Filters








574 Hits in 1.7 sec

TREC and TIPSTER experiments with inquery

James P Callan, W.Bruce Croft, John Broglio
1995 Information Processing & Management  
INQUERY is a probablistic information retrieval system based upon a Bayesian inference network model.  ...  This paper describes recent improvements to the system as a result of participation in the TIPSTER project and the TREC-2 conference.  ...  This research was partially supported by the NSF Center for Intelligent Information Retrieval at the University of Massachusetts, Amherst.  ... 
doi:10.1016/0306-4573(94)00050-d fatcat:byry5vcvfbgetpxvzyfjc7z6oy

Similarity searching in the CORDIS text database

Euripides G. M. Petrakis, Kostas Tzeras
2000 Software, Practice & Experience  
Bruce Croft at University of Massachusetts, Joel Remde at Bellcore Communications Research Inc. and Chris Buckley at Cornell University for kindly providing us with the software for INQUERY, LSI and SMART  ...  Karin Mölle, employee at GMD-IPSI, for her valuable contribution in this work. Mrs.  ...  Figure 5 : 5 Precision-Recall diagram of SMART, LSI and INQUERY for retrievals on field GEN INFORMATION. Figure 5 5 illustrates the precision-recall diagram for searching on field GEN INFORMATION.  ... 
doi:10.1002/1097-024x(20001110)30:13<1447::aid-spe344>3.3.co;2-3 fatcat:agdotbipzfghnjtrd53tbphpba

Modeling score distributions for combining the outputs of search engines

R. Manmatha, T. Rath, F. Feng
2001 Proceedings of the 24th annual international ACM SIGIR conference on Research and development in information retrieval - SIGIR '01  
Experiments show that this model fits TREC-3 and TREC-4 data for not only probabilistic search engines like INQUERY but also vector space search engines like SMART for English.  ...  ACKNOWLEDGEMENTS The origins of this work can be traced back to a discussion at the University of Glasgow.  ...  We will show examples of the modeling on queries from INQUERY and SMART from TREC-3.  ... 
doi:10.1145/383952.384005 dblp:conf/sigir/ManmathaRF01 fatcat:op2n2arpvrdndlb6y5s3epsnia

Similarity searching in the CORDIS text database

Euripides G. M. Petrakis, Kostas Tzeras
2000 Software, Practice & Experience  
Bruce Croft at University of Massachusetts, Joel Remde at Bellcore Communications Research Inc. and Chris Buckley at Cornell University for kindly providing us with the software for INQUERY, LSI and SMART  ...  Karin Mölle, employee at GMD-IPSI, for her valuable contribution in this work. Mrs.  ...  Figure 5 : 5 Precision-Recall diagram of SMART, LSI and INQUERY for retrievals on field GEN INFORMATION. Figure 5 5 illustrates the precision-recall diagram for searching on field GEN INFORMATION.  ... 
doi:10.1002/1097-024x(20001110)30:13<1447::aid-spe344>3.0.co;2-c fatcat:zyvbsywb4zb2dg6p2vz4aa7zkq

The effect multiple query representations on information retrieval system performance

Nicholas J. Belkin, C. Cool, W. Bruce Croft, James P. Callan
1993 Proceedings of the 16th annual international ACM SIGIR conference on Research and development in information retrieval - SIGIR '93  
These different formulations were grouped, and the groups, and combinations of them, were used as searches against the TREC test collection, using the INQUERY probabilistic inference network retrieval  ...  Five independently generated Boolean query formulations for ten different TREC topics were produced by ten different expert online searchers.  ...  combination of groups 1-5, and looked at retrieval performance.  ... 
doi:10.1145/160688.160760 dblp:conf/sigir/BelkinCCC93 fatcat:ghw3ahiaavd7loqayktbwpayzm

Iterative exploration, design and evaluation of support for query reformulation in interactive information retrieval

N.J Belkin, C Cool, D Kelly, S.-J Lin, S.Y Park, J Perez-Carballo, C Sikora
2001 Information Processing & Management  
We report on the progressive investigation of techniques for supporting interactive query reformulation in the TREC Interactive Track task.  ...  discuss what we believe has been learned through this series of investigations about effective term suggestion, usable and useful interface design, and the relationships between these two in support of the TREC  ...  Understanding, Use and Usefulness of Relevance Feedback in RU-INQUERY(1=Not at All; 5=A Great Deal) * One case missing.  ... 
doi:10.1016/s0306-4573(00)00055-8 fatcat:as2xccuojncytbqq3bj5nl433m

Passage retrieval based on language models

Xiaoyong Liu, W. Bruce Croft
2002 Proceedings of the eleventh international conference on Information and knowledge management - CIKM '02  
Results from the INQUERY search engine, which is not based on a language modeling approach, are also given for comparison.  ...  TREC-45 is a heterogeneous collection that is composed of all data from TREC disk 4 and 5. Statistics of the test collections are given in table 2.  ...  INQUERY consistently achieves its best passage retrieval results with window size 350. On AP and TREC-45, LM behaves similarly to INQUERY while RM achieves its best results with window size 150.  ... 
doi:10.1145/584792.584854 dblp:conf/cikm/LiuC02 fatcat:folf4uegrvcy5jiseudlvfevb4

Page 57 of Library & Information Science Abstracts Vol. , Issue 7 [page]

1995 Library & Information Science Abstracts  
In the Ist project 5 separate Boolean queries for each of the 50 TREC routing pe eee ge arty a er Using the INQUERY retrieval system, these queries were combined, and used to produce 5 separate retrieval  ...  Original abstract-amended. 8119 Large test collection experiments of na operational, interactive system: OKAPI at TREC. S. E. Robertson, S. Walker M. M. Hancock-Beaulieu.  ... 

