Low cost artificial intelligence Internet of Things based water quality monitoring for rural areas

Abstract

Safe drinking water is quite possibly one of the most difficult-to-find issues on our earth intently influencing the well-being and cleanliness of humanity, animals, and plants. The smart speed of industrialization and the conspicuous significance of horticulture with cultivating pesticides have impelled water contamination to a colossal degree. Safe water is likewise worried about its verdure. In this article, a low-cost and low-power is proposed to screen the nature of water that can be utilized in all networks wherever consumption of water is typically provided by means of the water dispersion framework. Planned Sensor hubs are viable with existing circulation organizations. Sensor nodes can collect the data for the electrochemical properties of water to the server. A developed algorithm using the Decision Tree and Naive Bayes methods identifies cloud predictions for safe drinkable water and alerts on divergence from a World Health Organization-specified safer range. For efficacy, the suggested method is scientifically evaluated in two villages, Mandalgarh and Bassi, Rajasthan, India, where drinking water is in short supply. Naive Bayes, Gradient Boosted Classifier, support vector machine, and artificial neural network models are applied to collected data of water quality and analyzed by the Naive Bayes. The obtained results efficiently with 0.56 F1-Score. The nodes of the distributed network can work in harsh conditions as well and low power consumption and information transmission viability are precisely estimated. The empirical results are verified by laboratory studies, and it is demonstrated that the method has a significant impact on the prevention of water-borne infections, especially in rural regions.