geographic identifier associated with the third geofence with the common event.

12. The method of claim 11, further comprising:

determining a third set of feature values for content associated with geofences enclosed within the third geofence;

determining a third event probability for the third geofence based on the third set of feature values; and

detecting an event within the geofence in response to the third event probability exceeding a second threshold event probability prior to associating the geographic identifier associated with the third geofence with the common event.

13. The method of claim 8, further comprising determining the threshold event probability based on historical event probabilities associated with the respective geofence.

14. The method of claim 8, wherein the trigger event comprises receipt of new content associated with the geofence from a social networking system.

15. The method of claim 8, further comprising:

selecting content associated with the event from the plurality of content;

aggregating the selected content into a content feed; and

sending the content feed to a remote user device.

16. The method of claim 15, further comprising indexing the piece of content in response to content receipt from the social networking system, comprising associating a unique content identifier with the content, wherein the content can be retrieved from the social networking system based on the unique content identifier;

wherein aggregating the selected content into a content feed comprises aggregating the unique content identifier into the content feed, wherein the content feed comprises a list of unique content identifiers; and

wherein sending the content feed to a remote user device comprises sending the list of unique content identifiers to the remote user device.

17. The method of claim 16, wherein indexing the content further comprises:

extracting feature values for the piece of content; and

storing the feature values for the piece of content in association with the unique content identifier.

18. The method of claim 8, wherein determining feature values from the plurality of content for each of a set of features comprises determining a post frequency value, comprising:

in response to receipt of a piece of content at a first time, determining whether a second piece of content associated with the geofence had been received within a predetermined time period before the first time.

19. The method of claim 8, wherein determining feature values from the plurality of content for each of a set of features further comprises determining a global popularity for the content of the plurality, comprising determining a count of secondary geofences, different from the first geofence, that are associated with content that substantially match content within the plurality of content.

20. The method of claim 19, wherein determining feature values from the plurality of content for each of a set of features further comprises determining a spread value for the content of the plurality, comprising identifying the secondary geofences and determining a geographic distance from the first geofence to each of the secondary geofences.