• May 03, 2016 News! IJIET Vol. 5, No. 10 has been indexed by EI (Inspec).   [Click]
  • Mar 13, 2017 News!Vol. 7, No. 5 has been indexed by Crossref.
  • Mar 10, 2017 News!Vol. 7, No. 5 issue has been published online!   [Click]
General Information
    • ISSN: 2010-3689
    • Frequency: Bimonthly (2011-2014); Monthly (Since 2015)
    • DOI: 10.18178/IJIET
    • Editor-in-Chief: Prof. Dr. Steve Thatcher
    • Executive Editor: Ms. Nancy Y. Liu
    • Abstracting/ Indexing: EI (INSPEC, IET), Electronic Journals Library, Google Scholar, Crossref and ProQuest
    • E-mail: ijiet@ejournal.net
Editor-in-chief
Prof. Dr. Steve Thatcher
University of South Australia, Australia
It is my honor to be the editor-in-chief of IJIET. The journal publishes good papers which focous on the advanced researches in the field of information and education technology. Hopefully, IJIET will become a recognized journal among the scholars in the filed of information and education technology.
IJIET 2012 Vol.2(3): 259-261 ISSN: 2010-3689
DOI: 10.7763/IJIET.2012.V2.124

Prediction of Chemical Oxygen Demand In Dondang River Using Artificial Neural Network

Anita Talib and Mawar Idati Amat

Abstract—The increase of human population every year results in increasing of water usage and demand. This phenomenon affects the overall water quality and change the flow of river ecosystems. Various residential and commercial building were built along Dondang river basin in Pulau Pinang. Irresponsible residents and factory manager treat the river as sites for illegal garbage dumping and waste disposal. As an effort to monitor river water quality in Malaysia, Alam Sekitar Malaysia (ASMA) collaborated with Department of Environment (DOE) Malaysia to measure water quality parameters every month. The data used for Dondang River was dated from January 1998 until December 2007 at three stations at different sites. Different variables in this model include Dissolve Oxygen (DO), Biological Oxygen Demand (BOD), Suspended Solid (SS), pH, ammonia (NH3), temperature, nitrate (NO3), total solid (TS) and phosphate (PO4).were used as inputs while the prediction is made for Chemical Oxygen Demand (COD). The number of hidden layer, processing elements, the value of learning rule and transfer were adjusted to achieve the minimum error and highest prediction accuracy. The prediction of COD were divided into training, cross validation and testing data. The best possible model estimation was choose from the model with higher generalization, less minimum square error and high r squared value. The model estimation then tested with Mean Sensitivity test to check the validity of the model and to determine the factors that affect the value of COD the most.  Sensitivity analysis showed that BOD is the most important variable that determine COD, followed by phosphate concentration, DO, susupended solids and temperature. The results are consistent with the effects of land use on the river water quality as Dondang River flows through mainly housing area and factories or industrial area.

Index Terms—Chemical oxygen demand, artificial neural networks, mean sensitivity test

A. Talib is with School of Distance Education, Universiti Sains Malaysia, 11800 Pulau Pinang, Malaysia (e-mail: anita@usm.my)
M. I. Amat is with Centre For Marine And Coastal Studies, Universiti Sains Malaysia, 11800 Pulau Pinang, Malaysia

[PDF]

Cite: Anita Talib and Mawar Idati Amat, "Prediction of Chemical Oxygen Demand In Dondang River Using Artificial Neural Network," International Journal of Information and Education Technology vol. 2, no. 3, pp. 259-261, 2012.

Copyright © 2008-2016. International Journal of Information and Education Technology. All rights reserved.
E-mail: ijiet@ejournal.net