×

Computer-implemented multiple entity dynamic summarization systems and methods

  • US 8,805,737 B1
  • Filed: 11/02/2010
  • Issued: 08/12/2014
  • Est. Priority Date: 11/02/2009
  • Status: Active Grant
First Claim
Patent Images

1. A computer-program product tangibly embodied in a non-transitory machine readable storage medium, and including instructions configured to cause a data processing apparatus to perform operations including:

  • accessing a set of data, wherein the set of data includes information about multiple transactions;

    computing a fraud score for each of the transactions, each of the fraud scores computed based on;

    a portion of data in the set of data that corresponds to that transaction; and

    a pre-determined fraud scoring model,wherein the fraud score represents a likelihood that a transaction falls within a pre-determined fraudulent category;

    identifying a plurality of entities, wherein each of the entities is associated with a subset of transactions from the multiple transactions, and wherein, in each of the subsets, the transactions are associated with a common parameter;

    generating a summary for each of the entities, wherein generating the summary includes identifying a fraudulent activity trend corresponding to the entity and wherein the fraudulent activity trend is determined using an analysis of the computed fraud scores;

    storing each of the generated summaries in a computer memory; and

    performing, on the computing device, a subsequent fraud detection on an additional transaction, wherein performing the subsequent fraud detection facilitates a decision for the additional transaction and includes;

    receiving additional transaction data that represents the additional transaction;

    associating the additional transaction with at least one of the identified entities; and

    computing a fraud score for the additional transaction, wherein computing the fraud score for the additional transaction is based on the additional transaction data, the pre-determined fraud scoring model, and the summary of the at least one entity associated with the additional transaction.

View all claims
  • 1 Assignment
Timeline View
Assignment View
    ×
    ×