Home > Published Issues > 2020 > Volume 11, No. 3, August 2020 >

Feature Selection Based on Euclid Distance and Neuro-fuzzy System

Seok-Woo Jang 1 and Sang-Hong Lee 2
1. Department of Software, Anyang University, Anyang-si, Republic of Korea
2. Department of Computer Science & Engineering, Anyang University, Anyang-si, Republic of Korea

Abstract—This article suggests the method to distinguish normal persons and a Parkinson’s disease patients by their sole pressure sensor data using NEWFM (Neural Network with Weighted Fuzzy Membership Functions). To make the features to be used as initial input data of NEWFM, the left and right sole pressure sensor data were extracted at the 1st step. In the 2nd step, the frequency scales of the characteristics extracted in the 1st step were divided into individual scales by the FFT (Fast Fourier Transform) using the Hamming method. In the final step, 1 to 15 dimensions were extracted as the characteristics from the values of the individual frequency scales produced in the 2st step by the PCA (Principal Component Analysis). The 75.90% in accuracy performance was acquired from the 8 dimensions with the highest performance, using them as the characteristics. 
 
Index Terms—Parkinson’s disease, gait, FFT, PCA, NEWFM

Cite: Seok-Woo Jang and Sang-Hong Lee, "Feature Selection Based on Euclid Distance and Neuro-fuzzy System," Journal of Advances in Information Technology, Vol. 11, No. 3, pp. 155-160, August 2020. doi: 10.12720/jait.11.3.155-160

Copyright © 2020 by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC BY-NC-ND 4.0), which permits use, distribution and reproduction in any medium, provided that the article is properly cited, the use is non-commercial and no modifications or adaptations are made.