DMCA
Graph-Based Processing of Logic Programs
Cached
Download Links
- [sunburn.informatik.uni-tuebingen.de]
- [www-info1.informatik.uni-wuerzburg.de]
- DBLP
Other Repositories/Bibliography
BibTeX
@MISC{Seipel_graph-basedprocessing,
author = {Dietmar Seipel},
title = {Graph-Based Processing of Logic Programs},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
A lot of practical recursive logic queries to a deductive database are expressed by linear recursive datalog programs with exactly one recursive rule, so-called linear sirups. We give a characterization of k-sided linear datalog sirups -- the concept was introduced by Naughton in [Na 87] -- based on a graph model and present a normal form to which every linear datalog sirup can be transformed. Normalized linear datalog sirups can be decomposed by a counting technique into k lower-dimensional linear sirups which can be evaluated in parallel by very efficient query evaluation algorithms. For certain sets of linear sirups defining the same intensional predicate we can also derive a decomposition approach from the graph model. Another technique based on the prefixes of compiled expansions transforms a set of chain rules with binary predicate symbols, a so-called CFG program, into a transition graph. Common subexpressions between different compiled formulas can be detected by the graph min...
Keyphrases
graph-based processing logic program linear datalog sirup graph model intensional predicate linear recursive datalog program k-sided linear datalog efficient query evaluation algorithm lower-dimensional linear sirup deductive database graph min so-called cfg program common subexpressions compiled expansion decomposition approach binary predicate symbol counting technique so-called linear sirup practical recursive logic query recursive rule certain set chain rule transition graph normal form linear sirup
