Stochastic apparatus and method for estimating credit card type when predicting interchange code to process credit card transactions

1. A computer-implemented method for predicting interchange codes corresponding to credit card transactions, the method comprising:

retrieving a historical completed transactions set from a database that corresponds to a plurality of merchants, where each completed transaction in the set comprises transaction features, a bank identification number (BIN), and a corresponding true interchange code;

extracting the transaction features from the completed transactions set;

transforming all BINs in the set into corresponding BIN features that comprise probabilities of a given BIN falling into each of the corresponding BIN features, where each of the probabilities are determined by analyzing the historical completed transactions set;

creating a first training set comprising all transaction features, all BIN features, and all true interchange codes associated with the historical completed transactions set;

training a random forest model using the first training set, wherein said training comprises rounding of all of the BIN features and non-discrete ones of the transaction features, and generating a second training set comprising rounded BIN features, rounded transaction features, discrete ones of the transaction features, and the true interchange codes;

training the random forest model using the second training set to generate a trained random forest model for prediction of the interchange codes;

generating a lookup table that maps each combination of a set of all rounded transaction features to a corresponding predicted interchange charge, wherein:

the corresponding predicted interchange charge comprises a weighted sum of interchange charges for a subset of all of the interchange codes;

a weight for each term of the weighted sum indicates a probability that an associated one of the set of all rounded transaction features will result in a corresponding interchange code with the subset;

probabilities for weights in the weighted sum are previously determined by executing the trained random forest model; and

outputs of the table responsive to specific rounded transaction features comprises a predicted interchange charge for a corresponding transaction; and

for new transactions, extracting and rounding corresponding transation-specific features and indexing the lookup table to obtain corresponding interchange charges.

2. The computer-implemented method as recited in claim 1 , wherein the transaction features further comprise a total transaction amount.

3. The computer-implemented method as recited in claim 2 , wherein the transaction features further comprise a tip percentage.

4. The computer-implemented method as recited in claim 3 , wherein the transaction features further comprise a type of transaction and an address verification system response.

5. The computer-implemented method as recited in claim 4 , wherein the transaction features further comprise a card network, a card type, and a merchant category code.

6. The computer-implemented method as recited in claim 1 , wherein the BIN features comprise regulated debit and debit.

7. The computer-implemented method as recited in claim 6 , wherein the BIN features further comprise rewards and high rewards.

8. A computer-readable storage medium storing instructions that, when executed by a computer, cause the computer to perform a method for using a computer system to predict interchange codes corresponding to credit card transactions, the method comprising:

retrieving a historical completed transactions set from a database that corresponds to a plurality of merchants, where each completed transaction in the set comprises transaction features, a bank identification number (BIN), and a corresponding true interchange code;

extracting the transaction features from the completed transactions;

transforming all BINs in the set into corresponding BIN features that comprise probabilities of a given BIN falling into each of the corresponding BIN features, where each of the probabilities are determined by analyzing the historical completed transactions set;

creating a first training set comprising all transaction features, all BIN features, and all true interchange codes associated with the historical completed transactions set;

training a random forest model using the first training set, wherein said training comprises rounding of all of the BIN features and non-discrete ones of the transaction features, and generating a second training set comprising rounded BIN features, rounded transaction features, discrete ones of the transaction features, and the true interchange codes;

training the random forest model using the second training set to generate a trained random forest model for prediction of the interchange codes;

generating a lookup table that maps each combination of a set of all rounded transaction features to a corresponding predicted interchange charge, wherein:

the corresponding predicted interchange charge comprises a weighted sum of interchange charges for a subset of all of the interchange codes;

a weight for each term of the weighted sum indicates a probability that an associated one of the set of all rounded transaction features will result in a corresponding interchange code within the subset;

probabilities for weights in the weighted sum are previously determined by executing the trained random forest model; and

outputs of the table responsive to specific rounded transaction features comprises a predicted interchange charge for a corresponding transaction; and

for new transactions, extracting and rounding corresponding transaction-specific features and indexing the lookup table to obtain corresponding interchange charges.

9. The computer-readable storage medium as recited in claim 8 , wherein the transaction features comprise a total transaction amount.

10. The computer-readable storage medium as recited in claim 9 , wherein the transaction features further comprise a tip percentage.

11. The computer-readable storage medium as recited in claim 10 , wherein the transaction features further comprise a type of transaction and an address verification system response.

12. The computer-readable storage medium as recited in claim 11 , wherein the transaction features further comprise a card network, a card type, and a merchant category code.

13. The computer-readable storage medium as recited in claim 8 , wherein the BIN features comprise regulated debit and debit.

14. The computer-readable storage medium as recited in claim 13 , wherein the BIN features further comprise rewards and high rewards.

15. A computer program product for predicting interchange codes corresponding to credit card transactions, the computer program product comprising:

a computer readable non-transitory medium having computer readable program code stored thereon, the computer readable program code comprising:

program instructions to retrieve a historical completed transactions set from a database that corresponds to a plurality of merchants, where each completed transaction in the set comprises transaction features, a bank identification number (BIN), and a corresponding true interchange code;

program instructions to extract the transaction features from the completed transactions;

program instructions to transform all BINs in the set into corresponding BIN features that comprise probabilities of a given BIN falling into each of the corresponding BIN features, where each of the probabilities are determined by analyzing the historical completed transactions set;

program instructions to create a first training set comprising all transaction features, all BIN features, and all true interchange codes associated with the historical completed transactions set;

program instructions to train a random forest model using the first training set, wherein said training comprises rounding of all of the BIN features and non-discrete ones of the transaction features, and generating a second training set comprising rounded BIN features, rounded transaction features, discrete ones of the transaction features, and the true interchange codes;

program instructions to train the random forest model using the second training set to generate a trained random forest model for prediction of the interchange codes;

program instructions to generate a lookup table that maps each combination of a set of all rounded transaction features to a corresponding predicted interchange charge, wherein:

the corresponding predicted interchange charge comprises a weighted sum of interchange charges for a subset of all of the interchange codes;

a weight for each term of the weighted sum indicates a probability that an associated one of the set of all rounded transaction features will result in a corresponding interchange code with the subset;

probabilities for weights in the weighted sum are previously determined by executing the trained random forest model; and

outputs of the table responsive to specific rounded transation features comprises a predicted interchange charge for a corresponding transaction; and

program instructions to, for new transactions, extract and round corresponding transaction-specific features and index the lookup table to obtain corresponding interchange charges.

16. The computer program product as recited in claim 15 , wherein the transaction features comprise a total transaction amount.

17. The computer program product as recited in claim 16 , wherein the transaction features further comprise a tip percentage.

18. The computer program product as recited in claim 17 , wherein the transaction features further comprise a type of transaction and an address verification system response.

19. The computer program product as recited in claim 15 , wherein the BIN features comprise regulated debit and debit.

20. The computer program product as recited in claim 19 , wherein the BIN features further comprise rewards and high rewards.