Simultaneous variable selection and outlier detection using a robust genetic algorithm

Predictors of Posttreatment Drinking Outcomes in Patients with Alcohol Dependence

“…Due to the large number of variables involved (a total of 69), it was necessary to use a mathematical technique able to perform a dimensional reduction of the number of variables that affected each of our output variables. Although the variable selection problem can always be solved by exhaustively examining all possible combinations of variables [34], there is a class of algorithms that attempt to mimic natural selection to arrive at a variable combination producing optimal results for a particular problem. These are known as GA and were introduced by Holland [35] in 1975.…”

Stiffness measurement of eggshell by acoustic resonance and PLS models

“…GA is an adaptive heuristic search algorithm which can be applied for the variable selection (Leardi et al, 2002;Llobet et al, 2004;Carneiro et al, 2007;Wiegand et al, 2009). In GA-PLS algorithm, the number of the genes at each chromosome is equal to the number of the samples (Leardi and Lup, 1998;Nørgaard et al, 2000).…”

Application of genetic algorithm for baseline optimization in standard addition voltammetry

“…Mostly, GAs are applied to select variables for a partial least squares modeling (PLSs) in many analytical methods like infrared spectroscopy [26][27][28][29][30], voltammetry [31][32][33], spectrophotometry [34] and for a bilinear least squares model in chromatography [35]. Also deconvolution of overlapping signals is one of solutions worked out by GAs in voltammetry [36].…”