Description
Guide: Dr. Ramesh Huda
The Indian Council of Medical Research is the apex body in India for the formulation, coordination and promotion of biomedical research.
I joined the NIIRNCD Lab as an Intern at the time of the inception of COVID-19. The project and the study I contributed to was mainly aimed at forecasting the number of deaths due to COVID-19, based on data that is pre-available. As ICMR was the national testing body of India during COVID-19, we used the number of tests administered to predict the number of deaths. The motive of the study was to aid in the decision making for handling the world wide outbreak of the deadly virus at the time.
My contributions:
— conducted a thorough Exploratory Data Analysis
— cleaned a raw dataset of 57,180 rows and 50 columns
— split the data into train (75%) and test (25%) sets
— implemented various Machine Learning algorithms
— developed a regression based supervised prediction model
— evaluated the performance according to R-squared score and mean absolute error
The final model (Random Forest Regressor) prognosticates the total number of deaths given the total number of tests with an accuracy of 99.98%, after hyperparameter tuning.
Being my first internship, this had a big impact in shaping my interest in Machine Learning. I learnt a lot about the practical implementation of Machine Learning, as well as its applications.