Title :
Letting ants labeling point features [sic.: for ´labeling´ read ´label´]
Author :
Schreyer, Michael ; Raidl, Günther R.
Author_Institution :
Inst. of Comput. Graphics & Algorithms, Vienna Univ. of Technol., Austria
fDate :
6/24/1905 12:00:00 AM
Abstract :
This paper describes an ant colony system (ACS) for labeling point features. A pre-processing step reduces the search space in a safe way. The ACS applies local improvement and masking, a technique that focuses the optimization on critical regions. Empirical results indicate that the ACS reliably identifies high-quality solutions which are in many cases better than those of a state-of-the-art genetic algorithm for point-feature labeling
Keywords :
artificial life; feature extraction; optimisation; ant colony system; critical regions; genetic algorithm; high-quality solution identification; local improvement; masking; optimization; point-feature labeling; preprocessing step; search space reduction; Ant colony optimization; Cities and towns; Computer graphics; Genetic algorithms; Knowledge based systems; Labeling; NP-hard problem; Space technology; Tagging; Visualization;
Conference_Titel :
Evolutionary Computation, 2002. CEC '02. Proceedings of the 2002 Congress on
Conference_Location :
Honolulu, HI
Print_ISBN :
0-7803-7282-4
DOI :
10.1109/CEC.2002.1004475