Hybrid classifier based on particle swarm optimization trained auto associative neural networks as non-linear principal component analyzer: Application to banking

摘要

This paper proposes a hybrid classifier consisting of two phases which work in tandem. In the first phase, particle swarm optimization trained auto associative neural network (PSOAANN) is executed in which weights of three layered of AANN are updated using particle swarm optimization (PSO). In this phase, dimensionality reduction takes place by treating the hidden nodes which should be less than the input nodes. The nonlinear principal components (NLPC) are drawn from hidden nodes as NLPCs. They are fed to the second phase where threshold accepting logistic regression (TALR) works as a classifier. The efficiency of the hybrid is analyzed on five banking datasets namely Spanish banks, Turkish banks, US banks and UK banks and UK credit dataset. All the datasets are analyzed using 10 fold cross validation (10 FCV). It turns out that the proposed hybrid yielded higher accuracies.