Finding similarities, aka as clustering, in air polution of multiple areas of India. The dataset comprises three types of air pollutant in India for specific cities. Techniques used: K-means clustering, Hierarchical Clustering, Affinity Propagation, Agglomerative Clustering, BIRCH Clustering, DBSCAN and Gaussian Mixture Model.
Jupyter Notebook
Clustering of air pollution
This project is based on a public dataset where air pollution in India listed concentrayion of three main pollutant in cities. Using feature engineering and machine learning techniques divided the dataset to three meaningful categories.