A generalized-constraint neural network model: Associating partially known relationships for nonlinear regressions☆

A theoretical and empirical study on unbiased boundary-extended crossover for real-valued representation

“…So we can conclude that the proposed crossover is a good choice. To demonstrate how the proposed methodologies work on real-world applications, we conducted a nonlinear regression [32] on a real-world dataset. We used the forest fires dataset first introduced in [59] and provided by the UCI repository of machine learning databases [12], which is widely used in data-mining research [9,52,63,85].…”

A knowledge-and-data-driven modeling approach for simulating plant growth: A case study on tomato growth

“…In general, there is no generic approach to designing the coupling connections. The actual configuration of the coupling is more problem dependent and can be quite complicated because it greatly depends on the form in which domain knowledge is available; domain knowledge may be presented in the form of constraint functions [ Table 1 in Hu et al (2009)], grammar (Todorovski and Džeroski, 2006), rules (Matsunaga et al, 2013), and even physically based models (Czop et al, 2011). Considering the wide variety of coupling connections in the KDDM approach, we limited our coupling connections to two simple and common coupling operators, namely, "superposition" and "composition" (Thompson and Kramer, 1994;Hu et al, 2009).…”

“…To take advantage of both the KDM and DDM approaches, studies on integrating these two types of modeling approaches have been conducted (Džeroski and Todorovski, 2003;Hu et al, 2009;Qu and Hu, 2011;Czop et al, 2011;Ran and Hu, 2014). Investigations on the successful application of this integrated approach especially deserve greater attention in the ecological sciences (Todorovski and Džeroski, 2006;Atanasova et al, 2008;Qu and Hu, 2009;Matsunaga et al, 2013).…”

Nonlinear System Identification Using Quasi-ARX RBFN Models with a Parameter-Classified Scheme

“…In contrast, a prior knowledge based nonlinear parameter estimation makes sense to interpret system properties meaningfully, especially with respect to the Complexity 3 quasi-ARX RBFN model as discussed later in Section 3. The useful prior knowledge can evolve a quasi-ARX model from a "black-box" tool into a "semianalytical" one [27], which makes some parameters interpretable by our intuition, just following the principle of application favorable in quasi-ARX modeling. Owing to this fact, nonlinear parameters are determined in terms of prior interpretable knowledge, and linear parameters are adjusted to fit the data.…”