Method, system and program product for matching of transaction records

Abstract

A method, system and program product comprise accessing a transaction records database. Unmatched records are collected into a first set. The first set at least comprises events and transactions. Probabilities of event matches of transactions originating from an event are calculated. The calculating uses at least defined features and stored probability distributions. A quality value for each of the event matches is calculated. The quality value is at least in part being determined by the probability of the event match. A second set of optimized event matches is determined using at least the quality values. Each of the optimized event matches at least comprises transactions deemed to have been generated by the event.

1 - 34 . (canceled) 
     
     
         35 . A self-pairing record database system, comprising:
 one or more computer processing units for modifying a database, stored in memory of the one or more computer processing units, from containing a plurality of unmatched entries into containing a corresponding set of matched entries, said modification comprising:   for each unmatched entry:
 generating a first one or more match sets each comprising a match between said unmatched entry and a second unmatched entry, and a second one or more match sets each comprising a match between said unmatched entry and each other unmatched entry in the database exclusive of said second unmatched entry, each match having a corresponding match probability; 
 normalizing a match probability between said unmatched entry and said second unmatched entry according to a difference between the match probabilities of the first one or more match sets and the second one or more match sets; and 
 modifying said unmatched entry and said second unmatched entry to indicate a matching association, responsive to the normalized match probability exceeding a threshold. 
   
     
     
         36 . The self-pairing record database system of  claim 35 , wherein said modification further comprises, for each unmatched entry, calculating a quality of matching between said unmatched entry and the second unmatched entry based on a sum of the match probabilities of the second one or more match sets. 
     
     
         37 . The self-pairing record database system of  claim 36 , wherein modifying said unmatched entry and said second unmatched entry to indicate a matching association is further responsive to the quality of matching exceeding a second threshold. 
     
     
         38 . The self-pairing record database system of  claim 37 , wherein said modification further comprises, for each unmatched entry, calculating an operational difference between a match between said unmatched entry and the second unmatched entry and a match between said unmatched entry and a third unmatched entry, the calculated quality of matching adjusted according to the calculated operational difference. 
     
     
         39 . The self-pairing record database system of  claim 35 , wherein generating the first one or more match sets each comprising a match between said unmatched entry and the second unmatched entry further comprises identifying, for each pair of said unmatched entry and each other unmatched entry, the match probability. 
     
     
         40 . The self-pairing record database system of  claim 35 , wherein said modification further comprises calculating a joint probability of matching as a product of the normalized match probability between said unmatched entry and said second unmatched entry, and a second normalized match probability between said second unmatched entry and said unmatched entry; and wherein modifying said unmatched entry and said second unmatched entry to indicate a matching association is further responsive to the joint probability of matching exceeding a second threshold. 
     
     
         41 . The self-pairing record database system of  claim 35 , wherein said modification further comprises storing the modified entries indicating the matching association to the database stored in memory of the one or more computer processing units. 
     
     
         42 . The self-pairing record database system of  claim 35 , wherein said modification further comprises modifying said unmatched entry to include an identifier of said second unmatched entry; and modifying said second unmatched entry to include an identifier of said unmatched entry. 
     
     
         43 . The self-pairing record database system of  claim 35 , wherein said modification further comprises extracting each entry of the database not identified as matched to any other entry. 
     
     
         44 . The self-pairing record database system of  claim 35 , wherein said modification further comprises extracting each entry of the database not comprising an identifier of another entry of the database. 
     
     
         45 . A method, comprising:
 for each unmatched entry in a database stored in memory of one or more computer processing units:
 generating, by the one or more computer processing units, a first one or more match sets each comprising a match between said unmatched entry and a second unmatched entry, and a second one or more match sets each comprising a match between said unmatched entry and each other unmatched entry in the database exclusive of said second unmatched entry, each match having a corresponding match probability; 
 normalizing, by the one or more computer processing units, a match probability between said unmatched entry and said second unmatched entry according to a difference between the match probabilities of the first one or more match sets and the second one or more match sets; and 
 modifying, by the one or more computer processing units, said unmatched entry and said second unmatched entry to indicate a matching association, responsive to the normalized match probability exceeding a threshold. 
   
     
     
         46 . The method of  claim 45 , further comprising, for each unmatched entry, calculating, by the one or more computer processing units, a quality of matching between said unmatched entry and the second unmatched entry based on a sum of the match probabilities of the second one or more match sets. 
     
     
         47 . The method of  claim 46 , wherein modifying said unmatched entry and said second unmatched entry to indicate a matching association is further responsive to the quality of matching exceeding a second threshold. 
     
     
         48 . The method of  claim 47 , further comprising, for each unmatched entry, calculating, by the one or more computer processing units, an operational difference between a match between said unmatched entry and the second unmatched entry and a match between said unmatched entry and a third unmatched entry, the calculated quality of matching adjusted according to the calculated operational difference. 
     
     
         49 . The method of  claim 45 , wherein generating the first one or more match sets each comprising a match between said unmatched entry and the second unmatched entry further comprises identifying, by the one or more computer processing units for each pair of said unmatched entry and each other unmatched entry, the match probability. 
     
     
         50 . The method of  claim 45 , further comprising calculating, by the one or more computer processing units, a joint probability of matching as a product of the normalized match probability between said unmatched entry and said second unmatched entry, and a second normalized match probability between said second unmatched entry and said unmatched entry; and wherein modifying said unmatched entry and said second unmatched entry to indicate a matching association is further responsive to the joint probability of matching exceeding a second threshold. 
     
     
         51 . The method of  claim 45 , further comprising storing, by the one or more computer processing units, the modified entries indicating the matching association to the database stored in memory of the one or more computer processing units. 
     
     
         52 . The method of  claim 45 , further comprising:
 modifying, by the one or more computer processing units, said unmatched entry to include an identifier of said second unmatched entry; and   modifying, by the one or more computer processing units, said second unmatched entry to include an identifier of said unmatched entry.   
     
     
         54 . The method of  claim 45 , further comprising extracting, by the one or more computer processing units, each entry of the database not identified as matched to any other entry. 
     
     
         54 . The method of  claim 45 , further comprising extracting, by the one or more computer processing units, each entry of the database not comprising an identifier of another entry of the database.