Institute of Electrical & Electronic Engineers
Now-a-days data visualization is one of the important fields in knowledge discovery. Human beings can perceive data up to three dimensions. In this paper the authors propose new approach in data set dimensionality reduction. They use classical principal component analysis transformation. Instead of rejecting features they generate new one by using nonlinear feature transformation. The values of transformation weights are changed evolutionary by using genetic algorithms. Results show better classification rates in smaller feature space. Visualization results also look better.