Method of operating artificial intelligence machines to improve predictive model training and performance

Abstract

A method of improving the training and performance of predictive models. A first method of operating an artificial intelligence machine produces predictive model language documents describing improved predictive models that generate better business decisions from raw data record inputs. A second method of operating an artificial intelligence machine including processors for predictive model algorithms produces and outputs better business decisions from raw data record inputs. Both methods enrich the raw data records their processors are fed by deleting data fields with data values that have little benefit in decision making, and that derive and add new data fields from information sources then available that do benefit in the decision making of the artificial intelligence machine through improved accuracies of prediction.

We claim: 
     
         1 . A computer-implemented method for classifying business records, comprising:
 receiving, at one or more processors, a business record corresponding to an entity and including a plurality of data fields storing business information;   accessing, via the one or more processors, a smart-agent predictive model corresponding to the entity;   generating, via the one or more processors, an enriched business record for consumption by the smart-agent predictive model by—
 accessing a predictive model language document embodying a data-enrichment descriptor describing a subset of the plurality of data fields for deletion and at least one combined data value; 
 deleting the subset of the plurality of data fields based on the data-enrichment descriptor; 
 calculating the at least one combined data value based on the data-enrichment descriptor, each of the at least one combined data values being calculated based on business information of at least two of the plurality of data fields; 
   producing, via the one or more processors, a prediction class output of the smart-agent predictive model based on the enriched business record;   accessing, via the one or more processors, at least one additional predictive model, each at least one additional predictive model being constructed according to one of: a neural network, data mining, clustering, case-based reasoning, a decision tree, a genetic algorithm, fuzzy logic, and rules and constraints;   generating, via the one or more processors, at least one additional predictive class output of the at least one predictive model; and   determining, via the one or more processors, a classification of the business record based at least in part on the prediction class output and the at least one additional predictive class output.   
     
     
         2 . The computer-implemented method of  claim 1 , wherein generating the enriched business record includes executing, via the one or more processors, a triple-DES decoding algorithm to decrypt the business information. 
     
     
         3 . The computer-implemented method of  claim 1 , wherein the predictive model language document embodying the data-enrichment descriptor is an encrypted XML document in a IFM format stored in non-transitory computer-readable media. 
     
     
         4 . The computer-implemented method of  claim 1 , wherein the smart-agent predictive model is embodied by an encrypted XML document in a IFM format stored in non-transitory computer-readable media. 
     
     
         5 . The computer-implemented method of  claim 4 , wherein the smart-agent predictive model stores one or more relationships between fields of the plurality of data fields in the form of a multi-level hash table tree. 
     
     
         6 . The computer-implemented method of  claim 5 , wherein the business record is a transaction record corresponding to a real-time request for approval of a transaction and the prediction class output represents a conclusion regarding whether the transaction is likely fraudulent. 
     
     
         7 . The computer-implemented method of  claim 1 , further comprising producing, via the one or more processors, a confidence output corresponding to the prediction class output of the smart-agent predictive model. 
     
     
         8 . The computer-implemented method of  claim 1 , wherein generating the enriched business record further includes generating at least one substitute datum by execution of at least one of a contextual dictionary algorithm and a context mining algorithm. 
     
     
         9 . The computer-implemented method of  claim 1 , further comprising failing to locate a matching model for the entity and, based on the failure to locate, building the smart-agent predictive model. 
     
     
         10 . The computer-implemented method of  claim 1 , wherein the smart-agent predictive model comprises a plurality of profiles, further comprising updating, via the one or more processors, the plurality of profiles based on the enriched business record. 
     
     
         11 . A server for classifying business records, comprising:
 one or more processors;   non-transitory computer-readable storage media having computer-executable instructions stored thereon, wherein when executed by the one or more processors the computer-readable instructions cause the one or more processors to—
 receive a business record corresponding to an entity and including a plurality of data fields storing business information; 
 access a smart-agent predictive model corresponding to the entity; 
 generate an enriched business record for consumption by the smart-agent predictive model by—
 accessing a predictive model language document embodying a data-enrichment descriptor describing a subset of the plurality of data fields for deletion and at least one combined data value; 
 deleting the subset of the plurality of data fields based on the data-enrichment descriptor; 
 calculating the at least one combined data value based on the data-enrichment descriptor, each of the at least one combined data values being calculated based on business information of at least two of the plurality of data fields; 
 
 produce a prediction class output of the smart-agent predictive model based on the enriched business record; 
 access at least one additional predictive model, each at least one additional predictive model being constructed according to one of: a neural network, data mining, clustering, case-based reasoning, a decision tree, a genetic algorithm, fuzzy logic, and rules and constraints; 
 generate at least one additional predictive class output of the at least one predictive model; and 
 determine a classification of the business record based at least in part on the prediction class output and the at least one additional predictive class output. 
   
     
     
         12 . The server of  claim 11 , wherein generating the enriched business record includes executing a triple-DES decoding algorithm to decrypt the business information. 
     
     
         13 . The server of  claim 11 , wherein the predictive model language document embodying the data-enrichment descriptor is an encrypted XML document in a IFM format stored in non-transitory computer-readable media. 
     
     
         14 . The server of  claim 11 , wherein the smart-agent predictive model is embodied by an encrypted XML document in a IFM format stored in non-transitory computer-readable media. 
     
     
         15 . The server of  claim 14 , wherein the smart-agent predictive model stores one or more relationships between fields of the plurality of data fields in the form of a multi-level hash table tree. 
     
     
         16 . The server of  claim 15 , wherein the business record is a transaction record corresponding to a real-time request for approval of a transaction and the prediction class output represents a conclusion regarding whether the transaction is likely fraudulent. 
     
     
         17 . The server of  claim 11 , wherein execution of the computer-readable instructions further causes the one or more processors to produce a confidence output corresponding to the prediction class output of the smart-agent predictive model. 
     
     
         18 . The server of  claim 11 , wherein generating the enriched business record further includes generating at least one substitute datum by execution of at least one of a contextual dictionary algorithm and a context mining algorithm. 
     
     
         19 . The server of  claim 11 , wherein execution of the computer-readable instructions further causes the one or more processors to build the smart-agent predictive model based on a failure to locate a matching model for the entity. 
     
     
         20 . The server of  claim 11 , wherein the smart-agent predictive model comprises a plurality of profiles and execution of the computer-readable instructions further causes the one or more processors to update the plurality of profiles based on the enriched business record.