This project involves the analyses of text data on users’ reviews of restaurants and sentiment analyses. It is based on a collection of reviews and accompanying star ratings from Yelp. We will examine the effectiveness of different sentiment ‘dictionaries’, and develop and evaluate classification models to help predict sentiment polarity (negative, positive).

In this project, I analyzed data from the year 2013-2015 of Lending Club with 100K observations. The goal is to develop different models to predict which loans are at risk of default. Different models - Decision Trees, Random Forest, Boosting models, linear models were built to predict the default rate of loans and devise an investment strategy.

An interactive map of 7 wonders of India built using Leaflet package

This Markdown file is the analysis of NOAA Storm Database