Multiclass Classification of Microarray Data Samples with Flexible Neural Tree

A reliable and precise classification of tumors is essential for successful diagnosis and treatment of cancer. But microarray data are often extremely asymmetric in dimensionality, such as thousands or even tens of thousands of genes and a few hundreds of samples. This paper proposes the multiclass Flexible Neural Tree (FNT) algorithm for cancer classification. Based on the pre-defined instruction/operator sets, a flexible neural tree model can be created and evolved. The FNT structure is developed using probabilistic incremental program evolution (PIPE) and the free parameters embedded in the neural tree are optimized by particle swarm optimization (PSO) algorithm. Empirical results on two well-known cancer datasets show competitive results with other methods.