PhD Student
ICCS, School of Informatics
University of Edinburgh
email: t.LASTNAME (at) ed.ac.uk
address: Informatics Forum
10 Crichton Street
Edinburgh EH8 9AB
Scotland, UK
phone: +44-131-650-4418 (office)
office: 3.38
I am a PhD student at the Institute for Communicating and Collaborative Systems (ICCS), University of Edinburgh investigating linguistically informed information retrieval in law. I work with Professor Jon Oberlander in the School of Informatics and Burkhard Schafer in the School of Law. My research uses deep text processing to extract predicate-argument style information applied in one branch of a split query expansion model.
I have returned from a 2.5 month programme of research at the Nara Institute of Science and Technology (NAIST) in summer 2009, and will engage in short visits to the Centre for Intelligent Information Retrieval at the University of Massachusetts, Amherst, in late 2009 and early 2010.
I developed an interest in law as an undergraduate at the University of California, Berkeley where I took my first legal course. My main interests are related to the following areas:
I am also interested in psychological aspects of communication, including sentiment and credibility. This follows my work in song lyric clustering that formed part of my MSc in Cognitive Science, completed at the University of Edinburgh in 2007. Here I used Kohonen self-organising maps to cluster songs by genre using 140 heterogenous features, and compared performance by tracking songs pairs through the clustering process. My interest in credibility relates to earlier work in public communications.
My research includes work with the NTCIR patent datasets. In the course of experimentation it came to my attention that out of the circa 33,000 patents identified as relevant to the 1000 sample queries in the NTCIR-6 patent collection, 1,396 patents are missing from the source collection. As a notice to fellow researchers, this error in the collection has been confirmed by the NTCIR organisers, and the list of missing documents can be found here.