Filters








6,795 Hits in 4.4 sec

Utilizing Focused Relevance Feedback

Elinor Brondwine, Anna Shtok, Oren Kurland
2016 Proceedings of the 39th International ACM SIGIR conference on Research and Development in Information Retrieval - SIGIR '16  
We present a novel study of ad hoc retrieval methods utilizing document-level relevance feedback and/or focused relevance feedback; namely, passages marked as (non-)relevant.  ...  The second method fuses retrieval scores produced by using relevant and non-relevant information separately.  ...  This paper is based upon work supported in part by the German Research Foundation (DFG) via the German-Israeli Project Cooperation (DIP, grant DA 1600/1-1), the Israel Science Foundation under grant no  ... 
doi:10.1145/2911451.2914695 dblp:conf/sigir/BrondwineSK16 fatcat:2nu4luba3bhoxnyhllnlkuy4pq

Artefact Retrieval: Overview of NLP Models with Knowledge Base Access [article]

Vilém Zouhar, Marius Mosbach, Debanjali Biswas, Dietrich Klakow
2022 arXiv   pre-print
In this paper, we systematically describe the typology of artefacts (items retrieved from a knowledge base), retrieval mechanisms and the way these artefacts are fused into the model.  ...  A lot of research has been devoted to devising and improving the way the knowledge base is accessed and incorporated into the model, resulting in a number of mechanisms and pipelines.  ...  utilize late fusion nor training-time knowledge base sources yet.  ... 
arXiv:2201.09651v1 fatcat:aeb5ljur4bbbvezgmtx6n3hxs4

A robust approach to optimizing multi-source information for enhancing genomics retrieval performance

Qinmin Hu, Jimmy Huang, Jun Miao
2011 BMC Bioinformatics  
Results: In the proposed approach, we first consider a common scenario for a metasearch system that has access to multiple baselines with retrieving and ranking documents/passages by their own models.  ...  These conclusions are very useful to direct the fusion work in the field of biomedical information retrieval.  ...  This motives us to consider a multi-source fusion approach in a metasearch system to utilize these meta-features.  ... 
doi:10.1186/1471-2105-12-s5-s6 pmid:21989123 pmcid:PMC3226256 fatcat:d3gstpmdbvaf3jjqvjht5yvuga

Concept-Based Information Retrieval Using Explicit Semantic Analysis

Ofer Egozi, Shaul Markovitch, Evgeniy Gabrilovich
2011 ACM Transactions on Information Systems  
Keyword-based retrieval may return inaccurate and incomplete results when different keywords are used to describe the same concept in the documents and in the queries.  ...  Information retrieval systems traditionally rely on textual keywords to index and retrieve documents.  ...  We show that combining concept-based relevancy of documents with that of passages in these documents performs best for ESA-based retrieval.  ... 
doi:10.1145/1961209.1961211 fatcat:tnco7za2vvedlhbksea4dy6lvi

RRF102: Meeting the TREC-COVID Challenge with a 100+ Runs Ensemble [article]

Michael Bendersky and Honglei Zhuang and Ji Ma and Shuguang Han and Keith Hall and Ryan McDonald
2020 arXiv   pre-print
In this paper, we report the results of our participation in the TREC-COVID challenge.  ...  from (a) lexical and semantic retrieval systems, (b) pre-trained and fine-tuned BERT rankers, and (c) relevance feedback runs.  ...  Acknowledgments We thank the TREC-COVID challenge organizers for their tremendous investment of time and effort in smooth execution of all the evaluation rounds.  ... 
arXiv:2010.00200v1 fatcat:wptmhsg3rnfatlf66qhgtsjml4

Combining Semantics, Context, and Statistical Evidence in Genomics Literature Search

Jay Urbain, Nazli Goharian, Ophir Frieder
2007 2007 IEEE 7th International Symposium on BioInformatics and BioEngineering  
Our results demonstrate improved search results in the presence of varying levels of semantic evidence, and higher performance using retrieval functions that combine document as well as sentence and passage  ...  When ranking documents based on the most relevant extracted passages, the results exceed the state-of-the-art by 13.89% as assessed by the TREC 2005 Genomics track collection of 4.5 million MEDLINE citations  ...  ., utilized a rule-based information extraction technique for identifying gene names in text [18] .  ... 
doi:10.1109/bibe.2007.4375738 dblp:conf/bibe/UrbainGF07 fatcat:axaenauynfcwxb5fkd45ljgc4q

From ``Identical'' to ``Similar'': Fusing Retrieved Lists Based on Inter-Document Similarities

A. Khudyak Kozorovitsky, O. Kurland
2011 The Journal of Artificial Intelligence Research  
Methods for fusing document lists that were retrieved in response to a query often utilize the retrieval scores and/or ranks of documents in the lists.  ...  We present a novel fusion approach that is based on using, in addition, information induced from inter-document similarities.  ...  This paper is based upon work supported in part by the Israel Science Foundation under grant no. 557/09, and by IBM's SUR award.  ... 
doi:10.1613/jair.3214 fatcat:aoqjbr7fivdztncvusdopxqwau

Learning Asymmetric Co-Relevance

Fiana Raiber, Oren Kurland, Filip Radlinski, Milad Shokouhi
2015 Proceedings of the 2015 International Conference on Theory of Information Retrieval - ICTIR '15  
Empirical evaluation demonstrates the merits of using the co-relevance estimate in various applications, including cluster-based and graph-based document retrieval.  ...  Several applications in information retrieval rely on asymmetric co-relevance estimation; that is, estimating the relevance of a document to a query under the assumption that another document is relevant  ...  This work has also been supported in part by Microsoft Research through its Ph.D. Scholarship Program.  ... 
doi:10.1145/2808194.2809454 dblp:conf/ictir/RaiberKRS15 fatcat:ywfesjip2fedzmdzhmwcw6j254

