32,971 Hits in 4.6 sec

Simple questions to improve pseudo-relevance feedback results

Giridhar Kumaran, James Allan
2006 Proceedings of the 29th annual international ACM SIGIR conference on Research and development in information retrieval - SIGIR '06  
This improvement is also spread across more queries compared to ordinary pseudo-relevance feedback, as suggested by geometric mean average precision.  ...  We explore interactive methods to further improve the performance of pseudo-relevance feedback. Studies [4] suggest that new methods for tackling difficult queries are required.  ...  The pseudo-relevance feedback mechanism is based on relevance models [5] .  ... 
doi:10.1145/1148170.1148305 dblp:conf/sigir/KumaranA06 fatcat:vvhecobz3ffyho6pclqwa2zv2a

The Simplest Thing That Can Possibly Work: Pseudo-Relevance Feedback Using Text Classification [article]

Jimmy Lin
2019 arXiv   pre-print
being made, this paper tries to tackle the decades-old problem of pseudo-relevance feedback with "the simplest thing that can possibly work".  ...  I present a technique based on training a document relevance classifier for each information need using pseudo-labels from an initial ranked list and then applying the classifier to rerank the retrieved  ...  Also included are results from NPRF, a recently-proposed neural approach to pseudo-relevance feedback [12] (row 19).  ... 
arXiv:1904.08861v1 fatcat:xwhkt7xty5bzxitscl6uzkkp2a


Kiduk Yang, Kelly Maglaughlin
1999 Text Retrieval Conference  
In this light, We posed the following question. • Is the subcollection retrieval results using pseudo-relevance feedback competitive to the initial retrieval results using the whole collection?  ...  For the TREC-8 ad-hoc task, we merged results of pseudo-relevance feedback to subcollections as in TREC-7. Our results were consistent with that of TREC-7.  ... 
dblp:conf/trec/YangM99 fatcat:6eavtn55dfglhizvzdp3gyrm5i

Query Expansion based on Pseudo Relevance Feedback from Definition Clusters

Delphine Bernhard
2010 International Conference on Computational Linguistics  
We apply the technique of pseudo relevance feedback to obtain expansion terms from definition clusters.  ...  The main difficulty lies in identifying relevant expansion terms in order to prevent query drift.  ...  Local feedback methods, such as Terrier-KL, only bring minor improvements for easy questions, but lead to slightly lower results for medium and hard questions.  ... 
dblp:conf/coling/Bernhard10 fatcat:bgrhaty3mfdwvje5uywxb5sel4

Using Google Translation in Cross-Lingual Information Retrieval

Xiaoning He, Peidong Wang, Haoliang Qi, Muyun Yang, Guohua Lei, Yong Xue
2008 NTCIR Conference on Evaluation of Information Access Technologies  
We use KL-divergence model to perform retrieval and Chinese character bigram as our indexing unit. Pseudo feedback was used trying to improve average precision.  ...  We achieved competitive results in the task.  ...  In IR4QA, the goal is to evaluate how good an IR system is at returning documents that are relevant to the information needs on average, given a set of natural language questions or question analysis results  ... 
dblp:conf/ntcir/HeWQYLX08 fatcat:yrifbhl77zcqfmalslcc27r7d4

GQE-PRF: Generative Query Expansion with Pseudo-Relevance Feedback [article]

Minghui Huang, Dong Wang, Shuang Liu, Meizhen Ding
2021 arXiv   pre-print
on the pseudo-relevance feedback.  ...  Query expansion with pseudo-relevance feedback (PRF) is a powerful approach to enhance the effectiveness in information retrieval.  ...  query [SEP] pseudo-relevance feedback) as negative samples.  ... 
arXiv:2108.06010v1 fatcat:e3sjq7d345fqdmn2apazmibi6e

Accurately extracting coherent relevant passages using hidden Markov models

Jing Jiang, ChengXiang Zhai
2005 Proceedings of the 14th ACM international conference on Information and knowledge management - CIKM '05  
In this paper, we present a principled method for accurately extracting coherent relevant passages of variable lengths using HMMs.  ...  This improvement agrees with our hypothesis that if we use short, accurate passages as pseudo feedback to estimate the relevance language model, the HMM method can automatically extend the passages to  ...  Accurate passage extraction not only allows an IR system to precisely point to the most relevant parts of a document, but can also potentially improve document ranking and relevance feedback by using short  ... 
doi:10.1145/1099554.1099631 dblp:conf/cikm/JiangZ05 fatcat:3eeoe526tngbna7glevpaubfrm

Estimation of statistical translation models based on mutual information for ad hoc information retrieval

Maryam Karimzadehgan, ChengXiang Zhai
2010 Proceeding of the 33rd international ACM SIGIR conference on Research and development in information retrieval - SIGIR '10  
pseudo-relevance feedback to further improve retrieval accuracy.  ...  As a principled approach to capturing semantic relations of words in information retrieval, statistical translation models have been shown to outperform simple document language models which rely on exact  ...  Results with Pseudo-Relevance Feedback Both statistical translation model and pseudo-relevance feedback are to capture word associations, so it would be interesting to see whether they are essentially  ... 
doi:10.1145/1835449.1835505 dblp:conf/sigir/KarimzadehganZ10 fatcat:lcivnnmokvgbfpfxggm7skj26u

