A new method to enhance the computational efficiency of data mining classification modelling techniques by introducing gain parameter
Noise is a form of a pollutant that is terrorizing the occupational health experts for many decades due to its adverse side-effects on the workers in the industry. Noise- Induced Hearing Loss (NIHL) handicap is one out of many health hazards caused due to excessive exposure to high frequency noise e...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Published: |
2011
|
| Subjects: | |
| Online Access: | http://eprints.uthm.edu.my/2876/ http://eprints.uthm.edu.my/2876/1/FRGS_0737.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Noise is a form of a pollutant that is terrorizing the occupational health experts for
many decades due to its adverse side-effects on the workers in the industry. Noise-
Induced Hearing Loss (NIHL) handicap is one out of many health hazards caused
due to excessive exposure to high frequency noise emitted from the machines. A
number of studies have been carried-out to find the significant factors involved in
causing NIHL in industrial workers using Artificial Neural Networks (ANN). Despite
providing useful information on hearing loss, these studies have neglected some
important factors. The traditional Back-propagation Neural Network (BPNN) is a
supervised Artificial Neural Networks (ANN) algorithm and widely used in solving
many real time problems in world. But BPNN possesses a problem of slow
convergence and network stagnancy. Previously, several modifications were
suggested to improve the convergence rate of Gradient Descent Back-propagation
algorithm such as careful selection of initial weights and biases, learning rate,
momentum, network topology, activation function and 'gain' value in the activation
function. This research proposed an algorithm known as GDAM which improving
the current working performance of Back-propagation algorithm by adaptively
changing the momentum value and at the same time keeping the 'gain' parameter
fixed for all nodes in the neural network. The performance of the proposed GDAM is
compared with 'Gradient Descent Method with Adaptive Gain (GDM-AG) (Nazri,
2007)' and 'Gradient Descent with Simple Momentum (GDM)' by performing
simulations on classification problems. The results show that GDAM is a better
approach than previous methods with an accuracy ratio of 1.0 for classification
problems. The efficiency of the proposed GDAM is further verified by means of
simulations on Noise-Induced Hearing loss WHL) audiometric data obtained from
Tenaga Nasional Berhad (TNB). The proposed GDAM shows improved prediction
results on both ears and will be helpful in improving the declining health condition of
industrial workers in Malaysia. At present, only few studies have emerged to predict NIHL using ANN but have failed to achieve high accuracy. The achievements made
by GDAM has paved way for indicating NIHL in workers before it becomes severe
and cripples him or her for life. GDAM is also helpful in educating the blue collared
employees to avoid noisy environments and remedies against exposure to excessive
noise can be taken in the future to prevent hearing damage. |
|---|