Patent №
US 7,849,037
Granted
—
Owner
—
Lab
—
AI components
2
kr · hardware
Assignment
None on record
Dataset
AIPD
2023_r1 edition
Application
11869699
A method for finding sequences of similar data (SDDs), which are similar to a target sequence of digital data, is invented. The method leverages a new category of signatures, called equivalence signatures, to characterize the SDDs. These signatures have the salient feature that, at worst, they change in a bounded manner when changes are made to the sequence of digital data and when used to find SDDs that are similar to a target SDD, they allow for a significant reduction in the number of SDDs to be compared with the target. This is an improvement over the state of the art wherein the cryptographic message digests used as signatures respond unpredictably to changes in the sequence of digital data and the comparison of a target SDD to a corpus of SDDs requires the computational expensive process of applying a complete search against the entire corpus.