Efficiency comparison of simulated annealing algorithm and Genetic Algorithm in Feature Selection

As of 1997, when a special issue on relevance including several papers on feature selection was published, few domains explored used more than 40 features.The situation has changed considerably in the past few years and, in this special issue, most papers explore domains with hundreds to tens of thousands of variables or features: new techniques are proposed to address these challenging tasks involving many irrelevant and redundant features. Feature selection has been proved to be important in improving the accuracy of classification. Genetic Algorithm is common used in a variety of situation and has been showed to have good result. But it cannot be ignored that it often takes much time when applying the Genetic Algorithm to the data sets with instances. In this study, we show that Simulated Annealing Algorithm is potentially to be more efficient than Genetic Algorithm in Feature Selection.