×

Efficient data infrastructure for high dimensional data analysis

  • US 7,870,114 B2
  • Filed: 06/15/2007
  • Issued: 01/11/2011
  • Est. Priority Date: 06/15/2007
  • Status: Active Grant
First Claim
Patent Images

1. A method for generating an inverted index for retrieval of data from a data structure, the method comprising:

  • generating a file model structure comprising a raw value data file, a mapped value file, and a dimension table, the generating comprising;

    populating the dimension table with mapped values and ranges of raw values of a dimension of a data structure, the ranges of raw values mapped to corresponding mapped values; and

    for raw values within the dimension of the data structure;

    populating the raw value data file with a raw value of the dimension; and

    populating the mapped value file with a mapped value corresponding to the raw value within the raw value data file based upon a mapping within the dimension table; and

    generating an inverted index comprising a recordID file and a Count-Offset file, the generating comprising;

    for mapped values within the dimension table;

    populating the recordID file with one or more record identifiers associated with a raw value mapped to a mapped value by falling within a range of raw values within the dimension table; and

    populating the Count-Offset file with a count value and an offset value of the mapped value based upon the populating of the recordID file.

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