CS 3891.04 / 5891.04 - Data Visualization

Overview Syllabus Schedule Assignments Project Resources


Syllabus

Course Description

Suppose you were provided a dataset, say a table of numbers, and you were interested in performing a set of analyses on the data. For instance, you might be interested to know if there exists trends, correlations, or outliers in the data. How does one conduct data analysis? One way is to perform statistical analysis, in order to test for these considerations. Yet statistical techniques might only tell a very small story about the data, for instance a single number and confidence bounds. How can we gain further insight about our data? A naive way is to look at the raw data directly, for instance in your favorite spreadsheet application. Certainly this will provide us with more detail, but is this useful for a higher-level understanding of the data?

Data visualization is concerned with the mapping of data into visual representations in order to support humans in gaining insight. In the above example, we considered one form of visualization, namely a tabular spreadsheet. Yet, there are other ways to visualize data that would more quickly lead to insight. For instance, we might map each data sample to a point, or summarize subsets of the data with bars. Certain visual representations are designed to leverage the human visual processing system for efficiently identifying patterns, and often these patterns reflect the analyses we care about. For instance, we might identify a correlation between two variables by plotting our data as a set of points, or identify an outlier by observing how a lone bar is really small or tall in a bar plot. This is the power of data visualization: the efficient processing of visual representations that can amplify one’s cognition for understanding data.

But when presented with a dataset, and a set of objectives we would like to address, how should one go about designing a visualization? What graphical marks should one choose? How should we customize the appearance of these marks? What is the set of operations that we should provide to the user so that they may interact with the visualization? The design space for creating visualizations is vast, especially as the complexity of the data increases. In this course, we are going to cover the fundamental aspects involved with effectively designing data visualizations.

Objectives

The set of topics that you will learn are broken down into two main sections: Visualization Basics and Visualization Techniques.

Visualization Basics

The first part of the course will be dedicated to the underpinnings of data visualization:

The first part of the course will also focus on how to write code for designing data visualizations. You will learn the basics behind D3, a Javascript library for developing visualizations for the web, and how to use D3 to map data to a core set of graphical marks and visual encodings.

Visualization Techniques

Equipped with an understanding of visualization basics, we will then delve into techniques. We will consider the following:

Prerequisites

CS 2201 or CS 2204, CS 2212 or Math 2410

More specifically, data visualization is inherently an interdisciplinary research field, as it relies on understanding the data at hand. That said, having a good understanding of data structures should enable you to grasp the types of data we will work with.

Now, another important aspect of visualization is space. Space is utilized in a myriad of ways in constructing data visualization. So you should be familiar with 2D geometric computing, as would be covered in linear algebra, in order to work with spatial transformations, scales, distances, and shapes.

Instructor

Matthew Berger

email: matthew.berger@vanderbilt.edu

office hours: TH 2:00-3:00, JH 379

Discussion

We will use Brightspace for any discussion related to the course: questions on lecture content, programming assignments, etc..

Course Format

The course will consist of lectures, a set of programming assignments, a project, and two exams: a mid-term and a final.

The lectures will typically be of two formats, what I call theory and practice. In the theory portions of lectures, I will discuss the basic concepts behind visualization; e.g. data representations, perceptual principles, design guidelines, and visualization techniques. The practice portions of lectures will complement the theory by covering how to actually implement the ideas that are discussed. The practice format will largely be focused on the details of D3, discussing how to use D3 for particular purposes, accompanied by sets of examples.

The lectures will draw from the textbooks mentioned below, as well as research papers. I expect everyone to participate in the lectures, as you will be expected to assess data visualizations during these times.

D3 will be the primary library that we will use for the programming assignments. If you are unfamiliar with web programming and Javascript, don’t fret! We will cover the basics as part of the course.

The exams are meant to test your knowledge on everything that is covered during the lectures. The mid-term will cover material for the (roughly) first half of the course, and the final will cover the (roughly) second half, with a small amount of material from the first half thrown in for good measure.

Last but not least, there will be a project associated with the course. You will work in teams of size 2-3. For the project, you will be expected to complete a design study: for a given dataset, you will create a set of objectives that comprise what you would like to learn about the data, a set of tasks that address these objectives, and design a visualization built around these tasks. More information is available on project.

Textbooks

Required

Visualization Analysis and Design by Tamara Munzner. This book will serve as the core for discussing visualization principles and techniques. Copies have been reserved in the book store; if not available, you can always purchase it online.

Interactive Data Visualization for the Web by Scott Murray. This book covers the basics behind D3, and its purpose is to get you to understand how to code data visualizations. If you are already familiar with D3, or are comfortable using the myriad of D3 resources on the web (see Resources), then you might not find this book necessary, otherwise I strongly recommend picking it up.

Optional

Design for Information by Isabel Meirelles. This book covers the basic principles in how to design visualizations for a variety of data types, along with abundant examples.

Making Data Visual by Danyel Fisher and Miriah Meyer. This book covers the why behind visualizing data, namely, how do tasks drive our decisions in designing visualizations.

Lectures

MW, 2:10-3:25, 258 Featheringill Hall

Lecture Slides

See schedule.

Assignments

See assignments.

Assignments will be posted at the beginning of class, and will be due at 11:59:59pm on their respective due dates. GitHub will be used to distribute assignments, and Brightspace will be used to submit assignments.

Resources

See resources.

Assessment

Grades

Your final grade will be numeric, and will be converted into a letter grade via the following:

Late Submission Policy

If an assignment is not submitted on its due date, it may still be submitted within the proceeding 24 hours, but the submission will receive a 15% penality. Beyond 24 hours, no credit for assignments will be given. No exceptions.

Regrading Policy

If you have disagreements with respect to grading of assignments, exams, or the project, you will have one week after having received it to dispute the grade. You may submit a written request for a reassessment, and I will then review the request, determine whether or not to regrade the assignment/exam/project, and perform the reassessment within one week of receiving the request.

Department and University Academic Policies

Academic Honesty

Studuents should adhere to the Vanderbilt Honor System. Cheating or plagiarizing will not be tolerated in this course.

Academic Integrity

More generally, students should act in accordance with the academic integrity policies of the university, please see Vanderbilt’s Academic Integrity for more information.

Privacy

All student data and information will be protected under FERPA laws. Please refer to the Vanderbilt Student Privacy Statement. Please take care to not disclose any private information during lectures and when submitting assignments.

Nondiscrimination and Anti-Harassment

Vanderbilt is committed to an environment that is free of discrimination and harassment of any kind. If you feel you are being sexually harassed, please see Project Safe. If you feel unsafe, taken advantage of in any way, or mentally/emotionally unwell, please reach out to the Student Care Network.

Subject to Change Statement

Information contained in the course syllabus, other than the general assessment, may be subject to change with advance notice, at the discretion of the instructor.

Acknowledgements

The design of this course was influenced by other data visualization courses from Alexander Lex, Alvitta Ottley, Hanspeter Pfister, and Carlos Scheidegger.