Method of classifying a bill

Abstract

Briefly, embodiments of a method of classifying a bill are disclosed.

1 . A method of classifying a bill comprising:
 applying to said bill to be classified one or more bill classification schemes, said one or more bill classification schemes being derived using one or more statistical based decision processes.   
     
     
         2 . The method of  claim 1 , wherein said bill comprises a medical bill. 
     
     
         3 . The method of  claim 2 , wherein said one or more bill classification schemes comprises a bill classification as potentially being produced by billing abuse or billing fraud. 
     
     
         4 . The method of  claim 3 , and further comprising: generating a confidence rating on said bill classification of said medical bill as being potentially produced by billing abuse or billing fraud. 
     
     
         5 . The method of  claim 2 , wherein said one or more bill classification schemes comprises a bill classification regarding point of origin. 
     
     
         6 . The method of  claim 5 , wherein the point of origin bill classification comprises a facility type bill classification. 
     
     
         7 . The method of  claim 1 , wherein said one or more statistical based decision processes comprises one or more machine learning techniques. 
     
     
         8 . The method of  claim 7 , wherein said one or more machine learning techniques includes an SVM-type process. 
     
     
         9 . The method of  claim 8 , wherein said SVM-type process comprises a bagged SVM-type process. 
     
     
         10 . The method of  claim 7 , wherein said one or more machine learning techniques includes active learning. 
     
     
         11 . The method of  claim 1 , wherein said one or more bill classification schemes comprise multiple classification schemes;
 and further comprising:   applying voting to the multiple outcomes of said multiple bill classification schemes to determine a preferred bill classification for said bill.   
     
     
         12 . The method of  claim 11 , wherein, prior to said applying multiple bill classification schemes, apparent errors in said bill are identified. 
     
     
         13 . The method of  claim 12 , wherein, prior to said applying multiple bill classification schemes, said apparent errors in said bill are adjusted in a manner so that quality control backing tracking is capable of being applied. 
     
     
         14 . The method of  claim 11 , and further comprising: generating a confidence rating on said preferred bill classification for said bill. 
     
     
         15 . The method of  claim 11 , wherein said multiple bill classification schemes are derived from multiple heterogeneous sets of training bills. 
     
     
         16 . An apparatus comprising:
 a computing platform;   said computing platform being adapted to classify a bill using one or more bill classification schemes, said one or more bill classification schemes being derived using one or more statistical based decision processes.   
     
     
         17 . The apparatus of  claim 16 , wherein said bill comprises a medical bill. 
     
     
         18 . The method of  claim 17 , wherein said one or more bill classification schemes comprises a bill classification as potentially being produced by billing abuse or billing fraud. 
     
     
         19 . The apparatus of  claim 18 , said computing platform being further adapted to generate a confidence rating on said bill classification of said medical bill as being potentially produced by billing abuse or billing fraud. 
     
     
         20 . The apparatus of  claim 17 , wherein said one or more bill classification schemes comprises a bill classification regarding point of origin. 
     
     
         21 . The apparatus of  claim 20 , wherein the point of origin bill classification comprises a facility type bill classification. 
     
     
         22 . The apparatus of  claim 15 , wherein said one or more statistical based decision processes comprises one or more machine learning techniques. 
     
     
         23 . The apparatus of  claim 22 , wherein said one or more machine learning techniques includes an SVM-type process. 
     
     
         24 . The apparatus of  claim 23 , wherein said SVM-type process comprises a bagged SVM-type process. 
     
     
         25 . The apparatus of  claim 22 , wherein said one or more machine learning techniques includes active learning. 
     
     
         26 . The apparatus of  claim 15 , wherein said one or more bill classification schemes comprise multiple classification schemes;
 said computing platform being further adapted to apply voting to the multiple outcomes of said multiple bill classification schemes to determine a preferred bill classification for said bill.   
     
     
         27 . The apparatus of  claim 26 , wherein, prior to said applying multiple bill classification schemes, apparent errors in said bill are identified. 
     
     
         28 . The apparatus of  claim 27 , wherein, prior to said applying multiple bill classification schemes, said apparent errors in said bill are adjusted in a manner so that quality control backing tracking is capable of being applied. 
     
     
         29 . The apparatus of  claim 26 , said computing platform being further adapted to generate a confidence rating on said preferred bill classification for said bill. 
     
     
         30 . The apparatus of  claim 26 , wherein said computing platform is further adapted to derive said multiple bill classification schemes from multiple heterogeneous sets of training bills. 
     
     
         31 . An article comprising: a storage medium having stored thereon instructions executable by a special purpose computing platform to classify a bill using one or more bill classification schemes, said one or more bill classification schemes being derived using one or more statistical based decision processes. 
     
     
         32 . The article of  claim 31 , wherein said bill comprises a medical bill. 
     
     
         33 . The article of  claim 32 , wherein said one or more bill classification schemes comprises a bill classification as potentially being produced by billing abuse or billing fraud. 
     
     
         34 . The article of  claim 33 , said storage medium having stored thereon further instructions executable by a special purpose computing platform to generate a confidence rating on said bill classification of said medical bill as being potentially produced by billing abuse or billing fraud. 
     
     
         35 . The article of  claim 32 , wherein said one or more bill classification schemes comprises a bill classification regarding point of origin. 
     
     
         36 . The article of  claim 35 , wherein the point of origin bill classification comprises a facility type bill classification. 
     
     
         37 . The article of  claim 31 , wherein said one or more statistical based decision processes comprises one or more machine learning techniques. 
     
     
         38 . The article of  claim 37 , wherein said one or more machine learning techniques includes an SVM-type process. 
     
     
         39 . The article of  claim 38 , wherein said SVM-type process comprises a bagged SVM-type process. 
     
     
         40 . The article of  claim 38 , wherein said one or more machine learning techniques includes active learning. 
     
     
         41 . The article of  claim 31 , wherein said one or more bill classification schemes comprise multiple classification schemes;
 said storage medium having stored thereon further instructions executable by a special purpose computing platform to apply voting to the multiple outcomes of said multiple bill classification schemes to determine a preferred bill classification for said bill.   
     
     
         42 . The article of  claim 41 , wherein, prior to said applying multiple bill classification schemes, apparent errors in said bill are identified. 
     
     
         43 . The article of  claim 42 , wherein, prior to said applying multiple bill classification schemes, said apparent errors in said bill are adjusted in a manner so that quality control backing tracking is capable of being applied. 
     
     
         44 . The article of  claim 41 , said storage medium having stored thereon further instructions executable by a special purpose computing platform to generate a confidence rating on said preferred bill classification for said bill. 
     
     
         45 . The article of  claim 41 , said storage medium having stored thereon further instructions executable by a special purpose computing platform to derive said multiple bill classification schemes from multiple heterogeneous sets of training bills.