Most demographic data, such as from the census, focuses focuses on objectively verifiable facts, but we think it is equally important to analyze trends in public perception. We picked the representative question "how much does the nation hate Chris?" and tried to analyze variations in public sentiment over time.
What it does
Plot several data sets related to the name "Chris," in order to form a qualitative plus/minus ranking.
How we built it
We scraped data from several sources, including using several outside APIs and web scraping through Beautiful Soup, to create several geoJSON and CSV files. We uploaded this to Esri and used their visualization capacity to form an interactive frontend.
Challenges we ran into
Getting enough data! There are not very many Chris themed data sets out there.
Accomplishments that we're proud of
Taking a silly idea with much more serious implications, and successfully executing it from start to finish.
What we learned
All of us gained experience designing a scalable implementation of our idea, and had to deal with data science issues such as data cleaning and verification for the first time.
What's next for Everyone Hates Chris
We want to scale this up into a more general sentiment analyzer - given a question, we wish to have a database of datasets from which we can automatically extract trends.