TY - GEN
T1 - A data mining based genetic algorithm
AU - Wu, Yi Ta
AU - An, Yoo Jung
AU - Geller, James
AU - Wu, Yih Tyng
PY - 2006
Y1 - 2006
N2 - Genetic algorithms (GAs) are considered as a global search approach for optimization problems. Through the proper evaluation strategy, the best "chromosome" can be found from the numerous genetic combinations. Although the GA operations do provide the opportunity to find the optimum solution, they may fail in some cases, especially when the length of a chromosome is very long. In this paper, a data mining-based GA is presented to efficiently improve the Traditional GA (TGA). By analyzing support and confidence parameters, the important genes, called DNA, can be obtained. By adopting DNA extraction, it is possible that TGA will avoid stranding on a local optimum solution. Furthermore, the new GA operation, DNA implantation, was developed for providing potentially high quality genetic combinations to improve the performance of TGA. Experimental results in the area of digital watermarking show that our data mining-Jbased GA successfully reduces the number of evolutionary iterations needed to find a solution.
AB - Genetic algorithms (GAs) are considered as a global search approach for optimization problems. Through the proper evaluation strategy, the best "chromosome" can be found from the numerous genetic combinations. Although the GA operations do provide the opportunity to find the optimum solution, they may fail in some cases, especially when the length of a chromosome is very long. In this paper, a data mining-based GA is presented to efficiently improve the Traditional GA (TGA). By analyzing support and confidence parameters, the important genes, called DNA, can be obtained. By adopting DNA extraction, it is possible that TGA will avoid stranding on a local optimum solution. Furthermore, the new GA operation, DNA implantation, was developed for providing potentially high quality genetic combinations to improve the performance of TGA. Experimental results in the area of digital watermarking show that our data mining-Jbased GA successfully reduces the number of evolutionary iterations needed to find a solution.
KW - Data mining
KW - Digital watermarking
KW - Evolutionary algorithm
KW - Genetic algorithm
UR - http://www.scopus.com/inward/record.url?scp=33750905521&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=33750905521&partnerID=8YFLogxK
U2 - 10.1109/SEUS-WCCIA.2006.2
DO - 10.1109/SEUS-WCCIA.2006.2
M3 - Conference contribution
AN - SCOPUS:33750905521
SN - 0769525601
SN - 9780769525600
T3 - Proc. - The Fourth IEEE Workshop on Software Technol. for Future Embedded and Ubiquitous Syst., SEUS 2006 andthe Second Int. Workshop on Collaborative Comput., Integr., and Assur., WCCIA 2006
SP - 55
EP - 60
BT - Proc. - The Fourth IEEE Workshop on Software Technol. for Future Embedded and Ubiquitous Systems, SEUS 2006 andthe Second Int. Workshop on Collaborative Computing, Integr., and Assurance, WCCIA 2006
T2 - 4th IEEE Workshop on Software Technologies for Future Embedded and Ubiquitous Systems, SEUS 2006 andthe 2nd International Workshop on Collaborative Computing, Integration, and Assurance, WCCIA 2006
Y2 - 27 April 2006 through 28 April 2006
ER -