Regularized estimation of mixture models for robust pseudo-relevance feedback

Tao Tao, ChengXiang Zhai
2006 Proceedings of the 29th annual international ACM SIGIR conference on Research and development in information retrieval - SIGIR '06  
Pseudo-relevance feedback has proven to be an effective strategy for improving retrieval accuracy in all retrieval models.  ...  However the performance of existing pseudo feedback methods is often affected significantly by some parameters, such as the number of feedback documents to use and the relative weight of original query  ...  The basic idea of pseudo feedback is to assume a certain number of top-ranked documents are relevant and learn from these documents to improve the query representation, thus to improve retrieval accuracy  ... 
doi:10.1145/1148170.1148201 dblp:conf/sigir/TaoZ06 fatcat:hcwpp2sahbcdtprxeixggxolqa

A Re-examination of IR Techniques in QA System [chapter]

Yi Chang, Hongbo Xu, Shuo Bai
2005 Lecture Notes in Computer Science  
The performance of Information Retrieval in the Question Answering system is not satisfactory from our experiences in TREC QA Track.  ...  In this article, we take a comparative study to re-examine IR techniques on document retrieval and sentence level retrieval respectively.  ...  It is natural to expand the query with pseudo-relevant feedback methods. Pseudo-relevant feedback is an effective method to make up for lack of keywords in a query.  ... 
doi:10.1007/978-3-540-30211-7_8 fatcat:uhsnvk6l3jezngesfrpzwjfroi

Utilizing sub-topical structure of documents for information retrieval

Debasis Ganguly, Johannes Leveling, Gareth J.F. Jones
2011 Proceedings of the 4th workshop on Workshop for Ph.D. students in information & knowledge management - PIKM '11  
Feedback in ad hoc IR task is shown to benefit from the use of extracted sentences instead of terms from the pseudo relevant documents for query expansion.  ...  This is particularly useful for QA and snippet generation tasks where the objective is to aggregate relevant and novel information from multiple documents satisfying user information need on one hand,  ...  Improving PRF of ad-hoc retrieval In query expansion for pseudo relevance feedback (PRF) it is wrongly assumed that a pseudo relevant document as a whole is relevant, which is generally not true as shown  ... 
doi:10.1145/2065003.2065019 dblp:conf/cikm/GangulyLJ11 fatcat:bywv22zyynbklalkdlbzn2xs5q

Temporal feedback for tweet search with non-parametric density estimation

Miles Efron, Jimmy Lin, Jiyin He, Arjen de Vries
2014 Proceedings of the 37th international ACM SIGIR conference on Research & development in information retrieval - SIGIR '14  
We explore this question in the context of tweet search and temporal feedback: starting with an initial set of results from a baseline retrieval model, we estimate the temporal density of relevant documents  ...  Temporal feedback improves over standard lexical feedback (with and without human judgments), illustrating that temporal relevance signals exist independently of document content.  ...  judgments, as opposed to the pseudo-relevance feedback methods described above.  ... 
doi:10.1145/2600428.2609575 dblp:conf/sigir/EfronLHV14 fatcat:7keb47pfyrcozfaex47te42hae

Toward whole-session relevance

Karthik Raman, Paul N. Bennett, Kevyn Collins-Thompson
2013 Proceedings of the 36th international ACM SIGIR conference on Research and development in information retrieval - SIGIR '13  
Pushpak Bhattacharya Aim: Using representative information about irrelevant documents, improve performance and stability of query retrieval via pseudo-relevance feedback.  ...  Overview: • Analogous to the concept of pseudo -relevance, proposed the concept of pseudo-irrelevance (a first approximation to true irrelevance) and a method of obtaining such pseudo-irrelevant documents  ... 
doi:10.1145/2484028.2484089 dblp:conf/sigir/RamanBC13 fatcat:7d36xyvelfhffluh6qh3pomyby

Reflections on "Aboutness" TREC-9 Evaluation Experiments at Justsystem

Sumio Fujita
2000 Text Retrieval Conference  
The results show that pesudo-relevance feedback is always effective while reference database feedback is effective only with very short queries.  ...  Experiments on the effects of supplemental noun phrase indexing, pseudo-relevance feedback and reference database feedback in view of the effect of various length of queries are reported.  ...  ACKNOWLEDGMENTS Our thanks to Mr. Toshiya Ueda and Mr. Tatsuo Kato for their assistance.  ... 
dblp:conf/trec/Fujita00 fatcat:knnau5ygrjfy7hs3mgh3c26rva

Query Expansion for IR using Knowledge-Based Relatedness

Arantxa Otegi, Xabier Arregi, Eneko Agirre
2011 International Joint Conference on Natural Language Processing  
Query expansion techniques like pseudo-relevance feedback rely on the target document set in order to bridge the gap between those words, but they might suffer from topic drift.  ...  The limitations of keyword-only approaches to information retrieval were recognized since the early days, specially in cases where different but closely-related words are used in the query and the relevant  ...  ., 2009b) , we explored a generic method to improve IR results using WordNetbased query expansion, and compared it to baseline query likelihood and pseudo-relevance feedback methods.  ... 
dblp:conf/ijcnlp/OtegiAA11 fatcat:tlbxl5nvv5ayrhp4fp7itrhuke
« Previous Showing results 1 — 15 out of 32,971 results