Applying computational intelligence methods to modeling and predicting common bean germination rates

Abstract

The relationship between seed germination rate and environmental temperature is complex. This study assessed the effectiveness of multi-layer perceptron (MLP) and Particle Swarm Optimization (PSO) techniques in modeling and predicting the germination rate of two common bean cultivars as a function of distinct temperatures. MLP was utilized to model the germination rate of the cultivars and PSO was employed to determine the optimum temperatures at which the beans germinate most rapidly. The outcomes derived from implementing the MLP were compared with those obtained by means of a traditional statistical method. The MLP provided more accurate results than the conventional statistical regression in predicting germination rate values regarding the two common bean cultivars. The optimum germination rate values derived from implementing the PSO model were more accurate than those obtained by using the conventional quadratic regression.