System And Method For Providing Document Classification Suggestions

Abstract

A system and method for providing document classification suggestions is provided. Clusters of uncoded documents are obtained. One or more of the uncoded documents in one such cluster are compared to a set of reference documents. Each reference document is assigned with a classification code. Those reference documents that are similar to the one or more uncoded documents are identified. Different types of the classification codes for at least a portion of the similar reference documents are identified and a count of the classification codes assigned to the portion of similar reference documents for each classification code type is obtained. A suggestion for classification of at least one of the one or more uncoded documents is provided based on the count of classification codes for each classification type and one of a presence and absence of each classification code type.

What is claimed is: 
     
         1 . A system for providing document classification suggestions, comprising:
 a database to store clusters of uncoded documents; and   a server comprising a central processing unit, memory, an input port to receive the clusters from the database, and an output port, wherein the central processing unit is configured to:
 compare one or more of the uncoded documents in one such cluster to a set of reference documents each assigned with a classification code; 
 identify those reference documents in the set that are similar to the one or more uncoded documents; 
 identify different types of the classification codes for at least a portion of the similar reference documents and obtaining a count of the classification codes assigned to the portion of similar reference documents for each classification code type; and 
 provide a suggestion for classification of at least one of the one or more uncoded documents based on the count of classification codes for each classification type and one of a presence and absence of each classification code type. 
   
     
     
         2 . A system according to  claim 1 , further comprising:
 a selection module to select the classification type that is present and has a highest count of classification codes as the suggested classification.   
     
     
         3 . A system according to  claim 1 , further comprising:
 a cluster classification module to classify the cluster of the one or more uncoded documents, comprising:
 a counter module to obtain a count of classification codes for each of the classification types based on the similar reference documents and the uncoded documents with assigned classification codes in the cluster; and 
 an assignment module to assign to the cluster the classification code of the classification type with a highest count of classification codes in the cluster. 
   
     
     
         4 . A system according to  claim 1 , further comprising:
 a display to provide the clusters along one or more spines based on a similarity of the uncoded documents.   
     
     
         5 . A system according to  claim 4 , further comprising:
 a determination module to determine the spine on which the cluster is placed; and   a label module to identify a label associated with the identified spine,   wherein the label is considered with the count of classification codes for each classification type and one of a presence and absence of each classification code type for the assignment of the classification code.   
     
     
         6 . A system according to  claim 4 , further comprising:
 a document identification module to identify the similar reference documents, comprising:
 a determination module to determine a measure for the spine on which the cluster is placed; 
 a comparison module to compare the measure with each of the reference documents; and 
 a selection module to select the reference documents most similar to the measure as the similar reference documents. 
   
     
     
         7 . A system according to  claim 1 , further comprising:
 a confidence module to determine a confidence level of the suggested classification and to provide the confidence level with the suggested classification.   
     
     
         8 . A system according to  claim 1 , further comprising:
 a document identification module to identify the similar reference documents, comprising:
 a determination module to determine a measure for the one or more uncoded documents in the cluster, wherein the measure comprises one of a center of the cluster, a sample of the uncoded documents, and the cluster center and the sample; 
 a comparison module to compare the measure with each of the reference documents; and 
 a selection module to select the reference documents most similar to the measure as the similar reference documents. 
   
     
     
         9 . A system according to  claim 1 , wherein the similar reference documents satisfy a threshold number. 
     
     
         10 . A system according to  claim 1 , further comprising:
 a receipt module to receive from a user, a location for injecting the similar reference documents into the cluster.   
     
     
         11 . A method for providing document classification suggestions, comprising:
 obtaining clusters of uncoded documents;   comparing one or more of the uncoded documents in one such cluster to a set of reference documents each assigned with a classification code;   identifying those reference documents in the set that are similar to the one or more uncoded documents;   identifying different types of the classification codes for at least a portion of the similar reference documents and obtaining a count of the classification codes assigned to the portion of similar reference documents for each classification code type; and   providing a suggestion for classification of at least one of the one or more uncoded documents based on the count of classification codes for each classification type and one of a presence and absence of each classification code type.   
     
     
         12 . A method according to  claim 11 , further comprising:
 selecting the classification type that is present and has a highest count of classification codes as the suggested classification.   
     
     
         13 . A method according to  claim 11 , further comprising:
 classifying the cluster with the one or more uncoded documents, comprising:
 obtaining a count of classification codes for each of the classification types based on the similar reference documents and the uncoded documents with assigned classification codes in the cluster; and 
 assigning to the cluster the classification code of the classification type with a highest count of classification codes in the cluster. 
   
     
     
         14 . A method according to  claim 11 , further comprising:
 displaying the clusters along one or more spines based on a similarity of the uncoded documents.   
     
     
         15 . A method according to  claim 14 , further comprising:
 determining the spine on which the cluster is placed; and   identifying a label associated with the identified spine,   wherein the label is considered with the count of classification codes for each classification type and one of a presence and absence of each classification code type for the assignment of the classification code.   
     
     
         16 . A method according to  claim 14 , further comprising:
 identifying the similar reference documents, comprising:
 determining a measure for the spine on which the cluster is placed; 
 comparing the measure with each of the reference documents; and 
 selecting the reference documents most similar to the measure as the similar reference documents. 
   
     
     
         17 . A method according to  claim 11 , further comprising:
 determining a confidence level of the suggested classification; and   providing the confidence level with the suggested classification.   
     
     
         18 . A method according to  claim 11 , further comprising:
 identifying the similar reference documents, comprising:
 determining a measure for the one or more uncoded documents in the cluster, wherein the measure comprises one of a center of the cluster, a sample of the uncoded documents, and the cluster center and the sample; 
 comparing the measure with each of the reference documents; and 
 selecting the reference documents most similar to the measure as the similar reference documents. 
   
     
     
         19 . A method according to  claim 11 , wherein the similar reference documents satisfy a threshold number. 
     
     
         20 . A method according to  claim 11 , further comprising:
 receiving from a user, a location for injecting the similar reference documents into the cluster.