Julian Giles

Hey! I'm Julian and I am a Freshman at Penn State Behrend. I'm majoring in DIGIT and minoring in Computer Science and Game Development. My favorite character from the Avatar series would be Toph.

Tyler Cameron

Hi, my name is Tyler Cameron. I am a current Junior in the DIGIT Program. I have a strong passion for photography and content creation. My goal is to end up in an position where passions become my life fulltime. Some other things I enjoy is music and nature/space.

Zachary Dominick

Hey, I’m Zack Dominick and I am a sophomore at Penn State Behrend. I am majoring in DIGIT with a minor in Music Technology. I began playing piano at the age of 5 and have been deeply passionate in music ever since. I also enjoy many aspects of videography and the visual arts.

Graesyn Tefft

Hi, I'm Graesyn Tefft and I'm a sophomore at Penn State Behrend working towards a major in DIGIT and a minor in Video Game Development. I'm passionate about practically any kind of 3D work. I hope to eventually work full time working on 3D games.

Erin Mooney

Hi! I'm Erin Mooney and I am currently a junior at Penn State Behrend. I am dual majoring in DIGIT and Arts Administration with a minor in French. I love doing any sort of artistic work.

Kyara Parrish

Hey, I'm Kyara Parrish and I am a second year junior at Penn State Behrend. I am majoring in Digital Media, Arts, and Technology. I am very passionate about content creation, and I am very into visualizations such as photography and videography.

Sources:

Episode Transcripts: Avatar: The Last Airbender "Fandom Wiki"

Python Web Scraper kindly provided by Dr. Elisa Beshero-Bondar

Character Photos are taken directly from the show and official art sources

Our fantastic webpage background was illustrated by Julian Giles

For our code, please refer to the following:

XSLT SVGs Scripts Image Mapping Xquery RNC

Our process

Our first step in the process was to acquire the transcripts for each episode. To accomplish this, we borrowed Dr. B's Python Web Scraper. Unfortunately, this produced more results than we were anticipating due to the inclusion of commentary episodes, so our first step in processing the files was removing the extraneous episodes from our collection. Another issue we immediately ran into was that the Avatar Wiki used a strange form of markdown for each episode which we had to convert into something more usable. To accomplish this, we used Find and Replace in the Oxygen XML editor. To ensure uniformity across all of the files, we used Relax NG to build a schema. The general format for our project included <description> tags which would describe the setting of the episode or scene, <speech> which was the actual spoken dialogue of each character, and <charAction> which would include "stage directions" of what either the context behind <speech> (i.e., "Aang said angrily) or the action that was happening at that time (i.e. Aang turns to face Katara). We also included metadata about the episode, such as director and producer.

At this point, we had to re-evaluate our original research questions because it was not going to be feasible to map character interactions. Instead, we decided to look at what characters had the most speeches and "character actions" as well as what the top adjectives used by each nation. This was all achieved through XQuery. The scripts we used are linked above. From this data, SVG graphs were generated to display our data. Some were made manually while others were generated using Pygal.

We also decided we wanted to include an interactive map of the Avatar world that showed where each episode took place. To accomplish this, we used more Xquery to dig through each episode and provide a location. To make the interactive map, we used an SVG image map overlay with links to HTML files containing the location data for each episode. These HTML files were then displayed in an IFrame next to the map.