This is one of my favorite lines from our paper: “By 0-dimensional features, we do not mean the…, but instead, the connected components of a graph.” I’m not sure why, but it was always something that stood out to me.

Funnily enough, although we plan to write another paper on the 1-dimensional features of a graph, I’m also working on another 1-D project with Dr. Rosen that specifically works with 1-D *functions*! These are just your typical line graphs that are found commonly for stocks, weather change, etc. We’re hoping to using persistent homology to simplify the line in a way that best preserves the original line in terms of error (by using the integral- wow! everything I learned in calculus is finally becoming useful) and in terms of user choice (we’re using mechanical turk for our study).

Our inspiration came from one of my friend’s projects where he represented some weather data as a tiny line graph and it looked, apparently, terrible! Dr. Rosen wondered if there was a better way to represent the data other than using rfft, gaussian, etc. This is where the project came from, and I was enlisted as one of the students on the project to complete a paper. My research partner, Chris, is now off working in industry (yay! but ew!) so I’ve taken the lead role for this project to complete the user study. This project is AWESOME and I’m really excited to see what we can do with it! I believe we’ll be submitting this project as a poster to Vis 2018.

Leave a Reply

Your email address will not be published. Required fields are marked *