We used several different data visualization softwares. These include: AntConc, SVG and seaborn via jupyter notebook.
We made two visualizations with AntConc. One for the first episode and one for the last episode. The main goal we had for these visualizations were to see what patterns changed or if any characters were mentioned less frequently as the show went on. AntConc is primarily designed for text analysis, so the types of visualizations we could generate were somewhat restricted. This meant we had to get creative to represent our data effectively.One issue that we faced was that there were too many words in the wordcloud that made it more difficult to read.We fixed this by reducing the word count analyzed from 200 to 50. This made the wordcloud much easier to read.
We made two SVGs for the first and fifth season. We found the data using XPath for the each episode of season 1 and 5. There were a total of 25 episodes (9 for season 1 and 16 for season 5). The goal of this comparison was to see if any characters had a change of frequency in dialogue. We found that Fry had more overall dialogue in both seasons when compared to Bender and Leela. All three character decreased their dialogue per episode by season 5. The main difficulty was figuring out how to recenter the text to appear on our site. The other difficulty was we had realized that season 1 and 5 had a different number of episodes which could account for why there Bender and Leela had increased dialogue. We averaged out the data instead to give more accurate results.
While Seaborn offers a high-level interface for creating attractive visualizations, we sometimes found ourselves limited by the library's built-in customization options. Achieving specific design or formatting requirements often required additional coding or workarounds.
We created 5 html pages and 2 CSS pages. While our work with HTML and CSS ran relatively smooth, there were some issues that we faced.One issue that we faced was that on one our laptops the "Data" heading text was overlapping with some of the other text.We were able to fix this by having some of the CSS simplified.
Despite these challenges, we were able to overcome them through collaboration, experimentation, and perseverance, ultimately producing insightful visualizations to support our project goals.