Research

I have worked as a research scientist at the Knowledge Management Research Lab of the German Research Center for Artificial Intelligence (DFKI) and as a research assistant at the University of Regina Department of Computer Science and Griffith University School of Computing and Information Technology. I also taught at the University of Toronto Department of Computer Science.

Contents

Research interests

  • computational linguistics
    • automatic summarization
    • information retrieval (IR)
    • information extraction (IE)
    • latent semantic analysis (LSA)
    • machine translation (MT)
  • technology and society
    • social informatics
    • Free Software
    • history and philosophy of technology
    • digital media and value theory
    • computers and law
    • copyright, patents, and "intellectual property"
  • document engineering
    • web standards
    • web internationalization
    • web accessibility
    • collaborative media
    • digital typography

Research projects

eFISK (DFKI, 2003–2004)

TBA

ICE (University of Toronto, 2002–2003)

A major problem with automatically-produced summaries in general, and extracts in particular, is that the output text often lacks textual coherence. Our goal with ICE (Improved-Coherence Extractor) was to improve the textual coherence of automatically produced extracts. We developed and implemented an algorithm which builds an initial extract composed solely of topic sentences, and then recursively fills in the lacunae by providing linking material from the original text between semantically dissimilar sentences. ICE differs in architecture from most other summarizers in that it measures semantic similarity with latent semantic analysis (LSA), a factor analysis technique based on the vector-space model of information retrieval. We believe that the deep semantic relations discovered by LSA can assist in the identification and correction of abrupt topic shifts in the summaries.

ICE was funded by the Natural Sciences and Engineering Research Council of Canada (NSERC).

Further information:

  • Tristan Miller. Latent semantic analysis and the construction of coherent extracts. In Nicolas Nicolov, Kalina Botcheva, Galia Angelova, and Ruslan Mitkov, editors, Recent Advances in Natural Language Processing III, volume 260 of Current Issues in Linguistic Theory (CILT), pages 277–286. John Benjamins, Amsterdam/Philadelphia, 2004.
  • Tristan Miller. Latent semantic analysis and the construction of coherent extracts. In Galia Angelova, Kalina Bontcheva, Ruslan Mitkov, Nicolas Nicolov, and Nikolai Nikolov, editors, Proceedings of the International Conference RANLP-2003 (Recent Advances in Natural Language Processing), pages 270–277, September 2003.
  • Tristan Miller. Generating coherent extracts of single documents using latent semantic analysis. Master's thesis, Department of Computer Science, University of Toronto, March 2003.

DELORES (Griffith University, 1999–2000)

DELORES (DEfeasible LOgic REasoning System) is a forward-chaining reasoning engine for defeasible logic, a less-expressive but more efficient non-monotonic logic. In contrast with most other non-monotonic logics, defeasible logic has linear complexity, allowing DELORES to execute large theories very quickly. DELORES's algorithm extends to general defeasible theories through the use of a pre-processing transformation eliminates all uses of defeaters and superiority relation. The transformation was designed to provide incremental transformation of defeasible theories, and systematically uses new atoms and new defeasible rules to simulate the eliminated features.

Financial support for DELORES was provided by the Australian Research Council.

Further information:

  • DELORES project home page
  • Michael J. Maher, Allan Rock, Grigoris Antoniou, David Billington, and Tristan Miller. Efficient defeasible reasoning systems. International Journal on Artificial Intelligence Tools, 10(4):483–501, 2001.
  • Michael J. Maher, Allan Rock, Grigoris Antoniou, David Billington, and Tristan Miller. Efficient defeasible reasoning systems. In Proceedings of the 12th IEEE International Conference on Tools with Artificial Intelligence, pages 384–392, 2000.
  • Tristan Miller. DELORES User's Manual. School of Computing and Information Technology, Griffith University, 2000.

WEBWEAVR-III (University of Regina, 1998)

WEBWEAVR-III is a research toolkit for normative decision support systems. The project, under continuous development for over a decade, is headed by Yang Xiang of the University of Guelph.

WEBWEAVR-III supports the construction of Bayesian networks, inference in standard and dynamic Bayesian networks and decomposable Markov networks, construction and verification of multiply sectioned Bayesian networks (MSBNs), inference in multi-agent MSBNs, and learning decomposable Markov networks. It is written in Java, and has a GUI and an extensive library to support development.

My contribution to the project involved the design and implementation of an efficient, robust algorithm for the random generation of directed acyclic graphs.

WEBWEAVR-III is funded by the Natural Sciences and Engineering Research Council of Canada (NSERC), the Institute for Robotics and Intelligence Systems (IRIS), and SaskTel.

Further information:

  • WEBWEAVR-III project home page
  • Yang Xiang and Tristan Miller. A well-behaved algorithm for simulating dependence structures of Bayesian networks. International Journal of Applied Mathematics, 1(8):923–932, 1999.

Publications

A somewhat outdated human-readable list of my research publications is available. You can also see the same list in BibTeX format.

Teaching

Thesis supervision

  • Elizabeth Wolf (Diplomarbeit, DFKI, 2005)

Courses

Retrieved from "http://en.nothingisreal.com/wiki/Research"