Academic Projects

I have performed a time series analysis of the stock prices of Tata Consultancy Services from 2002 to 2021. I have started by visualising the data. And then I fitted models like an autoregressive integrated moving average (ARIMA) model, Vector autoregression (VAR), SARIMA (seasonal ARIMA) model, Unobserved components model (UCM), and Dynamic Factor models

This is a Shiny Dashboard on the World Happiness and Corruption (2015-2020) data set, which I made for my DG1107: Visualisation course under my professor, Dr. Sourish Das, during the 1st semester of my M.Sc. in Data Science at the Chennai Mathematical Institute

This project is dependent on the MIMIC-III database, which encompasses clinical data from patients admitted to critical care units at the Beth Israel Deaconess Medical Center between 2001 and 2012. I conducted a comparative analysis of Machine Learning, Deep Learning, and Natural Language Processing models applied to discharge text notes to predict unplanned hospital readmissions.

The MNIST database (Modified National Institute of Standards and Technology database) is a large database of handwritten digits that is commonly used as a first test for new classification algorithms. We follow this tradition to investigate the performance of Artificial Neural Networks of different complexity on MNIST. However, since MNIST is too easy for accessing the full power of modern machine learning algorithms (Convolutional Neural Network) we will extend our analysis to the recently introduced, harder Fashion-MNIST dataset.

I conducted a performance analysis of two classifiers and two ensemble classifiers: one utilizing bagging with Decision Trees and the other employing boosting with Decision Stumps. This analysis was carried out on the ”Bank Marketing Data Set”. The primary objective of this classification task was to predict whether clients would subscribe to a term deposit.

I’ve done these exploratory data analysis using R while pursuing my B.Sc. Statistics degree from Presidency University, Kolkata in forms of assignments under our Professor Dr. Atanu Kumar Ghosh. I have broken down these assignment done over the years into 6 reports for easier read. Please check them out:- Chicago econ autompg Abalone SENIC birthwt

Code implementation of Deep Painterly Harmonization (Luan et al., 2018), an innovative algorithm for seamless integration of photos into paintings that enhances local statistical consistency, enabling superior results compared to traditional photo compositing and global stylization techniques.