Data Science & Machine Learning
Projects

CheXNet with COVID-19 (reproduce-chexnet-covid)

This project explores the addition COVID-19 to an existing multilabel classification algorithm of thoracic pathologies to see potential viability of the model in detecting the disease in chest x-rays. View the GitHub repository here.

Early Flu Detection using Social Media

In this project, we explored the potential of using social media as a tool in detecting influenza spikes outside of flu season by monitoring and capturing tweets mentioning relevant query terms, and analyzing sentiment using the NLTK library to rate polarity as positive, neutral, or negative. View the GitHub repository here.

Cancer Mortality Analysis

For this project, we conducted an analysis of demographic factors as indicators of cancer mortality rates. The dataset utilized was reduced to focus on California, South Caroline, and Illinois. The model fitting was done in R using AIC and BIC with comparisons using RMSE, R2.

Medicare Drug Spending Dashboard

A dashboard developed to provide insights on sudden changes in cost of prescription drugs under Medicare. The dashboard was created using aggregated Medicare drug spending data from 2011-2015. Hadoop MapReduce was used to aggregate drug cost totals for both generic and brand name drugs of the same type.