Sampling Methods in Genetic Programming Learners from Large Datasets: A Comparative Study

in Springer series Advances in Intelligent Systems and Computing, vol. 529International audienceThe amount of available data for data mining and knowledge discovery continue to grow very fast with the era of Big Data. Genetic Programming algorithms (GP), that are efficient machine learning techniques, are face up to a new challenge that is to deal with the mass of the provided data. Active Sampling, already used for Active Learning, might be a good solution to improve the Evolutionary Algorithms (EA) training from very big data sets. This paper present a review of sampling techniques already used with active GP learner and discuss their ability to improve the GP training from very big data sets. A method in each sampling strategy is implemented and applied on the KDD intrusion detection problem using very close parameters. Experimental results show that sampling methods outperforms results obtained with full dataset but some of them cannot be scaled to large datasets