Assessment of Water Quality Physicochemical Parameters Using Association Rule Mining for River Narmada

Objective: The objective of this study is to analyze the various water quality parameters of the Narmada River and to find hidden relationships between them so that it can suggest some decision plans or policies to predict or classify the water quality.

Background: Increasing demand for surface water supplies has caused enormous pressure on freshwater ecosystems worldwide to plug the gap in demand for water, many regions of the world have relied heavily on groundwater to meet needs. There have been huge changes in river water quality during the last 10 years. This may be due to the great involvement of human activities and industrial waste.

Methods: In this study, we find an approach to water quality management through Association or correlation studies between various water quality parameters. The Data Mining Technique called Association Rule Mining (Apriori Algorithm) is used to find and extract some rules or relationships between various water quality parameters for the Narmada River at Harda and Hoshangabad districts of Madhya Pradesh.

Findings: The study explores the features of WEKA and its function by importing dataset samples and experimenting with learning datasets. The experiment is limited to a pre-given dataset, the Water Quality Data of Handia (District Harda) and Hoshangabad, M. P., from 1990 to 2010 of River Maa Narmada. Nine parameters were selected for the experiments, categorized into four classes based on pollutant index: A, B, C, and D. The results show that the WEKA Explorer runs an environment for association and parameter setting during experiments. The study found some basic interesting rules, such as the need for regular water treatment before consumption, the need for regular water treatment, and the need for regular water treatment.

The study analysed the relationship between water quality parameters using association rules. It was found that a decrease in NH3-N concentration leads to poor water quality at one level. For a constant pH value, BOD is strongly related to DO. The study concluded that water quality improves when BOD concentration decreases for the same pH value. The results showed that BOD concentration is higher in the Hoshangabad District than in the Harda District, resulting in poor surface water quality at one level. The study also found a relationship between NH3_N and O_PO4, with a decrease in NH3_N causing water quality to decline by one level. The study concluded that industrial centers' inclusions and DO concentrations have a reverse relation, with a decrease in DO increasing water pollution. The study recommends further research on other important water quality parameters to understand their effects on rivers.

Application: This research presents a model with actual data both for spatial and temporal patterns and the benefits of employing data mining techniques towards the improvement of water quality management plans. These results conclude that there is an urgent need for strict regulatory monitoring for water quality maintenance in the river system in Hoshangabad District.