Paul Thomas

I'm now a postdoctoral fellow at the CSIRO. I'm interested in information retrieval, particularly metasearch, search over heterogenous data, and evaluation.

I'm still associated with the Department of Computer Science at the Australian National University.

IR (and friends)

I coordinate a fortnightly discussion group for people at ANU and nearby who work in information retrieval and related areas.

Publications

Full-text versions of papers provided via this site are author versions. In the case of ACM publications (SIGIR, CIKM, etc) they are posted here by permission of ACM for your personal use; in the case of Springer publications (IR, etc) they are posted here by permission of Springer for your personal use. Not for redistribution.

In press

• Cécile Paris, Stephen Wan, and Paul Thomas. Focused and aggregated search: A perspective from natural language generation. To appear in Information Retrieval.

2010

• Paul Thomas, Katherine Noack, and Cécile Paris. Evaluating interfaces for government metasearch. To appear in Proc. IIiX.
• Paul Thomas and David Hawking. 2010. Metasearch tools for desktop search. To appear in Proc. SIGIR workshop on desktop search.
• Paul Thomas and Milad Shokouhi. 2010. Evaluating server selection for federated search. In Proc. ECIR.

2009

• Judy Kay, Paul Thomas, and Andrew Trotman (eds). 2009. Proceedings of the fourteenth Australasian Document Computing Symposium.
• Tom Rowlands, Paul Thomas, and Stephen Wan. 2009. Web indexing on a diet: template removal with the sandwich algorithm. In Proc. Australasian Document Computing Symposium.
• Kristian Balog, Arjen P de Vries, Pavel Serdyukov, Paul Thomas, and Thijs Westerveld. 2009. Overview of the TREC 2009 Entity Track. In Proc. TREC.
• Paul Thomas and David Hawking. 2009. Server selection methods in personal metasearch: a comparative empirical study. Information Retrieval 12:581-604.
• Cécile Paris, Nathalie Colineau, Paul Thomas, and Ross Wilkinson. 2009. Stakeholders and their respective costs-benefits in IR evaluation. In Proc. SIGIR workshop on the Future of IR Evaluation.
• David Hawking, Paul Thomas, Tom Gedeon, Tim Jones, and Tom Rowlands. 2009. New methods for creating testfiles: Tuning enterprise search with C-TEST. In Proc. SIGIR workshop on the Future of IR Evaluation.
• Paul Thomas and Milad Shokouhi. 2009. SUSHI: Scoring scaled samples for server selection. In Proc. SIGIR.
• Milad Shokouhi, Leif Azzopardi, and Paul Thomas. 2009. Effective query expansion for for federated search. In Proc. SIGIR.
• David Hawking, Tom Rowlands, and Paul Thomas. 2009. C-TEST: Supporting novelty and diversity in testfiles for search evaluation. In Proc. SIGIR workshop on redundancy, diversity and interdependent document relevance.
• Stephen Wan, Paul Thomas, and Tom Rowlands. 2009. Web indexing on a diet: template removal with the sandwich algorithm. Technical report, CSIRO ICT Centre.
• Paul Thomas. 2009. Quality of language models for distributed information retrieval. Technical report, CSIRO ICT Centre.

2008

• Paul Thomas. 2008d. Server characterisation and selection for personal metasearch. SIGIR Forum 42(2), pp108-109.
• Krisztian Balog, Ian Soboroff, Paul Thomas, Peter Bailey, Nick Craswell, and Arjen P. de Vries. 2008. Overview of the TREC 2008 Enterprise Track. In Proc. TREC.
• Peter Bailey, Nick Craswell, Ian Soboroff, Paul Thomas, Arjen de Vries, and Emine Yilmaz. 2008. Relevance assessment: are judges exchangeable and does it matter? In Proc. SIGIR.
• Rob McArthur, Paul Thomas, Andrew Turpin, and Mingfang Wu. Proceedings of the thirteenth Australasian Document Computing Symposium.
• Paul Thomas and David Hawking. 2008. Experiences evaluating personal metasearch. In Proc. IIiX.
• Paul Thomas. 2008c. Implementation of PIS. Technical report, ANU Department of Computer Science.
• Paul Thomas. 2008b. Server characterisation and selection for personal metasearch. PhD thesis, Australian National University.
• Paul Thomas. 2008a. Generalising multiple capture-recapture to non-uniform sample sizes. In Proc. SIGIR (poster).

2007

• Peter Bailey, Paul Thomas, and David Hawking. 2007. Does brandname influence perceived search result quality? Yahoo!, Google, and WebKumara. In Proc. ADCS (short paper).
• Paul Thomas and David Hawking. 2007. Evaluating sampling methods for uncooperative collections. In Proc. SIGIR.
• Paul Thomas and Tom Rowlands. 2007. Estimating the value of automatic disambiguation. In Proc. SIGIR (poster).

Earlier

• Paul Thomas and David Hawking. 2006. Evaluation by comparing result sets in context. In Proc. CIKM.
• David Hawking and Paul Thomas. 2005. Server selection methods in hybrid portal search. In Proc. SIGIR.
• Mingfang Wu, Paul Thomas, and David Hawking. 2005. TREC 14 enterprise track at CSIRO and ANU. In Proc. TREC.
• Craig Wills and Paul Thomas. 1999. Exploiting a network charging model to reduce web costs. In Proc. AusWeb.

Data

• Division of .GOV and .GOV2 corpora by server: hosts in .GOV and .GOV2; id-to-host map for .GOV and .GOV2.
• List of homepages in .GOV.
• Search interface finder run over the above.

Etc

• Talk at SIGIR, 2009-07-21: "SUSHI: Scoring scaled samples for server selection".
• Seminar at MSR, 2009-07-13: "Information retrieval for real-world tasks". (Also: a very similar talk at ANU, 2010-03-04.)
• HAIL seminar, 2008-09-09: "Relevance assessment...".
• Seminar at USyd, 2008-07-02 (also a version at MSR, 2008-10-20, and Sheffield, 2008-10-23): "Personal metasearch".
• HAIL seminar, 2008-07-01: "Evaluating information retrieval in context".
• Poster at SIGIR, 2007-07-24.
• Seminar at ANU, 2007-05-14: "Sampling random documents from uncooperative search engines".
• Seminar at UMass, 2006-11-03 (also at CMU, 2006-11-10): "Personal metasearch".
• Seminar at Glasgow, 2006-10-23 (also at ANU, 2006-10-04): "Evaluating information retrieval in context".
• HCSnet next-generation search workshop, 2006-09-21.
• Poster at ANU open day, 2006-08-26.
• Personal metasearch. ACM SIGIR 2006 doctoral consortium. (Also: talk from the consortium, notes from the talk.)
• Poster at HCSNet, 2005-12-13.
• Seminar at ANU, 2005-10-12 (also at USyd, 2005-11-16, and CSIRO ICT Centre, Sydney, 2005-11-17): "Problems in personal information retrieval".

Contact

paul.thomas@csiro.au