Paper Title
Machine Learning and Computer Vision for Water Quality Improvement in Sustainable Cities

Abstract
Water is a necessity for human survival; therefore, mechanisms must be put in place to rigorously test the quality of water made available for drinking in towns and city-articulated supplies, as well as the rivers, creeks, and seashores that surround our cities. Preventing water-borne disease outbreaks and improving overall quality of life require the availability of high-quality water. Water quality issues have long been a problem in urban areas, causing illness, poisoning, disease outbreaks, and even human deaths. For human health, economic development, and forecasting pollutant impact on aquatic ecosystems, monitoring freshwater quality, as well as the risk of unexpected contamination, is critical.Even though technology for automated sampling and continuous analysis of water physicochemical parameters has advanced significantly, real-time warning capabilities against rapidly developing unknown chemical hazards remain limited. Traditional chemical analysis systems are ineffective at identifying unknown chemical mixtures and their additive and/or synergetic effects on biological systems. Acute exposures to chemical compounds that impact the neurological system and potentially enter freshwater sources unintentionally or deliberately can only be accurately analysed using appropriate functional biological models, according to neurotoxicology.Biological early warning systems(BEWS), which can continuously monitor behavioural and/or physiological parameters of appropriate aquatic bio-indicator species, have historically been expected to fill the gap and supplement conformist water quality test strategies in this regard. Changes in sub-lethal neurobehavioral persona have been shown to be physiologically significant endpoints for water quality sensing. Keywords - Biological Monitoring; Water Quality Assessment; Artificial Neural Network (ANN), Support Vector Machine (SVM); Japanese Medaka.