Determining themes

Abstract

Determining themes is disclosed. Reputation data extracted from at least one data source is received. The reputation data includes a plurality of user-authored reviews. The presence of a first keyword is detected in a first review. The presence of a second keyword that is different from but associated with the first keyword is detected in a second review. A sentiment for a theme is determined based on the detected presence of the first and second keywords. A report that indicates the sentiment for the theme is provided as output.

What is claimed is: 
     
         1 . (canceled) 
     
     
         2 . A system, comprising:
 a processor configured to:
 receive reputation data extracted from at least one data source, wherein the reputation data includes a plurality of user-authored reviews; 
 determine that a first item is present in a first review in the plurality of user-authored reviews; 
 assign to the first item an identifier that is generated based at least in part on a string from which the first item was derived; 
 in response to determining that an identifier assigned to a second item is the same as the identifier assigned to the first item, count the first item and the second item as a single item; 
 determine that a third item, different from but associated with the first item, is present in a second review in the plurality of user-authored reviews; 
 determine a sentiment for a theme based on the presence of the first and third items; and 
 provide as output a report that indicates the sentiment for the theme; and 
   a memory coupled to the processor and configured to provide the processor with instructions.   
     
     
         3 . The system of  claim 2 , wherein the identifier for the first item is generated based at least in part on a sentence from which the first item was derived. 
     
     
         4 . The system of  claim 2 , wherein the identifier for the first item is generated based at least in part on a clause. 
     
     
         5 . The system of  claim 2 , wherein the identifier for the first item is generated at least in part by performing a hash of a string. 
     
     
         6 . The system of  claim 5 , wherein determining that the identifier assigned to the second item is the same as the identifier assigned to the first item comprises comparing one or more hash codes. 
     
     
         7 . The system of  claim 2 , wherein determining that the first item is present includes performing natural language processing. 
     
     
         8 . The system of  claim 2 , wherein the first and second reviews are reviews of an entity. 
     
     
         9 . The system of  claim 8 , wherein the processor is further configured to determine whether the first item is present in an ontology associated with the entity. 
     
     
         10 . The system of  claim 2 , wherein the processor is further configured to select an ontology associated with an entity being reviewed in the first review. 
     
     
         11 . The system of  claim 2 , wherein the processor is further configured to select, from a plurality of ontologies, an ontology that is associated with an industry associated with an entity being reviewed in the first review. 
     
     
         12 . A method, comprising:
 receiving reputation data extracted from at least one data source, wherein the reputation data includes a plurality of user-authored reviews;   determining that a first item is present in a first review in the plurality of user-authored reviews;   assigning to the first item an identifier that is generated based at least in part on a string from which the first item was derived;   in response to determining that an identifier assigned to a second item is the same as the identifier assigned to the first item, counting the first item and the second item as a single item;   determining that a third item, different from but associated with the first item, is present in a second review in the plurality of user-authored reviews;   determining a sentiment for a theme based on the presence of the first and third items; and   providing as output a report that indicates the sentiment for the theme.   
     
     
         13 . The method of  claim 12 , wherein the identifier for the first item is generated based at least in part on a sentence from which the first item was derived. 
     
     
         14 . The method of  claim 12 , wherein the identifier for the first item is generated based at least in part on a clause. 
     
     
         15 . The method of  claim 12 , wherein the identifier for the first item is generated at least in part by performing a hash of a string. 
     
     
         16 . The method of  claim 15 , wherein determining that the identifier assigned to the second item is the same as the identifier assigned to the first item comprises comparing one or more hash codes. 
     
     
         17 . The method of  claim 12 , wherein determining that the first item is present includes performing natural language processing. 
     
     
         18 . The method of  claim 12 , wherein the first and second reviews are reviews of an entity. 
     
     
         19 . The method of  claim 18 , further comprising determining whether the first item is present in an ontology associated with the entity. 
     
     
         20 . The method of  claim 12 , further comprising selecting an ontology associated with an entity being reviewed in the first review. 
     
     
         21 . The method of  claim 12 , further comprising selecting, from a plurality of ontologies, an ontology that is associated with an industry associated with an entity being reviewed in the first review. 
     
     
         22 . A computer program product embodied in a non-transitory computer readable storage medium and comprising computer instructions for:
 receiving reputation data extracted from at least one data source, wherein the reputation data includes a plurality of user-authored reviews;   determining that a first item is present in a first review in the plurality of user-authored reviews;   assigning to the first item an identifier that is generated based at least in part on a string from which the first item was derived;   in response to determining that an identifier assigned to a second item is the same as the identifier assigned to the first item, counting the first item and the second item as a single item;   determining that a third item, different from but associated with the first item, is present in a second review in the plurality of user-authored reviews;   determining a sentiment for a theme based on the presence of the first and third items; and   providing as output a report that indicates the sentiment for the theme.