Further exploration of the hybrid Fuzzy-Rough Dendritic Cell immune classifier

The performance of the Fuzzy-Rough Dendritic Cell immune Algorithm relies on its data pre-processing phase which is based on the use of Rough Set Theory (RST). However, the developed approach presents an information loss as data should be discretized beforehand. Therefore, in this paper, we propose a new data pre-processing phase for the Fuzzy-Rough hybrid approach based on Fuzzy Rough Set Theory (FRST) to allow dealing with real-valued data with no data quantization beforehand. Results show that applying FRST, instead of RST, is more convenient for the Fuzzy-Rough Dendritic Cell Algorithm data pre-processing phase yielding much better performance in terms of accuracy.