Document classification system with user-defined rules

Abstract

Machines, systems and methods for classifying documents, the method comprising: classifying a document from among a plurality of documents in a first class, in response to applying statistical analysis to data associated with the document; classifying the document in a second class, in response to determining that a rule from among a plurality of rules applies to the document, wherein a proposed rule is added to the plurality of rules, in response to determining that application of the proposed rule to one or more of the plurality of documents to which the rule is applicable does not diminish accuracy of overall classification for the plurality of documents.

What is claimed is: 
     
         1 . A method for classifying documents, the method comprising:
 classifying a document from among a plurality of documents in a first class, in response to applying statistical analysis to data associated with the document; and   classifying the document in a second class, in response to determining that a rule from among a plurality of rules applies to the document,   wherein a proposed rule is added to the plurality of rules, in response to determining that application of the proposed rule to one or more of the plurality of documents to which the rule is applicable does not diminish accuracy of overall classification for the plurality of documents.   
     
     
         2 . The method of  claim 1 , wherein the proposed rule is not added to the plurality of rules, in response to determining that application of the proposed rule to one or more of the plurality of documents to which the rule is applicable diminishes accuracy of overall classification for the plurality of documents. 
     
     
         3 . The method of  claim 1 , wherein a scoring mechanism is utilized when a proposed rule is added to the plurality of rules such that a favorable score is assigned to the proposed rule, in response to determining that application of the proposed rule to one or more of the plurality of documents to which the rule is applicable enhances accuracy of overall classification for the plurality of documents. 
     
     
         4 . The method of  claim 3 , wherein a first score is assigned to a first proposed rule, wherein the first score is more favorable than a second score assigned to a second proposed rule, in response to determining that the first rule enhances the accuracy of the overall classification for the plurality of documents more than the second rule. 
     
     
         5 . The method of  claim 1 , wherein a scoring mechanism is utilized when a proposed rule is added to the plurality of rules such that an unfavorable score is assigned to the proposed rule, in response to determining that application of the proposed rule to one or more of the plurality of documents to which the rule is applicable diminishes accuracy of overall classification for the plurality of documents. 
     
     
         6 . The method of  claim 5 , wherein a first score is assigned to a first proposed rule, wherein the first score is less favorable than a second score assigned to a second proposed rule, in response to determining that the first rule diminishes the accuracy of the overall classification for the plurality of documents more than the second rule. 
     
     
         7 . The method of  claim 1 , wherein a scoring mechanism is utilized when a proposed rule is added to the plurality of rules such that a neutral score is assigned to the proposed rule, in response to determining that application of the proposed rule to one or more of the plurality of documents to which the rule is applicable neither enhances nor diminishes accuracy of overall classification for the plurality of documents. 
     
     
         8 . The method of  claim 1 , wherein a scoring mechanism is utilized when a proposed rule is added to the plurality of rules such that:
 a favorable score is assigned to the proposed rule, in response to determining that application of the proposed rule to one or more of the plurality of documents to which the rule is applicable enhances accuracy of overall classification for the plurality of documents, and   an unfavorable score is assigned to the proposed rule, in response to determining that application of the proposed rule to one or more of the plurality of documents to which the rule is applicable diminishes accuracy of overall classification for the plurality of documents,   wherein the score assigned to the proposed rule after the proposed rule is added to the plurality of rules is re-evaluated periodically to determine whether a more favorable or less favorable score is to be assigned to the rule based on analysis of data obtained from applying the rule to one or more of the plurality of documents after a period of time has elapsed since the rule was added.   
     
     
         9 . The method of  claim 1 , wherein a data structure is implemented to include N indicators associated with a rule that is applicable to N corresponding documents, wherein a respective indicator provides information about whether application of the rule to a corresponding document has enhanced or diminished the classification for the corresponding document. 
     
     
         10 . The method of  claim 9 , wherein the score associated with the rule is improved, in response to determining that the application of the rule to a portion of the N documents has enhanced the classification of said portion of the N documents.