TY - GEN
T1 - Density-based evolutionary outlier detection
AU - Banerjee, Amit
PY - 2012/8/20
Y1 - 2012/8/20
N2 - A novel density-based distance measure and an outlier detection method using evolutionary search are presented in this paper. A fitness function based on nearest neighbor distances is proposed and the genetic recombination operators are designed to achieve a balance of exploration and exploitation in the nearest neighborhood space. The methodology is tested on datasets of varying sizes (small to moderate) and dimensionalities and performance is compared to existing evolutionary methods for outlier detection. Copyright is held by the author/owner(s).
AB - A novel density-based distance measure and an outlier detection method using evolutionary search are presented in this paper. A fitness function based on nearest neighbor distances is proposed and the genetic recombination operators are designed to achieve a balance of exploration and exploitation in the nearest neighborhood space. The methodology is tested on datasets of varying sizes (small to moderate) and dimensionalities and performance is compared to existing evolutionary methods for outlier detection. Copyright is held by the author/owner(s).
UR - http://www.scopus.com/inward/record.url?scp=84865044892&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84865044892&partnerID=8YFLogxK
U2 - 10.1145/2330784.2330904
DO - 10.1145/2330784.2330904
M3 - Conference contribution
AN - SCOPUS:84865044892
SN - 9781450311786
T3 - GECCO'12 - Proceedings of the 14th International Conference on Genetic and Evolutionary Computation Companion
SP - 651
EP - 652
BT - GECCO'12 - Proceedings of the 14th International Conference on Genetic and Evolutionary Computation Companion
T2 - 14th International Conference on Genetic and Evolutionary Computation, GECCO'12
Y2 - 7 July 2012 through 11 July 2012
ER -