Assessing the relationship between pollution sources and water quality parameters of Sungai Langat Basin using association rule mining

Abstract

Due to water pollution, the water quality in Sungai Langat Basin is declining. This study was conducted in this regard to identify the sources of pollution and to analyse the relationship between sources of pollution and water quality parameters. The Malaysian Department of Environment (DOE) has provided data on pollution sources and water quality. Both data are combined and pre-processed through association rule mining for further modelling processes. Apriori algorithm was used to generate rules in finding any relationships between sources of pollution and water quality parameters. Water quality experts had analysed and validated the generated rules. The analysis of expert acceptance showed that 65% of the formulated rules were agreed by the experts while 35% of the rules were disagreed. Furthermore, the findings showed that all experts agreed that there was a relation between NH3 -N values ranging from 0.9 to 2.7 and the sewage as a source of pollution. Besides, as agreed by all experts, either effluent, sewage or livestock is the pollution source of Sungai Langat.