Computer-Implemented System And Method For Providing Concept Classification Suggestions Based On Concept Similarity

Abstract

A computer-implemented system and method for providing concept classification suggestions based on concept similarity are provided. One or more uncoded concepts and a plurality of reference concepts, each reference concept associated with a classification code, are obtained. At least one of the uncoded concepts is compared to the reference concepts and reference concepts similar to the at least one uncoded concept are identified based on the comparison. One of the classification codes is suggested for the at least one uncoded concept, which includes providing a number of the classification codes associated with the similar reference concepts and a number of the similar reference concepts for each of the classification codes.

What is claimed is: 
     
         1 . A computer-implemented system for providing concept classification suggestions based on concept similarity, comprising:
 a processor to execute code, comprising:
 a concept module to obtain one or more uncoded concepts and a plurality of reference concepts, each reference concept associated with a classification code; 
 a comparison module to compare at least one of the uncoded concepts to the reference concepts and to identify the reference concepts similar to the at least one uncoded concept based on the comparison; and 
 a suggestion module to suggest one of the classification codes for the at least one uncoded concept, comprising a number module to provide a number of the classification codes associated with the similar reference concepts and a number of the similar reference concepts for each of the classification codes. 
   
     
     
         2 . A system according to  claim 1 , further comprising:
 an association module to associate a visual representation with each of the classification codes; and   a display module to display the visual representation associated with the classification code of each of the similar reference concepts.   
     
     
         3 . A system according to  claim 2 , wherein the visual representation comprises at least one of a symbol, shape, and color different from the visual representations of the remaining classification codes. 
     
     
         4 . A system according to  claim 2 , further comprising:
 a list module to display representations of the uncoded concepts in a list; and   a receipt module to receive a user selection of the at least one uncoded concept,   wherein the visual representations associated with the similar reference concepts are displayed together with the representation of the at least one uncoded concept.   
     
     
         5 . A system according to  claim 1 , further comprising:
 a cluster module to obtain a plurality of clusters comprising the uncoded concepts;   a selection module to select one of the uncoded concepts in one of the clusters;   a neighborhood module to compare the selected concept to a neighborhood of the reference concepts similar to the selected concept;   an analysis module to analyze the neighborhood; and   a machine-generated suggestion module to provide a machine-generated suggestion of one of the classification codes for the selected uncoded concept based on the analysis.   
     
     
         6 . A system according to  claim 5 , wherein the selected uncoded concept comprises the at least one uncoded concept and the neighborhood of the reference concepts similar to the uncoded concept comprises one or more of the reference concepts similar to the at least one uncoded concept. 
     
     
         7 . A system according to  claim 5 , further comprising:
 a distance module to determine a distance between the selected concept and each of the reference concepts in the neighborhood; and   a generation module to generate the machine-generated suggestion, comprising at least one of:
 a minimum distance module to identify the reference concept with the closest distance to the uncoded concept and assign the classification code of the reference concept with the closest distance as the machine-generated suggestion; 
 a maximum count module to calculate an average of the distances between the uncoded concept and the reference concepts associated with each of the classification codes and assign the classification code with the closest average distance as the machine-generated suggestion; and 
 a weighted count module to count the reference concepts in the neighborhood for each of the classification codes, weigh each count based on the distance between the reference concepts with the classification code and the uncoded concept, and assign the classification code with the highest weighted count as the machine-generated suggestion. 
   
     
     
         8 . A system according to  claim 5 , further comprising:
 a confidence level module to provide a confidence level for the machine-generated suggestion.   
     
     
         9 . A system according to  claim 1 , further comprising:
 a cluster module to obtain a plurality of clusters comprising the uncoded concepts;   a spine module to compare the clusters and organize the clusters along spines, each spine comprising a vector, based on the comparison; and   a display module to display the clusters and the spines.   
     
     
         10 . A system according to  claim 9 , further comprising:
 a similarity module to provide one or more of the reference concepts similar to at least one of one or more of the clusters and one or more of the spines; and   a classification module to provide one or more classification suggestions for the at least one of the one or more clusters and the one or more spines based on the reference concepts similar to the at least one of the one or more clusters and the one or more spines.   
     
     
         11 . A computer-implemented method for providing concept classification suggestions based on concept similarity, comprising the steps of:
 obtaining one or more uncoded concepts and a plurality of reference concepts, each reference concept associated with a classification code;   comparing at least one of the uncoded concepts to the reference concepts and identifying reference concepts similar to the at least one uncoded concept based on the comparison; and   suggesting one of the classification codes for the at least one uncoded concept, comprising providing a number of the classification codes associated with the similar reference concepts and a number of the similar reference concepts for each of the classification codes,   wherein the steps are performed on a suitably programmed computer.   
     
     
         12 . A method according to  claim 11 , further comprising:
 associating a visual representation with each of the classification codes; and   displaying the visual representation associated with the classification code of each of the similar reference concepts.   
     
     
         13 . A method according to  claim 12 , wherein the visual representation comprises at least one of a symbol, shape, and color different from the visual representations of the remaining classification codes. 
     
     
         14 . A method according to  claim 12 , further comprising:
 displaying representations of the uncoded concepts in a list; and   receiving a user selection of the at least one uncoded concept,   wherein the visual representations associated with the similar reference concepts are displayed together with the representation of the at least one uncoded concept.   
     
     
         15 . A method according to  claim 11 , further comprising:
 obtaining a plurality of clusters comprising the uncoded concepts;   selecting one of the uncoded concepts in one of the clusters;   comparing the selected concept to a neighborhood of the reference concepts similar to the selected concept;   analyzing the neighborhood; and   providing a machine-generated suggestion of one of the classification codes for the selected uncoded concept based on the analysis.   
     
     
         16 . A method according to  claim 15 , wherein the selected uncoded concept comprises the at least one uncoded concept and the neighborhood of the reference concepts similar to the uncoded concept comprises one or more of the reference concepts similar to the at least one uncoded concept. 
     
     
         17 . A method according to  claim 15 , further comprising:
 determining a distance between the selected concept and each of the reference concepts in the neighborhood; and   generating the machine-generated suggestion, comprising at least one of:
 identifying the reference concept with the closest distance to the uncoded concept and assigning the classification code of the reference concept with the closest distance as the machine-generated suggestion; 
 calculating an average of the distances between the uncoded concept and the reference concepts associated with each of the classification codes and assigning the classification code with the closest average distance as the machine-generated suggestion; and 
 counting the reference concepts in the neighborhood for each of the classification codes, weighing each count based on the distance between the reference concepts with the classification code and the uncoded concept, and assigning the classification code with the highest weighted count as the machine-generated suggestion. 
   
     
     
         18 . A method according to  claim 15 , further comprising:
 providing a confidence level for the machine-generated suggestion.   
     
     
         19 . A method according to  claim 11 , further comprising:
 obtaining a plurality of clusters comprising the uncoded concepts;   comparing the clusters and organizing the clusters along spines, each spine comprising a vector, based on the comparison; and   displaying the clusters and the spines.   
     
     
         20 . A method according to  claim 19 , further comprising:
 providing one or more of the reference concepts similar to at least one of one or more of the clusters and one or more of the spines; and   providing one or more classification suggestions for the at least one of the one or more clusters and the one or more spines based on the reference concepts similar to the at least one of the one or more clusters and the one or more spines.