A flexible nonlinear approach to represent cause-effect relationships in FCMs

► A flexible non-linear function is proposed to represent the influence between FCM's concepts and enrich the behavior of the system. ► Various influence interrelations between concepts are captured due to the adjustable character of the generalized logistic function. ► Higher accuracy and higher convergence rates are achieved through the proposed model when combined with various learning methods. ► Experimental comparisons, based on known as well as random problems, are presented.