Kinship contextualization

Muhammad A. Norozi, Paavo Arvola
2013 Proceedings of the 36th international ACM SIGIR conference on Research and development in information retrieval - SIGIR '13  
In this study we hypothesize that the context of an XML-element originated from its preceding and following elements in the sequential ordering of a document improves the quality of retrieval.  ...  Utilizing this context in scoring is called contextualization.  ...  This organization of document gives an intuitive starting point for manipulating text passages at the established hierarchy levels of text documents.  ... 
doi:10.1145/2484028.2484111 dblp:conf/sigir/NoroziA13 fatcat:iwkcznwt2fbdrjqomwlpm77i2a

The cluster hypothesis in information retrieval

Oren Kurland
2013 Proceedings of the 36th international ACM SIGIR conference on Research and development in information retrieval - SIGIR '13  
, fusion, federated search, query expansion, microblog retrieval, relevance feedback, adversarial search • Concluding notes The ad hoc retrieval task • Ranking the documents in a corpus by their relevance  ...  for ad hoc information retrieval • Graph-based methods for ad hoc retrieval that utilize inter-document similarities • Additional tasks/applications • Search results visualization, query-performance prediction  ...  of relevant documents in cluster c of k documents for which ; is the highest is the precision@k attained by cluster-based retrieval with strategy #1.  ... 
doi:10.1145/2484028.2484192 dblp:conf/sigir/Kurland13 fatcat:3lw6kkfzs5gsvmsvkpwlynck2u

Exploring Classic and Neural Lexical Translation Models for Information Retrieval: Interpretability, Effectiveness, and Efficiency Benefits [article]

Leonid Boytsov, Zico Kolter
2021 arXiv   pre-print
We study the utility of the lexical translation model (IBM Model 1) for English text retrieval, in particular, its neural variants that are trained end-to-end.  ...  Using Model 1 we produced best neural and non-neural runs on the MS MARCO document ranking leaderboard in late 2020.  ...  In that, the sparsified model-executed on a CPU-is more than 10 3 times faster than BERT-based rankers, which run on a GPU. It is 5 × 10 3 × faster in the case of passage retrieval.  ... 
arXiv:2102.06815v2 fatcat:bg74b25ks5e4lk7za25j6s6ace

Contextualization using hyperlinks and internal hierarchical structure of Wikipedia documents

Muhammad Ali Norozi, Paavo Arvola, Arjen P. de Vries
2012 Proceedings of the 21st ACM international conference on Information and knowledge management - CIKM '12  
We tested several variants of contextualization and verified notable improvements in comparison with the baseline system and gold standards in the retrieval of full documents and focused elements.  ...  This paper proposes various methods of accumulating evidence from the context, and measures the effect of contextual evidence on retrieval effectiveness for document and focused retrieval of hyperlinked  ...  Aside evaluating element retrieval, passage retrieval evaluation is also supported in INEX. In this study we use the data provided by the 2009 INEX ad-hoc track.  ... 
doi:10.1145/2396761.2396855 dblp:conf/cikm/NoroziAV12 fatcat:gmpjpuxrm5b5hljmsgdp6sf7di

Relevance feedback with too much data

James Allan
1995 Proceedings of the 18th annual international ACM SIGIR conference on Research and development in information retrieval - SIGIR '95  
no other documents contain the term, so it cannot be useful in retrieving additional relevant documents. judged.Y We believe, nonetheless, that it is important to consider feedback methods that utilize  ...  (The "data fusion" effect is noticed here just as it was with passage feedback.)  ... 
doi:10.1145/215206.215380 dblp:conf/sigir/Allan95 fatcat:54a7fcyqsrcphahneg6vgo6uvi

Pruning the Index Contents for Memory Efficient Open-Domain QA [article]

Martin Fajcik, Martin Docekal, Karel Ondrej, Pavel Smrz
2021 arXiv   pre-print
Furthermore, previous work often comes with a massive index of external documents that scales in the order of tens of GiB.  ...  Specifically, it proposes the novel R2-D2 (Rank twice, reaD twice) pipeline composed of retriever, passage reranker, extractive reader, generative reader and a simple way to combine them.  ...  The second reader is based on Fusion-In-Decoder (FiD) (Izacard and Grave, 2020b).  ... 
arXiv:2102.10697v2 fatcat:rbkfghoainb37asci2etnhyzse

Discriminating capabilities of syllable-based features and approaches of utilizing them for voice retrieval of speech information in Mandarin Chinese

Berlin Chen, Hsin-min Wang, Lin-shan Lee
2002 IEEE Transactions on Speech and Audio Processing  
Good approaches for better utilizing such capabilities, including fusion with the word-and character-level information and improved approaches to obtain better syllable-based features and query expressions  ...  Index Terms-Confidence measure, retrieval of speech information, syllable-based features, term association matrix. the Ph.D. degree in electrical engineering from  ...  Based on this assumption, the database of speech documents to be retrieved as described in Section III were automatically divided into a total of 8152 passages based on the silence boundaries.  ... 
doi:10.1109/tsa.2002.802541 fatcat:jogtcvpn2vb5ranx44jf2czqiq
« Previous Showing results 1 — 15 out of 6,795 results