Filters








27,284 Hits in 7.0 sec

Query performance prediction for IR

David Carmel, Oren Kurland
2012 Proceedings of the 35th international ACM SIGIR conference on Research and development in information retrieval - SIGIR '12  
retrieval scores can potentially help predict query performance.  ...  , divided into 10 partitions of roughly equal size SIGIR 2012 Tutorial: Query Performance Prediction Content enhancement using missing content analysis SIGIR 2012 Tutorial: Query Performance Prediction  ...  Results were moderate, probably due to the sparseness of the training data, which over-represents the lower end of the performance values We discussed three frameworks for query-performance prediction  ... 
doi:10.1145/2348283.2348540 dblp:conf/sigir/CarmelK12 fatcat:ppzmf3ihkzbrje4ny33h6elave

Learning to Select a Ranking Function [chapter]

Jie Peng, Craig Macdonald, Iadh Ounis
2010 Lecture Notes in Computer Science  
enhanced if an appropriate ranking function is selected for each individual query.  ...  The ranking function which performs the best on this identified training query set is then chosen for the unseen query.  ...  For our future work, we plan to investigate other query features.  ... 
doi:10.1007/978-3-642-12275-0_13 fatcat:7lrhhv67bfb25gixtze2y5ineu

Predicting query performance in microblog retrieval

Jesus A. Rodriguez Perez, Joemon M. Jose
2014 Proceedings of the 37th international ACM SIGIR conference on Research & development in information retrieval - SIGIR '14  
Query Performance Prediction (QPP) is the estimation of the retrieval success for a query, without explicit knowledge about relevant documents.  ...  QPP would be of great benefit in the context of microblog retrieval, as AQE was the most widely deployed technique for enhancing retrieval performance at TREC.  ...  The QueryTermIdf predictor utilizes the IDF values of query terms for making estimations of retrieval performance.  ... 
doi:10.1145/2600428.2609540 dblp:conf/sigir/PerezJ14 fatcat:x762ayb2nbfudku7dvo2w4ka6u

Measuring ranked list robustness for query performance prediction

Yun Zhou, W. Bruce Croft
2007 Knowledge and Information Systems  
Our initial motivation for measuring ranking robustness is to predict topic difficulty for content-based queries in the ad-hoc retrieval task.  ...  Though our focus is on prediction under the ad-hoc retrieval task, we observe an interesting negative correlation with query performance when our technique is applied to named-page finding queries which  ...  Generally speaking, current prediction methods extract features of retrieval and compute the performance score for each query by using the features to estimate the query performance.  ... 
doi:10.1007/s10115-007-0100-8 fatcat:a3wpp4hv7fbkzozkkksq5cnjxy

Selective Application of Query-Independent Features in Web Information Retrieval [chapter]

Jie Peng, Iadh Ounis
2009 Lecture Notes in Computer Science  
The approach is based on an estimate of the divergence between the retrieved document scores' distributions prior to, and after the integration of a query-independent feature.  ...  The application of query-independent features, such as PageRank, can boost the retrieval effectiveness of a Web Information Retrieval (IR) system.  ...  We thank Craig Macdonald & Ben He for their helpful comments and feedback on the paper.  ... 
doi:10.1007/978-3-642-00958-7_34 fatcat:qmmhwqyh2rgt7l4w5jxkr52iam

Query Performance Prediction for Information Retrieval Based on Covering Topic Score

Hao Lang, Bin Wang, Gareth Jones, Jin-Tao Li, Fan Ding, Yi-Xuan Liu
2008 Journal of Computer Science and Technology  
We present a statistical method called Covering Topic Score (CTS) to predict query performance for information retrieval.  ...  We compare CTS with previous state-of-the-art methods for query performance prediction including clarity score and robustness score.  ...  Acknowledgements We thank Jian Zhang from Purdue University for his help about statistics. We also thank all anonymous reviewers for their kind help to improve the paper.  ... 
doi:10.1007/s11390-008-9155-6 fatcat:2aiutvxd4fbdfel63qnm7k7n4a

Query efficiency prediction for dynamic pruning

Nicola Tonellotto, Craig Macdonald, Iadh Ounis
2011 Proceedings of the 9th workshop on Large-scale and distributed informational retrieval - LSDS-IR '11  
In this work, we investigate the causes for inefficient queries, identifying reasons such as the balance between informativeness of query terms, and the distribution of retrieval scores within the posting  ...  Dynamic pruning strategies are effective yet permit efficient retrieval by pruning -i.e. not fully scoring all postings of all documents matching a given query.  ...  In contrast, most post-retrieval predictors require the scores or contents of retrieved documents (e.g. to estimate a language model) or passes over the posting lists to estimate the number of documents  ... 
doi:10.1145/2064730.2064734 fatcat:cfx45kl4fbfdvios4ic7n3rmmi

Visual Localization Using Semantic Segmentation and Depth Prediction [article]

