Below are the winners for the 2018 Data Visualization Competition. Click on the thumbnail image or title to view the full visualization. You can also view all the fantastic entries or read the competition details if you are curious.
“This visualization displays racial demographic information regarding students from the University of Illinois at Urbana-Champaign with data provided by the Division of Management Information. I created the visual out of curiosity surrounding how students have racially identified on this campus over the last few decades given the historical context of enrollment at public universities in the United States. Users are able to get a view into how racial diversity has evolved through the years while also seeing the inclusion of broader race categories in recent years (Hawaiian/Pacific Islander and Multiracial). The data also chooses to refrain from classifying international students under the American standard race categories and provides an all-encompassing international category. The intention of this visual is to get users thinking about racial identity (which is self-reported by students) and how individuals have come to understand that for themselves and those around them.
This work was advised by Wade Fagen-Ulmschneider who provided technical feedback on the visual and published it through his Data Driven Discoveries page. Kara Landolt and Tina Abraham provided conceptual feedback and also tested the visual prior to publication.”
“My interactive webapp contains different visualizations that allow students to explore UIUC’s course data, so they will be able to make informed decisions when choosing which courses to register for. Students can browse through courses to see the overall grade distribution for a course as well as grade distributions for different individual sections of that course. Students may also search for professors to find grade distributions for courses they have previously taught. Another feature of the webapp allows students to search for courses satisfying several general education categories. In addition, the webapp also lets students search for all courses belonging to a particular subject. Finally, please note that all visualizations are interactive and hovering over visualizations with your mouse will trigger more detailed information to be displayed.”
“Does money buy electoral victory? An investigation into electoral spending in the US, and the correlations between (1) the generally accepted political leaning of a district and the overspending by one party and (2) overspending by one candidate and electoral victory.
“The National UFO Research Center (NUFORC) collects and serves over 80,000 reports of UFO sightings from more than 100 years. The dataset contains the time, location, duration, shape of the sighting.
Features of the Interactive Tableau Dashboard includes:
- The color of the visualization is set to Black to match the theme of the dataset
- USA Map: The sightings are represented based on:
- Exact location (Lat-long)
- The size of the sighting is matched using Tableau Shapes
- The size of the data point is directly proportional to the time of the sighting.
- If you hover over the map, the tooltip will show details regarding individual sighting
- Shapes of the sighting
- Time of the sighting
- Year of the sighting”
“Our visualization is for understanding a vision based end-to-end imitation learning autonomous rover, with only one single camera as its perception input.
The autonomous rover is modeled based on end-to-end imitation learning approach, which currently reaches L3 level in autonomous driving industries. However, few companies achieve L4 (a higher autonomous level) with end-to-end training.
This visualization helps us understand what knowledge is really learned within this “black box,” for developer to potentially reinforce the badly-handled situations through hard sample minings. Through visualizing the intermediate feature maps of the deep model, we found out that the convolutional kernels have learnt quite some obstacle detection and image segmentation capabilities even though we were not specifically training for those tasks. In addition to helping developer analyze thoroughly on the robustness of deep models, this visualization also potentially helps accelerate other deep learning applications. End-to-end labeled data are relatively easy to obtain while data for other tasks such as image segmentation and object detection models requires a lot of time to hand label. From the visualization we believe that end-to-end model have very high potential to be fine grained to a specialized model.”