Automated Groundwater Quality Index Classification Using Adaptive Aquila Optimizer with Deep Learning

Water is consumed for a various purposes, together with agriculture, industrial, and drinking usage. The water quality has directly influences both the environment and human health. For effective management of water, the water quality index (WQI) remains as a critical indication. Current WQI methods evaluate water quality with many classifiers. Therefore, various approaches offer many elucidations for the same water property that contributed a significant amount of uncertainty in the accurate water quality categorization. The study aims to assess the performances of the water quality index (WQI) method for correctly classifying coastal water quality utilizing a newly devised technique. This study performs an Automated Groundwater Quality Index Classification using Adaptive Aquila Optimizer with Deep Learning (AGQIC-AAODL) technique on Thiruvallur district, India. In the presented AGQIC-AAODL technique, a three-stage process is involved such as data normalization, prediction, and AAO based hyperparameter tuning. At the primary stage, data normalization is performed for scaling the data into uniform format. The AGQIC-AAODL method applies multi-head bidirectional gated recurrent unit (MHBGRU) model for classification process. At the final stage, the AAO based hyperparameter tuning is applied for hyperparameter tuning process. The simulation outputs of the AGQIC-AAODL method were experimented on the WQI dataset. The overall comparison analysis reported the superior accomplishment of the AGQIC-AAODL method by means of various measures.