S-type random 2 satisfiability logic integrated in discrete Hopfield neural network

Random 2 Satisfiability (RAN2SAT) has overcome a lack of interpretation by introducing a flexible structure and generating a variety of solutions that converge to global minimum solutions. It is one of the most important non-systematic logics in the field of artificial neural network (ANN) research, discussing the problem of knowledge in logical rules. Defining The structure in RAN2SAT occurs at random, Also, the possibility of receiving literal (negative, positive) is distributed by uniform distribution without taking into account the distribution for the real-world dataset. A non-systematic SAT with the ability to best extract information from the probability distribution function that represents the real datasets is needed to find. The research aims to find the best RAN2SAT structure representing the real dataset in a Discrete Hopfield Neural Network (DHNN) by integrating the Binomial distribution function for real datasets and RAN2SAT in (DHNN). The new non-systematic logic is motivated by a probabilistic approach that studies the distribution probability for each variable. The predetermined relation between the real dataset before implementing it in DHNN and then determining the best structure for SRAN2SAT logic aid in exploring an optimal solution space. Hence, the proposed network will be more capable of capturing information from real SRAN2SAT datasets. Therefore, the proposed model will be more useful in real-world applications since SRAN2SAT consists of flexible combinations representing data effectively. The simulation's outcome shows successful integrated SRANSAT in HNN with minimal RMES energy.