Together with researchers from different disciplines, I develop algorithms and tools that combine natural language processing and user interface design to support research efforts in the digital humanities.
This work analyzes the use of poetic language using computational language techniques. One discovery is that quotations shared in social media have stylistic discourse patterns that differ from standard prose. Another discovery is an algorithm for interseting word embeddings to produce poetic terms that reside metaphorically at the intersection of two conceptual frames.
The WordSeer project is a text analysis environment that combines visualization, information retrieval, sensemaking and natural language processing to make the contents of text navigable, accessible, and useful.
This project draws on 20 years of experience of teaching Data Visualization. This includes coorganizing a panel called Vis, The Next Generation: Teaching Across the Researcher-Practitioner Gap, at IEEE Infoviz in October 2015, and co-organizing two subsequent workshops: Pedagogy in Data Vis Workshop 2016 and Pedagogy in Data Vis Workshop 2017.
This also encompasses innovating in the design of educational exercises.