This project analyzes the National UFO Reporting Center (NUFORC) data. We built this webpage to host all of the visualizations we created so other people can access this information and gain a better understanding of the UFO sighting dataset.
The data set is a single, scrubbed data set that includes the geocoded and time standardized data for around 88,000 UFO sightings. Each of these reports include geographic and time data as well as a field containing the text of the report, which is anywhere from a few words to a few paragraphs. The original dataset can be found here.
The globe visualization on the landing page visualizes the 500 centroids generated when we ran the K-means algorithm over all of the latitude and longitude points and specified that we wanted 500 clusters. The height of each plotted centroid represents the number of sightings in that centroid's cluster. The higher number of sightings in that area, the greater the height of that point.
Histograms that visualize specific aspects of the data such as UFO sightings by year, frequency of UFO sighting shapes, UFO sighting density in America, and UFO sightings by specific states can be found on the More Data page.