Variations in the Number of Layers and the Number of Neurons in Artificial Neural Networks: Case Study of Pattern Recognition

F. Arifin; H. Robbani; T. Annisa; N.N.M.I. Ma'Arof

doi:10.1088/1742-6596/1413/1/012016

← Back to search

Variations in the Number of Layers and the Number of Neurons in Artificial Neural Networks: Case Study of Pattern Recognition

Open

F. Arifin, H. Robbani, T. Annisa, N.N.M.I. Ma'Arof

2019 Journal of Physics: Conference Series Vol. 1413 Issue 1 Conference paper Cited by 31 SDG 17SDG 9SDG 16

Abstract

This paper presents the analyst the number of layers and the number of neurons in the hidden layer of the Artificial Neural Network. In this study, case studies were taken in the recognition of alphabet patterns and shape patterns. First, the number of layers is varied to get the best number of layers. Furthermore, the number of neurons is varied to get the best number of neurons. The results showed that the best number of layers was 1-5 layers in the hidden layer, with validation values from the recognition system 96-100%. While the best number of neurons is obtained with 19 neurons, with an average accuracy percentage of 81%. © Published under licence by IOP Publishing Ltd.

Affiliations

Department of Electronic Engineering, Engineering Faculty, Universitas Negeri Yogyakarta, Yogyakarta, Indonesia; Department of Engineering Technology, Faculty of Technical and Vocational, Universiti Pendidikan Sultan Idris, Tanjong Malim Perak, Malaysia

View full text (DOI) →

Research at a Glance

Research at a Glance

Topics & SDG Alignment

Collaboration

Author Profile (Selected)

References Overview

Journal & Source

Metadata & Integrity