id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_yeyqio5mnjehhhco2g5avg33ii	A. Echeverría	Evolving linear transformations with a rotation-angles/scaling representation	2012	7	.pdf	application/pdf	6710	719	71	the Covariance Matrix Adaptation Evolution Strategy (CMA-ES) is applied to the optimization of linear in the second part of the paper to represent linear transformations by means of rotation angles and scaling factors, based on the Singular Value Decomposition theorem (SVD). In the second method, transformations are coded as rotation matrices and Second method: rotation angles and scaling matrix With regard to the remaining domains, CMA-ES diagonal obtains significatively better results than KNN on the original data This table adds to Table 3 a new column for our classification rate (percentage) with KNN (k = 1) for the transformed data when scaling factors and rotation angles are directly the results as expected: KNN obtains a very bad classification accuracy on the original data (9.2%) and when a diagonal matrix is used domains, both synthetic and real, and the results show that, in general, both diagonal and square matrices found by CMA-ES either	./cache/work_yeyqio5mnjehhhco2g5avg33ii.pdf	./txt/work_yeyqio5mnjehhhco2g5avg33ii.txt