Alexey Gulyuk1,2,Akhlak-ul Mahmood1,Paul Westerhoff3,Yaroslava Yingling1
North Carolina State University1,STEPS research center2,Arizona State University3
Alexey Gulyuk1,2,Akhlak-ul Mahmood1,Paul Westerhoff3,Yaroslava Yingling1
North Carolina State University1,STEPS research center2,Arizona State University3
The problem of reusing available natural resources and, particularly, removing various chemical pollutants from water sources is a topic that currently receives a lot of attention. Utilization of various agents (nanoparticles, hydrogels, or solvents) opens new paths to recovery the dangerous wastewater pollutants like hydrogen, phosphorus, or heavy metals.<br/><br/>The very first step for developing water cleaning strategies requires assessing full<br/>chemical composition of a target water source. In practice, extensive water quality<br/>analysis involves analyzing enormous amounts of data and requires rigorous data<br/>preparation and several preprocessing steps.<br/><br/>Here we want to present our vision of how the combination of experimental and ML- derived data can help to facilitate the increased accuracy of data analysis for further advances in clean water sustainability. Particularly, we focus on assembly, analysis, and completion of the water chemical composition dataset, which became one of cornerstones of the project. Data enrichment tools utilized in this work rely on Machine Learning algorithms and enable elements of Convergence Informatics, thus advancing data-driven research with the end goal of finding the most effective water treatment compounds and agents.