99,899 Hits in 4.3 sec

Combining Multiple Evidence from Different Relevance Feedback Methods

Joon Ho Lee
1997 Database Systems for Advanced Applications '97  
We then describe the effect of combining the multiple query vectors on retrieval effectiveness.  ...  Recent work suggests that significant improvement in retrieval performance can be achieved by combining multiple representations of an information need.  ...  Acknowledgments This work was supported in part by the NSF Center for Intelligent Information Retrieval at the University of Massachusetts at Amherst.  ... 
doi:10.1142/9789812819536_0044 fatcat:hfkar5l2czf23b3bjzr4ojen6m

Dublin City University at CLEF 2006: Cross-Language Speech Retrieval (CL-SR) Experiments [chapter]

Gareth J. F. Jones, Ke Zhang, Adenike M. Lam-Adesina
2007 Lecture Notes in Computer Science  
We explored alternative translations methods with document retrieval based the combination of the multiple document fields using the BM25F field combination model.  ...  BabelFish free online translation service combined with domain-specific translation lexicons gathered automatically from Wikipedia.  ...  System Parameters Our retrieval system requires a number of parameters to be set for the term weighting, field combination, and PRF components.  ... 
doi:10.1007/978-3-540-74999-8_100 fatcat:icdb526fqjffbdzsg2gkylkdom

Dublin City University at CLEF 2005: Cross-Language Speech Retrieval (CL-SR) Experiments [chapter]

Adenike M. Lam-Adesina, Gareth J. F. Jones
2006 Lecture Notes in Computer Science  
We explored alternative translations methods with document retrieval based the combination of the multiple document fields using the BM25F field combination model.  ...  BabelFish free online translation service combined with domain-specific translation lexicons gathered automatically from Wikipedia.  ...  System Parameters Our retrieval system requires a number of parameters to be set for the term weighting, field combination, and PRF components.  ... 
doi:10.1007/11878773_87 fatcat:wp752kbvorg7hbvtudkasc2c6y

A Comparison of Score, Rank and Probability-Based Fusion Methods for Video Shot Retrieval [chapter]

Kieran Mc Donald, Alan F. Smeaton
2005 Lecture Notes in Computer Science  
In this paper we investigate a range of fusion methods for combining based on multiple visual features (colour, edge and texture), for combining based on multiple visual examples in the query and for combining  ...  Using three TRECVid collections and the TRECVid search task, we specically compare fusion methods based on normalised score and rank that use either the average, weighted average or maximum of retrieval  ...  At their most basic, late fusion methods combine the scored and ranked retrieval results from dierent systems/models/features in order to improve upon the best individual retrieval result.  ... 
doi:10.1007/11526346_10 fatcat:u6cm5yb2mrh7paimr2xml4toya

Toward a Robust data fusion for document retrieval

Daqing He, Dan Wu
2008 2008 International Conference on Natural Language Processing and Knowledge Engineering  
Our results show that the new methods can significantly outperform CombMNZ in combining retrieval results that are low and diverse.  ...  This paper describes an investigation of signal boosting techniques for post-search data fusion, where the quality of the retrieval results involved in fusion may be low or diverse.  ...  First, the combination can happen at the query side, where query terms from multiple sources are combined to generate a single query to be entered to a retrieval system (e.g., Belkin and his colleagues  ... 
doi:10.1109/nlpke.2008.4906754 dblp:conf/nlpke/HeW08 fatcat:h5fdxxavkzdenjprbptdjlyfmu

Combining Image and Structured Text Retrieval [chapter]

D. N. F. Awang Iskandar, Jovan Pehcevski, James A. Thom, S. M. M. Tahaghoghi
2006 Lecture Notes in Computer Science  
In this paper we present our approach of combining evidence from a contentoriented XML retrieval system and a content-based image retrieval system using a linear evidence combination approach as part of  ...  Two common approaches to retrieving images from a collection are retrieval by text keywords, and retrieval by visual content.  ...  The merging (or also known as fusion) phase deals with combining the evidence from multiple retrieval systems.  ... 
doi:10.1007/978-3-540-34963-1_40 fatcat:rwhe2fyr7rctvf3w4jgatlbdh4

Aggregation of Multiple Pseudo-Relevance Feedbacks for Image Search Reranking

Wei-Chao Lin
2019 IEEE Access  
Image retrieval effectiveness can be improved by pseudo relevance feedback (PRF), which automatically uses top-k images of the initial retrieval result as the pseudo feedback.  ...  In particular, different combinations of the original retrieval result with the result of PRF and the result of PRF by pseudo positive and negative feedbacks, using a strategy based on the Borda count,  ...  Another post-retrieval approach is based on combining multiple search results by search result aggregation.  ... 
doi:10.1109/access.2019.2942142 fatcat:jqucuyx4vrcqpg7jzv7vyv3ndu

Study on Merging Multiple Results from Information Retrieval System

Hiromi Ozaku, Masao Utiyama, Hitoshi Isahara, Yasuyuki Kono, Masatsugu Kidode
2002 NTCIR Conference on Evaluation of Information Access Technologies  
We participated in Web retrieval tasks in order to investigate the integration of multiple retrieval results.  ...  On the other hand, if the systems use Rocchio's feedback method, ranked results are shown to be the best.  ...  The multiple systems are called meta-search systems. The conventional meta-search systems need to combine the returned lists of pages from different search engines.  ... 
dblp:conf/ntcir/OzakuUIKK02 fatcat:o76vghwedvadriu3jajjrnsrtq

