Detecting events from features derived from multiple ingested signals

Abstract

The present invention extends to methods, systems, and computer program products for detecting events from features derived from multiple signals. In one aspect, an event detection infrastructure determines that characteristics of multiple 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 signals indicate a possible event of interest to one or more parties. A validator then determines that characteristics of one or more other 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.

What is claimed: 
     
         1 . A method comprising:
 receiving a first normalized signal;   deriving first one or more features of the first normalized signal;   determining that the first one or more features do not satisfy conditions to be identified as an event;   receiving a second normalized signal;   deriving second one or more features of the second signal;   aggregating the first one or more features with the second one or more features into aggregated features; and   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;   validating the possible event as an actual event based on the second one or more features.   
     
     
         3 . The method of  claim 1 , further comprising including the first normalized signal in a signal sequence;
 determining that the second normalized signal has sufficient temporal similarity to the first normalized signal;   determining that the second normalized signal has sufficient spatial similarity to the first normalized signal; and   including the second normalized signal in a signal sequence that contains the first normalized signal.   
     
     
         4 . The method of  claim 3 , 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. 
     
     
         5 . The method of  claim 4 , wherein deriving features of the signal sequence comprises deriving one or more of: a percentage, a count, a histogram, or a duration. 
     
     
         6 . The method of  claim 1 , wherein the first normalized signal is one of: a social post with geographic content, a social post without geographic content, an image from a 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. 
     
     
         7 . The method of  claim 6 , wherein the second normalized signal is a different one of: a social post with geographic content, a social post without geographic content, 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. 
     
     
         8 . The method of clam  1 , wherein deriving first one or more features of the first normalized signal comprises deriving the first one or more features from a first single source probability assigned to the first normalized signal;
 wherein deriving second one or more features of the second normalized signal comprises deriving the first one or more features from a second single source probability assigned to the second normalized signal;   wherein aggregating the first one or more features with the second one or more features into aggregated features comprises aggregating the first single source probability and the second single source probability into a multisource probability;   wherein detecting an event from the aggregated features comprises detecting an event from the multisource probability.   
     
     
         9 . A method, the method comprising:
 receiving a normalized signal including time, location, context, and content;   forming a signal sequence including the normalized signal;   receiving another normalized signal including another time, another location, another context, and other content;   determining that there is sufficient temporal similarity between normalized signal and the other normalized signal;   determining that there is sufficient spatial similarity between the normalized and the other normalized signal; and   including the other normalized signal in the signal sequence based on the sufficient temporal similarity and the sufficient spatial similarity   
     
     
         10 . The method of  claim 9 , wherein determining that there is sufficient temporal similarity between the normalized signal and the other normalized signal comprises determining that the time and the other time are within a specified time of one another. 
     
     
         11 . The method of  claim 9 , wherein determining that there is sufficient spatial similarity between the normalized and the other normalized signal comprises determining that the location and the other location are within a specified distance of one another. 
     
     
         12 . The method of  claim 9 , wherein determining that there is sufficient spatial similarity between the normalized and the other normalized signal comprises determining that the location and the other location are within a specified number of geo cells of one another. 
     
     
         13 . The method of  claim 9 , further comprising determining that the other normalized signal is not a duplicate of the normalized signal prior to including the other normalized signal in the signal sequence. 
     
     
         14 . The method of  claim 9 , further comprising:
 deriving one or more features of the signal sequence based on the normalized signal and the other normalized signal;   detecting an event from the derived one or more features.   
     
     
         15 . The method of claim of  claim 14 , wherein deriving one or more features of the signal sequence comprise deriving a multisource probability for the signal sequence. 
     
     
         16 . The method of  claim 15 , wherein deriving a multisource probability indicating the probability of the normalized signals in the signal sequence indicate a specified type of event. 
     
     
         17 . A method, the method comprising:
 accessing a signal sequence of normalized signals, normalized signals included in the signal sequence having a sufficient temporal similarity to one another and having a sufficient spatial similarity to one another;   extracting features from the signal sequence; and   detecting an event based on the extracted features.   
     
     
         18 . The method of  claim 17 , further comprising prior to detecting the event:
 detecting that the extracted features do not indicate the event;   adding an additional normalized signal to the signal sequence;   extracting further features from the signal sequence based on the additional normalized signal; and   wherein detecting an event based on the extracted features comprises detecting an event based on the further extracted features.   
     
     
         19 . The method of  claim 17 , further comprising deriving a multisource probability from the extracted features; and
 wherein detecting an event based on the extracted features comprises detecting an event based on the multisource probability.   
     
     
         20 . The method of  claim 17 , wherein extracting features of the signal sequence comprises deriving one or more of: a percentage, a count, a histogram, or a duration.