Genetic programming-based regression for temporal data

Abstract

Various machine learning techniques exist to perform regression on temporal data with concept drift occurring. However, there are numerous nonstationary environments where these techniques may fail to either track or detect the changes. This study develops a genetic programming-based predictive model for temporal data with a numerical target that tracks changes in a dataset due to concept drift. When an environmental change is evident, the proposed algorithm reacts to the change by clustering the data and then induce nonlinear models that describe generated clusters. Nonlinear models become terminal nodes of genetic programming model trees. Experiments were carried out using seven nonstationary datasets and the obtained results suggest that the proposed model yields high adaptation rates and accuracy to several types of concept drifts. Future work will consider strengthening the adaptation to concept drift and the fast implementation of genetic programming on GPUs to provide fast learning for high-speed temporal data.

Implementation of new hybrid evolutionary algorithm with fuzzy logic control approach for optimization problems

“…To determine this problem, look at the performance and quality of algorithms. It is common to refer to genetic algorithms as evolutionary algorithms or evolutionary computation as, evolutionary strategies [1], learning classifier systems [2], evolutionary programming [3], differential evolution [4], genetic programming [5], and estimation of distribution algorithms as evolutionary algorithms or evolutionary computation [6]. These algorithms share the same conceptual framework for simulating individual structure evolution but differ in issue description, selection method, and estimation of distribution algorithms.…”