International Journal of Applied Science and Technology

ISSN 2221-0997 (Print), 2221-1004 (Online) 10.30845/ijast

Comparison Between Multiple Linear Regression And Feed forward Back propagation Neural Network Models For Predicting PM10 Concentration Level Based On Gaseous And Meteorological Parameters
Ahmad Zia Ul-Saufie, Ahmad Shukri Yahya, Nor Azam Ramli, Hazrul Abdul Hamid


Air pollution is a major issue that has been affecting human health, agricultural crops, forest and ecosystem. Local environmental or health agencies often need to make daily air pollution forecasts for public advisories and for input into decisions regarding abatement measures and air quality management. Forecasts are usually based on statistical relationships between weather conditions and ambient air pollution concentrations. Multiple linear regression models have been widely used for this purpose, and well-specified regressions can provide reasonable results. The aim of this study is to determine the best technique between Multiple Linear Regression (MLR) and Feedforward Backpropagation Artificial Neural Network (ANN) models for predicting concentration in Pulau Pinang. Multiple regression models and neural networks are examined for Seberang Jaya, Pulau Pinang with the same independent variables, enabling a comparative study of the two approaches. Model comparison statistics using Prediction Accuracy (PA), Coefficient of Determination (R2), Index of Agreement (IA) , Normalised Absolute Error (NAE) and Root Mean Square Error (RMSE) show that ANN is better than MLR.


Full Text: PDF