System and method of detecting mortgage related fraud

A system for detecting fraud, the system comprising:

• a storage configured to receive mortgage data associated with an applicant and at least one entity related to processing of the mortgage application;
• and
• a processor configured to apply the mortgage data to a first model that is based on data from a plurality of historical mortgage transactions associated with the at least one entity,
• wherein the applied first model includes at least one cluster associated with the plurality of historical mortgage transactions of the at least one entity and
• wherein the at least one cluster is based at least in part on respective data values of at least one data field of each of the plurality of historical mortgage transactions associated with the at least one entity, and
• wherein to apply the mortgage data to a first model, the processor is configured to compare at least a portion of the mortgage data with the at least one cluster;
• wherein the processor is further configured to:
• determine a first score based at least partly on a result of applying the mortgage data to the first model;
• generate data indicative of fraud based at least partly on the first score;
• and
• output the data indicative of fraud.

A system for detecting fraud, the system comprising:

• means for storing mortgage data associated with an applicant and at least one entity related to processing of the mortgage data;
• and
• means for electronically processing the stored data, said processing means comprising:
• means for applying the mortgage data to a first model that is based on data from a plurality of historical mortgage transactions associated with the at least one entity and determining a first score based at least partly on a result of applying the mortgage data to the first model,
• wherein the first model includes at least one cluster associated with the plurality of historical mortgage transactions of the at least one entity,
• wherein the at least one cluster is based on respective data values of at least one data field of each of the plurality of historical mortgage transactions associated with the at least one entity, and
• wherein applying the mortgage data to a first model comprises comparing at least a portion of the mortgage data with the at least one cluster;
• and
• means for generating data indicative of fraud based at least partly on the first score and outputting the data indicative of fraud.

A computerized method of generating models for detecting fraud, the method comprising:

• receiving data indicative of a plurality of historical mortgage transactions
• receiving data identifying a first portion of the historical mortgage transactions as having been fraudulent;
• and
• identifying a second portion of the historical mortgage transactions as having been non-fraudulent;
• executing a machine learning program on at least one processor to generate a first model based on data from a plurality of mortgage transactions, wherein each of the mortgage transactions is associated with an applicant and at least one entity related to processing of the mortgage transaction based on the data identifying the transactions as fraudulent and non-fraudulent and based on the data indicative of the historical mortgage transactions;
• storing the first model in a first computer readable medium;
• executing a machine learning program on the at least one processor to generate a second model based on data indicative of a portion of the received plurality of historical mortgage transactions associated with the at least one entity,
• wherein executing the machine learning program on the at least one processor to generate the second model comprises determining on the at least one processor at least one cluster and at least one value of the cluster associated with the at least one entity,
• wherein the at least one cluster is determined based on respective data values of at least one data field of each of the portion of the plurality of historical mortgage transactions associated with the at least one entity;
• and
• storing the second model in a second computer readable medium.