Prediction of Air Pollutants Using Supervised Machine Learning

Air pollution is a severe problem in areas where population density is high such as metropolitan cities. Various types of emissions caused by people's actions, such as transportation, power, and fuel use, are affecting air quality. The machine learning in its components has become the key focus areas of all companies from the early stage of startups to major platform vendors. Machine learning is a field in which an artificial intelligence device collects sensor data and learns to act. One of those factors, the proposed research work has chosen to learn and forecast the air quality index with the opportunity to adapt to the machine learning algorithms. Air Quality Index is a device used for determining the air quality and the causes of air pollution. Air pollution is widely regarded as a concern in India. The highest possible values that are considered fit for public medical services are relatively higher than the daily air quality quantified attributes. The situation is particularly bad in larger cities like Delhi, where the AQI has reached an all-time high of 999. A few of the air pollution control processes are carried out by the federal government and state governments. Many researchers implement some of the algorithms we use, but none of them compares their performance across all six algorithms in a single study under similar conditions and using the same data. This paper portrays different strategies followed in forecasting the Air Quality Index using supervised machine learning procedures. The supervised machine learning technique (SMLT) was used to analyze a dataset and capture multiple pieces of information, including variable recognition, uni-variate analysis, bi-variate and multivariate analysis, missing value treatments, and data analysis. Moreover, the Estimation of the air quality index can be accomplished based on pollutants causing effect viz., PM10, PM2.5, SO2, CO, and NO2. These are the components used in supervised machine learning procedureres to compare the air quality of an environment. The main goal is to create a machine learning model and investigate the air quality index by predicting the best results from various machine learning algorithms based on their precision. We used logistic regression, decision tree, support vector machine, random forest tree, Nave Bayes theorem, and K-nearest neighbor as six basic machine learning algorithms.