Event detection through text analysis using trained event template models

Abstract

A system and method for detecting events based on input data from a plurality of sources. The system may receive input from a plurality of sources containing information about possible events. A method for event detection involves pre-processing and normalizing a data input from a plurality of sources, extracting and disambiguating events and entities, associate event and entities, correlate events and entities associated from a data input to results from a different data sources to determine if an event has occurred, and store the detected events in a data storage.

What is claimed is: 
     
         1 . A method comprising:
 identifying, by a computer, in a data stream received from a data source, one or more features matching a model feature in an event model stored in a non-transitory machine-readable storage of an event concept store, wherein at least one feature in the one or more features is an event candidate corresponding to the event model, wherein at least one feature in the one or more features is an entity;   assigning, by the computer, a weight to each respective entity according to the event model corresponding to the event candidate;   determining, by the computer, an event probability score based on one or more weights respectively assigned to the one or more entities;   associating, by the computer, the event candidate with each respective entity into a first record of a spatial-temporal grouping responsive to the computer determining that the event probability score satisfies an event likelihood threshold score of the event model;   storing, by the computer, the first record containing the event candidate, each respective entity associated with the event candidate, and the event probability score into a non-transitory machine-readable spatial-temporal grouping memory;   receiving, by the computer, a tagged source item comprising one or more tags that identify one or more features of an event model comprising a corresponding weight;   generating, by the computer, a new event model based on the received tagged source item;   storing, by the computer, the new event model in the event concept store.   
     
     
         2 . The method of  claim 1 , further comprising:
 normalizing, by the computer, the data of the data stream into a format recognized by at least one computing device.   
     
     
         3 . The method of  claim 1 , further comprising:
 extracting, by the computer, each of the respective entities identified in the data stream; and   disambiguating, by the computer, each of the respective entities.   
     
     
         4 . The method of  claim 1 , wherein the associating the event candidate with each respective entity into the first record of the spatial-temporal grouping comprises:
 generating, by the computer, the first record comprising the event candidate, each respective entity associated with the event candidate identified in the data stream, and the event probability score.   
     
     
         5 . The method of  claim 1 , further comprising:
 receiving, by the computer, a plurality of data streams from a plurality of data sources;   identifying, by the computer, one or more entities in each respective data stream according to one or more event models corresponding to one or more event candidates.   
     
     
         6 . The method of  claim 1 , wherein the spatial-temporal memory stores a plurality of records of spatial-temporal groupings, wherein each respective record is associated with each respective data source, wherein each respective record comprises the event candidate, a set of one or more entities identified in the associated data source, and the event probability score for each respective record. 
     
     
         7 . The method of  claim 1 , further comprising:
 determining, by the computer, a validation score for the first record based on a comparison of each of the event probability scores in each respective record containing the event candidate;   storing, by the computer, the event candidate of the first record as a new verified event into non-transitory machine-readable storage memory of a verified event store responsive to determining the validation score for the first record satisfies a validation threshold.   
     
     
         8 . The method of  1 , further comprising:
 generating, by the computer, a notification indicating a new verified event was identified in the data source according to a set of notification criteria stored in non-transitory machine-readable storage memory.   
     
     
         9 . The method of  8 , further comprising:
 transmitting, by the computer, the notification to a remote electronic device upon generating the notification.   
     
     
         10 . The method of  claim 1 , wherein the event concept store comprises an in-memory database. 
     
     
         11 . A system comprising:
 an event concept store comprising a memory storing one or more event models, wherein an event model corresponds to an event candidate and comprises a threshold event score and a set of one or more features comprising a corresponding weight;   an entity and topic extraction processor configured to extract a set of entities and a set of topics from a data stream and disambiguate each topic and each entity;   an event extraction processor configured to identify each of the features of each event model that occur in the data stream, to determine an event score for one or more event candidates comprising an identified feature using the corresponding event model and to extract the event candidate upon an event score satisfying the threshold event score of the event model;   a spatial-temporal event processor configured to associate each topic and entity extracted from each of the data streams with each of the event candidates extracted from each of the data streams and to form a spatial-temporal event grouping comprising one or more records, wherein a record comprises the event candidate and the associated topic or entity of a data stream;   a training processor configured to build an event model, to update the event model in the event concept store based at least in part on one or more new features identified in a tagged source item comprising a corresponding weight, to generate a new event model based on the tagged source item, and to store the new event model in the event concept store.   
     
     
         12 . The system of  claim 11 , further comprising:
 a verified event store storing one or more verified events;   an event validation processor configured to verify that each of the candidate events correspond to a real-world event, to determine a verification score for each event candidate in the spatial-temporal grouping based on a comparison of each of the records in the spatial-temporal grouping, and to store the event candidate as a verified event in the verified event store when the event candidate satisfies a verification threshold.   
     
     
         13 . The system of  claim 11 , wherein a verified event store comprises a verified event query processor configured to receive a query from a remote computing device, to process the query, and to return a query result to the remote computing device. 
     
     
         14 . The system of  claim 11 , wherein a verified event store comprises a verified event notification processor configured to transmit a verified event notification to a remote computing device based on one or more subscription rules. 
     
     
         15 . The system of  claim 11 , wherein the training processor is configured to determine whether a source item is tagged. 
     
     
         16 . The system of  claim 11 , further comprising:
 a data normalizer configured to normalize one or more source items into the data stream adequate for computer-automated processing corresponding to each source item.   
     
     
         17 . The system of  claim 11 , wherein the event concept store comprises an in-memory database. 
     
     
         18 . The system of  claim 11 , wherein the data stream comprises social media information. 
     
     
         19 . The system of  claim 11 , wherein the entity and topic extraction processor, the event extraction processor, the spatial-temporal event processor, and the training processor are at least partially distributed across a plurality of computers. 
     
     
         20 . The system of  claim 11 , wherein at least two of the entity and topic extraction processor, the event extraction processor, the spatial-temporal event processor, and the training processor operate in parallel with each other.