Noise-induced hearing loss prediction in Malaysian industrial workers using gradient descent with adaptive momentum algorithm

Mohd Nawi, Nazri and Rehman, M. Z. and Ghazali, Mohd Imran (2011) Noise-induced hearing loss prediction in Malaysian industrial workers using gradient descent with adaptive momentum algorithm. International Review on Computers and Software , 6 (5). pp. 740-749.

Full text not available from this repository.

Abstract

In the past 30 years Malaysian Industry has progressed a lot and emerged as a key player in maintaining the economical stability. Despite providing efficient solutions for Malaysian economical progress most of the major industries face a nemesis in the form of Noise. The high frequency noise emitted from the machines is causing adverse side-effects to the health of the blue-collared employees. One of the common injury caused due to exposure to high frequency noise is known as Noise-Induced Hearing Loss (NIHL). Several studies have been conducted to study the major factors involved in causing NIHL in humans using Artificial Neural Networks. But these studies have neglected some important factors that play a major role in causing hearing loss. Therefore, this study is carried-out to predict NIHL in workers using age, work-duration, and noise exposure as the main factors involved in the hearing loss of a human worker. Gradient Descent with Adaptive Momentum (GDAM) algorithm is proposed to predict the hearing loss on both ears. GDAM shows improved prediction results for both ears. The Mean Square Error (MSE) calculated for Left Hearing Loss (LHL) and Right Hearing Loss (RHL) was found to be 2.18x10-3 and 2.30x10-3 respectively. The overall accuracy achieved for NIHL prediction in human workers was close to 99.01 percent for left and right ears respectively.

Item Type:Article
Subjects:T Technology > TD Environmental technology. Sanitary engineering > TD891-894 Noise and its control
Divisions:Faculty of Computer Science and Information Technology > Department of Software Engineering
ID Code:3041
Deposited By:Normajihan Abd. Rahman
Deposited On:21 Feb 2013 14:24
Last Modified:22 Jan 2015 08:38

Repository Staff Only: item control page