Phrase-based data classification system

Abstract

A method of classifying data is disclosed. Text data items are received. A set of classes into which the text data items are to be classified is received. A phrase-based classifier to classify the text data items into the set of classes is selected. The phrase-based classifier is applied to classify the text data items into the classes. Here, the applying includes creating a controlled vocabulary pertaining to classifying the text data items into the set of classes, building phrases based on the text data items and the controlled vocabulary, and classifying the text data items into the set of classes based on the phrases.

1 . (canceled) 
     
     
         2 . A system comprising:
 one or more processors;   one or more modules implemented by the one or more processors, the one or more modules configured to, at least:   construct a controlled vocabulary of words relevant for categorizing a set of data items into a set of classes;   extract sets of phrases from the set of data items that are composed of words in the controlled vocabulary;   classify different combinations of the sets of phrases into different classes of the set of classes;   map each data item of the set of data items onto the set of classes based on a correspondence between sets of phrases in the data item to the different combinations of the sets of phrases extracted from the set of data items; and   customize a service of a back-end system based on the mapping of each data item to the set of data items.   
     
     
         3 . The system of  claim 2 , wherein the extracting of the sets of phrases from the controlled vocabulary includes selecting a predetermined percentage of the most common data items included in the set of data items. 
     
     
         4 . The system of  claim 2 , wherein the classifying of the different combinations of the sets of phrases into different classes of the set of classes includes receiving input from a crowdsourcing system. 
     
     
         5 . The system of  claim 2 , wherein the mapping of each data item of the set of data items onto the set of classes includes removing words from each data item that are not included in the controlled vocabulary. 
     
     
         6 . The system of  claim 2 , wherein the different combinations of the sets of phrases include all possible phrases used in the different classes. 
     
     
         7 . The system of  claim 2 , wherein the set of data items is a set of user-specified job titles and the set of classes is a set of standardized job titles. 
     
     
         8 . The system of  claim 7 , wherein the service is a targeted advertising service and wherein the customizing of the service includes generating targeted advertising for a user based on the mapping. 
     
     
         9 . A method comprising:
 constructing a controlled vocabulary of words relevant for categorizing a set of data items into a set of classes;   extracting sets of phrases from the set of data items that are composed of words in the controlled vocabulary;   classifying different combinations of the sets of phrases into different classes of the set of classes;   mapping each data item of the set of data items onto the set of classes based on a correspondence between sets of phrases in the data item to the different combinations of the sets of phrases extracted from the set of data items; and   customizing a service of a back-end system based on the mapping of each data item to the set of data items.   
     
     
         10 . The method of  claim 9 , wherein the extracting of the sets of phrases from the controlled vocabulary includes selecting a predetermined percentage of the most common data items included in the set of data items. 
     
     
         11 . The method of  claim 9 , wherein the classifying of the different combinations of the sets of phrases into different classes of the set of classes includes receiving input from a crowdsourcing system. 
     
     
         12 . The method of  claim 9 , wherein the mapping of each data item of the set of data items onto the set of classes includes removing words from each data item that are not included in the controlled vocabulary. 
     
     
         13 . The method of  claim 9 , wherein the different combinations of the sets of phrases include all possible phrases used in the different classes. 
     
     
         14 . The method of  claim 9 , wherein the set of data items is a set of user-specified job titles and the set of classes is a set of standardized job titles. 
     
     
         15 . The system of  claim 14 , wherein the service is a targeted advertising service and wherein the customizing of the service includes generating targeted advertising for a user based on the mapping. 
     
     
         16 . A non-transitory machine-readable medium embodying a set of instructions that, when executed by a processor, cause the processor to perform operations, the operations comprising:
 constructing a controlled vocabulary of words relevant for categorizing a set of data items into a set of classes;   extracting sets of phrases from the set of data items that are composed of words in the controlled vocabulary;   classifying different combinations of the sets of phrases into different classes of the set of classes;   mapping each data item of the set of data items onto the set of classes based on a correspondence between sets of phrases in the data item to the different combinations of the sets of phrases extracted from the set of data items; and   customizing a service of a back-end system based on the mapping of each data item to the set of data items.   
     
     
         17 . The non-transitory machine-readable medium of  claim 16 , wherein the extracting of the sets of phrases from the controlled vocabulary includes selecting a predetermined percentage of the most common data items included in the set of data items. 
     
     
         18 . The non-transitory machine-readable medium of  claim 16 , wherein the classifying of the different combinations of the sets of phrases into different classes of the set of classes includes receiving input from a crowdsourcing system. 
     
     
         19 . The non-transitory machine-readable medium of  claim 16 , wherein the mapping of each data item of the set of data items onto the set of classes includes removing words from each data item that are not included in the controlled vocabulary. 
     
     
         20 . The non-transitory machine-readable medium of  claim 16 , wherein the different combinations of the sets of phrases include all possible phrases used in the different classes.