ENTROPIC FEATURE FOR SEQUENCE PATTERN THROUGH ITERATED FUNCTION SYSTEMS

Authors
ROMANROLDAN R BERNAOLAGALVAN P OLIVER JL
Citation
R. Romanroldan et al., ENTROPIC FEATURE FOR SEQUENCE PATTERN THROUGH ITERATED FUNCTION SYSTEMS, Pattern recognition letters, 15(6), 1994, pp. 567-573
Citations number
9
Categorie Soggetti
Computer Sciences, Special Topics","Computer Science Artificial Intelligence
Journal title
Pattern recognition letters → ACNP
ISSN journal
01678655
Volume
15
Issue
6
Year of publication
1994
Pages
567 - 573
Database
ISI
SICI code
0167-8655(1994)15:6<567:EFFSPT>2.0.ZU;2-T
Abstract
Entropy and relative entropy am proposed as features extracted from sy mbol sequences. Firstly, a proper Iterated Function System is driven b y the sequence, Producing a fractal-like representation (CSR) with a l ow computational cost. Then, two entropic measures are applied to the CSR history of the CSR and theoretically justified. Examples are inclu ded.