Selection of Most Relevant Features from High Dimensional Data using IG-GA Hybrid Approach?

Feature selection is considered a problem of global combinatorial optimization in machine learning, which reduces the number of features, removes irrelevant, noisy and redundant data, and results in acceptable classification accuracy. In the past few decades, researchers have developed large amount of feature selection algorithms. These algorithms are designed to serve different purposes, are of different models, and all have their own advantages and disadvantages. Although there have been intensive efforts on surveying existing feature selection algorithms, to the best of our knowledge, there is still not a dedicated repository that collects the representative feature selection algorithms to facilitate their comparison and joint study. To fill this gap, in this work, an IG-GA hybrid approach with MRMR evaluation function is presented for high dimensional data set.