Water Quality Analysis of Major Rivers of India Using Machine Learning

Water is one of the most important natural resources. It is very crucial for the existence of life and nourishment of all living organisms living on earth. It is observed that over the years the river water quality is degrading at a very rapid rate because of the toxic waste and contaminants. This has made river water unsuitable for any usage. It becomes very important to analyze the water quality of various rivers as this river water is used for drinking and domestic purpose, irrigation and aquatic life as well as fish and fisheries. In-order to understand the quality of water that whether it clean or not we have to study and analyze various water quality parameters like Biochemical Oxygen Demand (BOD), temperature, Potential of Hydrogen (pH), Dissolved Oxygen (DO), and conductivity and to understand about the quality of water of various rivers of India that whether it is clean or not, a classification model using three different classifier is presented in the study. We used J48, LMT, and Naïve Bayes classification algorithm in-order to classify the water quality data. The WEKA tool was used for analyze the collected data of various rivers then classify as clean or not clean. Further the Exploratory Data Analysis (EDA) of the collected data was performed using python programming language in jupyter notebook and some of the python libraries which was used include NumPy, Pandas, matplotlib, seaborne. In this work we have studied 15 papers from various publishers and created a summary to study about how and why the water is classified as clean or not clean using various machine learning algorithms.