Query Transformations for Result Merging

Abstract : This paper describes Carnegie Mellon University's entry at the TREC 2014 Federated Web Search track (FedWeb14). Federated search pipelines typically have two components: (i) resource-selection, and (ii) result-merging. This work documents experiments to modify queries to merge results in the federated-search pipeline. Approaches from previous attempts at solving this problem involved custom query- document similarity scores or rank-combination methods. In this document, we explore how term-dependence models and query expansion strategies influence result-merging.

[1]  W. Bruce Croft,et al.  A Markov random field model for term dependencies , 2005, SIGIR '05.

[2]  Quoc V. Le,et al.  Distributed Representations of Sentences and Documents , 2014, ICML.

[3]  Jeffrey Dean,et al.  Efficient Estimation of Word Representations in Vector Space , 2013, ICLR.

[4]  Yue Liu,et al.  ICTNET at Federated Web Search Track 2014 , 2014, TREC.

[5]  Jeffrey Dean,et al.  Distributed Representations of Words and Phrases and their Compositionality , 2013, NIPS.

[6]  Massimo Melucci,et al.  University of Padua at TREC 2014: Federated Web Search Track , 2014, TREC.

[7]  João Magalhães,et al.  NovaSearch at TREC 2013 Federated Web Search Track: Experiments with rank fusion , 2013, TREC.

[8]  CHENGXIANG ZHAI,et al.  A study of smoothing methods for language models applied to information retrieval , 2004, TOIS.

[9]  Luo Si,et al.  A semisupervised learning method to merge search engine results , 2003, TOIS.

[10]  Yoshua Bengio,et al.  Neural Probabilistic Language Models , 2006 .

[11]  Milad Shokouhi,et al.  Federated Search , 2011, Found. Trends Inf. Retr..

[12]  Mandar Mitra,et al.  ISI at the TREC 2013 Federated task , 2013, TREC.

[13]  Yoshua Bengio,et al.  A Neural Probabilistic Language Model , 2003, J. Mach. Learn. Res..

[14]  Hugo Zaragoza,et al.  The Probabilistic Relevance Framework: BM25 and Beyond , 2009, Found. Trends Inf. Retr..

[15]  Jimmy J. Lin,et al.  CWI and TU Delft at TREC 2013: Contextual Suggestion, Federated Web Search, KBA, and Web Tracks , 2013 .

[16]  Milad Shokouhi,et al.  Robust result merging using sample-based score estimates , 2009, TOIS.

[17]  Richard A. Harshman,et al.  Indexing by Latent Semantic Analysis , 1990, J. Am. Soc. Inf. Sci..

[18]  Djoerd Hiemstra,et al.  Overview of the TREC 2014 Federated Web Search Track , 2013, TREC.

[19]  Lukás Burget,et al.  Recurrent neural network based language model , 2010, INTERSPEECH.