Huanhuan Fan, Yuhao Zhou, Ang Li, Shuang Gao, Jijunnan Li, Yandong Guo
2020 arXiv   pre-print
We apply semantic consistency evaluation to rank the image retrieval results and a practical clustering technique to reject estimation outliers.  ...  In addition, we demonstrate a substantial performance boost achieved with a combination of multiple feature extractors.  ...  SCW is then calculated for all query-retrieval pairs, and only those with high SCW scores will proceed to subsequent steps.  ... 
arXiv:2005.11922v1 fatcat:ia63euutmvdt3e54s4on4oik2y

Automatic document prior feature selection for web retrieval

Jie Peng, Craig Macdonald, Iadh Ounis
2008 Proceedings of the 31st annual international ACM SIGIR conference on Research and development in information retrieval - SIGIR '08  
This paper aims to investigate whether the retrieval performance can be further enhanced by selecting the best document prior feature on a per-query basis.  ...  We present a novel method for selecting the best document prior feature on a per-query basis.  ...  For a given query and its corresponding top retrieved documents, we propose to estimate the divergence between the retrieved document scores distribution prior to, and after the integration of the document  ... 
doi:10.1145/1390334.1390490 dblp:conf/sigir/PengMO08 fatcat:sz5uc2if2rbrjmhbgy3fmbypoy

Ranking robustness

Yun Zhou, W. Bruce Croft
2006 Proceedings of the 15th ACM international conference on Information and knowledge management - CIKM '06  
We compare the robustness score with the clarity score method which is the state-of-the-art technique for query performance prediction.  ...  We find that the clarity score is barely correlated with query performance on the GOV2 collection while the correlation between the robustness score and query performance remains significant.  ...  Generally speaking, these methods extract features of retrieval and compute the performance score for each query by using the features to estimate the query performance.  ... 
doi:10.1145/1183614.1183696 dblp:conf/cikm/ZhouC06 fatcat:feavtrzshjbbxjttxuwbkqld5q

Predicting the Usefulness of Collection Enrichment for Enterprise Search [chapter]

Jie Peng, Ben He, Iadh Ounis
2009 Lecture Notes in Computer Science  
Query Expansion (QE) often improves the retrieval performance of an Information Retrieval (IR) system.  ...  In this paper, we propose the use of query performance predictors to selectively apply CE on a per-query basis.  ...  Conclusions We have proposed the use of query performance predictors to selectively apply CE on a per-query basis for document search within an enterprise.  ... 
doi:10.1007/978-3-642-04417-5_41 fatcat:zp55sx7rebeidkibi3z6ll7elq

A study of selective collection enrichment for enterprise search

Jie Peng, Craig Macdonald, Ben He, Iadh Ounis
2009 Proceeding of the 18th ACM conference on Information and knowledge management - CIKM '09  
performance predictors.  ...  However, it is not always clear for which queries the collection enrichment technique should be applied.  ...  For example, for the clarity score predictor [3] , a lower query performance score on the external resource means a higher similarity between the query language model and the external collection's language  ... 
doi:10.1145/1645953.1646286 dblp:conf/cikm/PengMHO09 fatcat:csyrexn7tngy7j3v27xidqwcni

Ranking using multiple document types in desktop search

Jinyoung Kim, W. Bruce Croft
2010 Proceeding of the 33rd international ACM SIGIR conference on Research and development in information retrieval - SIGIR '10  
In addition, we show that type prediction performance and search effectiveness can be further enhanced by combining existing methods of type prediction using discriminative learning models.  ...  Predicting which types of documents a user is looking for in the context of a given query is a crucial part of providing effective desktop search.  ...  Query performance prediction methods can also be used for type prediction by assigning a higher score for the collection with higher predicted performance.  ... 
doi:10.1145/1835449.1835461 dblp:conf/sigir/KimC10 fatcat:zuc63g2vpfbknhujz7cvs5ntxa

Information preservation in static index pruning

Ruey-Cheng Chen, Chia-Jung Lee, Chiung-Min Tsai, Jieh Hsiang
2012 Proceedings of the 21st ACM international conference on Information and knowledge management - CIKM '12  
We evaluated the proposed approach on three different test corpora, and the result shows that our approach is comparable in retrieval performance to state-of-the-art methods.  ...  When efficiency is of concern, our method has some advantages over the reference methods and is therefore suggested in Web retrieval settings.  ...  ., postings, from a production retrieval system. At the cost of sacrificing some degree of retrieval accuracy, this practice has been shown to enhance both disk usage and query throughput [4] .  ... 
doi:10.1145/2396761.2398673 dblp:conf/cikm/ChenLTH12 fatcat:eappm2ofkvb3rirhcdmseqyvwe

Joint Ranking for Multilingual Web Search [chapter]

Wei Gao, Cheng Niu, Ming Zhou, Kam-Fai Wong
2009 Lecture Notes in Computer Science  
Ranking for multilingual information retrieval (MLIR) is a task to rank documents of different languages solely based on their relevancy to the query regardless of query's language.  ...  A probabilistic graphical model is trained for the joint relevance estimation.  ...  Existing techniques usually combine query translation and monolingual retrieval to derive a relevancy score for each document.  ... 
doi:10.1007/978-3-642-00958-7_13 fatcat:rrr7y4xoh5bezcq63lc2xzftzi
« Previous Showing results 1 — 15 out of 27,284 results