Filters








47 Hits in 8.7 sec

Microsoft Cambridge at TREC 2002: Filtering Track

Stephen E. Robertson, Steve Walker, Hugo Zaragoza, Ralf Herbrich
2002 Text Retrieval Conference  
Adaptive Filtering Okapi systems At the Microsoft Research laboratory in Cambridge, we are developing an evaluation environment for a wide range of information retrieval experiments.  ...  Okapi at TRECs 1-10 A summary of the contributions to TRECs 1-7 by the Okapi team, first at City University London and then at Microsoft, is presented in [6] .  ...  For TREC 2002, we developed a new Okapi/Keenbow component called the Basic Filtering Dogsbody (BFD).  ... 
dblp:conf/trec/RobertsonWZH02 fatcat:ygznvl3q3rbhxe4w5rlyf6egxy

Microsoft Cambridge at TREC-12: HARD track

Stephen E. Robertson, Hugo Zaragoza, Michael J. Taylor
2003 Text Retrieval Conference  
We took part in the HARD track, with an active learning method to choose which document snippets to show the user for relevance feedback (compared to baseline feedback using snippets from the top-ranked  ...  2002 filtering topics.  ...  Overview The present team at Microsoft Cambridge may be regarded as the descendant of the Okapi team, working first from City University London and then from Microsoft.  ... 
dblp:conf/trec/RobertsonZT03 fatcat:fxj4akgifvcuhmngq2qfqkxrr4

The TREC 2002 Filtering Track Report

Stephen E. Robertson, Ian Soboroff
2002 Text Retrieval Conference  
The TREC-11 filtering track measures the ability of systems to build persistent user profiles which successfully separate relevant and non-relevant documents in an incoming stream.  ...  This report describes the track, presents some evaluation results, and provides a general commentary on lessons learned from this year's track.  ...  Acknowledgements We give our thanks to all the people who have contributed to the development of the TREC filtering track over the years, in particular David Lewis, David Hull, Karen Sparck Jones, Chris  ... 
dblp:conf/trec/RobertsonS02 fatcat:fwhyepunqfgndikadr55ztwbwu

Toshiba KIDS at NTCIR-3: Japanese and English-Japanese IR

Tetsuya Sakai, Makoto Koyama, Masaru Suzuki, Toshihiko Manabe
2002 NTCIR Conference on Evaluation of Information Access Technologies  
Toshiba participated in the Japanese monolingual track and the English-Japanese cross-language track at NTCIR-3, and achieved the highest retrieval performances among the DESCRIPTION runs.  ...  At NTCIR-2, Toshiba (the first author) collaborated with Microsoft Research Cambridge to participate in the same track [10] .  ...  Introduction Toshiba participated in the Japanese monolingual track and the English-Japanese cross-language track at NTCIR-3. Three automatic DESCRIPTION runs were submitted to each track.  ... 
dblp:conf/ntcir/SakaiKSM02 fatcat:feoa3rk72zcdlop7plvuurwkyu

Data-Intensive Question Answering

Eric Brill, Jimmy J. Lin, Michele Banko, Susan T. Dumais, Andrew Y. Ng
2001 Text Retrieval Conference  
There is a separate report in this volume on the Microsoft Research Cambridge submissions for the filtering and Web tracks (Robertson et al., 2002) .  ...  Introduction Microsoft Research Redmond participated for the first time in TREC this year, focusing on the question answering track.  ... 
dblp:conf/trec/BrillLBDN01 fatcat:nxcjqfhxsvgwnn6jcl7scpskeu

Margin-based local regression for adaptive filtering

Yiming Yang, Bryan Kisiel
2003 Proceedings of the twelfth international conference on Information and knowledge management - CIKM '03  
Examining this approach together with a Rocchio-style classifier on the TREC 2001 and TREC 2002 benchmark data sets in adaptive filtering, we obtained significant improvements in performance (measured  ...  using F β = 0.5) over the baseline system that did not adapt the threshold over time, and the best result ever reported on the TREC 2002 benchmark corpus for adaptive filtering evaluations.  ...  These approaches yielded strong performance for the Microsoft/Cambridge system in the TREC-2000 evaluations for adaptive filtering, but not nearly as strong the results in the subsequent years of TREC.  ... 
doi:10.1145/956900.956902 fatcat:2y654pzrqrecxp6qxxbhpo23fi

Margin-based local regression for adaptive filtering

Yiming Yang, Bryan Kisiel
2003 Proceedings of the twelfth international conference on Information and knowledge management - CIKM '03  
Examining this approach together with a Rocchio-style classifier on the TREC 2001 and TREC 2002 benchmark data sets in adaptive filtering, we obtained significant improvements in performance (measured  ...  using F β = 0.5) over the baseline system that did not adapt the threshold over time, and the best result ever reported on the TREC 2002 benchmark corpus for adaptive filtering evaluations.  ...  These approaches yielded strong performance for the Microsoft/Cambridge system in the TREC-2000 evaluations for adaptive filtering, but not nearly as strong the results in the subsequent years of TREC.  ... 
doi:10.1145/956863.956902 dblp:conf/cikm/YangK03 fatcat:ctyh6rng7zcshl54a2ni6kiniu

