Queripidia - Query-specific Wikipedia Construction
Abstract: We all turn towards Wikipedia with questions we want to know more about, but eventually find ourselves on the limit of its coverage. Instead of providing "ten blue links", my goal is to answer any web query with something that looks and feels like Wikipedia. I am developing algorithms to automatically retrieve, extract, and compile a knowledge resource for a given web query. For a very early web demo with some example queries see: http://ciir.cs.umass.edu/~dietz/queripidia/
Build a Shrine to Yourself
Abstract: Am I good enough? Is this the right topic? Am I in the right place? Am I passionate enough? If I switch now, will my career be over? - I have a lot of self doubts. I went through several advisors, topics, and research labs before I graduated age 34. Maybe I did everything wrong. Yet, some people believe that I am a successful researcher. How did I trick people? This is a talk about how to build a shrine to yourself.
Laura Dietz is a post-doctoral researcher / research scientist working with Bruce Croft at the Center for Intelligent Information Retrieval (CIIR) at the University of Massachusetts. Before that she was working with Andrew McCallum. She obtained her doctoral degree with a thesis on topic models for networked data from Max Planck Institute for Informatik in early 2011, being supervised by Tobias Scheffer and Gerhard Weikum. Between Masters and PhD, she worked at a Fraunhofer Institute until it got shut down. She worked on her PhD in Darmstadt, Berlin, Saarbruecken, Cambridge (UK), and Amherst.
35 members attended, including undergraduate students, graduate students, faculty, research scientists, and staff.