Fuzzy based genetic neural networks for the classification of murder cases using Trapezoidal and Lagrange Interpolation Membership Functions

This paper describes the construction of a decision system to be used by judges who is about to pass sentence in murder cases. Classification models of murder cases based on fuzzy neural network with random weights and fuzzy neural network with Genetic Algorithm based weights are designed. A simulation program in C++ has been deliberated and developed for analyzing the consequences. Results show that the fuzzy neural networks increase the rate of convergence in comparison with conventional neural networks with backpropagation algorithm. That the fuzzy neural networks for classification of murder cases using Trapezoidal Membership Function outperform Lagrange Interpolation and Gaussian Membership Function is also reported. Comparative studies are carried out for a number of networks and configurations.

Graphical abstractFigure optionsDownload full-size imageDownload as PowerPoint slideHighlights► Lagrange Interpolation and Trapezoidal Membership Functions. ► Fuzzy neural network. ► Criminal law ► Genetic Algorithm.