Passage retrieval based on language models

Xiaoyong Liu, W. Bruce Croft
2002 Proceedings of the eleventh international conference on Information and knowledge management - CIKM '02  
Results from the INQUERY search engine, which is not based on a language modeling approach, are also given for comparison.  ...  TREC-45 is a heterogeneous collection that is composed of all data from TREC disk 4 and 5. Statistics of the test collections are given in table 2.  ...  INQUERY consistently achieves its best passage retrieval results with window size 350. On AP and TREC-45, LM behaves similarly to INQUERY while RM achieves its best results with window size 150.  ... 
doi:10.1145/584853.584854 fatcat:lwy3dn7m4ng6jgswfakbyrcmsm

Applications of Score Distributions in Information Retrieval [chapter]

R. Manmatha
2003 Language Modeling for Information Retrieval  
This model fits not only probabilistic search engines like INQUERY but also vector space search engines like SMART and also LSI search engines.  ...  Figure 5 shows recall-precision plots for combining INQUERY and SMART on TREC-3 data.  ...  The model has been shown to work for a large number of search engines on TREC-3 and TREC-4 data including probabilistic search engines like INQUERY and vector space search engines like SMART.  ... 
doi:10.1007/978-94-017-0171-6_8 fatcat:ya6dxlbfqbawplrlgbyclrob3i

Structured queries, language modeling, and relevance modeling in cross-language information retrieval

Leah S. Larkey, Margaret E. Connell
2005 Information Processing & Management  
Two probabilistic approaches to cross-lingual retrieval are in wide use today, those based on probabilistic models of relevance, as exemplified by INQUERY, and those based on language modeling.  ...  INQUERY, as a query net model, allows the easy incorporation of query operators, including a synonym operator, which has proven to be extremely useful in cross-language information retrieval (CLIR), in  ...  2001 TREC 2002 TREC 4 TREC 5 Corpus Language Arabic Arabic Spanish Spanish Number of Documents 383,872 383,872 57,868 172,823 Size of Corpus (MB) 567 567 200 336 Mean doc len (words  ... 
doi:10.1016/j.ipm.2004.06.008 fatcat:b7phj2x7jfbtjkef7rj3gama6m

Accurate user directed summarization from existing tools

Mark Sanderson
1998 Proceedings of the seventh international conference on Information and knowledge management - CIKM '98  
The summarizer that was evaluated was based on a set of existing statistical techniques that had been applied successfully to the INQUERY retrieval system.  ...  Rank 5 10 20 50 100 250 500 1000 R (%) 64 62 55 56 53 54 50 47 P (%) 78 81 80 79 76 76 75 79 F (%) 70 70 65 66 62 63 60 59Table 5: Changes in summary accuracy at different This material is based on work  ...  In both SUMMAC and TREC, a document was judged relevant if at least one sentence within it was regarded as relevant to the topic.  ... 
doi:10.1145/288627.288640 dblp:conf/cikm/Sanderson98 fatcat:tdj5wdztkrahnl33wbrng63m6a

Comparing the performance of database selection algorithms

James C. French, Allison L. Powell, Jamie Callan, Charles L. Viles, Travis Emmitt, Kevin J. Prey, Yun Mou
1999 Proceedings of the 22nd annual international ACM SIGIR conference on Research and development in information retrieval - SIGIR '99  
The testbed is a decomposition of the TREC TIPSTER data into 236 subcollections.  ...  The databases from our testbed were ranked using both the gGlOSS and CORI techniques and compared to a baseline derived from TREC relevance judgements.  ...  Figure 4 :Figure 5 : 45 Figures 4, 5 and 6 also show h o w all of the approaches discussed here approximate RBR.  ... 
doi:10.1145/312624.312684 dblp:conf/sigir/FrenchPCVEPM99 fatcat:zuuo7k63kbdv5l5tjgdpxk4wri

Partial replica selection based on relevance for information retrieval

Zhihong Lu, Kathryn S. McKinley
1999 Proceedings of the 22nd annual international ACM SIGIR conference on Research and development in information retrieval - SIGIR '99  
Long queries are automatically created from TREC topics using InQuery query generation techniques 5 , which consist of terms, phrases and proximity operators.  ...  E ectiveness of Partial Replica Selection In this section, we e v aluate the e ectiveness of our replica selection approach using the InQuery retrieval system 5 , and the 2 GB TREC volumes 2+3 collection  ... 
doi:10.1145/312624.312662 dblp:conf/sigir/LuM99 fatcat:vppfmxqnmfgobdpg5pnt7d76e4

Robust result merging using sample-based score estimates

Milad Shokouhi, Justin Zobel
2009 ACM Transactions on Information Systems  
We use precision at 5 and 10 for comparing the performance of merging algorithms.  ...  TREC topics 201-250 (long) and their relevance judgments have been used as queries. Collections use an identical retrieval model (INQUERY).  ... 
doi:10.1145/1508850.1508852 fatcat:m5ergw3svvaelncjjrtpm7suxi
« Previous Showing results 1 — 15 out of 574 results