Linear Kernel Hopfield Neural Network Approach in Horn Clause Programming.

A linear Kernel is a machine computationally efficient, also it has the ability to work by the analysis-data in the high-dimensional feature space with complex structure random. The purpose for the study in this paper is to introduce a new vision into the linear kernel machines by integrated with Hopfield Neural Network for doing the logical programming by using horn clause logic to become new network called Linear Kernel Hopfield Neural Network (LKHNN). The benefit by integrating kernel machine with Hopfield network to reduce the arithmetic burden by intelligently determining the pattern of memory embedded in high-dimensional space feature. The new network (LKHNN) able to formulation estimates for the state of neurons. The LKHNN and HNN simulation are performed using Dev C ++ program. The outcomes from simulator show the effectiveness of LKHNN in doing logic program by optimization horn clauses problem and improve efficiency to find the global solution. The robustness of LKHNN and HNN in doing logic programming by using horn clause are evaluated based on global minima ratio(zM), Hamming distance(HD) and computational time(CT). [ABSTRACT FROM AUTHOR]