Introduction to Data Science

PPOL670

Spring 2021 Georgetown University

Course Outline

Course Schedule

Week	Date	Topic	Assignment
1	January 26	Work Flow and Reproducibility
2	February 2	Introduction to Programming in R
3	February 9	Reproducibility in Practice
4	February 16	Data Wrangling in R	Problem Set 1 Assigned
5	February 23	Data Visualization	Problem Set 1 Due
6	March 2	Web Scraping	Problem Set 2 Assigned
7	March 9	Geospatial Data	Problem Set 2 Due
8	March 16	Text as Data	Problem Set 3 Assigned
9	March 23	Introduction to Statistical Learning	Problem Set 3 Due
	March 30	Spring Break; No class
10	April 6	Applications in Supervised Learning (Regression)	Problem Set 4 Assigned
11	April 13	Applications in Supervised Learning (Classification)	Project Proposal Due; Problem Set 4 Due
12	April 20	Interpretable Machine Learning	Problem Set 5 Assigned
13	April 27	Applications in Unsupervised Learning	Problem Set 5 Due
14	May 4	Project Presentations
Final	May 18	Final Project Due (9:00 PM)

---

Virtual Classroom

Link to Virtual Zoom Classroom (Tuesdays 6:30pm - 9:00pm)

• Meeting Id: 924 0032 5980
• Pass code: 54193548

Virtual Office Hours (Wednesdays 9:00am - 11:00am)

Recurrent Zoom link can also be found on Canvas. Please contact the professor/TA through Slack if any of the Zoom links break.

---

Syllabus

Access an online version/.pdf version.

---

Readings

List of all required and suggested readings

---

Project

Data science is an applied field and therefore, it is important that you understand how to conduct a complete analysis from collecting data, to cleaning and analyzing it, to presenting your findings. Toward this end, students are required to complete an independent data science project, applying concepts learned throughout the course. The project is composed of three parts: a project proposal, an in-class presentation, and a project report.

More information regarding the final project will be circulated during class on Week 8

Overview

Project Overview

Presentation Rubric

Final Project Rubric

Examples of Successful Projects

• Matteo Lieb - "Poverty Rates and the Impact of Minimum Wages
• Frances Chen - “Predicting Social Distancing in the Age of COVID-19”
• Julia Eigner - “Predicting Maternal Mortality”
• Kristina Costa - “Analyzing FEMA Pre-Disaster Mitigation Grants”
• Liping Wang - “Female Labor Force Participation”

---

Installation

The following are installation instructions for R and RStudio.

R Software

To install R, download R from CRAN via the following:

• Windows: https://cran.r-project.org/bin/windows/base/ (Links to an external site.)
• Mac: https://cran.r-project.org/bin/macosx/

To install RStudio, download from the following (scroll to the bottom):

• https://www.rstudio.com/products/rstudio/download/ (Links to an external site.)

Video walkthroughs:

• Video Walkthrough - [Put together by former TA (Nathan Lovin)]
• RStudio Videos
  • Install R
  • Install RStudio

---

Contact

Eric Dunford (Professor)

• Office: Bedroom (formerly 404 Old North)
• Office Hours: Wednesdays 9:00am to 11:00am (EST) or by appointment
• Email: eric.dunford@georgetown.edu

Maddie Pickens (Teaching Assistant)

• Office Hours: Book via Calendarly
• Email: mp1595@georgetown.edu

Class

• Slack: class communications will primarily take place through Slack. Please follow the invite link to be added to the class Slack channel.

---

Lecture Materials

Asynchronous lecture materials will go live approximately one week prior to the scheduled synchronous meeting date.

Week 1: Work Flow and Reproducibility

• Lecture Slides

  • Table of Contents
    • About this Course
    • Reproducibility
    • Interacting with R

Week 2: Introduction to Programming in R

• Asynchronous Material

• Lecture Slides

  • Table of Contents
    • Objects
    • Data Structures
    • Indexing
    • Operators
    • Functions
    • Packages
    • Importing and Exporting Data
    • But where is my data exactly?

• Practice

  • In breakout rooms, let’s practice creating an R project, subsetting and creating data structures in R, and importing/exporting data.

