Projects

FireMind

Machine learning is adept at distinguishing patterns which evade human faculties, and there’s a wealth of historically available data about wildfire. Training machine learning models is slow, but once a trained model is available, making a prediction is fast; on the order of milliseconds. FireMind uses deep neural networks and historical wildfire data to model how wildfire spreads. Read More on how it works and Lessons Learned.

Social Network Analysis of Interdisciplinary Research Centers

My current project focuses on using social network analysis to discover patterns of collaboration in interdisciplinary research organizations, and then guide qualitative research into how and why successful collaborations develop. Looking at the substantive outcomes of interdisciplinary research, rather than ambitions or effort, shows that real interdisciplinarity is very hard. Achieving successful interdisciplinary research requires building whole new modes of thinking, an effort which is rarely fully vested in by participants with many immediate scholarly concerns. I’ve also extended this work by using natural language processing to model article abstracts. See some blog posts (1 2 3) and Scientometrics 101 on github.

Better Book Reviews

I care deeply about books. I’ve always been a reader, and since 2011 I’ve reviewed all the books I’ve read on Goodreads. But not all book reviews are created equal. My review of John Haidt’s The Righteous Mind is by far the highest impact thing I’ve written. I used linear regression to find that I should read more popular books (pass) and included images (doable). A follow on project used a recommendation engine built on surprise to discover books which are not widely read, but which would be liked in they were.