Conceptual Information Compression and Efficient Pattern Search

Angelova, Galia, and Mihov, Stoyan. "Conceptual Information Compression and Efficient Pattern Search." Serdica Journal of Computing 2.4 (2008): 369-402. <http://eudml.org/doc/11473>.

@article{Angelova2008,
abstract = {This paper introduces an encoding of knowledge representation statements as regular languages and proposes a two-phase approach to processing of explicitly declared conceptual information. The idea is presented for the simple conceptual graphs where conceptual pattern search is implemented by the so called projection operation. Projection calculations are organised into off-line preprocessing and run-time computations. This enables fast run-time treatment of NP-complete problems, given that the intermediate results of the off-line phase are kept in suitable data structures. The experiments with randomly-generated, middle-size knowledge bases support the claim that the suggested approach radically improves the run-time conceptual pattern search.},
author = {Angelova, Galia, Mihov, Stoyan},
journal = {Serdica Journal of Computing},
keywords = {Data Compaction and Compression; Finite Automata; Applications; Efficiency; Semantic Networks; knowledge representation},
language = {eng},
number = {4},
pages = {369-402},
publisher = {Institute of Mathematics and Informatics Bulgarian Academy of Sciences},
title = {Conceptual Information Compression and Efficient Pattern Search},
url = {http://eudml.org/doc/11473},
volume = {2},
year = {2008},
}

TY - JOUR
AU - Angelova, Galia
AU - Mihov, Stoyan
TI - Conceptual Information Compression and Efficient Pattern Search
JO - Serdica Journal of Computing
PY - 2008
PB - Institute of Mathematics and Informatics Bulgarian Academy of Sciences
VL - 2
IS - 4
SP - 369
EP - 402
AB - This paper introduces an encoding of knowledge representation statements as regular languages and proposes a two-phase approach to processing of explicitly declared conceptual information. The idea is presented for the simple conceptual graphs where conceptual pattern search is implemented by the so called projection operation. Projection calculations are organised into off-line preprocessing and run-time computations. This enables fast run-time treatment of NP-complete problems, given that the intermediate results of the off-line phase are kept in suitable data structures. The experiments with randomly-generated, middle-size knowledge bases support the claim that the suggested approach radically improves the run-time conceptual pattern search.
LA - eng
KW - Data Compaction and Compression; Finite Automata; Applications; Efficiency; Semantic Networks; knowledge representation
UR - http://eudml.org/doc/11473
ER -