pamelot317.github.io

Pamela Patterson’s Data Science Portfolio

This portfolio is a compilation of notebooks and projects I created for data analysis or for exploration of machine learning algorithms.

Carbon Market trading - data cleaning and analysis - paid project

This project was contracted by Water Climate Trust. They were preparing for work at the 2018 and 2019 United Nations Climate Change Conferences (COP24 and COP25). The objective was to use the Clean Development Mechanism (CDM) project data available through the UN to uncover patterns and issues with the carbon market trading stystem. I did initial research and gathered the raw data, then I cleaned the data to make it easier to use. I then analyzed it and created data visualizations.

Part 1 of the data cleaning can be found here

Part 2 of the data cleaning and analysis can be found here

This was a group project for a graduate data science course taken at UC Davis. We obtained bike share usage data, including stations used, trip duration, and member status for 3 years: 2014, 2015, and 2016. We compared this data to weather data, neighborhood data (crime, income), and proximity of bike share stations to Caltrain stations. Our goal was to gain insights into the new bike share program in order to inform future expansion efforts.

Machine Learning Problems and Projects

These are problems, projects, and assignments I worked on while taking graduate level coursework.

K-nearest neighbors, linear regression, and risk

See homework assignment with both conceptual and data analysis problems here.

Linear and Logistic Regression, sklearn, Lasso

See homework assignment with both conceptual and data analysis problems here.

Ordinary Least Squares Fit, Loss functions, Linear Discriminant Analysis (LDA), Support Vector Machines (SVMs)

See homework assignment with both conceptual and data analysis problems here.

Cross validation and more SVMs

See homework assignment with both conceptual and data analysis problems here.

Random Forests and more SVMs

See homework assignment with both conceptual and data analysis problems here.

Project: Stock Market prices and the news

This project involved scraping websites for data and gathering multiple data sources. We then chose our method of analysis and assessed its effectiveness. See the project here for more detail. Supporting notebooks are here and here.

Collection of data science/analysis problems and projects from my graduate level coursework.

Assignments from data science course

These assignments showcase skills such as image processing, web scraping, data cleaning, working with geospatial data, and data analysis. Assignment 1, Assignment 2, Assignment 3, Assignment 4, Assignment 5, Assignment 6

Data analysis projects using R

Fitting a Model for Abalone Age

Logistic Regression Model to Predict Probability of House Sparrow Survival

Other Graduate Level Coursework

UC Davis

Mathematical Statistics (2 courses)

Applied Statistics (2 courses)

Probability Theory

Machine Learning

Data Science

Computational Statistics

Topological Data Analysis

Georgia Tech

Computability, Complexity,and Algorithms

Machine Learning