Effects of Adaptive Neuro-Fuzzy Logic Systems Trained with Heuristic Methods to Classification Problems

ANFIS is a decision-making mechanism that is a combination of artificial neural networks and fuzzy logic problems. In this rule-based system, inputs are passed through the ANFIS layered structure and an output value is generated. This structure is used in many areas such as classification, estimation, dynamic system identification in recent years. Parameters determined by heuristic methods as usual. Heuristic methods are divided into many categories such as swarm-based, physics-based and chemistry based and aim to find optimum candidate solutions. In this study, we tried to comparatively show that the effect on classification problems of ANFIS trained with Cricket Algorithm with Chaos Map, and The Whale Optimization Algorithm which are heuristic methods. It is seen that the studies performed on the known datasets provide an increase in the accuracy of the trained network. Another important aspect of the work is that the Cricket Algorithm with Chaos Map is being used for the first time in ANFIS training. It is anticipated that this will give the researchers an idea