Projects
As a high school English teacher transitioning to data analysis, I have completed projects in Python, Tableau, and SQL. This page showcases some examples of my work.
Why do people get heart attacks, and what are the most common warning signs?
This project analyses whether age, gender, and various symptoms correlate with contracting heart disease.
Skills Displayed: Python, Jupyter Notebooks, Exploratory Data Analysis
How quickly is our planet warming?
This project searches for errors in the data such as empty rows, duplicates, and outlier, then uses R's ggplot graphing library to show the rate at which climate change is accelerating.
Skills Displayed: R Markdown, Tidyverse
Data Cleaning, Linear and Polynomial Regression
Where are there the most unique species of animals: near New York City, or further upstate?
This project uses maps to examine where there are the animals, and looks at whether there are correlations between the presence of endangered species and the wealth or population of their human neighbours.
Skills Displayed: GeoJSON, MathPlotLib, incorporating data from multiple sources
Do most monsters in D&D tend to be easier or harder to defeat? Is there a relationship between a creature's strength and agility, between its intelligence and charisma? This analysis uses R Markdown to look at trends within the fantasy tabletop game
Skills Displayed: R Markdown, Correlation Matrix, , Data type conversion
What percentage of Amazon reviews are positive or negative? Are there particular words or phrases that correspond with an emotional tone?
This project uses the Natural Language Tool Kit to prepare product reviews for analysis, then examines which words and phrases correspond with positive and negative sentiments.
Skills Displayed: Tokenisation and Lemmatisation, the VADER method from NLTK,
Why is it so difficult for computer models to anticipate how stocks will perform? Do different types of Machine Learning algorithms perform significantly better or worse at predicting how a stock will perform?
This project uses 'SciKitLearn' Python Library, training and testing models to predict how Google stocks will do in the future.
Skills Displayed: Data Cleaning, Support Vector Regression, Random Forest Regression, Periodicity Analysis
Can a bank predict which customers are most likely to default on their loans? Which factors contribute the most to whether or not a borrower is likely to default?
This project uses a logistic regression Machine Learning method to combine multiple variables into one model, then uses metrics like precision and recall to assess the reliability of the algorithm.
Skills Displayed: Switching between Data Types, Seaborn, Scikitlearn, Logistic Regression
