No abstract

Classification of imbalanced data has been reported to require modification of standard classification algorithms and lately has attracted a lot of attention due to practical applications in industry, banking and finance. The aim of the paper is to examine algorithms known from literature when two modifications are introduced: MapReduce to parallelize computations and Relief to select most valuable attributes. Both modifications are needed in Big Data area. Also two new algorithms are considered.

The paper investigates a Gene Expression Programming (GEP)-based ensemble classifier constructed using the stacked generalization concept. The classifier has been implemented with a view to enable parallel processing with the use of Spark and SWIM — an open source genetic programming library. The classifier has been validated in computational experiments carried out on benchmark datasets. Also, it has been inbvestigated how the results are influenced by some settings. The paper is an extension of a previous paper of the authors.

No abstract