Evolutionary Feature Selection for Big Data Classification: A MapReduce Approach

Table 2

Parameter specification for all the methods involved in the experimentation.

Algorithm	Parameters
CHC	value for -NN: 1
	Trade-off between reduction and accuracy: 0.5
	Proportion of flipped bits in restarting process: 0.05
	Population size: 40
	Number of evaluations: 1000
Naive Bayes	Lambda: 1.0 (default)
Logistic Regression	Iterations: 100 (default)
	StepSize: 1.0 (default)
	miniBatchFraction: 1.0
SVM	Regularization parameter: 0.0; 0.5
	Iterations: 100 (default)
	StepSize: 1.0 (default)
	miniBatchFraction: 1.0