Flower Pollination Inspired Algorithm on Exchange Rates Prediction Case

Abstract

Flower pollination algorithm is a bio-inspired system that adapts a similar process to genetic algorithm, that aims for optimization problems. In this research, we examine the utilization of the flower pollination algorithm in linear regression for currency exchange cases. The solutions are represented as a set that contains regression coefficients. Population size for the candidate solutions and the switch probability between global pollination and local pollination have been experimented with in this research. Our result shows that the final solution is better when a higher size population and higher switch probability are employed. Furthermore, our result shows the higher size of the population leads to considerable running time, where the leaning probability of global pollination slightly increases the running time.