Abstract

The TREC-2004 Genomics track evaluation experiments at Patolis Corporation are described with a focus on the document length issues in different retrieval models such as TF*IDF or probabilistic language modeling approaches. In the genomics ad hoc retrieval task, combination of pseudo-relevance feedback and reference database feedback is applied. For the triage sub-task, we trained a SVM classifier using leave-one-out-cross-validation, and calibrated parameters to be optimal against the training set.

Citations

1086	Making large-scale SVM learning practical - Joachims - 1999
728	Relevance feedback in information retrieval - Rocchio - 1971
684	A language modeling approach to information retrieval - Ponte, Croft - 1998
498	A study of smoothing methods for language models applied to information retrieval - Zhai, Lafferty
370	Okapi at trec-3 - Robertson, Walker, et al. - 1994
313	M.: Pivoted document length normalization - Singhal, Buckley, et al. - 1996
299	Learning to Classify Text Using Support Vector Machines. Methods, Theory and Algorithms - Joachims - 2002
289	Some simple effective approximations to the 2-poisson model for probabilistic weighted retrieval - Robertson, Walker - 1994
220	Information retrieval as statistical translation - Berger, Lafferty - 1999
158	A Hidden Markov Model Information Retrieval System - Miller, Leek, et al. - 1999
101	Twenty-One at TREC-7: Ad-hoc and Cross-language Track - Hiemstra, Kraaij - 1999
94	Experiments using the Lemur toolkit - Ogilvie, Callan - 2001
53	Model-based feedback in the kl-divergence retrieval model - Zhai, Lafferty
21	Applying support vector machines to the TREC2001 batch filtering and routing tasks - Lewis - 2001
18	Overview of Patent Retrieval Task at NTCIR-4 - Fujii, Iwayama, et al. - 2004
13	Tagging Gene and - Tababe, Wilbur - 2002
4	Overview of the third ntcir workshop - Kando - 2002
2	Notes on Phrasal Indexing---JSCB Evaluation Experiments at NTCIR - Fujita - 1999
1	Reflections on “Aboutness”— TREC-9 Evaluaton Experiments at Justsystem - Fujita
1	More reflections on “Aboutness”—TREC-2001 Evaluaton Experiments at Justsystem - Fujita
1	Revisiting the Document Length Hypotheses --NTCIR-4 CLIR and Patent Experiments at Patolis - Fujita