Systems and methods for improving computation efficiency in the detection of fraud indicators for loans with multiple applicants
First Claim
1. A computer-implemented method comprising:
- receiving, from a first information source, applicant data associated with a loan application having a plurality of applicants, wherein the applicant data comprises at least a name, a social security number, and a street address for each of the plurality of applicants;
querying one or more of a public database and a private database with at least a portion of the applicant data;
receiving, from a second information source that is independent of the first information source, a plurality of independent information in response to the querying;
determining, with one or more computer processors in communication with a memory, based at least in part on a comparison of the applicant data with at least a portion of the plurality of independent information, a first validity indication of the applicant data;
creating, with the one or more computer processors, disambiguated applicant records responsive to the first validity indication by one or more of;
performing data cleansing on one or more of the applicant data and the plurality of independent information to eliminate one or more name variations; and
adding metadata record to one or more of the applicant data and the plurality of independent information;
determining, with the one or more computer processors, relationships among the disambiguated applicant records by one or more of;
creating a core join data structure with at least a portion of all available disambiguated applicant records;
splitting the core join data structure into persisted parts; and
clustering one or more of the persisted parts and the disambiguated applicant records;
scoring, with the one or more computer processors and based at least in part on the determined relationships among the disambiguated applicant records, at least one parameter of the applicant data;
determining, with the one or more computer processors, one or more indicators of fraud based on the scoring of the at least one parameter; and
outputting, for display, one or more indicators of fraud.
1 Assignment
0 Petitions
Accused Products
Abstract
Certain embodiments of the disclosed technology include systems and methods for increasing efficiency in the detection of identity-based fraud indicators associated with a loan application. A method is provided that includes receiving applicant data associated with a loan application having a plurality of applicants. The application data includes at least a name, a social security number, and a street address for each of the plurality of applicants. The method includes querying one or more databases with the applicant data; receiving a plurality of independent information in response to the querying; determining a validity indication of the applicant data; creating disambiguated applicant records; determining relationships among the disambiguated applicant records; scoring at least one parameter of the applicant data; determining one or more indicators of fraud based on the scoring; and outputting, for display, one or more indicators of fraud.
81 Citations
20 Claims
-
1. A computer-implemented method comprising:
-
receiving, from a first information source, applicant data associated with a loan application having a plurality of applicants, wherein the applicant data comprises at least a name, a social security number, and a street address for each of the plurality of applicants; querying one or more of a public database and a private database with at least a portion of the applicant data; receiving, from a second information source that is independent of the first information source, a plurality of independent information in response to the querying; determining, with one or more computer processors in communication with a memory, based at least in part on a comparison of the applicant data with at least a portion of the plurality of independent information, a first validity indication of the applicant data; creating, with the one or more computer processors, disambiguated applicant records responsive to the first validity indication by one or more of; performing data cleansing on one or more of the applicant data and the plurality of independent information to eliminate one or more name variations; and adding metadata record to one or more of the applicant data and the plurality of independent information; determining, with the one or more computer processors, relationships among the disambiguated applicant records by one or more of; creating a core join data structure with at least a portion of all available disambiguated applicant records; splitting the core join data structure into persisted parts; and clustering one or more of the persisted parts and the disambiguated applicant records; scoring, with the one or more computer processors and based at least in part on the determined relationships among the disambiguated applicant records, at least one parameter of the applicant data; determining, with the one or more computer processors, one or more indicators of fraud based on the scoring of the at least one parameter; and outputting, for display, one or more indicators of fraud. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12)
-
-
13. A system comprising:
-
at least one memory for storing data and computer-executable instructions; and at least one processor configured to access the at least one memory and further configured to execute the computer-executable instructions to; receive, from a first information source, applicant data associated with a loan application having a plurality of applicants, wherein the applicant data comprises at least a name, a social security number, and a street address for each of the plurality of applicants; query one or more of a public database and a private database with at least a portion of the applicant data; receive, from a second information source that is independent of the first information source, a plurality of independent information in response to the querying; determine, with the at least one processor, based at least in part on a comparison of the applicant data with at least a portion of the plurality of independent information, a first validity indication of the applicant data; create, with the at least one processor, disambiguated applicant records responsive to the first validity indication by one or more of; performing data cleansing on one or more of the applicant data and the plurality of independent information to eliminate one or more name variations; and adding metadata record to one or more of the applicant data and the plurality of independent information; determine, with the at least one processor, relationships among the disambiguated applicant records by one or more of; creating a core join data structure with at least a portion of all available disambiguated applicant records; splitting the core join data structure into persisted parts; and clustering one or more of the persisted parts and the disambiguated applicant records; score, with the at least one processor and based at least in part on the determined relationships among the disambiguated applicant records, at least one parameter of the applicant data; determine, with the one or more computer processors, one or more indicators of fraud based on the score of the at least one parameter; and output, for display, one or more indicators of fraud. - View Dependent Claims (14, 15, 16, 17, 18, 19)
-
-
20. One or more computer readable media comprising computer-executable instructions that, when executed by one or more processors, configure the one or more processors to perform the method of:
-
receiving, from a first information source, applicant data associated with a loan application having a plurality of applicants, wherein the applicant data comprises at least a name, a social security number, and a street address for each of the plurality of applicants; querying one or more of a public database and a private database with at least a portion of the applicant data; receiving, from a second information source that is independent of the first information source, a plurality of independent information in response to the querying; determining, with one or more computer processors in communication with a memory, based at least in part on a comparison of the applicant data with at least a portion of the plurality of independent information, a first validity indication of the applicant data; creating, with the one or more computer processors, disambiguated applicant records responsive to the first validity indication by one or more of; performing data cleansing on one or more of the applicant data and the plurality of independent information to eliminate one or more name variations; and adding metadata record to one or more of the applicant data and the plurality of independent information; determining, with the one or more computer processors, relationships among the disambiguated applicant records by one or more of; creating a core join data structure with at least a portion of all available disambiguated applicant records; splitting the core join data structure into persisted parts; and clustering one or more of the persisted parts and the disambiguated applicant records; scoring, with the one or more computer processors and based at least in part on the determined relationships among the disambiguated applicant records, at least one parameter of the applicant data; determining, with the one or more computer processors, one or more indicators of fraud based on the scoring of the at least one parameter; and outputting, for display, one or more indicators of fraud.
-
Specification