Open AccessOptimizing Adaptive Neuro Fuzzy Inference System (ANFIS) parameters using Cuckoo Search (Case study of world crude oil price estimation)
Author(s) -
Mohamad Yusak Anshori,
Dinita Rahmalia,
Teguh Herlambang,
Denis Fidita Karya
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1836/1/012041
Subject(s) - adaptive neuro fuzzy inference system , cuckoo , cuckoo search , gradient descent , neuro fuzzy , computer science , inference system , artificial intelligence , artificial neural network , machine learning , fuzzy logic , fuzzy control system , ecology , biology , particle swarm optimization
There are some methods that have found for estimating data and one of them is Adaptive Neuro Fuzzy Inference System (ANFIS). In estimation using ANFIS, there are some initial parameters such as premise parameters (nonlinear) and consequent parameters (linear) which should be fixed to be trained forward and backward by gradient descent. In this research with case study of world crude oil price estimation, initial ANFIS parameters will be optimized by Cuckoo Search method. Cuckoo Search uses reproduction strategy i.e. laying their eggs in the other bird’s nest. When the eggs are hatched, their chicks are fed by other birds. In Cuckoo Search method, initial ANFIS parameters is represented as bird nest position. Based on simulation, Cuckoo Search method can optimize initial ANFIS parameters giving the best estimation both of training data and testing data in world crude oil price estimation.