MVGPC+: An Integrated Approach to Machine-Learning Classification

C. Aranha and H. Iba (Japan)

Keywords

Genetic Programming, Classification, Bio-informatics,Financial data, SVM, k-NN, and GP.

Abstract

We propose a new evolutionary framework called MGVPC+ for the sake of solving classification problems. This approach is an extension of our earlier framework MVGPC, in which a majority voting genetic programming classifier was established with multiple GP runs. In this paper, we integrate GP-based classifiers and other machine learning techniques, e.g., SVM and k-NN methods. The effectiveness of our approach has been confirmed by the experimental results with several classifier task in terms of improved performance and reduced computational burden.

Important Links:



Go Back