In the context of water quality analysis, input data would consist of water quality parameters collected over time from different monitoring stations or sampling points of a water resource. The Association Rule Extraction System is developed to analyze patterns and relationships in water parameters based on various metrics like biological, chemical, and physical characteristics that can be further used in feature selection for deep learning technologies. Three association mining techniques, Apriori, FP-growth, and Improved FP-growth using adjacency list, are used to extract the set of rules with strong relationships among the water parameters. We used this model to mine association rules using a dataset of 1200 samples and 22 important water parameters. The association rules mined using Apriori and FP-growth algorithms are mapped with association rules that have been found using improved FP-growth algorithm with the confidence of 0.8 and the lift greater than 1, so that confidence in the rules found using IFP can be assured. Learning these interconnectedness of water parameters, would help in observing unknown patterns and assessing quality of water. Also wastewater plants can treat water more effectively so that level of pollution can be controlled before being released into water bodies as well as healthy aquatic environments can be monitored successfully.