Week 3: Reproducibility in Practice

• Asynchronous Material

• Lecture Slides

  • Table of Contents
    • Iteration
    • Control Statements
    • R Markdown
    • Generating Reproducible Examples

• Practice:

  • Breakout 1: Piecing together the code
  • R Markdown Walkthrough (download .Rmd)
  • Breakout 2: Generate a Reproducible Example

Week 4: Data Wrangling in R

• Asynchronous Material

• Lecture Slides

  • Table of Contents
    • Data Wrangling
    • Tidy Data
    • Manipulating Data
    • Piping
    • Joining Data
    • Reshaping Data

• Practice:

  • Breakout 1: Manipulating conflict event data
  • Breakout 2: Practice joining and reshaping data

Week 5: Data Visualization

• Asynchronous Material

• Lecture Slides

  • Table of Contents
    • The Components of Data Visualization
    • Grammar of Graphics

• Practice:

  • Walkthrough: let’s visualize the Gapminder dataset in a live coding walkthrough.
    • Download the notebook (.rmd as a .zip)
    • Answers to the notebook.

• Third Party Visualization Packages that build off of ggplot2.

  • ggthemes: for great theme and color schemes.
  • scico: for beautiful continuous color schemes.
  • gghighlight: for highlighting values or subgroups in a ggplot2 plot.
  • ggrepel: for adding text labels to ggplot2 plots.
  • GGally: for generating standard (but more complicated to construct) ggplot2 plots, such as correlation heatmaps and pairs plots.

Week 6: Web Scraping

• Asynchronous Material

• Lecture Slides

  • Table of Contents:
    • Functions
    • Websites
    • Scraping Content
    • Building a Scraper
    • Legality
    • Dates

• Discussion:

  • Data Acquisition using R packages (wrappers on APIs)

• Practice:

  • Breakout 1: Scraping Wikipedia for information on Presidential Debates

Week 7: Geospatial Data

• Asynchronous Material

• Lecture Slides

• Practice:

  • Breakout: Mapping Conflict in Nigeria
  • Answers to the Breakout

• Discussion:

  • Retrieve country-level and subnational-level shape files for free from diva-gis.org.
  • Explore other popular mapping packages in R:
    • ggmap: ggmap is an R package that makes it easy to retrieve raster map tiles from popular online mapping services like Google Maps and Stamen Maps and plot them using the ggplot2 framework
    • leaflet: leaflet is one of the most popular open-source JavaScript libraries for interactive maps. It’s used by websites ranging from The New York Times and The Washington Post to GitHub and Flickr, as well as GIS specialists like OpenStreetMap, Mapbox, and CartoDB.
    • tidycensus: tidycensus is an R package that allows users to interface with the US Census Bureau’s decennial Census and five-year American Community APIs and return tidyverse-ready data frames, optionally with simple feature geometry included.

Week 8: Text as Data

• Asynchronous Material

• Lecture Slides

  • Table of Contents
    • Strings
    • Tidy Text
    • Sentiment Analysis
    • Topic Models

• Practice:

  • Breakout: Exploring Inaugural Presidential Speeches
    • Download the inaug_speeches.csv dataset used in the breakout.
    • Breakout Answers

Week 9: Introduction to Statistical Learning

• Asynchronous Material

• Lecture Slides

• Table of Contents

  • Statistical Learning
  • Supervised Learning
  • Cross-Validation
  • Pre-Processing Data
  • caret API

• Practice:

  • Walkthrough: Data Pre-processing with recipes
    • Download the credit_data.csv dataset used in the walkthrough.

Week 10: Applications in Supervised Learning (Regression)

• Asynchronous Material

• Lecture Slides

  • Table of Contents
    • Regression Methods
    • K Nearest Neighbors
    • Regression Trees
    • Bagging and Random Forest

• Practice:

  • Walkthrough: Walkthrough using regression methods with caret
    • Download the london_bike_sharing.csv dataset used in the walkthrough.

Week 11: Applications in Supervised Learning (Classification)

• Asynchronous Material

• Lecture Slides

  • Table of Contents:
    • Classification
    • Logistic Regression
    • K-Nearest Neighbors
    • Classification Trees
    • Support Vector Machines

• Practice:

  • Walkthrough: Walkthrough using classification methods with caret
    • Download the health-coverage.csv dataset used in the walkthrough.

Week 12: Interpretable Machine Learning

• Asynchronous Material

• Lecture Slides

  • Table of Contents:
    • Interpretable Machine Learning
    • Variable Importance
    • Partial Dependence Plots
    • Individual Conditional Expectation Plots
    • Global Surrogate Models

• Practice:

  • Walkthrough: Walkthrough using interpretablility methods
    • Download the health-coverage.csv dataset used in the walkthrough.

Week 13: Applications in Unsupervised Learning

• Asynchronous Material

• Lecture Slides

  • Table of Contents:
    • Unsupervised Learning
    • K-Means Clustering
    • Hierarchical Clustering

• Exploratory Data Analysis in the Social Sciences

• Practice:

  • Walkthrough: Walkthrough using unsupervised learning
    • Download the vdems_civil_liberties.csv dataset used in the walkthrough.