Wan Ishak, Wan Hussain and Siraj, Fadzilah and Othman, Abu Talib (2001) Multi-backpropagation network in medical diagnosis. In: Prosiding Seminar Kepintaran Buatan (AIS 2001), 1-3 November 2001, Universiti Utara Malaysia. (Unpublished)

PDF Restricted to Registered users only Download (177kB)

Abstract

Backpropagation (or backprop) algorithm is one of the well-known algorithms in neural networks. It is capable to deal with various types of data and also nble to model a complex decision system. Some problem domains involve a large amount of data. The bigger the number of input or hidden units is, the more complex the model would be. Hence, reducing the network complexity would be an advantage to the network. This paper proposes a multibackpropagtion network that reduces the size of a large backpropagation network. The network is divided into several smaller networks, which act as a specialized network. As a result, the raining time would be reduced.

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	Neural Network, Multi Network,Backpropagation, Medical Diagnosis
Subjects:	Q Science > QA Mathematics > QA76 Computer software
Divisions:	College of Arts and Sciences
Depositing User:	Prof Madya Fadzilah Siraj
Date Deposited:	15 Aug 2011 04:03
Last Modified:	15 Aug 2011 04:03
URI:	https://repo.uum.edu.my/id/eprint/3436

Actions (login required)

View Item