DocumentCode :
2188968
Title :
A Similarity Measure Using Smallest Context-Free Grammars
Author :
Cerra, Daniele ; Datcu, Mihai
Author_Institution :
Remote Sensing Technol. Inst., German Aerosp. Center (DLR), Wessling, Germany
fYear :
2010
fDate :
24-26 March 2010
Firstpage :
346
Lastpage :
355
Abstract :
This work presents a new approximation for the Kolmogorov complexity of strings based on compression with smallest Context Free Grammars (CFG). If, for a given string, a dictionary containing its relevant patterns may be regarded as a model, a Context-Free Grammar may represent a generative model, with all of its rules (and as a consequence its own size) being meaningful. Thus, we define a new complexity approximation which takes into account the size of the string model, in a representation similar to the Minimum Description Length. These considerations result in the definition of a new compression-based similarity measure: its novelty lies in the fact that the impact of complexity overestimations, due to the limits that a real compressor has, can be accounted for and decreased.
Keywords :
computational complexity; context-free grammars; string matching; CFG; Kolmogorov complexity; complexity approximation; complexity overestimations; compression-based similarity measure; context-free grammars; dictionary; generative model; minimum description length; string model; Clustering algorithms; Context modeling; Data compression; Data mining; Data structures; Dictionaries; Entropy; Image coding; Information theory; Remote sensing; Compression-based Similarity Measure; Context Free Grammars; Kolmogorov Complexity;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Data Compression Conference (DCC), 2010
Conference_Location :
Snowbird, UT
ISSN :
1068-0314
Print_ISBN :
978-1-4244-6425-8
Electronic_ISBN :
1068-0314
Type :
conf
DOI :
10.1109/DCC.2010.37
Filename :
5453478
Link To Document :
بازگشت