System and methods for extracting facts from unstructured text

Abstract

A system and method for extracting facts from unstructured text files are disclosed. Embodiments of the disclosed system and method may receive a text file as input and perform extraction and disambiguation of entities, as well as extract topics and facts. The facts are extracted by comparing against a fact template store and associating facts with events or topics. The extracted facts are stored in a data store.

What is claimed is: 
     
         1 . A method comprising:
 extracting, by a server, an entity identifier and a topic identifier from a first portion of unstructured text in a text file, the unstructured text comprising a plurality of strings;   comparing, by the server, the strings to data stored in a data structure, the data comprising a fact template model containing a keyword related to a fact identifier and a keyword weight associated with the fact identifier;   extracting, by the server, the fact identifier from a second portion of the unstructured text of the text filed based upon comparing the strings to the data comprising the fact template model;   associating, by the server, the entity identifier with the topic identifier and the fact identifier; and   determining, by the server, a score based on the associating, wherein the score is indicative of a degree of accuracy of the extracting of the fact identifier, wherein the score is based on a spatial distance between the first portion and the second portion.   
     
     
         2 . The method of  claim 1 , wherein the spatial distance is determined via a token. 
     
     
         3 . The method of  claim 1 , wherein the text file comprises a plurality of co-occurring entity identifiers, wherein the score is based on comparing the co-occurring entity identifiers. 
     
     
         4 . The method of  claim 1 , wherein the fact template model includes metadata. 
     
     
         5 . The method of  claim 4 , wherein the metadata includes a count of a number of times a sentence structure corresponding to the fact template model is repeated across a plurality of electronic documents comprising the text file. 
     
     
         6 . The method of  claim 4 , wherein the metadata comprises the score. 
     
     
         7 . The method of  claim 1 , where at least two of the extracting, the comparing, the extracting, the associating, and the determining are distributed among a plurality of computers. 
     
     
         8 . The method of  claim 1 , further comprising:
 determining, by the server, whether the score exceeds a predetermined threshold;   in response to the score exceeding the predetermined threshold, storing, by the server, the fact identifier in a second data structure.   
     
     
         9 . The method of  claim 1 , wherein the spatial distance is determined via a natural language processing technique. 
     
     
         10 . A device comprising:
 a processor;   a memory storing a set of instructions executable by the processor to perform a method comprising:
 extracting, by the processor, an entity identifier and a topic identifier from a first portion of an unstructured text in a text file, wherein the unstructured text comprises a plurality of strings, wherein the unstructured text comprises a second portion; 
 comparing, by the processor, the strings to a data stored in a data structure, wherein the data comprises a fact template model which identifies a keyword related to a fact identifier and a keyword weight associated with the fact identifier; 
 extracting, by the processor, the fact identifier from the second portion based on the comparing; 
 associating, by the processor, the entity identifier with the topic identifier and the fact identifier; 
 determining, by the processor, a score based on the associating, wherein the score is indicative of a degree of accuracy of the extracting of the fact identifier, wherein the score is based on a spatial distance between the first portion and the second portion. 
   
     
     
         11 . The device of  claim 10 , wherein the spatial distance is determined via a token. 
     
     
         12 . The device of  claim 1 , wherein the text file comprises a plurality of co-occurring entity identifiers, wherein the score is based on comparing the co-occurring entity identifiers. 
     
     
         13 . The device of  claim 1 , wherein the fact template model includes metadata. 
     
     
         14 . The device of  claim 13 , wherein the metadata includes a count of a number of times a sentence structure corresponding to the fact template model is repeated across a plurality of electronic documents comprising the text file. 
     
     
         15 . The device of  claim 13 , wherein the metadata comprises the score. 
     
     
         16 . The device of  claim 10 , where at least two of the extracting, the comparing, the extracting, the associating, and the determining are distributed among a plurality of computers. 
     
     
         17 . The device of  claim 10 , further comprising:
 determining, by the processor, whether the score exceeds a predetermined threshold;   in response to the score exceeding the predetermined threshold, storing, by the processor, the fact identifier in a second data structure.   
     
     
         18 . The device of  claim 10 , wherein the spatial distance is determined via a natural language processing technique.