My projects
Data Scientist
I’m interested in building data-driven applications using machine learning and artificial intelligence.
Projects
1. The Creation of an Automated Serverless Data Pipeline for Analysis of Financial News Sentiment with Dash
Gathering news data through a serverless pipeline on AWS with lambda functions and creating an interactive dashboard to visualize it. Blog.
2. k-means and PCA to Visualize Clusters in National Health Survey Data
Dimensionality reduction by PCA and k-means clustering to visualize patterns in data from diet, physical examinations, and hospital laboratory reports. Blog. Code.
3. Predicting the Status of Water Pumps in Tanzania
This project demonstrates the application of feature engineering, hyperparameter tuning, and predictive modelling techniques to a multi-class classficiation problem. Using data from Taarifa and the Tanzanian Ministry of Water, I predicted the status of water pumps based on three categories: functional, needs repairs, does not work. This dataset describes a number of variables about what kind of pump is operating, when it was installed, and how it is managed. A smart understanding of which waterpoints will fail can improve maintenance operations and ensure that clean, potable water is available to communities across Tanzania. Code.
4. Picasso Painted It
This app was developed with a team of data scientists, front-end, and back-end engineers as a project for my first build week at Lambda School in ~4 days. The user inputs a photo which gets processed by a deep learning style transfer algorithm to be rendered in a style of a Picasso painting. Site.
5. Exploring 26 Years of Medicaid Data
I wanted to gain a better understanding of drug utilization rates in the U.S, so I analyzed prescription rates in a large subset of medicaid drug data. This project demonstrates various cleaning, exploration, and visualization techniques which I performed on all of the available drug utilization data from Medicaid.gov. Blog. Code.
6. Tweetmate
This app was developed with a team of data scientists, front-end, and back-end engineers as a project for my second build week at Lambda School in ~4 days. It allows the user to gain a deeper understanding of their Twitter persona through a psychographic analysis of text. Site.