Extending test collection pools without manual runs

Gaya K. Jayasinghe, William Webber, Mark Sanderson, J. Shane Culpepper
2014 Proceedings of the 37th international ACM SIGIR conference on Research & development in information retrieval - SIGIR '14  
Information retrieval test collections traditionally use a combination of automatic and manual runs to create a pool of documents to be judged.  ...  By combining a simple voting approach with machine learning from documents retrieved by automatic runs, we are able to identify a large portion of relevant documents that would normally only be found through  ...  Figure 4 : 4 Kendall's τ correlation of IR system rankings with varying number of automatic systems in the pool.  ... 
doi:10.1145/2600428.2609473 dblp:conf/sigir/JayasingheWSC14 fatcat:b27qe5hkt5hbfhac22hqf6qgom

The Exploration and Analysis of Using Multiple Thesaurus Types for Query Expansion in Information Retrieval

Rila Mandala, Takenobu Tokunaga, Hozumi Tanaka
2000 Journal of Natural Language Processing  
modern information retrieval system can be re-stated as to push the relevant documents to the top of the retrieved documents rank.  ...  The task of information retrieval system is to extract relevant documents from a large collection of documents in response to user queries(Salton and McGill 1983).Most modern information retrieval systems  ...  Acknowledgments The authors would like to thank the anonymous referees for useful comments on the earlier version of this paper.We also thank Chris Buckley (SabIR Research) This research is partially  ... 
doi:10.5715/jnlp.7.2_117 fatcat:5qvzw5ycejblxbz5dtaalpu7a4

WIDIT in TREC 2004 Genomics, Hard, Robust and Web Tracks

Kiduk Yang, Ning Yu, Adam Wead, Gavin La Rowe, Yu-Hsiu Li, Christopher Friend, Yoon Lee
2004 Text Retrieval Conference  
The basic approach of WIDIT was to leverage multiple sources of evidence, combine multiple methods, and integrate the strengths of man and the machine.  ...  The key component of WIDIT for the Web track was an interactive system tuning process called "Dynamic Tuning", which optimizes the fusion formula that combines the contributions of multiple sources of  ... 
dblp:conf/trec/YangYWRLFL04 fatcat:74oxu6cxtnaqldkfcd7lq7if4u

XML Information Retrieval from Spoken Word Archives [chapter]

Robin Aly, Djoerd Hiemstra, Roeland Ordelman, Laurens van der Werff, Franciska de Jong
2007 Lecture Notes in Computer Science  
In this paper the XML Information Retrieval System PF/Tijah is applied to retrieval tasks on large spoken document collections.  ...  Furthermore simple ways of combining different metadata kinds are shown to be beneficial in terms of mean average precision.  ...  With this scenario the following questions need to be addressed: (i) how to efficiently store these multiple annotations, (ii) how to combine and rank retrieval scores based on them and (iii) how to select  ... 
doi:10.1007/978-3-540-74999-8_97 fatcat:hoqftbe7sfb2to3umnil7yao6u

WHU-NERCMS at TRECVID2017: Instance Search Task

Dongshu Xu, Jiamei Lan, Xiaoyu Chai, Yiyue Chen, Xiao Wang, Jiaqi Li, Longxiang Jiang, Chao Liang
2017 TREC Video Retrieval Evaluation  
And the result fusion module is to combine the results of person search and scene retrieval to obtain the ranking result at first and then to optimize it to get the final result.  ...  In interactive INS task, we selected effective expansion of person retrieval and scene retrieval ranking result and deleted the obviously wrong shots.  ...  ,Ltd for providing us the source of face recognition. Thank BBC for providing programme material. Thank all the former INS members who accumulated a great deal of experiences for us.  ... 
dblp:conf/trecvid/XuLCCWLJL17 fatcat:kfkc3vdxtvdyhehwse2zflyqr4

Towards More Effective Techniques for Automatic Query Expansion [chapter]

Claudio Carpineto, Giovanni Romano
1999 Lecture Notes in Computer Science  
We then argue, based on their variation in performance on individual queries, that the set of ranked terms suggested by individual distributional methods can be combined to further improve mean performance  ...  Techniques for automatic query expansion from top retrieved documents have recently shown promise for improving retrieval effectiveness on large collections but there is still a lack of systematic evaluation  ...  Acknowledgments This work has been carried out within the framework of an agreement between the Italian PT Administration and the Fondazione Ugo Bordoni.  ... 
doi:10.1007/3-540-48155-9_10 fatcat:rqqc734gjzcf5pvbxmve6xqma4

Extreme video retrieval

Alexander G. Hauptmann, Wei-Hao Lin, Rong Yan, Jun Yang, Ming-Yu Chen
2006 Proceedings of the 14th annual ACM international conference on Multimedia - MULTIMEDIA '06  
In either case, as humans search and find relevant results, the system can invisibly re-rank its previous best guesses using a number of knowledge sources, such as image similarity, text similarity, and  ...  Experimental evidence shows a significant improvement using the combined extremes of human and machine power over either approach alone.  ...  We plan to also conduct an examination of the types of errors humans make when looking at a list of results but failing to see the relevant shots (misses) as well as the situations where humans are erroneously  ... 
doi:10.1145/1180639.1180721 dblp:conf/mm/HauptmannLYYC06 fatcat:53ae6kpcdbdjxepgu6wms6xwoa
« Previous Showing results 1 — 15 out of 99,899 results