×

Method for identifying sub-sequences of interest in a sequence

  • US 8,046,173 B2
  • Filed: 03/06/2008
  • Issued: 10/25/2011
  • Est. Priority Date: 06/02/2004
  • Status: Active Grant
First Claim
Patent Images

1. A tangible, machine-readable media, comprising:

  • code adapted to analyze a data sequence, consisting essentially of a plurality of symbols, based on a grammar comprising at least an initial grammar, wherein the grammar defines at least the plurality of symbols contained within the data sequence;

    code adapted to partition the data sequence into a plurality of sub-sequences and to calculate a statistical heuristic for each sub-sequence of the analyzed data sequence;

    code adapted to compare a selected statistical heuristic of a selected sub-sequence with one or more reference conditions, and to yield a termination result if the selected statistical heuristic is beyond a threshold defined by the one or more reference conditions or a non-termination result otherwise;

    code adapted to update the grammar and the data series with a symbol representing the selected sub-sequence based upon the non-termination result of the comparison; and

    code adapted to identify the selected sub-sequence as a sequence of interest based upon the termination result of the comparison.

View all claims
  • 2 Assignments
Timeline View
Assignment View
    ×
    ×