Linear networks, artificial neural network construction

The method of Lydersen [28] is a GCM of this type to estimate the critical temperature, Tc. Other approaches to non-linear GCMs include the model of Lai et al. [29] for the boiling point, Tby and the ABC approach [30] to estimate a variety of thermodynamic properties. Further, artificial neural networks have been used to construct nonlinear models for the estimation of the normal boiling point of haloalkanes [31] and the boiling point, critical point, and acentric factor of diverse fluids [32]. 

---