Indexing, browsing, and searching of digital video

Alan F. Smeaton
2005 Annual Review of Information Science and Technology  
The VMR application is a retrieval system based on the VMR video mail retrieval project at Cambridge University.  ...  TREC and the video retrieval track within TREC are described in more detail in the next section of this chapter.  ... 
doi:10.1002/aris.1440380109 fatcat:lpwg36agnze4vl7ehkjuiv3pbu

CMIC@TREC 2009: Relevance Feedback Track

Kareem Darwish, Ahmed El-Deeb
2009 Text Retrieval Conference  
This paper describes CMIC's submissions to the TREC'09 relevance feedback track.  ...  Both techniques attempt to topically diversify these 5 documents as much as possible in an effort to maximize the probability that they contain at least 1 relevant document.  ...  The IR group at Microsoft Research, Cambridge, kindly provided us with 2,500 search results for each query [4] .  ... 
dblp:conf/trec/DarwishE09 fatcat:qusjixara5htpfqfc63opup2ae

Report on INEX 2009

T. Beckers, S. Geva, W.-C. Huang, T. Iofciu, J. Kamps, G. Kazai, M. Koolen, S. Kutty, M. Landoni, M. Lehtonen, V. Moriceau, P. Bellot (+17 others)
2010 SIGIR Forum  
Link-the-Wiki Track Investigating link discovery between Wikipedia documents, both at the file level and at the element level.  ...  (Section 4), the Entity Ranking track (Section 5), the Interactive track (Section 6), the QA track (Section 7), the Link the Wiki track (Section 8), and the XML Mining track (Section 9).  ...  Relevance assessments were collected using the Book Search System, available at http: //www.booksearch.org.uk, developed by Microsoft Research Cambridge, which allowed participants to search, browse, read  ... 
doi:10.1145/1842890.1842897 fatcat:46evgkszirdm3grr6fqrtuyqtm

Statistical source expansion for question answering

Nico Schlaefer, Jennifer Chu-Carroll, Eric Nyberg, James Fan, Wlodek Zadrozny, David Ferrucci
2011 Proceedings of the 20th ACM international conference on Information and knowledge management - CIKM '11  
and 67% on TREC questions. iv First of all, I would like to thank my advisor Eric Nyberg for his support and guidance throughout my studies at Carnegie Mellon.  ...  questions and from 59% to 64% on TREC factoid questions.  ...  In the TREC 2002 Novelty track [Harman, 2002] , the approach outperformed a query expansion algorithm that augmented query terms with their hyponyms.  ... 
doi:10.1145/2063576.2063632 dblp:conf/cikm/SchlaeferCNFZF11 fatcat:whoy62klazctbdo4p57wbevkdu

Processing Social Media Messages in Mass Emergency: A Survey [article]

Muhammad Imran, Carlos Castillo, Fernando Diaz, Sarah Vieweg
2015 arXiv   pre-print
These challenges can be mapped to classical information processing operations such as filtering, classifying, ranking, aggregating, extracting, and summarizing.  ...  Processing social media messages to obtain such information, however, involves solving multiple challenges including: handling information overload, filtering credible information, and prioritizing different  ...  The performance is evaluated under experimental conditions different from the TREC track, making it difficult to compare with other results.  ... 
arXiv:1407.7071v3 fatcat:e7mcvae5freddaus7ndolygeti

Perceiving and Using Genre by Form – An Eye-Tracking Study

Malcolm Clark, Ian Ruthven, Patrik O'Brian Holt
2010 Libri  
Perceiving and using genre by form -an eye-tracking study. Available from OpenAIR@RGU. [online].  ...  Perceiving and using genre by form -an eye-tracking study. Libri, 60 (3), pp. 268-280. This paper reports on our approach to the analysis of genre recognition using eyetracking.  ...  Without the kind permission to use their eye-tracking equipment we could not have collected such useful data as we have for this study.  ... 
doi:10.1515/libr.2010.023 fatcat:4g6yczurgnh5nnaferm3ncxwpm

Search the Audio, Browse the Video—A Generic Paradigm for Video Collections

Arnon Amir, Savitha Srinivasan, Alon Efrat
2003 EURASIP Journal on Advances in Signal Processing  
The amount of digital video being shot, captured, and stored is growing at a rate faster than ever before.  ...  The SDR track was followed by the current video track, started in 2001, with emphasis on CBIR [14] . In 2002, the second video track promoted a new approach to video search.  ...  Examples of challenging topics and queries can be found in the search task of the NIST TREC first video track, held in 2001.  ... 
doi:10.1155/s111086570321012x fatcat:wpdlika5wjdu7dnk635s64gcwq

Introduction to information retrieval

2009 ChoiceReviews  
Boldi et al. (2002) and Shkapenyuk and Suel (2002) provide more recent details of implementing large-scale distributed web crawlers.  ...  The Robots Exclusion Protocol standard is described at http://www.robotstxt.org/wc/exclusion.html.  ...  A number of recent evaluations, including INEX, some TREC tracks, and NCTIR have adopted a graded notion of relevance with documents divided into 3 or 4 classes, distinguishing slightly relevant documents  ... 
doi:10.5860/choice.46-2715 fatcat:ruwoe46pgzcupjygnwbnit4z3u
« Previous Showing results 1 — 15 out of 47 results